Volume 4 - The IMP Journal

Transcription

THE
IMP Journal
Industrial Marketing and Purchasing
VOLUME 4 2010
Forum for research into
business interaction,
relationships and
networks.
Editor
Håkan Håkansson
BI Norwegian Business School
Executive Board
Håkan Håkansson
Annalisa Tunisini
Pete Naude
University of Urbino
Manchester Business School
Editorial Board
Helén Anderson James Anderson Luis Araujo
Björn Axelsson
Enrico Baraldi
Keith Blois
Roberta Bocconcelli
Anna Dubois
Geoff Easton
David Ford
Lars-Erik Gadde Simone Guercini
Aino Halinen
Lars Hallén
Debbie Harrison
Stefan Henneberg
Lars Huemer
Tibor Mandjak
Lars-Gunnar Mattsson Stefanos Mouzas
Kristian Möller Ann-Charlott Pedersen
Thomas Ritter
Robert Salle
Asta Salmi
Judit Simon
Ivan Snehota
Robert Spekman
Annalisa Tunisini
Alexandra Waluszewski
Ian Wilkinson Louise Young
Jönköping International Business School
Northwestern University
Lancaster University
Stockholm School of Economics
Uppsala University
Urbino University
Chalmers University of Technology
University of Bath, School of Management
Chalmers University of Technology
Florence University
Turku School of Economics and Business Administration
Mälardalen University, School of Business
Corvinus University
Helsinki School of Economics
Norwegian University of Science and Technology
Copenhagen Business School
EM Lyon
Helsinki School of Economics
Corvinus University
University of Lugano
University of Virginia
University of Urbino
Uppsala University
University of New South Wales
University of Western Sydney
Judith Zolkiewski
Publisher
IMP Group, ISSN 0809-7259, www.impjournal.org
The IMP Journal
Volume 4
Contents
Business Relationships and Networks: Reflections on the IMP tradition
Björn Axelsson
3
Manager - Researcher Relationships
Geoff Easton
31
Temporal Profiles of Activities and Temporal Orientations of Actors as Part of Market
Practices in Business Networks
Per Andersson and Lars-Gunnar Mattsson
57
Analysing Business Interaction
David Ford, Lars-Erik Gadde, Håkan Håkansson, Ivan Snehota and Alexandra Waluszewski
82
A Critical Episode Analysis of the Dynamics of the Interaction Atmosphere in a New
Product Development Relationship
Patrick Lynch and Thomas O’Toole
106
The emergence of a successful business network – What was the role of public policy?
Tommy Shih
131
Distribution network evolution – challenges for practice and theory
Lars-Erik Gadde
160
Distribution network dynamics: Evolution in the PC distribution network
Kajsa Hulthén and Lars-Gunnar Mattsson
170
Opportunities and constraints for intermediaries in distribution: The challenge of variety
Leif-Magnus Jensen
194
Organizing a network within the network The case of MC Elettrici
Chiara Cantù, Francesca Montagnini and Roberta Sebastiani
220
Produced by IMP Group in cooperation with
Catholic University of the Sacred Heart
The IMP Journal
Volume 4, number 1
1
A Letter From The Editor
Welcome to a new issue of IMP Journal – the first in volume 4. First, a big thank you
to David who managed to take us through the first three volumes of the IMP Journal.
The change of editor does not signal any change of orientation or content. We will
continue to publish articles dealing with interaction in business settings and we are
especially interested in empirically based articles. Informative cases about business
interactions taking place in different geographic, technological or social contexts are
highly appreciated.
In this issue we are pleased to present four articles from some of the most senior IMP
researchers. The ambition is to present some pictures of problems and issues
currently dealt with by these researchers. Together they provide an interesting
mixture of where the research has taken us. In one way or another all are struggling
with the basic “interaction” issue. In the first contribution Björn Axelsson discusses
how it has been possible to use the models developed within IMP in interactions with
students, executives and companies. He takes the two basic models – the interaction
model from 1982 and the ARA model from 1995 – as starting points and discusses
what he sees as major strengths and major weaknesses of the two models. By using
some short cases he tries to demonstrate the kinds of problems that will be identified
by using the models and the kinds of solutions to which they point. The article
becomes an example of the struggling user of business research even when the user
is highly competent. This kind of use, which I think we all have experienced, has also
been an important input in the research process and has clearly affected what has
emerged over time.
The same basic theme comes back in the second article by Geoff Easton. He uses
an interaction approach to analyse the relationship between researchers and
managers in a number of research projects mainly related to IMP. The emerging
picture, in a very empirically rich article, is a good example of interaction studies.
There is a very large variation in how both sides see each other and act in relation to
each other. There is no surprise in the finding that the researchers are not a
homogeneous group – they differ in attitudes as well as experience and position
within the research organisation. But there is the same large variation on the
management side. There we also have large differences in attitudes and ambitions
as well as in how they systematically approach the researchers. There are some who
are very systematic and conscious of how they approach the researchers. The
majority see it as quite a marginal task. Still, the interviewed managers are probably
more research oriented than average due to the fact that they have been part of at
least one research project. The conclusion for us researchers is that we have to
develop a much more systematic interaction with managers if we want to establish a
more interested counterpart.
The IMP Journal
Volume 4, number 1
2
This will probably take a long time and we can use the third contribution by Per
Andersson and Lars-Gunnar Mattsson to gain some ideas about how we can get time
to work for us. In their contribution they focus on the temporal profiles of business
activities and the temporal orientation of business actors in order to add to the
understanding of dynamics in business and relationships between companies. They
base their discussion on earlier IMP research on time and their starting point is that
all business activities have a unique temporal profile. This includes dimensions such
as synchronisation, sequencing, timing and others. All actors’ ways of perceiving and
relating to time are depicted in their temporal orientation. This is assumed to be
related to their network theory and network orientation. These concepts are
discussed in relation to three empirical illustrations. In the conclusion part the
importance of time is especially connected to six problem areas: mergers,
internationalisation, business cycles, horizons in financial markets vs. “real markets”,
implementations of information and communication systems and logistics.
In the fourth contribution Ford et al., discuss the basic problem of interaction – how it
should be conceptualised and modelled. The article is an attempt to find a new way
to frame interaction where, due to the existence of “substance”, it leaves traces in
different ways. It makes time and place important dimensions of interaction. In the
suggested model three aspects of time – specialisation, path dependence and coevolution – and three of space – interdependency, heterogeneity and jointness – are
related to the activity, actor, and resource layers of the interaction respectively. This
way of conceptualising business interaction will hopefully make this phenomenon
both more general and more specific; general in terms of making it easier to relate to
other theoretical models and more specific by relating interaction to some specific
features of time and place.
Enjoy the issue!
Oslo
Håkan Håkansson
The IMP Journal
Volume 4, number 1
3
Business Relationships and Networks:
Reflections on the IMP tradition
Björn Axelsson
Stockholm School of Economics, P.O. Box 6501, SE-113 83 Stockholm, e-mail: bjorn.axelsson@hhs.se
Abstract
This paper provides a reflective view of some of the developments of the IMP- oriented research. It
focuses on the two core models, the interaction model for descriptions and analyses of buyer- supplier
relationships and the ARA- network model for analyses of business networks. The discussion takes as its
point of departure the three major tasks of a typical academic professional: research, education of students
and practitioner training. The discussion addresses some of the strengths and weaknesses of the IMPapproach as reflected by the two models and in relation to the three tasks. It utilises several illustrations to
demonstrate the ways in which the two models may support analyses of real events. But it also points to
some criticism as well as potential ways to address some critical issues. It is not meant to be a general
evaluation of good or bad but rather demonstrating some of the important contributions from the models and
also some of the critical issues where continued progress would be desirable. The discussion is solely
created by one researcher with a 30 year long experience from being involved with this research in all three
fields of practice.
1. Introduction
The present article is an
attempt to provide some personal
reflections on the IMP research tradition
in business markets. As a scholar, I
have had a relation to the IMP tradition
for around 30 years. This means that I
have been involved in the three major
duties of scholars working in academia:
educating students, doing research and
supporting people in business practice.
The latter task has primarily taken the
form of my working as a consultant in
training,
educating
and
advising
business managers. The reflections
presented below will take as their point
of departure my own experiences in
utilizing the IMP framework. What are
the core observations and findings?
What has worked? How has it been
possible to make use of the ideas and
theories put forward in this area of
research? What has been accepted and
proved to be of value in the three
contexts? What has caused confusion?
What has been difficult about utilizing
the framework? What weaknesses have
been addressed and resolved, and what
weaknesses seem to remain? These are
some of the questions underlying this
article. I cannot address them all in
detail. Therefore, the questions will
primarily serve as a general background
to the text.
It should also be pointed
out that I will use a rather limited number
of references to the IMP research as
such. The background and the general
development of the IMP Group and its
thinking have already been dealt with by
others (see a.a. Mattsson & Johanson,
2005).
Furthermore, it should also
be stated that I will primarily emphasize
what I consider the two main models in
the IMP tradition: the early interaction
model (Håkansson ed., 1982) and the
business network model, foremost the
ARA model (Activities Resources
Actors, Håkansson & Snehota, 1995,
Axelsson & Easton, 1992). These have
also been vital parts of my own research
practice.
The early findings in this
research tradition pointed to the natural
and frequent appearance of long-term
The IMP Journal
Volume 4, number 1
relations1 in industrial, business-tobusiness, markets. This was strongly
addressed in Håkansson ed.’s early
book from 1982, especially under the
heading “We Challenge” (in Chapter 1).
This notion of the significance of
business relations has also been put
forward in other research traditions,
such as the Relationship Marketing
tradition (RM, e.g. Christopher et al.,
2002). Thus, the IMP tradition is not
alone in addressing these topics. Such
alternative approaches have both
similarities
and
differences.
For
instance, Ford et al. (2002, pp 104-105
and Mattsson, 1997) addressed some of
the similarities among and differences
between the RM and IMP schools of
thought. In Pels et al. (2009), there is,
furthermore, a comparison between
several schools of thought, including the
IMP tradition, all of which have
addressed business relationships. In the
present text, I will solely focus on the
IMP tradition and foremost on the two
models mentioned above. I believe that
these models and their implications are
two highly significant features of this
tradition. Let me first discuss the interaction model and thereafter the ARA
model.
2. The interaction model
My general assumption is
that the reader is familiar with the
interaction model. Still, a few words may
be needed for general orientation.
The model consists of four
main groups of variables:
1.
The
interaction
process
as
characterized by what is exchanged
(products, information, social, financial
content) as well as by prioritized issues,
involved actors on both sides. Interaction processes take part in relationships between actors. Relationships are,
e.g., characterized as being new/old,
having a history of small or major
adaptations made and so on. The inter1
The concepts “relation” and “relationship” are used.
Relation refers to relations in a broad and general
sense. Relationship refers to more specific relations
e.g. between a buyer and seller.
4
action process and the relationship are
at the core of the model.
2. The characteristics of the involved
parties. This has to do with the marketing and purchasing strategies of the
involved parties, their respective organizational designs, the product and
process technology, the involved individuals and their aims, and some additional features that may have an impact
on interaction processes.
3. The atmosphere that surrounds and is
part of the relationship. This could be
expressed in terms of the degree of
cooperation/conflict, power/dependency,
between the parties.
4. The environment is the outer context
in which the interaction takes place. This
could be characterized as a stable/
dynamic setting, domestic/inter-national
in scope.
The development of the
interaction model made it possible, in
rather unified ways, to describe but also
to aggregate and systematically find
similarities among and differences
between various b2b relationships and
situations. Instead of looking at each
relationship and situation as totally
unique, it was possible to see patterns
of similarity and difference: Why does it
work like this in this situation, but not like
this in that situation? What should we
expect when confronted by situation X?
This contribution was in line with one of
the basic tasks of science: to analyse
phenomena and see patterns, to confirm
one’s hypotheses or be surprised.
Although it seems as if today’s writers,
working in line with this tradition, do not
adhere to this to the same degree as
before, I still believe this model, and
what it brings with it, is one of the
outstanding aspects of the IMP tradition.
I will use two examples to elucidate this
observation.
2.1 Two illustrations
The illustrations below
have been used in lecturing and training
as a way to address the postulate:
“relationships, yes of course, but what
kind of relationships?”
The IMP Journal
Volume 4, number 1
The IT company and its
new sales organization
The big IT company had
recently chosen to carry
out all its marketing and
sales efforts by introducing
a separate marketing unit
operating as a cost centre.
It did so in order to “get in
order” all the visits paid to
customers by a variety of
its people. Top management of the firm thought
that behaviour had previously been like in the
“wild west” – nobody really
knew what anyone else
was doing. The company
offered
both
standard
products and advanced
customer-specific problem
solutions. The latter had
developed greatly of late.
These
products
were
developed and produced
by separate design and
production units.
The episode in focus refers
to a situation in which the
innovative production unit
(the producer of customerspecific solutions) demanded an explanation from
the present joint sales
organization. This is what
their representative said:
“Our ideas and suggestions for new products,
new solutions to customer
problems, end up in nothing. Our sales organization does not sell and
seems to be unknown to
our old customer contacts.
And – we receive very little
input
via
our
sales
organization”.
The innovative production
unit even demanded that
the sales organizations
demonstrate how they
5
operate the market. They
explicitly asked the sales
units to reveal their contacts at the customers:
What kind of professionals
do they meet, how intensive are these contacts,
and so forth. This is what
their representative continued to say:
“We doubt that the sales
units have the right contacts to support our mission. We think that their
organization
and
their
interaction patterns with
the customers are well
suited to the sales of our
company’s standard products, but not to the products we want supported.”
The sales units, on their
part, argued that they
certainly have the contacts
asked for but they cannot
spend a lot of time on
“endless discussions about
a number of loose ideas
that do not result in any
orders”.
This is a situation from real
life. It is probably quite a common one
experienced by many sales organizations. It brings us back to an early
IMP- article by Håkansson and Östberg
(1975),
”Industrial
marketing:
an
organisational problem?”, which is one
of the earliest IMP-related texts. It
emphasizes that business marketing is
very much about organizing relationships, e.g. creating relevant interfaces
between parties using organizational
means. But the interaction model can
take us further in understanding the
origins and the overt outcome of a
situation like the one described.
Using the terminology from
the interaction model, this is an episode
that takes place in a specific exchange
and relationship context. It is clear that
this episode is part of and connected to
exchange processes with several in-
The IMP Journal
Volume 4, number 1
volved actors (various customers to the
supplier). The episode is also embedded
in a context of previous episodes and,
with experiences following from that, the
creation of expectations from all parties
involved. Furthermore, it is influenced by
some of the characteristics of the parties
and it takes place in various externally
defined contexts. The situation, thus,
relates to all four variables of the model.
But what support does the interaction
model give in providing a principal
understanding of this?
Some of the most outstanding features seem to be the following. First, the supplier has a strategy to
market and sell two families of products
that may or may not attract the same
customer. It seems as if the demands on
interaction needed to manage the
necessary information exchange and to
solve problems differs a great deal
between the products. The company
also operates under economic pressure.
It needs a marketing and sales
organization that does not cost too
much. This takes us to the next, the
second, explanatory variable. The
company
has
decided
on
an
organizational design of its sales unit
that tends to streamline and standardize
too much. In this way, the unit’s
efficiency seems to be outshining its
effectiveness. The company has lost
much of its ability to create new
business. The new organizational
design thus hinders some of the
business performance. This is especially
true for the customer-specific segment.
Three variables of the model – the
supplier’s strategy, the customer’s
usage situation and needs, and the
supplier’s organizational design – seem
to be the most significant parts of the
model at play in this situation.
The influence of other
variables
in
understanding
the
fundaments of this episode is more
vague. But one can easily understand
that changes in contact persons,
changes in interaction patterns, etc.,
may have influenced the atmosphere
between parties and, perhaps, also
triggered changes among competitors
(parts of the environment), e.g. by
6
mobilizing them to try to acquire customers from the focal firm.
The second illustration is
similar, but also different…
”The impossible delivery”
During the past five years,
the Publisher has concentrated on its purchases of
printing services. Today it
only has two suppliers. As
a result, it has improved its
prices (volume discounts).
At one of the two, the
Printer, the Publisher now
takes 40 percent of its total
production.
One of the most important
markets for the Publisher is
university textbooks. An
ambitious author and an
over-committed
project
leader at the Publisher
have taken too much time
to complete their job. The
manuscript is not ready to
print by the 1st of August,
instead it will be ready on
the 20th. The Publisher had
promised its customers,
the course responsible at
the universities, that the
book would be available in
good time for courses starting on September 1st.
Representatives of the
Publisher inform the representatives of the Printer
that this is a tricky situation, but are certain that
the supplier will do their
best to print the book in
about ten days. The reaction is very negative. The
production manager at the
Printer argues “this is
impossible” and declares “I
refuse”. What to do?
A dialogue starts inside the
Publisher’s
organization.
Participating in the dialo-
The IMP Journal
Volume 4, number 1
gue is their CEO. The customers at the universities
are contacted and receive
a positive answer. “This
will be all right in the end”.
New contacts are made
with the Printer. This time
the contacts are at a
higher level in the hierarchy. Tough words as well
as
kind
ones
are
exchanged. There are also
threats. Within a few hours
comes the final answer
from the Printer. “Yes, we
will make it, but only this
time, and we want your
guarantee that this will not
happen again”.
This is also an illustration
from real life, and again it is an episode
embedded in an exchange and relationship context. Also in this case, we
can imagine the potential influence of all
four groups of variables. This episode is
embedded in one (several) relationship(s). What happens seems to have a
great deal to do with characteristics of
the actors involved. There is an
atmosphere of expectations created,
and we see an influence of external
(market) conditions.
Some of the outstanding
features of this episode are the
following: First and foremost, the
atmosphere of the two parties is at play,
especially in terms of power and dependency. The Publisher clearly utilizes its
power in relation to the Printer. But that
possibility has been created by, among
other things, the buying firm’s strategy of
sourcing from a limited number of
suppliers,
thus
concentrating
its
purchases. Therefore, the second significant variable is the sourcing strategy of
the buyer to focus on fewer suppliers
(and the supplier’s sales strategy to
allow more business to go to one
specific customer). In this way, the two
parties have become more dependent
on each other – and that is one of the
fundaments for all this to happen. A third
variable activated is the internal
organization of the parties, foremost the
7
interplay between the operative and
strategic levels. We see that both parties
seem to have a mechanism in place
whereby episodes can be escalated to
involve higher decision levels, thus
making extraordinary measures possible. The interaction context outside the
focussed relationship also influences the
process, partly by the initial failure of an
indirect contact to the Publisher, the
Author, to deliver on time. It also shows
the influence of the end users, the
Universities and the students, who need
to have access to the books at a defined
time. Thus, in this illustration, we can
identify that at least four of the variables
of the model have been at play. Analogous to the discussion from the first
illustration, one can also imagine the
presence of other variables of the
model. The situation is different, and so
are some of the variables at play, but
the model supports the analysis.
2.2 Reflections on the interaction
model
The two illustrations serve
to point out how the model can be used
as a means to deepen our understanding of the ways in which exchange
processes and relationships are conditioned by important elements of the
other variables. The reverse is also true;
it helps in analysing the ways in which
exchange processes condition other
aspects (e.g., the ability to implement a
certain product strategy). Business
relationships are complex phenomena,
and we need to be able to distinguish
one kind from the other and to analyse
various situations with some accuracy.
The interaction model is therefore, to my
mind, a helpful tool when one wishes to,
among other things:
- Analyse and understand specific
events within a relationship (e.g., specific efforts to change the content of the
exchange processes, e.g. changed
intentions to value creation).
- Analyse and understand the relationship as such; is it a strong or fragile
one, are there power-dependence
imbalances, etc.? Based on such an
The IMP Journal
Volume 4, number 1
analysis, important inferences may be
drawn.
- Understand specific drivers underlying
(variables influencing) interaction patterns as well as behaviour and responses to various initiatives
- Understand and – perhaps – prescribe
relevant and matching contact interfaces
and interaction patterns between two
parties.
The model serves as a
way to synthesize numerous potential
situations and patterns of behaviour.
This is very attractive to me and, in my
opinion, makes the model unique.2
It has not been a core
issue to try to become normative among
IMP scholars. But based on empirical
observations and the conceptualization
provided by the model, I believe we are
able to address, in meaningful ways,
questions such as the following:
- What should we expect from suitable
interaction patterns in a relation built
around a standard solution (simple
product/service) in contrast to a complex
and – perhaps – individually forged one?
- What should we expect from interaction patterns in a relation built around
the supply of major production facilities
or production equipment in contrast to a
regularly delivered component?
- What should we expect in terms of
interaction processes (information exchange, safe-guarding behaviour, etc.)
when the customer has a single-source
strategy in contrast to multi-sourcing?
- What should we expect from
interaction processes in a relation
characterized by genuine distrust (in the
short- and in the long-term) in contrast
to a genuinely trustful relation?
- What should we expect in an exchange
process where there are no previous
experiences of doing business between
the parties, in contrast to highly repetitive business?
- What should we expect from an
interaction process that is primarily
2
It is true that there are alternative formulations of the
model, e.g. Turnbull & Valla (1986) or Halinen (1994).
But I connect the efforts to systematically look at and
compare relationships with the original model.
8
based on personal trust between the
involved individuals, in contrast to one
primarily based on corporate- wide performance?
In the introduction, I set out
to discuss the findings based on this
model from several points of departure,
foremost a student’s, a practitioner’s and
a researcher’s perspective. Below
follows a brief discussion that starts from
these perspectives. An overview is summarized in Table 1, below. Following this
summary is a more general discussion,
using this framework, of some of the
developments that have occurred and
some that are ongoing.
To my mind, the interaction
model reveals its core strengths when it
is used as a tool to investigate ex-change processes and analyse episodes and
relationships. To me, this has been very
helpful in lecturing to students as well as
practitioners. In relation to both categories, the model has presented a
framework and provided a means for
carrying on relevant discussions. It
creates a generalized preparedness to
deal with upcoming or experienced
situations. It therefore takes us much
further than mere examples and
anecdotes. In addition to these categories of users, there are some experienced problematic features of the
model (as I see it). Among non-analytic
practitioners, one expressed view is that
the model is “overdoing” a very natural
thing: doing business and handling
business relationships! Another practitioner remark is summarized in the following general statement: “It still doesn’t
give much guidance in making decisions
and taking actions!”
Among students, one remark is that the interaction model
appears to be old-fashioned, as most
applications still seem to be from
manufacturing industries. This view is
also strengthened by the fact that some
of the important variables are “production technology” and “product technology”. In practice, business exchanges are increasingly about services and
broader market offerings, making these
concepts look old-fashioned. The other
The IMP Journal
Volume 4, number 1
remark made by students is that students tend miss the explicit role of
individuals and their interpretations,
decisions and actions. I have frequently
heard the view expressed that the model
is very structural and more or less deterministic.
Researchers have also
addressed the model’s strengths and
weaknesses. They seem to appreciate
how the model allows them to systematically analyse business relationships in
an interesting way. And there is a
recognition that the model has helped in
opening doors to new and relevant
questions/phenomena. The criticism
from this group appears to be that many
9
researchers focussing on business
relationships seem to prefer more
“open” views on relationships. The
model steers the view of relationships in
a specific direction and, thus, hinders
researchers from seeing other dimensions than those included in the model.
Furthermore, it may be too closely connected to a manufacturing industry
framework, as pointed out in the students’ views, mentioned above. This is
summarized in Table 1.
2.3 Some developments
along these lines
The
discussion
above
Table 1
A summary of the discussion on experiences of utilizing the interaction model within the
IMP tradition. The pros and cons have been discussed in the text and are significantly
shaped by the author’s own experience. This is true for all three experience domains.
Experience Students
domain
Interaction
model
Pros:
- Accuracy in
understanding business
relationships.
- Realizing how the
relationship notions
influence our view of
other phenomenon.
Practitioners
Research
Pros:
Pros:
- Framework for
- Systematically analyse
analysing specific
business relationships.
business contexts.
- Seeing phenomenon
- Appreciate the
in new light.
emphasis of the
atmosphere concept,
Cons:
including power relations, - Prefer more “open”
etc.
views on relationships.
Cons:
- Too closely connected
to a manufacturing
- Appears old-fashioned Cons:
(manufacturing
- Feels like “overdoing” a industry framework.
very natural thing.
industries).
- Missing explicit view of - Not much explicit
“individuals’”
guidance in making
decisions and taking
interpretations and
action.
actions.
The IMP Journal
Volume 4, number 1
pointed to some ways in utilizing the
interaction model to generalise about
expected patters of interaction in various
situations. In this section I want to take
one
step
further
and
illustrate
developments along these lines with
some studies published. It is meant as
illustrations not a comprehensive list of
all relevant studies.
Already in the IMP book
from 1982, various authors discussed
themes that had emerged from the
interaction model, e.g. patterns of
interaction caused by the production
technologies exercised (Johanson, ibid),
the role of the individual’s personal
contact
network
(Cunningham
&
Turnbull, ibid.). One study along these
lines that came later is Hallén and
Seyed-Mohamed (1991), where the
impact of production technology on the
interaction pattern was investigated
using statistical analysis of sample data.
They found a statistically significant
outcome for interaction patterns and this
variable. One recent study is the
development of a classification scheme
for services based on how the service is
used by the customer: as a component,
as a working method (instrumental) or
as a semi-finished (raw material) service
(Wynstra, Axelsson, & van der Valk,
2006). From this point of departure, the
researchers identified “ideal” interaction
patterns for such services (dialogue
orientation, contact patterns). The
deduced interaction patterns have also
been tested to determine whether or not
they are a necessity for effective interaction patterns to prevail (van der Valk,
2007). The results are quite convincing
from a normative point of view. It is a
theory test carried out using the case
study method (34 business relation
cases were used to test the theory). It
gives a strong argument for the
proposed patterns of interaction as a
necessary condition for well-functioning
(effective) relationships.
There are also other
initiatives that could be argued to
address topics like this. Such themes
include ways to handle relationships in
order to develop trust. The purpose of
10
this section was to illustrate some
developments and neither to provide a
comprehensive documentation nor a
state-of-the art positioning. Still, I think
the early orientation of the IMP research
as manifested in the Håkansson ed
1982- study and the above venue of
example studies does not seem to have
been frequent on the agenda during
later years.
The interaction model has
been highly fruitful. As pointed out
above, it has provided ways for students
and business managers to build a nuanced view of the kind of situation or
relationship they are looking into. There
should, however, still be a great deal
more to say about this. Thus, I think
much more of this kind of structured
analyses would be of great value in the
future. In developing such themes, many
other bases of knowledge could be
utilized, e.g. lessons learnt about
power/dependence, trust- creation, interaction in relation to b2b- services, and
so on.
2.4 General reflections and notes on
potential problems associated with
the interaction model
Over the years, there has
been some questioning and criticism of
the model (and the IMP approach as
such). This has already been addressed
to some extent from the point of view of
the three segments above. The following
is the author’s own synthesis.
Three such critical issues
seem to be rather common: 1. The
model’s relatively limited focus on
individuals involved in business and
their interpretations and actions, and the
implications of this. 2. The model
provides a view of being static; business
practice is often highly dynamic, which is
not adequately reflected by the model
and applications carried out. 3. The
environment outside of the dyad is not
dealt with using sufficient sensitivity.
These three topics will be considered
individually.
The IMP Journal
Volume 4, number 1
1. The critical points of view on the
role of the individual and her
interpretations, etc., are actually
put forward in the basic model.
The characteristics of the involved parties distinguish the organizational level and the individual. Therefore, this claim
should probably be considered
more a problem of application. It
may be that we have seen
relatively few studies focussing
on how the actions of either party
are interpreted by involved
actors. One exception is Klint
(1985). He indentified a number
of critical events (episodes) that
took place in relationships
between buyers and sellers in
the pulp and paper business.
Respondents were asked about
their interpretations of these
episodes based on a frame of
reference that defined a relationship context3. One could
think of a few other studies along
these lines. Still, it is probably
the case that the bulk of studies
based on the interaction model
have looked more into interaction
patterns as outcomes of structural variables such as production technology. These variables
as a basis for analyses may
seem less relevant in many
contexts, e.g. many interaction
processes involving services
firms. A study of research in
sales management by Åge
(2009) points to this issue;
interaction processes are key to
quality in services. This is due to
the importance of staff performance in various service
encounters that involve actually
meeting customers. Bringing the
right people into the interaction
process, i.e. designing the
supplier-customer interfaces, is
3
It could be argued that he did not make use of the
whole interaction model and therefore would not be a
totally relevant example of how the model has been
used. But the focus of the study; the episodes in
relationships, interpretations and outcome on perceived
demonstrations of trust, power, etc have the kind of
business and analyses “flavour” I am thinking of.
11
not only a matter of the right
position, role and general
competence. It is also largely a
matter people on both sides of
the relation getting along well
together – of being able to pick
up the signals of the other party
even when they are not expressed in written or spoken
language. It may be that the IMP
school has underplayed this
aspect, perhaps not so much in
the construction of the model,
but when judging on the basis of
most applications of the model.
2. The next criticism, or claim, is
that the interaction model is
static, whereas business practice, interaction processes and
relationships are highly dynamic.
There are gives and takes, steps
forward as well as steps backward, suggestions for alternative
solutions and prices, negotiations, changes in partners, conflict and cooperation, etc. The
dynamism aspect does not seem
to be apparent. This may to
some extent be a model design
problem. When we wish to
illustrate dynamism, we tend to
use, e.g., circular or spiral
technique (e.g., vicious circles)
or feedback loops. There are no
such techniques in place in the
visualization of the model. Therefore it does not look dynamic.
However, there are several dynamic concepts included, such
as the emphasis on processes
(interaction, exchange, etc.),
power/ dependency, and (degree
of) dynamism in the environment.
There is some early work
pointing to a certain kind of
dynamism, e.g. Ford’s (1980)
discussion of business relationships passing through various
stages of development, though it
may be criticized for not strongly
emphasizing that there is no selfevident movement from one
stage to the other and that the
The IMP Journal
Volume 4, number 1
processes may lose their direction. This discussion is still about
interaction, and it is dynamic in
nature. In the future, it may be a
relevant priority to somehow
bring more dynamism into the
model, both also to apply it to
contexts and situations that are
truly dynamic – and to focus on
the dynamism aspect! This is
definitely an important feature of
business practice, and it would
be satisfying to see it convincingly captured in future research.
3. The relationship context is a very
broad category in the original
model. What impact it could have
on interaction processes and
relationships is to some extent
addressed in the 1982 book.
There is, e.g., a discussion on
interaction processes and the
environment; whether it is a
dynamic or stable environment,
or whether the market is
homogenous or heterogeneous.
But that is a rather brief
discussion, and when it is
obvious that other relationships
often significantly impact on a
focussed relationship, it is clear
that one needs to take a closer
look at the environment. The
interaction model does not
provide that connection. (Models
need to be simplified and focus
on something.)
In contrast to the two critical
topics raised above, this one is
hardly caused by lack of
applications. Quite the contrary,
there are a huge number of
studies on business networks.
But in those studies, IMP
researchers have gone beyond
the interaction model and moved
into the ARA model (or other
potential frameworks). In a way it
could be argued that studies in
line with the ARA model have
even replaced studies in line with
the interaction model. This could
12
be seen as a move towards
becoming more accurate in
defining and scrutinizing the
fourth group of variables: the
environment of the relationship.
The connection between the
interaction and the ARA models
deserves more attention than
this, and thus I will return to this
issue later on.
Before moving on to the
ARA
model,
some
summarizing
thoughts should be presented. The
interaction model has been very helpful.
It has clearly supported both students’
and managers’ thinking about relationships. It can be used, in more principal
and generalized terms, to understand
the specific interaction context. It has
also guided researchers into interesting
projects. Still, our understanding is not
yet complete, and it would be desirable
to see more work in this field that builds
on the model. In the discussion above,
some such areas have been pointed at.
3. The network model – bringing the
environment in
A major move in the IMP
research was when researchers started
taking a closer look at more than one
relationship at a time, thus shifting their
attention more strongly towards the
external context of the specific
relationship. Based on the notion that
relationships (dyads) often are conditioned by relationships outside the dyad
– and that the dyad has an impact on
relationships outside the dyad, the
environment was looked upon as
various kinds of network settings. This
relates to the third critical reflection
above. What is outside the dyad could,
and in many cases perhaps should
preferably, be considered embedded in
contexts of – somehow – connected
relationships.
The key model developed
for analysing business networks is the
aforementioned ARA model. Even if it is
assumed that the reader is well aware of
this model, some words of characteri-
The IMP Journal
Volume 4, number 1
zation are needed as a background to
the coming discussions and illustrations.
The framework builds on
the three layers of substance in a
relationship and/or in networks: activities, actors and resources. Actors could
be firms or individuals as well as units in
a firm. They could also be other
organizations, such as governmental
bodies and other stake-holders. Actors
carry out activities of various kinds
(production, distribution, communication)
to perform functions, i.e. to create value.
In doing so, they need to have access to
resources
such
as
knowledge,
production equipment, energy. Ongoing
operations lead to the creation of
connections between activities among
several actors, thus causing activity
networks to emerge. But it is also the
case that the existence of activity
networks makes it possible to connect
activities and thus causes activities to
emerge. Therefore, networks are
undergoing change, but frequently with
a specific pattern as their point of
departure. Ongoing operations likewise
call for the mobilization of resources,
thus causing related resource networks
to emerge. It is, again, true that the
resource layer is an important prerequisite or context for new resources.
The performing of activities and
mobilization of resources involves actors
of various kinds, thus causing networks
of actors to emerge. These networks
(activity, resource, and actor networks)
are more or less overlapping and
connected. Researchers may choose to
put the primary focus on, e.g., the
resource dimension (cf. Baraldi & Vedin,
2006), the activity dimension (cf.
Waluszewski, 1996) or the actor
dimension (Liljegren, 1988). In doing so,
the other dimensions are still in
operation, but placed in the background.
Naturally, researchers could also put
equal weight on and strive to explicitly
integrate the three layers.
This is the basic frame.
Based on empirical studies and
analyses
of
business
networks,
researchers have noticed that network
settings can differ and can play significant roles in various ways in business.
13
There has been a great deal of
conceptual thinking that often has
generated relevant and interesting
knowledge. The ARA model has been
appreciated for the structure it brings to
descriptions and analyses, but also
criticized. Later on we will return to
some of the principal topics of such
criticism. It should also be emphasized
that alternative network models do exist.
Some of them emphasize what have
been called “strategic networks”, which
are often actor-oriented and focus on
top levels in the firms and which may
include explicitly declared (formalized)
strategic alliances between two or more
cooperating firms (Thorelli, 1986 and
more recently Gulati et al., 2000). Other
studies are more concentrated on the
level of individuals and tend to either
focus on the social capital of the
individual, allowing, e.g., the individual
entrepreneur to mobilize resources to
benefit ongoing ventures, or on
structures
and measurements
of
network formations. Studies of this kind
tend to emphasize the strong tie/weak
tie notion (Granovetter, 1973) and the
filling of “structural holes” (missing
connections) by linking relationships (cf.
Burt, 1992). To my mind, much of the
variance in network studies and foci
boils down to two fundamentally different views and definitions of networks
and their resulting varied implications.
Networks could be defined
as:
A. “Alliance networks” that constitute
networks of co-operating firms that have
explicitly decided to join forces as a
group. An alliance network is thus a
defined group of two or more partners
who, to some degree and in some parts
of their activities, will have common
goals and a joint view of the network.
Such networks can be defined as “A
number of actors acting together in
order to achieve common goals”.
Selected members create
alliance networks for specific purposes.
The other alternative view is the following.
The IMP Journal
Volume 4, number 1
B. “Emerging networks”. We find
networks emerging from co-operation
between
individual
actors
(firms,
organizations, or individuals). As a result
of ongoing activities involving various
actors in a field of activities,
relationships and network structures
evolve. These structures can be
distinguished by various themes, e.g. an
industry. There is thus a web of
interlinked connections, but no clearly
defined boundaries excluding certain
actors from the network.
Each firm cooperates with a number of
firms and other organizations. To some
extent, the partners involved in a focal
firm’s network can be identical to those
of another network, and to some extent
they can differ. The buyers and sellers,
around which the network is formed, can
be active in different stages of a typical
value added chain. However, they are
all to some extent connected. It may
also be possible to distinguish a number
of more or less overlapping networks. A
firm may be part of many networks, for
example within an application area,
while an individual may be part of
numerous networks of professionals.
This means that there are no clear
boundaries of the network. There is thus
no such thing as the Network. One
actor’s view of the network may differ
from the views of other actors. The
actors involved may even have
significantly different goals for their
network activities. This kind of network
can thus be defined as “A number of
connected exchange relationships” (cf.
Cook & Emerson, 1978).
The ARA framework is
based on the latter definition. It deals
with networks that are ever–expanding
(connected exchange relationships),
with no clear boarders defining who is
inside (e.g., a member of the alliance)
and who is outside. It emphasizes the
connectedness and emerging aspect.
As new exchange relationships becomes relevant to include in the study it
could be done – and thereby new actors
as well as resources may become parts
14
of the study and analysis. It is, thus, not
a matter of pre-planned and defined
structures, although there is recognition
of the existence of so-called “locomotive”-dominated networks, i.e. networks dominated by one or a few actors
(which are not necessarily able to
govern the network in hierarchical
ways).
In what follows, my focus
will be entirely on the ARA framework,
as this is the core IMP domain in this
regard. Let us first, as we have done
above in the discussion of the
interaction model, look at some illustrations and some of the ways in which the
ARA model may serve as a tool to
improve our understanding of the various expressions of the business
network phenomenon.
Below are four illustrations
of business networks. They have been
documented during different time
periods and relate to different industries
and businesses. Three of them have in
common that they build on texts
published in various forums and that
they illustrate significant aspects of
business networks. The first illustration
shows how networks can sometimes
become strong barriers to change. It is a
case study in a doctoral dissertation
(Laage-Hellman, 1989). The second
case has not been published, but has
served as a case in lecturing business
networks (as with the two cases used to
discuss the interaction model). The third
illustration shows that relations and
networks can be a very strong force in
mobilizing actors and resources to
achieve a certain goal. It builds on a
book chapter on networks and business
renewal (Blombäck, 20002). The fourth
illustration shows how a company can
actively utilize its international network
to promote a certain development. It
builds on a case study prepared for
educational purposes. It thus has
similarities with the first illustration, but is
at the same time highly different in other
respects.
ASP – bringing new technology
to the market
The IMP Journal
Volume 4, number 1
A steelworks in central Sweden,
Söderfors, working in cooperation with the equipment producer
ABB (one of ABB’s business
areas at the time), developed a
new tool-steel based on powder
metallurgy. It was a development
process that took more that ten
years and entailed considerable
investments, not least in terms of
the time spent by several
qualified people in a number of
involved companies, throughout
the process. The steel was
called ASP, based on the initials
of key actors involved. It was
meant as a material for the
production of cutting tools like
drills, twisters and the like, to be
used in industrial applications. It
should replace the, by then,
predominant conventional speed
steel. The most important actors
in a system such as this are the
following: the steel producers
who make the material, the tool
producers who make the tools
from the material, the distributors
of the tools and the final users,
predominantly in the car production industry. When car producers make motor-blocs and other
metallic components, they need
to carry out numerous cutting
operations. To do this, they use
machines (robots) to which they
have attached several tools
(drills). In this case, ABB played
a role in the development of the
new ASP material as an equipment producer and also as a
provider of equipment to the
steel producer.
ASP is a high-alloyed material
and the production process is
very complicated. This makes
the ASP steel more expensive to
produce than the conventional
material. But when the material
is in place, and the toolmakers
take on their role in the value
creation process, they have an
easier job. As a result, the final
tool need not be more than 20
15
percent more expensive than the
conventional. This additional cost
should be measured against the
advantages.
Pilot tests had shown that the
new material was more robust
and that the structure of this new
material was more fine-grained,
thanks to the powder metallurgy.
As a result, when it was used, it
deteriorated in a smoother and
more even process. The material
could be used twice as long in
the cutting processes, and thus
withstand twice as many cutting
operations (cutting hours) as
conventional steel. This was
good news. Who could resist a
product that was twice as good
in the key selling point at only 20
percent higher costs? There
should be a margin for increased
profits for all parties involved!
Perhaps a final market price that
is 30 percent above the present
price of these products – and
that offers twice as good a
performance – would be the final
result.
The market introduction of this
new material was not at all as
easy as expected. The natural
first stake-holder to turn to would
be the tool producers. They
would get a wonderful new
material. However, it turned out
that they were not enthusiastic at
all. If the material can be used
for twice the number of operations, we would lose half our
sales. How can we compensate
for that? If only a portion of the
customers turn to the new
material, we would have to run
two production lines and account
for
stock-keeping
of
two
assortments; that would be
costly and complicated. This is a
new material and someone
needs to convince the big users
in the automotive industry; we
don’t have such a market
organization. The management
The IMP Journal
Volume 4, number 1
of Söderfors then turned to the
end users. The most nearby
producers, Volvo and Saab, had
no interest at the time. They had
so many other options for
improvement. Next stop was
Germany
and
German
producers. They also had many
other
options.
They
were
impressed by the product, but
saw no real advantage in using
it. These tools did not represent
significant
costs
in
their
production relative to many other
activities. They did not like the
idea of becoming dependent on
one small supplier from far away.
They anticipated a heavy
process in convincing various
internal stake-holders. All-in-all,
the market was not interested.
In a next phase, when rethinking
this product and its potential
value, a new idea emerged. Why
not use the material to run the
operations faster? When the
speed of cutting and twisting was
increased by 30 percent, it
turned out that they needed to
consume the same number of
twisters and cutters. That was
good news for the tool producers. But even more so for the
car producers. They could now,
in some of their applications,
reduce the number of workstations as well as the number of
people operating the workstation.
In this way, they could, e.g.,
reduce the number of workstations from 21 to 14 in one
specific application. That made a
huge difference! The machines
used needed refurbishment or
earlier replacement, which also
made the machine producers
happy. There were, thus, at least
“four winners” (steel makers, tool
producers, car producers and
machine equipment producers).
Based
on
this,
Söderfors
gradually started to work this
product into the market, but it
took a lot of changes in
16
organization, reallocating people
with specific expertise. (Based
on Laage-Hellman, 1989).
In this case, it is clear that
there was a very cemented structure of
activities, resources as well as actors.
The activity patterns had been developed and maintained over a long time
period. Resources and activities had
been allocated and trimmed and the
roles of the involved actors specified
and step-by-step adapted to gradual
changes. These created “hard” structures that even “the best product in the
world” had a difficult time defeating. But
given that there were several winners,
they made it at last. In this case, the
ARA model provides strong analytical
support. It is possible to trace all
activities internal to each actor and
between all actors; moreover, the
resources invested and used in the old
methods as well as those needed in the
new method could be identified. Thereby
one can look into what changes are
needed by actors in the entire network
and, based on that, imagine what
benefits must be achieved in order for
this to be successful. Existing actors as
well as needed changes inside presently
involved as well as actors to become
involved could be analysed. This was
not done in the actual situation, however.
Again, this story says
something about the impact of network
structure. On the actor level, this
network is less dependent on specific
individuals and more on the structures of
activities as well as constellations of
resources built over time. Exchanging
for example some individuals would
probably not make much difference in
this context. So much has been build in
into the basic activity structures and
related resources. Any change needs to
be tailored to result in significant
improvements involving at least two but
preferably more “winners” than that.
The second illustration is
slightly different.
The IMP Journal
Volume 4, number 1
Kinnarps in France
Kinnarps is a major family
business and a leading producer of office equipment. The
basic idea is to deliver functional
and
ergonomically
excellent systems for offices.
”Personal energy” is the
visionary idea in their market
communication. The unique
selling point is that the company takes responsibility for
making a diagnosis of each
office and providing a good
”systemic” solution primarily
based on the company’s product lines (chairs, tables, etc.).
Some twenty years ago,
Kinnarps had no export to
France and no other international customers outside Scandinavia. The market in Scandinavia was very good, and the
management team had no
plans for any further market
extension than that. In spite of
this,
Kinnarps,
somehow,
received an order from a
Norwegian firm (Norsk Veritas
AS) situated in France. The
request was met and it turned
out to be the first step into the
French market. Norsk Veritas
had become familiar with
Kinnarps’ products from operations with customers in Sweden. Now someone from its
French office wanted such
furniture there.
Torsten Ekstrand, then in
charge of the Swedish Chamber of Commerce in Paris,
happened to see the furniture
on a visit to this Norwegian
company. He liked the items so
much that he also bought
Kinnarps’ equipment for the
office of the Chamber of
Commerce. Wooden furniture in
offices was quite unusual in
France,
and
the
French
17
business partners of the Chamber showed a great deal of
interest in them. ”Maybe it
would be a good idea to start
selling these pieces of furniture
in a systematic way”, Ekstrand
thought. He was – anyway –
soon to retire from his job at the
Chamber.
Ekstrand made contact with the
founder and CEO of Kinnarps.
This key individual agreed to
the proposal to appoint Ekstrand as Kinnarps’ representative (agent) in the French
market. Ekstrand hired a
French native to support him.
They got going very soon.
There were many exhibitions
and trade fares in a short time,
and Kinnarps became represented. Suddenly, there was a
major order from Renault, the
French firm that by then was in
a merging process with Volvo,
the Swedish automotive company. People from Renault had
seen Kinnarps furniture on a
visit to Göteborg.
In the early 1990s, everything
looked very promising, when a
dramatic change took place. The
recession that came at the same
time as the Gulf War in 1990/1
led to a dramatic reduction in
Kinnarps’ sales in France. In a
few years, Ekstrand retired as
CEO of the French operation and
Kinnarps had to take over (insource) the operation. It turned
out that, despite the initial
successes, it was difficult to
manage the French market.
Kinnarps’ furniture was something genuinely new, but it didn’t
really seem to be attractive
enough. Gradually, people at
Kinnarps realized that:
• In France, normally only top
managers have their offices
equipped with wooden furniture; ordinary employees
don’t have things that nice,
The IMP Journal
•
•
•
Volume 4, number 1
There were also great doubts
as to whether this furniture
would be destroyed by coffee, there is a very widespread habit in France of
bringing your own coffee to
the workplace,
Furthermore the sizes of
Kinnarps’ furniture did not fit
perfectly with French standards.
In addition, the French customers were used to higher
discounts on list prices than
in the Scandinavian context.
Later on, the situation on the
French market improved again.
Some years back, Kinnarps
launched a new ”European
adapted” series of furniture,
which is expected to solve some
of the problems mentioned. The
involvement in the market has
deepened. Kinnarps is establishhed in Paris, Lille and Lyon. Still,
there is considerable uncertainty
as to what to do in order to gain
a bigger market- share in France.
This case demonstrates a
situation characterized by much less
hard structures of activities, actors and
resources. The products in focus are not
equally strictly embedded in related
activity structures as in the ASP case.
There are, partly as a consequence of
this, much more opportunities for
change, for finding new customers and
for influencing customer behaviour.
Furthermore, the events and resulting
network connections have developed
largely as a result of seemingly random
events, i.e. through serendipity. The
actors are looking for business opportunities. Decisions and actions come to
the surface and they act on the opportunities they perceive and evaluate as
“fit for purpose”. The network will, as in
the ASP- case, put some limits to acting.
But in this situation it is a much broader
space and there are many more
18
opportunities to mobilise others and find
fruitful options to act.
The illustration has in
common with the previous illustration
that what has happened before acts as
some kind of relationship sediment
(Agndal & Axelsson, 2004) on which to
build actual and future business. Because someone from Norsk Veritas had
seen Kinnarps’ furniture when visiting
customers in Sweden, someone from
the French part of the company wanted
to bring Kinnarps to France. Because
Renault and Volvo had their joint
venture, people from Renault came to
see Kinnarps’ furniture at Volvo in
Sweden, and they wanted Kinnarps in
France too. Such previous developments create an arena for new events to
take place. In this illustration, we can
also notice a much stronger focus on the
actions of specific individuals. Individuals really seem to make a difference!
Individuals, acting within the frame of
present and future business structures
and events, are able to mobilize other
actors and resources to make a change.
Still this seems to be within rather
narrow confines. But what is described
could be understood as actions within
activity structures that are not very strict
and structured, but still established as
structures. The ARA framework also
seems helpful in analysing this situation.
The third network illustration is – again – different.
The
Freeloader
networking to mobilize
case
–
Freeloader was an IT company
founded by two partners, Mark
Pincus and Sunil Paul. Their idea
was to create an instrument for
the Internet market. They wanted
to develop an off-line browser
that loads your favourite websites even when you are not
logged on to the Internet. This
was by then a new feature.
They really entered a networking
process. First of all, they made
contact with anyone in their
surroundings, a number of good
The IMP Journal
Volume 4, number 1
friends, whom they thought could
be of help in realizing the idea.
Mark and Sunil contacted – via
such “contacts” or acquaintances
– a multimedia company, who
helped them build a prototype of
the product. The next step was
to convince some other friends
who happened to be in the
Government to find respectable
test focus groups. This helped –
in turn – the firm to raise venture
capital to proceed building their
fame. The next step was to
approach some people in finance
they knew and thereby to attract
the interest of some Venture
Capitalists. These – in turn –
became bridges to major Internet
and media actors.
Freeloader used an advanced
strategy when entering the PR
community.
According
to
Anonymous (1998), the company
targeted wire services knowing
that other media (newsprint, etc.)
often follow in their footsteps.
Pintus was, apparently, also a
good story-teller, and he actively
contacted and met with reporters. The message Freeloader
used was that their offer was the
latest thing in the market and
also “the next thing” in the
market. When Pintus ran out of
stories to tell about Freeloader’s
promising product, he chose to
based his arguments in the story
of another company’s similar
product, which would soon be on
the market. All this gave results
when, for example, during two
hectic weeks, Freeloader was
featured in US News & World
Report, Newsweek, Investor’s
Business Daily and Information
Week.
The importance of good PR and
hence connections in the media,
as well as the deliberate strategy
of positioning and building the
image of Freeloader, is some-
19
how illustrated by the following
quotations:
“It was positioned as a savvy
happening in a hot space”. “The
ones with better press are
getting big bucks, acquired, or
venture-capital backing. It bears
no relationship to the quality of
the product or the quality of the
team.”
Eight months after founding the
company, the two founders cashed out 38 million dollars from
the buyer Individual without
having proved revenues of any
sort. However, the acquisition
was preceded by a “bidding
race” between several interested
parties. When Freeloader got the
first bid from Individual, they
wanted more and hence turned
down the offer. By receiving
promises from financial institutions to underwrite private
placements,
Freeloader
put
themselves in a better position
for negotiations. The company
also took every chance of mentioning that they had been
offered an acquisition when
meeting other potential buyers,
hence stirring up the surface and
playing actors off each other. It
turns out that Individual, who
finally bought Freeloader, experienced the situation as a “ ..kind
of rush”.
Further, as incentives for acquiring the company, the chief
financial officer of Individual
mentions: “….they had a very
good brand recognition…I think
we thought it was good for our
business. We were willing to
deal”. The Freeloader product
proved to be no real commercial
success in a market that changed very fast and the company
divested after less than a year’s
operations in the new parent
company. Individual did let go of
the Freeloader product. It was
The IMP Journal
Volume 4, number 1
no longer the hype of the market
and the market niche developed.
The investments needed for
becoming a major actor in the
field were too high for Individual,
who hence decided to stay in
their old core business. (The
entire description builds on
Blombäck, 2002).
This case illustrates that
sometimes, in some networks and in
certain general environments such as, in
this case, the early IT/Internet hype,
networking efforts can mobilize significant amounts of resources as well as
generate activities and involve actors. In
terms of substance, such as established
activity structures of, e.g., ongoing
operative activities, there was very little
in this case. There was no cemented
structure of activities. The same applies
to resources. Huge amounts of
resources were mobilized, but they
brought very little substance in terms of
resources dedicated to specific aims. A
rather large number of actors were
mobilized for the mission. But in this
gradually emerging network, they were
not yet deeply committed to the mission.
All such observations point to a situation
with a nascent, emerging network that
may have to stabilize. We can also see
that it is the two focussed network actors
(individuals) who actually perform all the
acts of mobilization. It is also clear that
this network builds heavily on the
personal contacts between involved
individuals.
These observations and
reflections point to structural aspects as
the reason for the emerging pattern of
behaviour. There also seem to be
“cultural” reasons (the hype-based
stress). Regarding the ARA model use
in a case like this, the focus would be on
the actors and the resources created. It
was a process of creating commitment
from various stake-holders and of
generating some resources for the short
term (e.g., to develop a prototype), but
also for the long term (e.g., to achieve a
financially strong position for the
company by attracting financially strong
owners). The activities were thus far
20
rather disperse, resembling a number of
campaigns (or actions) serving several
purposes along the way. Analysing the
activities as activity structures seems,
however, to be less meaningful at this
nascent stage of network development.
The ARA framework may
help us understand why some actors are
– and why these specific actors were –
so successful in mobilizing other actors
and in making them commit resources.
But it may not add so much to our
understanding of a networking process
like the one described.
The fourth example complements the other ones. It is a network of
structures similar to the first case (ASP).
It has much more of a planned and
conscious action view than the second
(Kinnarps).
Like
the
third
one
(Freeloader), it demonstrates deliberate
efforts to network in order to achieve the
progress wanted. Thus there will be
some similarities and some differences
in relation to each of the previously
presented examples.
ABB Robotics’ entry into Japan.
Positioning vis-à-vis competitors
In 1982, ABB Robotics decided
to establish itself in Japan. If it
was to become one of the
world’s leading robotic producers, it was deemed necessary
to be present in that market, as it
represented around 40 percent
of worldwide sales. Japanese
suppliers were beginning to
export robots and ABB Robotics
realized that competing with
them in their home market could
create valuable advantages. It
would be possible to get a better
idea of their way of doing
business, and this would be of
use in competing with them
when they entered the North
American
and
European
markets. Another advantage of
deploying activities in Japan was
that
Japanese
companies
abroad, for example car transplants, the fastest (almost only)
growing new production fac-
The IMP Journal
Volume 4, number 1
tories, often bring their suppliers
with them from home. ABB
Robotics would then be in a
better position than other robot
producers who were not active in
Japan. The possibility to “strike
back” more easily at the Japanese in their home market if they
became too aggressive in
Europe and in the US did not
play a major role in the decision.
ABB Robotics established a joint
venture with the Swedish trading
company Gadelius, which had a
very long tradition in Japan. It
had until then been the only way
of entering this protected market
since it, because of history, was
considered a Japanese company. Gadelius, however, had
limited experience in the industry
automation business and also a
limited number of contacts with
actors in the field. In spite of that,
the development looked promising and the operations were
considerably enlarged. Thanks to
major efforts involving high costs
for the operation, they managed
to establish some customer
relationships. However, in the
second half of the 1980s, the
positive development halted, and
ABB Robotics faced a worldwide
economic crisis. The management decided to close down the
activities in Japan.
The crisis was solved thanks to
an impressive rationalization
programme. But some trends in
the world market, including the
development of Japanese transplants in the US and Japan,
accelerated. In late 1980s, ABB
Robotics again decided they had
to be present in Japan. In 1989
they managed to sign an
agreement with X, one of the
largest users of robots as well as
a producer itself, in Japan. This
seemed to be the best alternative available. The old contacts were no longer possible.
21
ABB’s product line seemed to be
complementary to that of Matsushita. Matsushita became responsible for marketing and selling ABB’s robots. Via Matsushita, ABB Robotics hoped to
reach the Japanese car producers and to get their robots
installed in their Japanese “prototype factories” to become duplicated as transplants in Europe
and the US. Some five years
later, the agreement still had not
been as successful as was
hoped. The market share in
Japan was still low and no
substantial effects had been
noticed in other parts of the
worldwide network involved in
this business. It is not unlikely
that Matsushita was focusing on
selling their own products now
that they had gained the
knowledge and experience they
were looking for. (Based on
Axelsson, Carlsson & Wynstra,
2003).
In this illustration, we see
how actors can try to calculate network
effects. By creating a specific relationship (with automakers in Japan), ABB
Robotics anticipated improved opportunities to reach other actors (Japanese
transplants in the US). This means
reallocating activities as well as resources and (internal) actors. This is a
very clear case of thinking strategically
about networks. In any network situation, there is a repertoire of potential
strategic moves. A network actor (e.g. a
company) could often identify what other
actor or actors it would like to reach. But
this may not always be possible. The
intended contact may not be interested
for various reasons. The focussed actor
may then, similar to his case, consider
other and sometimes indirect ways of
getting there. ABB Robotics wanted to
strengthen its position in the US and EU
markets, particularly by finding ways to
reach Japanese transplants in these
regions. This time they saw one route to
go via the original factories in Japan and
as a second step reach the intended
The IMP Journal
Volume 4, number 1
aim. In general terms actors could
consider other potential moves, e.g.,
trying to find a new partner (as in this
case), integrate more strongly with one
or several existing ones, dissolve or
disconnect one or several actors, bypass an existing contact, create links to
specific partners where direct contacts
are not possible, block an actor to
prohibit that one from interfering (cf.
Smith & Laage-Hellman, 1992). Such
strategic moves could most likely be
utilized in many network contexts, but
they were particularly explicit in this
case. The number of potential
opportunities and relevant strategies are
due to network structures at hand and
other features, such as the degree of
dynamism in the specific net.
3.1 Reflections on the network model
Observations of business
relationships
have
improved
our
understanding of several phenomena in
business
markets.
They
have
demonstrated domino effects – that one
action somewhere in the network may
have consequences for another, sometimes nearby and sometimes distant, but
connected, relation. They have shown
that actions are sometimes strengthened by activities in other relationships
and sometimes weakened. There are
many such observations and, as
illustrated above, structural features of
networks can provide an improved
understanding of specific actions and
contexts.
The ARA model is a helpful
support in analysing potential moves in
line with this. It should be, for example,
a helpful tool in understanding complex
situations and dependencies and
analysing questions such as:
- What should we expect in terms of
activity, resource and actor fit when it
comes to creating change, e.g., introducing a new product or technology in a
dense, well- structured, network?
- What should we expect when the
product or service exchanged is not a
critical one and not strongly dependent
22
on surrounding resource and activity
structures?
- What should we expect when trying to
create changes in a nascent and newly
created network?
- What should we expect from interdependencies of actors and resources
when a limited number of key individuals
dominate the network?
- What should we expect when a network consists of few or many relevant
actors?
- What should we expect when a network is strongly dominated by one key
resource that is controlled by one actor/
firm?
The framework has helped
us understand some other significant
aspects of business. By gathering
experiences and reflecting on them,
contributions have been made to topics
such as processes of mobilizing others
(e.g., Waluszewski, 1996), processes of
trust- creation and processes of identity
and position building (e.g., Henders,
1992). It has also helped in identifying
some general network dilemmas – or
paradoxes – reflecting the dynamic
dimension of networks and relationships, e.g. the following paradoxes (Ford
et al., 2003):
- “A company’s relationships are the
basis of its operations, growth and
development and are at the heart of
its survival. But these relationships
may also tie into its current ways of
operating and restrict its ability to
change. Thus the paradox facing
managers is that a network is both
the source of freedom for a company
and the cage that imprisons it”.
- “A company’s relationships are the
outcomes of its own decisions and
actions. But the paradox is that the
company is itself the outcome of
those relationships and of what has
happened in them. Thus it is possible
to analyse a company’s position in a
network from the premise that the
company determines its relationships, or that it is determined for
them. Both situations exist simul-
The IMP Journal
Volume 4, number 1
taneously and both premises are
equally valid”.
- “Companies try to manage their
relationships
and
control
the
networks that surrounds them to
achieve their own aims. This ambition
is one of the key forces in developing
networks. But the paradox is that the
more that a company achieves this
ambition of control, the less effective
and innovative will be the network”.
I will not use this text to
dwell on these dilemmas. They do
express much of the experience generated by significant numbers of network
studies. The kind of questions and
dilemmas indicated above should be
possible to address and analyse in
systematic ways by using the ARA
model for the basic analysis.
When, in parallel to the
previous section, we look into the
reactions of the three segments discussed before – students, practitioners
and researchers – the following points of
view emerge. (The text that follows will
be summarized in Table 2 below).
Using the ARA model,
students get a complementary and more
realistic view of markets in contrast to
the dominant market framework from
micro-economics. It also becomes obvious that there is a connectedness
between various actions and decisions.
This should be an eye-opener for those
who have not thought of this before.
There seem to be some disadvantages
too. To many students, the model has
been viewed as self-evident (simplistic)
and not worth an intellectual journey.
Another criticism has been that utilizing
the concepts may be considered merely
a “translation” exercise, assigning new
words to things they see or read about
that are expressed using other terms.
Practitioners
seem
to
appreciate at least a “sniff” of the model
when applying it. It gives them a helpful
tool for obtaining a “map” (that’s the
“sniff”, drawing a picture of the network)
and for “seeing” things in a different
way. It also helps them focus on some
key connections and distinguish the
23
important from the less important. The
cons mentioned by this segment seem
to be that carrying out a full ARA
analysis is only attractive to people who
are highly motivated. It is rather demanding. It also seems as if they either find
it trivial or too complicated.
When using the model,
researchers have a method for analysing business on a level where e.g.
internal activities, resources and actors
connects to external (suppliers, customers) and their respective actors,
resources and activities – and not just
involving the most nearby actors to
whom it has direct contact but several
more in the relevant business net4. In
doing so several phenomena are seen
in a different light. That is positive, as it
sheds new light on various phenomena.
The cons are among others considered
to be the following: ARA is sometimes
criticized for providing a structural and
static view of something that is highly
dynamic in nature. Furthermore, it is
sometimes considered to have problems
in terms of reducing complexity, e.g.
some “arbitrary” measurements such as
defining relevant network boarders
(which actors, what activities, what resources to include), business horizons,
etc. This could be summarized as follows in Table 2.
There are, however, also
several other topics, critical and positive,
on a general level that have been
addressed or need to be addressed in
the future. Below follows my synthesis
44
A distinction could be made between the “net” and
the “network” as a concept. One interpretation of this is
that the network consists of all connected exchange
processes while the net consists of a selection made to
extract the ones, for the specific study or purpose, most
relevant parts of the total network.
The IMP Journal
Volume 4, number 1
24
Table 2
A summary of the discussion about experiences from utilizing the network (ARA) model
within the IMP tradition. The pros and cons have emerged in the text and are
significantly shaped by the author’s own experience.
Experience
domain
Students
Practitioners
Research
ARA network model
Pros:
- Complementary view
of markets.
- The underpinning of
connectedness
between various
actions and decision.
Pros:
- Tool for obtaining a
“map” and “seeing”
things.
- Helps distinguish
important from less
important (critical direct
and indirect
relationships).
Pros:
- A method for systematic
analyses.
- Seeing phenomena in a
different light.
Cons:
- Structural and static
Cons:
view of something
dynamic.
- Evident, simple, (not
- Problems in reducing
worth an intellectual
Cons:
journey).
- The full ARA analysis complexity, e.g. some
“arbitrary” measurements
- Utilizing the concepts too demanding.
(network boarders,
a “translation” exercise. - Either trivial or too
business horizons).
complicated.
of the situation.
General reflections and
notes on potential problems associated
with the ARA model
Over the years, there has
been some questioning and criticism of
the ARA model (as well as of the IMP
approach as such). Four such views
seem to be more common: 1. It is highly
structural and deterministic in view.
Connected to this, the ARA model,
similar to the criticism of the interaction
model, seems to provide a static view. A
common question is “where are the
interpretations of the involved actors”?
2. It deals primarily with production
industries. How useful is this when
today’s business climate is dominated
by services? 3. It discusses functions
and resources, but rarely translates this
into monetary forms. 4. How can it be
synchronized with the interaction
model? These four topics will be dealt
with one at a time.
1. The first point is, again, a
reflection of the suggested relatively limited focus on individuals
involved in business, their
interpretations and actions and
the implications of this. This is
thus also repeated in judging the
ARA approach in studying
business relationships, but in the
context of networks. The actor
level has often considered actors
as companies, not so much as
individuals (in companies). The
activity level has been studied as
structures of activities inside and
in- between firms, not so much
as actions by individuals. As a
consequence, it has been
proposed that the model should
be complemented by a fourth
layer, namely individuals’ acting,
interpreting processes in networks (Holmlund et al., 2008).
This is similar to the discussions
on the interaction model above.
The ARA model itself does not
imply such analyses, but many
applications of it have gone in a
structural direction. One exception, which is not unique, is
Waluszewski (1990), where a
The IMP Journal
Volume 4, number 1
highly dynamic development
process is constructed by a lot of
notifying of involved individuals.
There are also network studies
that have focused on actors’
“network pictures” (e.g., Naudé
et al., 2004, Ford, D. and
Redwood, M. (2005) and Ramos
& Ford, 2006), which in a way
brings in individuals’ views and
interpretations. This is true also
for Welch, C. and Wilkinson, I.
(2002). This constitutes steps
towards satisfying some interpretative issues among acting
actors. Still, it is probably true
that it would be beneficial if, in
future studies, the ARA model
were to evoke applications that
are more interpretative and
dynamic in style.
2. IMP research had its origin in
manufacturing industries. The
industry totally dominated the
early studies and probably the
models and concepts developed
as well. This is probably also the
case for the ARA model. But it is
not farfetched to use it in, e.g.,
the case of Freeloader above. In
Axelsson & Wynstra (2002) the
model is utilized as a general
framework for discussion and
analyzing the buying of business
services. The argument is that
services consists of activities and
carrying out activities demand
resources as well as actors.
There may, however, still be a
need to apply the model in many
more studies on services Industries. In this regard, perhaps the
topics discussed in the previous
reflection (dynamism, interpretations, action focus) should stand
out as even more urgent. That
remains to be seen.
3. Some critical remarks concern
the general lack of economic
figures. Compared to some other
authors focussing on value creation in business-to-business contexts, e.g. Anderson and Narus
(2004), this seems to be less
emphasized in IMP. Anderson
25
and Narus’s value calculation
approach
puts
considerable
effort into trying to identify the
various value elements at stake
and, as a second step, estimate
the economic worth of these
elements to various users. The
IMP tradition seems, most often,
to stop after having identified
changed – improved – functions
and the level of improvement in
terms of, e.g., time savings or
increased customer satisfaction
with the business relationship.
One could argue that the
improved functions are the real
values and that the rest, the
translation into money, is merely
calculation. Still, money brings
perceived accuracy and it may
be a good idea to put more
emphasis on economic estimation of the values created. The
same applies to cost and profitability aspects of various customer relationships (customer
profitability, e.g., Storbacka,
1994 and 2004) and cost estimations of various ways to handle
customer relationships (costs of
sales organizations, cost of
marketing, and resulting performance in terms of income generation). This would also be a way
to emphasize the strategic
dimension of relationship management. In business practice,
when talking to CEOs and CFOs,
having a toolbox of financials is
valuable. Admittedly, there are
some exceptions. One such
example is Håkansson and
Lind’s (2004) study on interorganizational accounting. This
connects to the broader field of
inter-organizational management
accounting. I think it would be
beneficial to see more of this for
the reasons discussed above.
4. Topics such as atmosphere –
whether or not there is a good,
cooperative relationship or whether there are differences in
strength and influence – have a
prominent role in the interaction
The IMP Journal
Volume 4, number 1
(dyadic) model. Such aspects
are, to my mind, much more
indirect in the ARA framework.
They should be integrated into
the the Actor layer. But that
aspect seems to be downplayed.
The substances of relationships
are, according to the ARA
framework, the activity links,
resource ties and actor bonds
(Håkansson & Snehota, 1995).
But actor bonds thee way the
framework is presented in this
source seem to me highly generic. There are bonds, social,
technological, economical and
this could easily connect to the
power/dependency, trust/distrust,
satisfaction/disappointment, from
the interaction model. But this
opportunity is not that obvious I
think. Thereby the two models
may look more different than
they are. The actor perceptions
and experiences of what goes on
in various relationships and
networks as well as trust, power/
dependency and other topics
should be integrated in the
model. There is nothing in the
model hindering this, but it
seems to be out of the picture for
some reason. I think, therefore,
that there is a “missing link”
between the interaction and ARA
models in this regard. Finding or
clarifying this link would be a
valuable contribution.
Like the interaction model,
the ARA network model has been a
valuable tool for both students and
practitioners. It has helped in pointing
out the roles of relations as such, it has
pointed out and created awareness of
various relationship-based phenomena
(such as domino effects, the impact of
indirect relationships). It has also helped
in carrying out stricter structural
analyses, as indicated in the ASP
illustration above. And, of course, it has
inspired a great number of researchers
and prompted many good studies.
Making students use the
tools (the concepts) in analysing various
26
network contexts has worked well, but
dealing with managers and encouraging
them dig into it has proven more difficult.
Still, they may appreciate the basic
messages and benefit from their –
potential – changed views. It is probably
the case that this conceptual framework,
which goes beyond a typical two-by-two
matrix in complexity, is too heavy to
digest. Researchers have been more
appreciative. The model has helped
bring about some systematic order. One
critical remark returns to points 1, 2 and
4 above. I think the increasing interest in
Actor Network Theory (ANT, Czarniawska & Hernes, 2005, Araujo, 2007,
Andersson et al., 2008) should be
viewed in light of this. The ANT school
of thought emphasizes the process –
dynamic – and processing dimension of
business networks as well as the
individual’s crucial role as actor. The
approach looks promising even though it
introduces a rather complex terminology. I would appreciate efforts to try to
build these dimensions into the ARA
framework in more distinct ways, i.e. in
line with the suggestions made by
Holmlund et al. (ibid.). Another piece of
work, in line with a view of gradually
correcting or complementing the existing
model, is Easton and Lenney’s (2009)
initiative to add commitments to the
model. When acting in a network, actors
commit themselves to future activities
and to allocating resources of various
kinds to specific aims. In doing so, they
make commitments, promises that form
expectations and are perceived and
evaluated in relation to the actors’
experience, track record and resources.
These two studies indicated points to my
mind into an interesting direction as it
introduces more of a human touch and
could be considered a link to the
interaction model where the actor layer
(including interpretation, trust, atmosphere) is more in the forefront.
4.
Conclusions and some brief
suggestions for future research
This has been a reflective
discussion on relations and networks in
business-to-business settings. The point
The IMP Journal
Volume 4, number 1
27
of departure is my own long experience
of working rather close to IMP research.
The discussion primarily focussed on
the two core models of IMP: the
interaction and the ARA models.
The discussion has pointed to several strengths of the two
models in capturing the complexity of
business-to-business. It has clearly helped both in pointing to relevant phenomena and in providing systematic analysis of them. The latter aspect seems
largely forgotten today as far as the
interaction model is concerned. That
stream of research, however, does not
yet seem to have been filled at all. But
some criticism and problems have also
been identified. It seems as if there is a
need for some vitalization, to find ways
of opening up new doors to phenomena
that are unknown, new or different in
shape. It seems as if both the interaction
and the ARA models would benefit from
bringing in the dynamism and the interpretative individual as actor in new and
improved ways. I have pointed to some
such studies that have already been
presented. I don’t think such a move and
increased efforts in bringing about such
studies requires totally new models.
Complements to today’s situation may
emerge from empirical studies with such
an orientation, which may, in turn, result
in slight refurbishment of the original
models. This would provide stability as
well as change. We would use what we
have and what is confirmed, which gives
stability. And we would try to bring in
and adapt the model to issues that have
received less attention thus far, thereby
causing some change. Isn’t this also in
line with some of the core findings from
studies of product development in business networks? Without some basic
stability, there is no robust change!
References
Axelsson, B. & Wynstra, F. (2002).
Buying Business Services.
Viley International.
Axelsson, B., Carlsson, M. & Wynstra,
F. (2004). ABB Robotics –
25 Years of Managing
Business
Relationships.
Stockholm
School
of
Economics.
Baraldi, E. & Strömsten, T. (2006).
Embedding and utilizing
low weight: Value creation
and resource configurations in the networks
around IKEA’s Lack table
and Holmen’s newsprint.
IMP Journal (1).
Blombäck, A. (2002). Networking, a
Juicy Fruit for Internet and
IT Brands – How Strategic
Networking
can
Build
Baloon Brands. In Agndal,
H. & Axelsson, B. (Eds.).
Networks and Business
Renewal. JIBS Research
Reports No 2002-4, 147170. Jönköping International Business School in
Agndal & Axelsson, 20002
Burt, R. S. (1992). Structural Holes: The
Social Structure of Compe-
Agndal, H. & Axelsson, B. (2004).
Internationalisation of the
firm. The influence of relational sediments. In Forsgren, M., Havila, V. &
Håkansson, H. Critical
Perspectives on Internationalisation. 437-456. London: Pegamon.
Anderson, J., Narus, J. & N. (2004).
Business Market Management:
Understanding,
Creating and Delivering
Value.
Upper
Saddle
River: Prentice Hall.
Andersson, P., Aspenberg, K. &
Kjellberg, H. (2008). The
Configuration of Actors in
Market Practice. Marketing
Theory. 8 (1), March 2008.
Araujo, L. (2007). Markets, MarketMaking and Marketing.
Marketing Theory. 7 (3),
211-26.
Axelsson, B. & Easton, G. (1992).
Business networks. A new
view of reality. Routledge.
The IMP Journal
Volume 4, number 1
tition. Cambridge, MA:
Harward University Press.
Christopher, M., Payne, A. & Ballantyne,
D. (2002). Relationship
Marketing. 2nd ed. Oxford:
Butterworth- Heinemann.
Cook, K. & Emerson, R. M. (1978). NonExchange Relations in
Networks. American Sociological Review, 43, 721739.
Cunningham, M. & Turnbull, P. (1982).
Inter-organizational
Personal Contact Patterns.
In Håkansson, H. (Ed.).
(1982). International Marketing and Purchasing of
Industrial
Goods.
An
Interaction Approach. Chichester: Wiley.
Czarniawska, B. & Hernes, T. (Eds.).
(2005). Actor Network
Theory and Organizing.
Malmö. Liber.
Easton, G. & Lenney, P. (2009). Actors,
resources, activities and
commitments.
Industrial
Marketing
Management.
38, 553-561.
Ford, D. I. (1980). The Development of
Buyer-Supplier Relationships in Industrial Markets.
European Journal of Marketing. 15 (5/6), 339-354.
Ford, D., Berthon, P., Brown, S., Gadde,
L-E.,
Håkansson,
H.,
Naudé, P., Ritter, T. &
Snehota, I. (2002). The
Business Marketing Course. Chicester: John Wiley.
Ford, D., Gadde, L-E., Håkansson, H, &
Snehota, I. (2003). Managing Business Relationships. (2nd Ed). Wiley
Ford, D. and Redwood, M. (2005).
'Making sense of network
dynamics through network
pictures: A longitudinal
case
study’.
Industrial
Marketing
Management,
34, 648–657.
Granovetter, M. (1973). The Strength of
Weak
Ties.
American
Journal of Sociology. 98
(6), 1360-1380.
28
Gulati, R., Nohria, N. & Zaheer, A.
(2000).
Strategic
Networks; Strategic Management Journal; 21, 203-215.
Hallén, L., Johanson, J. & SeyedMohamed, N. (1991). Interfirm Adaptation in Business Relationships. Journal of Marketing. 55 (2),
29-37.
Halinen,
A.
(1994).
Exchange
Relationships in Professional Services. Turku
School of Economics and
Business Administration.
Series A-6,1994.
Henders, B. (1992). Positions in
Industrial Networks. Marketing Newsprint in the UK.
PhD Diss. Uppsala University.
Holmlund, M., Edvardsson, B. &
Strandvik, T. (2008). From
the sellers offering to the
buyers needing in a Business-to-business setting.
Paper presented at the
IMP-Conference in Uppsala.
Håkansson,
H.
(Ed.).
(1982).
International
Marketing
and Purchasing of Industrial Goods. An Interaction
Approach.
Chichester:
Wiley.
Håkansson, H. & Snehota, I. (1995).
Developing Relationships
in Business Networks.
London: Routledge.
Håkansson, H. & Östberg, C. (1975).
Industrial Marketing – An
Organisational
Problem.
Industrial Marketing Management. 4, 113-23.
Håkansson, H. & Lind, J. (2004).
Accounting and network
co-ordination. Accounting,
organization and society.
29, 51-72.
Johanson,
J.
(1982).
Production
Technology and UserSupplier Interaction. In
Håkansson,
H.
(Ed.).
(1982). International Marketing and Purchasing of
The IMP Journal
Volume 4, number 1
Industrial Goods. An Interaction Approach. Chichester: Wiley.
Klint,
M.
(1985).
Mot
en
Konjunkturanpassad Kundstrategi: Om den sociala
relationoens roll vid marknadsföring av massa och
papper.
(Towards
a
Marketing Strategy Allied
to the Business Cycle: On
the Role of the Social
Relation in Marketing Pulp
and
Paper)
Uppsala:
Department of Business
Studies, Uppsala University (Doctoral Thesis No
30).
LaageHellman,
J.
(1989).
Technological
Development
in
Industrial
Networks. Acta Universitatis Upsaliensis. No 16.
Uppsala Universitet.
Liljegren, G. (1988). Interdependens och
dynamik I långsiktiga relationer. Stockholm: MTC /
EFI.
Mattsson, L-G. (1997). Relationship
Marketing and the Markets-as-Networks
Approach – A Comparison
Between Two Evolving
Research Traditions. Journal of Marketing Management. 13, 447-461.
Mattsson, L-G. & Johanson, J. (2005).
Discovering Market Networks. European Journal
of Marketing.
Naudé, P., Mouzas, S. & Henneberg, S.
(2004). Network Pictures –
Concepts and Representations. Paper presented at
the IMP- conference in
Copenhagen.
Pels, J., Möller, K. & Saren, M. (2009).
Do we really understand
business marketing? Getting beyond thhe RM and
BM matrimony. Journal of
Business & Industrial Marketing. 24 (5/6), 322-336.
Ramos, C. & Ford, D. (2006). Network
Pictures: Conceptual De-
29
velopment and Preliminary
Findings. Paper presented
at the IMP- conference in
Milan.
Smith, P.C. & Laage- Hellman. J.
(1992). Small Group Analysis in Industrial Networks. In Axelsson, B., &
Easton, G. (1992). Industrial Networks. A New View
of Reality. London: Routledge.
Storbacka, K. (1994). The Nature of
Customer
Relationship
Profitability: Analysis of
Relationships and Customer Bases in Retail
Banking. Research Reports No 55, Svenska
Handelshögskolan.
Doctoral Dissertation of
HANKEN Swedish School
of Economics and Business Administration, Helsinki.
Storbacka, K. (2005). Create Your
Future by Investing in Your
Customers. Velocity. 1 (1st
quarter), 2004.
Thorelli, H. B. (1986). Networks,
Between Markets and
Hierarchies. Strategic Management Journal. 7, 3751.
Turnbull, P. & Valla, J-P. (1986).
Strategies in International
Industrial Marketing. London: Croom Helm.
Waluszewski, A. (1990). Framväxten av
en ny mekanisk massateknik. En utvecklingshistoria. (The Emergence
of a New Mechanical PuldTechnology. A Development
Story).
Uppsala
Universitet. Uppsala University. PhD Diss.
Waluszewski, A. (1996) The Glulam
Case in Developing Relationships in Business Networks. London: Routledge.
Van der Walk, W. (2007). Buyer-Seller
Interaction Patterns During
Ongoing
Service
Ex-
The IMP Journal
Volume 4, number 1
change. Diss. Rotterdam
School of Management.
Welch, C. & Wilkinson, I. (2002) 'Idea
Logics
and
Network
Theory in Business Marketing',
Journal
of
Business To Business
Marketing. 9, 27-48.
Wynstra, F., Axelsson, B. & van der
Valk, W. (2006). An
application-based classification
to
understand
buyer- supplier interaction
in
business
services.
International Journal of
Services Industries Management. 17 (5).
Åge. L-J. (2009). Business Manoeuvring
– A Grounded Theory of
Complex Selling Processes.
Diss.
Stockholm
School of Economics.
30
The IMP Journal
Volume 4, number 1
31
Manager - Researcher Relationships
Geoff Eastona
a
Department of Marketing, Lancaster University Management School, Lancaster University, Lancaster LA1 4YX
G.Easton@lancaster.ac.uk
Acknowledgments
This research was funded by an International Study Fellowship as a member of the Advanced
Institute of Management Research (AIM), which in turn was financed by the UK Economic and Social
Research Council. Thanks are due to all those who helped me to make the contacts that led to the
interviews and especially to the interviewees who were so cooperative when called on at short notice.
Abstract
Very little is known about what happens when researchers and managers are involved in
research situations. This paper reports on a project which sought to begin to explore these processes
and structures through an open interview study of 114 interviewees, managers and academics, of
varying degrees of seniority. The results were analysed employing a broad analytical framework using,
implicitly, various IMP concepts and conceptualising research episodes and relationships as socioeconomic exchanges but also extending them to networks beyond the research interaction dyad. One
major finding was the importance, in researching managers and organisations, of the issue of
normativity.
1. Introduction
The special issue of the British
Journal of Management (Starkey and
Madan (2001)) on the nature and
purpose of management research
fuelled a fire that has been burning
since management education and
research first began. It was given a
new twist and a new language through
its linkage to contemporary interests in
knowledge creation (modes 1 and 2
(Gibbons et al (1994)). It was however
largely focused on the rigour /
relevance issue.
Unfortunately this complex
debate has been largely conducted at
a philosophical or theoretical level
without much recourse to empirical
data (Starkey and Madan (2001),
Hatchuell (2001), Grey (2001), Das
(2003), Caswill and Wensley (2007),
and Fincham and Clark (2009)). There
is very little evidence that anything like
an empirical study of what might be
called the sociology of management
research has been carried out with the
possible exception of the book of
invited chapters edited by Löwstedt
and Stjernberg (2006). However even
the chapter by Stjernberg on
interviewing refers, in the main, to
general normative texts on the process
rather citing any empirical research. A
UK Advanced Institute of Management
International Fellowship provided an
opportunity to carry out some
preliminary and innovative research
which was intended to frame the
scope of the topic and surface the
issues involved.
The form of the research, its
organisation and the research setting,
were influenced by the work of the
Industrial Marketing and Purchasing
group (IMP). Around 30 years ago a
number
of
European,
largely
Scandinavian, researchers began
researching
Business-to-Business
(B2B) markets, informed by the
observation that buying and selling
organisations were often, but not
always, involved in relationships rather
than transactions and networks rather
than atomistic markets (Axelsson &
Easton (1992), Ford, D. (2002),
Wilkinson, I. and Young, L. (2002)).
The IMP Journal
Volume 4, number 1
The IMP group offered a
number of major advantages as both a
source of theory and a research
setting. Firstly there is a high
concentration
of
members
in
Scandinavia (notably Norway, Sweden
and Denmark) who were prepared to
help in arranging the research project.
Secondly English is used as a second
language by both managers and
academics in these countries. Finally
there now exists a body of theoretical
and empirical transdisciplinary work
that is as large and coherent as any in
marketing or management and one
that has helped to provide the
analytical basis for this project.
2. Objectives
The overall objective of the
study was to provide initial insights into
the
processes
through
which
researchers and managers become
involved with each other when
engaged in acts of research. These
were always likely to be initial insights
since this is, as far as I can determine,
the first study of its kind, at least in the
field of research into manager researcher behaviour. As a result, the
objective is necessarily open ended. It
was planned that the research
settings, methods, processes and the
disciplines of the researchers involved
in the study would be diverse and that
such diversity would help to provide a
first broad but also rich picture of the
nature of manager – researcher
interactions and relationships.
There was a secondary
objective, which was present from the
beginning of the research, and that
was to discover how, if at all,
normativity
was
a
feature
of
management
research
activity.
Specifically,
were
researchers
expecting and willing to provide help,
advice and consultancy to managers
and did managers have similar
expectations. This was an issue raised
in the knowledge creation and
dissemination debates in the mid
1990s discussed above but about
32
which there remains little empirical
evidence.
3. Research Methods
The primary research method
was in depth qualitative interviews with
both researchers and managers. The
sample of respondents was nonrandom partly because of the difficulty
of setting up interviews with busy
managers and partly due to time
constraints. Instead, a form of cluster
sampling was used in which IMP
colleagues in key Scandinavian
management
schools acted as
research
coordinators,
making
contacts and setting up interviews with
both researchers and managers. The
management schools concerned were
(SSE),
Chalmers
University
of
Technological (CUT), Gothenburg
University School of Economics and
Commercial Law (GU), Jönköping
International Business School (JIBS),
Norwegian School of Management (BI)
and Copenhagen Business School
(CBS).
The sample was obviously
biased in that the initial academics
contacted were either IMP colleagues
or in a proximate department.
However it was possible to extend the
range of academics involved on arrival
at the research sites by “snowballing”
and thereby provide a somewhat more
balanced sample. A similar problem
arose with managers in that only those
who already had good relationships
with academics were likely to be
contacted and agree to participate.
The interviews were semi
directed
and
began with
the
interviewees
describing
their
background
and
research
experiences. One or two particular
research projects were then chosen
and detailed descriptions of the
process and content were elicited. Out
of these descriptions more general
themes
emerged
and
specific
questions were only asked where a
topic of particular interest had not
arisen naturally.
The IMP Journal
Volume 4, number 1
As an experiment in the
interviews an effort was made to coproduce knowledge. There was no
attempt to maintain distance from the
interviewees and academic and
managerial
experiences
were
exchanged between the interviewer
and the interviewee in the early
interviews. In addition information
about the research findings were
introduced into the discussion in the
later interviews. This experiment
worked well and led to broader and
deeper “structured conversations”. The
interviews were carried out in the
period mid February to the end of May
2004.
The achieved data set involved
114 interviewees of whom 21 were
managers although a further 48
academic researchers had substantial
business experience. Clearly the main
category
of
respondents
was
academic due to problems of access.
However while the “researcher with
managerial experience” group were
treated as researchers in the
interviews they often reflected upon
both their roles which produced a
richer picture than would otherwise
have been possible.
The academics included a
University Principal, deans of Business
Schools, professors and lecturers at
various levels and PhD students. The
managers
were
mostly
middle
managers. The interviews lasted about
an hour on average. The interviews
were digitally recorded and analysed
using the HyperResearch software
package, which allows sound extracts
to be coded for content analysis. Forty
three active codes were created and
used to code over 3600 sound
extracts.
The data were not collected as
a random sample and even if this was
the case each interview had its own
unique identity. Even the interviews
within a category, for example Ph.D.
students, varied. Clearly there was no
sense
in
tabulating
responses
numerically. The strength of the data
set lies in the capture of key issues
and the ways in which they are
33
presented through the language of the
respondents. What has been done is
to present the results through the
population of a broad analytical
framework by means of salient
quotations. In general what this
process does is to demonstrate both
the variability and richness of the
respondents’ perceptions of, or
attitudes towards, a particular issue.
4. Analytical Framework
IMP research is focused on
buyer – seller relationships and
networks of those relationships. In the
broad framework used to report on this
study, which was partly induced from
the data, researcher and manager
interactions and relationships replace
seller and buyer interactions and
relationships.
A simple actor substitution
would make managers sellers of
information and researchers buyers.
However the situation is somewhat
more subtle and complex than that. In
the original IMP Interaction Approach
(IMP
group
(1982)),
exchange
episodes were characterised as
involving exchanges of products /
services,
information,
financial
resources and exchanges of a social
nature. Clearly product exchanges are
very unlikely to be involved in manager
researcher relationships. However
research which involves a consultancy
element may be regarded as a service
exchange. In this case but in this case
only there may be a financial
exchange.
Information
exchange
would seem to be central to research
episodes and of course it need not be
just a one way process, manager to
researcher. Social exchange almost
always plays a major part in
interactions and this is particularly true
for manager -researcher relationships.
More broadly actors enter into
exchanges in expectation of a rich mix
of various kinds of returns about which
both can have very different attitudes
and beliefs. In this case researchers
receive information, knowledge and
understandings which they can
The IMP Journal
Volume 4, number 1
convert into academic papers, use to
obtain research grants, employ as the
basis for their teaching all of which
allow them to be employed and gain
promotion. However they will usually
also value the processes by which the
knowledge
is
produced
(social
exchange) as well as those that occur
while working with the knowledge in
ways which provide intellectual
personal affective benefits. The costs
that they incur are the opportunity
costs of doing other things in their
academic
life
(teaching,
administration) or in their non
academic life (leisure time, alternative
employment).
For managers the opportunity
costs are the time they could be
spending on their managerial tasks
and while this cannot be calculated it
can be experienced in terms of
commitments neglected. The benefits
are also much less clear. The
economic ones are information which
may help to improve their own, or their
organisation’s
efficiency
or
effectiveness. If the involvement is
more protracted, as in action research
or consultancy, then the costs go up
but so too do the potential benefits. In
addition managers can experience and
value
social
and
psychological
rewards as a result of sharing their
experiences,
concerns
and
understandings with a researcher.
Simply being chosen as a respondent
can elicit affective rewards.
The research process itself varies in
form and content and of particular
importance to the research question is
the
normative
effect
of
such
involvement and the variety of modes
of researcher involvement.
Each manager – researcher
relationship,
however
brief,
is
embedded in a network of other more
or less influential relationships and
contexts,
both
personal
and
institutional. Academics work with
academics and managers exist within
complex organisational forms. Also the
social
network
that
connects
researchers and managers and makes
such exchanges possible provides the
34
intermediate context. More distant
from the action are the institutional
settings within which researchers
(universities,
research
institutes,
consultancies, academic disciplines)
and managers (departments, SBUs,
holding companies) are embedded.
Country cultures, industrial structures
and institutions also afford differences.
While Scandinavian countries have a
great many similarities there are also
crucial differences particularly in the
structure of management education
and in governmental influence.
It should be emphasised that
this is a framework for initial analysis
in what can only be regarded as a
scoping study. It was partly created
prior to the research but was also
heavily influenced by the findings. It
attempts to identify the key aspects of
researcher – manager relationships
and what influences them. Further
research is required to discover the
particular mechanisms that operate in
different situations.
Initial analysis has been carried
out but the data set is of such size and
richness that the results reported here
are necessarily preliminary in nature
and more work needs to be done. The
results are structured in line with the
analytical framework described above
as well as identifying key issues within
each concept in line with the codes
used. Bracketed anonymous and
edited quotations are used to illustrate
particular points.
5. Findings
5.1
Researchers
5.1.1
Attitudes
Experiences
and
Researchers’ motivations for
carrying out, and attitudes towards,
research were predictable. For
example thinking is a major part of
what academics do. In the next 4
sections all the quotes are from
researchers, from PhD students to full
professors.
The IMP Journal
Volume 4, number 1
“I really want a job where I can sit and
think.”
Researchers
also
valued
academic freedom and being allowed,
within reason, to research what they
wanted to research.
“Nobody was interested .. except me.”
The variety of academic
research was also important for some
respondents.
“We try to work with everything all the
time”
“I like broad reading..I like to learn”
“I cannot focus ….I enjoy tackling new
topics”
Most were enthusiastic and
excited about what they did
“I am so excited about it how important
our theories are.”
“He reintroduced me to that work and I
love it”.
35
“How are elements connected…”
“The main reward
knowledge. I know
enough for me”
is increased
more. That’s
“I feel overwhelmed by what I don’t
know.”
“I couldn’t understand the dynamics of
it. I would like to study it more.”
“I believe that research is also mental.
You have to like it (theory)”
Position and money seemed
much less important.
“I have done my chasing for new
titles. If I don’t get a professorship I’ll
survive. Research is what makes my
life interesting”
“Consultancy?
smart!”
I haven’t been that
“Talking to companies is a waste of
my time. If I wanted to make money I
would not have come to academic
life”
“I am very interested in what I study. I
love my work!”
The constraints are mostly
familiar
“I do it for the pleasure of it”
“We don’t have the time”
and interested too.
“It’s a very interesting history, a kind of
fairy tale”
“I have never been bored”
“It’s also a matter of where the money
is”
However
some
research
results can upset interested parties
“You must think about your PhD and
what’s your own interest because it is
now so deeply involved and
motivated”
“They called the dean!”
Of course knowledge is the
basic output of research and
respondents were quite clear about its
importance and motivational power.
It is clear that the many of the
researchers
interviewed
enjoyed
researching
managers
and
organisations.
“I wanted to understand what the hell
was happening”
“I would die as an academic if I didn’t
meet real people”
“We will not send our people to you
(your business school)”
The IMP Journal
Volume 4, number 1
“I would like to work continuously with
business”
“Now I would be able to do what I
actually like to ..work with companies”
“I don’t want to sit in a tower and look
down on companies”
“Real research and not just seminars
and articles”
“I like being very much involved with
organisations”
This appears to be partly to do
with the social nature of research and
knowledge acquisition but also the
desire to help, if possible.
“I get so much from them I doubt what
I give back”
“Bridge building (to industry) has
always been part of my objective”
“It’s not proper real research. As an
academic you miss some important
points. It is nice once in a while to be
ahead of them. It gives (us) some
respect.”
“It’s not just to write articles but it’s to
go out and help people”
5.1.2
Researchers
Normativity
and
The term normativity means
the quality of being normative and in
this section refers to the attitude
towards normativity of the researchers
interviewed; their views as to whether
they should give help to businesses.
First some comments from those
reflecting a normative turn of mind.
“That’s my job as well. It’s not only to
write articles or whatever it is but you
go out and try to help people.”
“I enjoy staying not only here but
spending a substantial amount of our
time in companies working together
36
with managers in the companies. Not
necessarily running around doing all
sorts of actions but working in close
collaboration. Staying a substantial
amount of time out there to learn how
to relate and to establish the right
relationships with the practitioners.
Learning the language of practical
management.”
“Yes but I think it is appealing if you
hold, keep yourself, don’t become the
complete consultant.”
“My view is that generally speaking,
and that goes for all Sweden, my view
is that the best researchers are also
the best consultants.”
“I guess that for almost all
researchers,
they
can’t
help
themselves but think they can change
the situation..that can be a problem…
as a (consumer) researcher with views
on the industry why am I better off
than a (consumer)… what better
solutions can I give? “
“I should not prejudice any one of my
respondents. I want my research to
be relevant for them.”
And some comments from
those who do not, in general, agree.
The first respondent doesn’t do
consultancy because it isn’t helpful for
their kind of research.
“I don’t consult, generally speaking…
Talking to companies is a waste of my
time… I have written the occasional
paper saying, you know, implications
for managers but I don’t if I can help
it… I am doing this (current research)
because it gives me a kick.”
The
second
and
third
respondents simply believe that they
are not competent to do so.
“I’ve never been interested in that
(consultancy). I’ve tried to avoid it. I
thought that it was really difficult to try
to tell people that have been working
The IMP Journal
Volume 4, number 1
37
with things and worked for 30 years,
what they should do.”
“This isn’t going to be published in a
way that will hurt you”
“Because it is impossible not to end up
with the conclusion that the world is so
complicated we can’t do anything,
because you have to do things. Then,
hopefully, eventually (our research
programme) will also provide not only
some questions but also some
answers.”
“No one other than me will use this
information”
It may be that the two are
largely incompatible at least according
to one respondent.
“I don’t want to mix my research with
my advising. I want to keep these
things separate. I don’t want to convert
my role from being a researcher to
something else. I think that there is to
some extent a conflict between these
two roles.”
However perhaps there is no
choice.
“So in that way I think that it is
impossible not to be normative. So
whatever we try to do, however we try
to do, we are always normative and
the question is if we manage to
realise that, because if we realise it
we can handle it better than if we
believe that we are not normative.”
5.1.3
Researchers and Ethics
Surprisingly few respondents
commented about the ethics of their
research methods but those who did
were
mainly
concerned
about
anonymity. Views ranged from those
with largely liberal views to those with
views
that
were
strikingly
conservative.
“I have promised that if I give named
quotations they would have the
opportunity to react to them”
“Only a very few of them didn’t want
their names mentioned”
“It’s OK as long as you don’t mention
names”
One respondent quoted one of
the managers he interviewed saying
rather wistfully.
“I hope we will never be ashamed of
inviting you to our house”
Another researcher expressed
their general concern by saying
“Do they really know what I am
doing?”
5.2 Managers
For managers, involvement in
research was generally viewed as a
very peripheral activity. Moreover their
motivations for getting involved were
diverse. A key motivator was specific
interest and the possibility of acquiring
valuable and applicable knowledge
and assistance.
Sometimes
strikes it lucky.
the
researcher
“The nature of my research topic was
really hot in the company… and the
way I did it was very involving… many
hundreds of people were involved”
Researcher
“But also upper management were
interested in what we were doing … I
had presentations about that in upper
management. I didn’t really have this
reverse mentoring in a one to one
situation. I had it more in the
management more as a resource as
an internal consultant. I had lots of
discussions with them… a change
agent… a bouncing board to new
ideas.”
Manager
The IMP Journal
Volume 4, number 1
“It could be something of value to me
in my job as a manager …this is
something I don’t have time to look
into stress reactions or whatever… if a
guy came along and said I want to do
stress reactions I would say wow
perfect”
Manager
At other times the way in is
more difficult to find and the rewards
have to been seen to be greater and
are likely to be a longer time coming
“I always get that question when I am
presenting what we are working with
“what’s the value of this? Can you
measure the value (of this piece of
research)?” And I usually pose the
question back to them can you
measure your value of being in this
room right now?”
Researcher
“But (the company) came and they
wanted to write a history…So what
has happened to ..with the original
culture? Has it disappeared? Does it
still live? What’s the continuity? Where
are the breaks in the development?
Historian
“So there are different motives. This
was one of most brilliant projects, (X’s)
managing directors’ participated in the
discussions and it was quite
extraordinary to sit there with them.
They suddenly said ah now I
understand why we still have these
difficulties.”
Historian
My first thinking is “is this really for my
organisation and for me?”. I then try to
ask the person what is your
background. What would you like to
see and why… if they say I would just
like to do this see this and this and go
back and do my paper I would usually
say no. I would like them to see more
in my company and get more involved
because I don’t believe they would
see the whole picture if they don’t …
more than just having a couple of
interviews It has to be in line with
38
what we are interested in …Usually
when I am involved in …. research
projects I always want to put my own
energy and meet the people and see
what they are really after and can we
do this in a good way together instead
of just sending them somewhere in
the organisation and say I have a guy
who would like to look at you I am
concerned about their time And that
giving back could just be a lecture or
a workshop that’s fine giving
something back.
Manager
Many managers simply value
the chance to reflect, learn about new
ideas,
and
have
thoughtful
discussions.
“I think you get wider views. It’s
always good getting people from a
university out there. I think you get a
little more than when you are sitting
here every day digging.”
Manager
“Gives a new insight into what I am
doing”
Manager
“I brought in some ideas from some
other companies”
Researcher
Other seeks
impartial view
an
outsider’s
“And then we have a discussion.. a
point of view how would you evaluate
it? My experience is that they find it in
a way stimulating to discuss their
company with people from the outside
who has nothing… I mean we are not
shareholders. We are not interested in
anything but doing a good job.”
Researcher
Managers
are
however
sometimes willing to simply be good
citizens
“We did it to be good citizens”
Manager
The IMP Journal
Volume 4, number 1
“I was a fairly young student so I would
say they did it to be good citizens.. to
help this young guy mostly but of
course it was an issue for them”
Senior researcher
“Because they want to be seen and to
contribute to the good of the thing…
my motivation to work together with
the university, because I think it is a
part of the society… Both daughters
…they have a need, are going in
university and they have a need not by
me but by other people who give time
for that”
Manager
However such involvement
clearly involves an opportunity cost
39
“He was a friend of the professor”
Researcher
Anonymity rarely came up as
an issue perhaps because
“(They) don’t seem to mind on the
whole and some (managers like it
(being quoted).
Researcher
5.3 Research Process
Interviews were the most
common form of interaction between
managers and researchers that came
up in this study and they involved
rather
complex
socio-economic
exchange processes.
“It is about measuring this quarter by
quarter syndrome. It has hit us as well.
I mean for instance say that I talk to
you today I don’t know whether that is
a plus or minus. Should be plus of
course but if you had been some
important person from a business
point of view people would have talked
about it.
Manager
“I sent out my topics beforehand”
Researcher
There are also some less
defensible motives that managers
admit to or researchers divine.
“We had a questionnaire for the
interviewer but we never used it!”
Researcher
“It’s a chance to get away from your
everyday life”
Researcher
Generally the researchers
used semi structured interviews and
were prepared to pursue new topics
as they emerged during the interview.
“CEOs, they have nobody to talk to”
Researcher
“More interested in talking about
themselves than their company”
Researcher
“They want to be “seen””
Researcher
“Being given the opportunity to say we
are
supporting
research”
Manager
“It came in from the top level!”
Manager
Preparation was sometimes
helpful; sometimes not.
“I looked up information on the
industry so I didn’t look stupid”
Researcher
“I try to follow up when they say
something interesting”
Researcher
“I tried to answer as honest as I could”
Researcher
“Conversation that they feel is very
relevant for them”
Researcher
“Asking them to tell stories”
Researcher
The IMP Journal
Volume 4, number 1
This often resulted in the
interview progressing to what many
researchers described as a dialogue.
“To go from questions and answers to
dialogue”
Researcher
“If you have a dialogue you would be
stupid if you didn’t learn something
from it”
Researcher
“Quite brief check list – more a
dialogue”
Researcher
40
There is usually a tension
between precision and richness with
researchers varying their approach
depending on the mix of exploration
and understanding they were seeking
at any point.
“My questions became more and more
precise”
Researcher
“(There are).. tensions between
opening up and trust and precision”
Researcher
But at least one respondent
argued
But dialogue is not always easy
to achieve
“In the beginning there can be a lot of
my own voice”
Researcher
“I talk too much”
Researcher
“If they have to talk about something
else then that’s the price we have to
pay”
Researcher
Further involvement could also
lead to the sharing of experiences and
the co- construction of knowledge
“You can tell stories about people and
they say it’s like you were a fly on the
wall”
Researcher
This kind of comment was, for
obvious reasons, largely confined to
senior researchers.
Researchers recognised that trust was
required for this process to take place
and that this takes time.
“It didn’t start out very open but it soon
got to that”
Researcher
“I disagree but I don’t say anything”
Researcher
“You can ask pointed and difficult
questions”
Interviewers took, or were
given, particular roles for example the
novice.
“Tell me what you know because I
don’t know nothing.”
Researcher
“I took the position of student not
understanding these types of diagram”
Researcher
Interestingly the gender role
was also mentioned
“I just play the silly woman”
Researcher
One novice woman researcher
told the story of how she was finding it
difficult to get useful responses from
the managers of small manufacturing
firms until she hit on the ploy of asking
to be shown around the factory before
the interview. Another female research
remarked
“Maybe I dress up a bit”
Researcher
There were sometimes status
and role issues
The IMP Journal
“He (the senior
competing
with
Researcher
Volume 4, number 1
researcher) was
the
manager!”
41
dual interviewing;
“It can be more relaxing”
Researcher
Most researchers were aware
that an interview should be a
rewarding
experience
for
the
respondent
“I relate my questions to their own
business”.
Researcher
and repeated interviewing
“Easier on the second interview, they
get to know you”
Researcher
Length of interview was a
factor that divided researchers.
“People love to be interviewed
because you show interest in their
opinions”
Researcher
“One hour if I am not done I go
anyway, I respect their time”
Researcher
“Personally I (also) find interviews
extremely rewarding”
Researcher
“They say I only have 45 minutes and
you sit there 3 and a half hours.”
Researcher
“We
(interviewers)
are
enthusiastic, like Oh really!”
Researcher
“If the person talks a lot it is difficult”
Researcher
both
Managers also took roles,
usually that of the curious expert
“They expect to give a point of view”
Researcher
“They also want to know what you
think”
Researcher
and their reasons for helping
were largely those described earlier.
However negative experiences could
be a problem.
“(He had a) really bad experience with
journalists - he was suspicious”
Researcher
A number of technical options
emerged, each with its own pros and
cons;
the recording of interviews;
“When I was taking notes it was time
for them to reflect”
Researcher
5.4 Modes of
intervention
involvement
and
A central question in this
research was the extent to which
researchers actually help managers
and their organisation during the
research process. Various modes
were identified and could be roughly
classified as to their degree. In the
interpersonal
interview
situations
intervention can take place even if it is
in its mildest form
““….and tried to not facilitate change
in that sense but I ask questions about
whether or not change had been
facilitated and in that sense, of course,
I provoked change”.
Researcher
“I think that we have been a little bit
provocative sometimes and said "why
do you?"”
Researcher
Of course not everyone agrees
I don’t think you can really change
people’s minds by the questions you
ask.”
The IMP Journal
Volume 4, number 1
Researcher
The phrase that was used most
often by managers was “a new
perspective”
“An outside person with fresh eyes; A
second opinion”
Researcher
However intervention can most
obviously occur when feedback is
provided.
I mean we were not expensive
consultants. We were actually honest
researchers with a kind of commitment
….We sent them a few pages saying
we perceive your strategy is like ……,
that was actually a second opinion, a
view from outside which is very often a
surprise for the enterprise. They don’t
get it (very often).”
Researcher
I told him I was going to do research in
the iron industry …I don’t know that
much about your company… and he
said OK but what’s the benefit for me. I
said he could see me as an outside
person
with
fresh
eyes.
Researcher
“In my own mind I am creating a
mirror to hold up to them, they can see
if their hair is a mess”
Researcher
A lot of the things that I am talking
about get taken up in their discussion
…One of the senior managers said “I
got this from (PhD student)”
Researcher
If this provides a better
understanding of the situation then it
sends a powerful message.
“They said several times that it helped
them understand, it helped them make
choices”
Researcher
And what came out of it, they took
something out of it, we cannot suggest
42
you anything at all but we can describe
the situation today as a product of the
past so you are on a route and there is
a path and all complications you are
facing today have roots in the past…
but we are not advising this with the
purpose of helping you … to help you
make choices… they said several
times that it helped them understand, it
helped
them
make
choices.
Business Historian
Although they started out by saying
this is nice because it is free, in the
aftermath they see the value of having
persons involved with the purpose of
seeing
patterns,
understanding
processes rather than sheer delivery
of results.”
Researcher
“They do say these are our problems
and we would like to have an
answer… how to deal with it … for
example in Sweden none of the
(industry) companies are very large…
one of the companies reaches say
around 100 people they break. They
had a problem; they were asking me,
they don’t understand how this
happens.”
Researcher
They see (business school) as some
sort of helper producing these kinds of
pictures but they are not waiting for
that…. It is more that we are a
catalyser being involved and helping.
Researcher
Prediction can be even more
useful.
“The short term consequences could
be that the customers get upset… the
long term consequences could be that
it works.”
Researcher
Advice is the next stage.
They were interested in what came out
of the reports and of course we were
giving some advice in general but they
were not asking for the specific, with
The IMP Journal
Volume 4, number 1
one exception where there was a
project we did for (nationalised
industry) where they specifically want
us to say how they could develop the
(programme).
Researcher
“They know that I am a researcher and
the next steps they will have to take for
themselves.”
Researcher
and then implementation of
that advice
“In this last meeting it ended up - now
we want to do something concrete. We
have to do something that will lead to
some
kind
of
action
and
implementation.””.
Researcher
But of course advice is not
always heeded
“They have a lot of opinions but they
still
have
all
the
answers!”
Researcher
Intervention can occur in single
one-off surveys but is much more
likely to happen when the research is
case based, when repeated visits to
the firm are required.
“What is needed is such relationships
are developed so you can really start
to try to find ways to help them.”
Researcher
“What am I really on about, that’s a
tricky role to take because it is action
research, I am in it, I am living it.
Researcher
The most interesting finding
was that some researchers were, in
effect, in an action research mode,
without
actually
labeling
or
acknowledging it as such.
Action
research
provides
wonderful access, the ability to take
part in a natural experiment, the
rewards of killing more than one bird
with the same stone and, often,
43
substantial funding. However it raises
the problem of objectives that may
clash and the need carefully to agree
and align actions. Researchers may
have to compromise in terms of the
data they collect, the theories they use
and where they publish.
“Action research …getting into the
dialogues and interventions, you know,
with the employees .. and the subject
of the research isn’t just you thinking
about it and then developing the
knowledge.”
Researcher
“And that was also a problem when I
started at (company). From the
beginning we wanted to have a
(particular theoretical) perspective …
soon we found out it is very difficult to
force on the (particular theoretical)
perspective and especially when you
are in a company. I mean they wanted
to do it from their point of view and so
we kind of changed it to a (different
theoretical) perspective instead.”
Researcher
“Still it was difficult to get something
out of it like a good article, a lot of that
research was very applied”
Researcher
“We always tried to get some openings
to write some paper. But it always
came in addition to the work”.
Researcher
“It was like a business way of giving
something back but in my head I have
my PhD.”
Researcher
“The consultancy outcomes were
articulated differently, the papers were
trying to catch the fine patterns in
another fashion”.
Researcher
“Because they are so involved in the
practical side of it they can’t theorise”
Researcher
The IMP Journal
Volume 4, number 1
44
Managers also recognise the
fact that researchers are not
consultants and have dual objectives,
work more slowly and don’t always
offer “simple solutions”.
“I used to work there”
Researcher
“The consultants are more focused to
determine what my needs are,
Researchers have a view of what they
need”
Manager
“I work there now as a PhD student or
participant observer”
Researcher
5.5 Access Networks
5.5.1 Cold calling
“I went back to my old company”
Researcher
“Using earlier contacts and moving
around!”
Researcher
and their own personal networks.
Almost
all
respondents
reported that access by cold calling
would only be difficult under certain
rare circumstances.
“I know the top managers in the
management and purchasing area.
Researcher
“If you ask top managers to give you
an hour of their time they will.”
Researcher
“If I hadn’t known these guys for years
they wouldn’t have supported the PhD
programme”
Researcher
“Direct is not a problem”
Researcher
setting,
Within the business school
However the practice was quite
different. Cold calling seemed to be
the exception. Instead researchers
used a vast variety of different
contacts networks.
level,
“A network of friends whose fathers
were business men who I could draw
on at that time”
Researcher
“Sweden is a small country”
Manager
family;
“My professor”
Researcher
“Met at a business school meeting
then
passed
on
internally”
Researcher
5.5.2 Access Routes
At
a
personal
researchers exploited friends;
supervisors;
colleagues
“I got (a colleague) to help me”
Researcher
“Inherited from senior researchers”
Researcher
current students
“Through MBAs links to CEOs”
Researcher
“My father’s firm”
Researcher
former colleagues;
“Mainly through my students in the
evening
classes
final
thesis”
Researcher
and past students
The IMP Journal
Volume 4, number 1
“Probably 60% of my respondents
graduated from the school”
Researcher
45
“That person (HRM) has no good
contacts in the line”
Researcher
but any kind of contact could
“A personal contact who was on the
MBA”
Researcher
Current or past research
projects also created contacts that
could be used
“He had been interviewed previously”.
Researcher
“(Professor) had been here before. He
heard about me and he wanted to
interview me”
Manager
“Spin - off from another research
programme”
Researcher
Alternatively
“Salesmanship - X sold these projects
and he is a very good salesman”
Researcher
5.5.3 Contacts within
through the organisation
and
Once a connection had been
made within a firm the identity of the
contact shaped the process. CEOs
could provide the necessary level of
authority.
work.
“I introduced her to people in my dept”
Manager.
However moving from firm to firm
depended on the power of those
concerned.
“We always introduced them into
suppliers”.
Manager
“Outside then inside the firm and
snowballing”
Researcher
“Continuing from one firm to another
through contacts also family firms
linking to (Business School).
Researcher
Another option was to use
retirees
“She is a very active, retired and
interested woman”
Researcher
5.5.4 Contacts through other
organisations
“Started with a letter to the CEO and it
worked its way down”
Researcher
Scandinavian countries also
have a vast range of business related
associations, representative bodies
and development agencies all of
whom can provide access to firms and
their managers
“We need a godfather who will say
that you are the people we want to
develop with”
Researcher
“I’m an industry person and a
representative on industry funding
body”
Researcher
“He should check with the owner who
has
given
permission”
Researcher
“Confederation of industry – indirect
(route) that can open doors”
Researcher
“(X) had all the business contacts
using the name of local development
The IMP Journal
Volume 4, number 1
organisation”
Researcher
“The French company (they) were the
most difficult”
Researcher
Similarly, events such as
conferences, seminars and exhibitions
provide fertile ground for making
contacts and some business school
departments use these modes of
operation quite deliberately
“Met someone at our conference”
Researcher
“Speeches
audience”
Researcher
46
with
an
“She went to London and she came
back and said when I was telling
about my research people didn’t really
believe it”
Researcher
“Cuban government… problem… all
lies”
Researcher
industrial
5.6 Researcher Networks
Consultancy also offered a possible
route.
“We knew him very well through
research and consultancy for many
years”
Researcher
Few
researchers
work
completely on their own. Cooperation
however can be both stimulating and
depressing.
5.6.1 PhD Supervisors
through
One of the most salient
relationships is that between research
supervisor and PhD student. Some
supervisor relationships are positive
and helpful.
A more passive approach
involved persuading the respondents
to come to the researcher. This can be
done by becoming famous.
“Doing interviews together.. getting
my impressions of the field.” Student
5.5.5
publicity
Contacts
“Read your article in a newspaper”
Researcher
“By writing a famous book”
Researcher
“We have enough
background”
Supervisor
of
a
shared
“I want them to go to the company on
their own”
Supervisor
“Becoming a media source”
Researcher
other
“He didn’t have to say very much. I got
the whole picture”
Student
Researching in other countries
provided
somewhat
different
experiences.
“He is my source of inspiration, he
challenges me and he gives me very
much room”
Student
5.5.6
cultures
Contacting
in
“Spain was so hard. I was so spoiled
in Sweden. I call 28 times”
Researcher
“We are as close to equal partners as
you can get and in that sense and a
good friend. And then he is my
supervisor
Student
The IMP Journal
Volume 4, number 1
“I get involved in everything he is
doing”
Supervisor
“Sometimes he has to hold me back”
Student
Others are less so and there
were a surprising number of these
relationships described.
“He asked me to do something I
wasn’t interested in”
Student
“I would
questions”
Student
probably
ask
different
47
“Changing supervisors. I didn’t get
along with supervisor”
Student
“He is a good researcher. (But)I had
difficulty in writing in a way that he
found satisfactory”
Student
“Perhaps I
supervisor.”
Supervisor
am
not
the
right
“My entry wasn’t regulated very much”
Student
“Not too fond of me digging too deep.”
Student
“It’s very important who you are to do
the interview. We have different
frames of references”
Student
“I don’t feel this has PhD potential. The
second supervisor was very angry
Student
“I had lots of room. He had 9 PhDs!”
Student
“The process I had been involved in
was not good”
Student
And, of course
“All doctoral supervisor relationships
are different. So very, very different”
Student
5.6.2 Colleagues
“When I am there I dominate too
much”
Supervisor
“I have certain
opinions”
Supervisor
prejudices,
some
“Some professors, it’s like normal
work. You work for the professor and
you write theses.”
Student
“The process I had been involved in
was not good”
Student
“They are all embedded in different
epistemological (paradigms)”
Student
“He has a really managerial style of
doing things”
Student
Few
researchers
work
completely on their own. Cooperation
with other researchers however can be
both stimulating and depressing.
5.6.4 Colleagues
Working with other researchers
has many of the same problems but
the power structure is different. In this
case synergy appears to be important.
Researchers can complement each
other.
“Working
with
people
complementary skills”
Researcher
with
“She’s not a manager she is a
researcher; a perfect combination”
Researcher
The IMP Journal
Volume 4, number 1
48
“I have a lot of consultancy friends
who I work with.
Researcher
Researcher
“I let other people collect the data”
Researcher
While the research involved
mainly B2B researchers working in the
IMP tradition a number of other
researchers in other more applied
management disciplines were also
interviewed. On the basis of this very
small sample there certainly seemed
to be large differences about what
constituted
good
research
and
publication.
In
particular
the
normativity of these disciplines was
much more apparent especially in
terms of publication and it is possible
to
envisage
a
continuum
of
management disciplines defined by
their normativity. At one end
management scientists accept that
providing
prescriptive
theoretical
solutions for managerial problems is
perfectly acceptable. At the other end
critical management studies questions
the role of management and would not
regard helping managers to manage
as legitimate. What became obvious
from interviews with researchers in
those disciplines closest to the
physical, technical and / or controllable
aspects
of
management
e.g.
operations management, logistics,
project management, was that the
normative mode dominated.
“If you are more confused it is great to
be put in a project”
Researcher
Learning from fellow PhD students…
is good”
Researcher
Colleagues can also help to
motivate
“Puts pressure on me. I am a slow
starter”
Researcher
“Of course I sit a lot on my own.
Pressure
from
others
helps”
Researcher
Social rewards
“Summer school I used to organise the
parties. Ever since we had this clique.”
Researcher
“(She) is a friend. I see that as added
value”
Researcher
“I was getting fed up working on my
own “
Researcher
But
problems
relationships
can bring
“He gets to get his name on the
publication and people read it.”
Researcher
“People have different working styles”
Researcher
and confusion
“I don’t have a formula
relationships
should
look
how
like”
5.7 Academic Disciplines
Engineering management;
“That was a technical project run at
that time… the guy was asked to
investigate between legislation and
some technical developments. He took
his time. It’s a great thing, those guys
have the time for it and the normal
people in their normal capacity - well
you know how it is. You are stretched
for time always. He talked to people
inside the company … and outside the
company, he summarised, concluded
and said this is the way to go and we
used that.”
Manager
Government policy;
The IMP Journal
Volume 4, number 1
“I mean when I was working for the
(foreign government) that was a
different story. I could see the results
almost immediately. We were working
with politicians, NGOs for social and
civil development.”
Researcher
Construction contracting
“I would say that the typical article we
can publish would look like this. It is
rather coolly descriptive 90% and then
we are more or less required by the
sort of journals.. in the conclusions we
draw
a
number
of
practical
implications which are based on not
very well thought out normative
assumptions but we are not doing
action research or anything like that.
Researcher
“In most of our projects in our
department, it’s applied together with
companies”
Researcher
Operations management
“Well, we go to them. We go to the
people we know. We say that we
believe there is interest in this area.
We see that you have the same
problem internally. Do you feel it
severe enough to finance our
research?”
Researcher
We can hardly ever say an exact
outcome but we can have, we can
state the ambition, we can state the
method, we can state what tools we
will use, then it is up to how clever we
are to deliver results that is good
payback for the money.”
Researcher
49
as much being involved and doing,
well, answering questions they would
have. So for me is more a basic
research
project.
But
different
disciplinary bases may be a problem.”
Researcher
In other disciplines attitudes varied
although even business historians
were not averse to helping out
managers.
“We designed the book in the way
they could understand themselves in
order to face these challenges”
Business Historian
5.8 Business Organisations
Clearly the organisations that
researchers are involved with vary
enormously. Some are very close to
business schools.
“More a part of their everyday life to
have
close
relationships
with
universities”
Manager
“They came to the seminar with 7
people…
high
up
managers”
Researcher
“Very open minded culture- they have
been very honest”
Researcher
“Attitudes to industrial PhD? – open to
this in (company)”
Manager
“The CEO has read two of my papers”
Researcher
However there were clearly
concerns about whether this openness
could continue as in the past.
Logistics
“I think what changed with (Research
Programme) actually from doing
projects for companies writing big
reports you know and stuff like that.
(Research programme) is more for me
doing research in companies but not
“We have open doors but now are
going to close some”.
Manager
Other researchers had less
good
experiences
with
the
organisations they research or seek to
The IMP Journal
Volume 4, number 1
research although this seemed to be
mostly in situations where action
research or consultancy was involved.
“They want the solutions or the tools to
get a solution”
Researcher
“They had their own set up framework.
Nobody could tell them what to do.”
Researcher
“Arrogant snobs. Didn’t have anything
to learn from anyone.”
Researcher
“They didn’t want to spend time
teaching the other companies how to
do business!”
Researcher
“They pay
interested”
Researcher
but
they
don’t
seem
Behaviour was also seen to
differ by type of organisation. In
particular SMEs were perceived to be
unfamiliar with research and also
varied, not surprisingly, in their
attitudes towards it. Some researchers
prefer researching in small firms.
“They don’t know everything. Big firms
think they know everything”
Researcher
“I have a fondness for them”
Researcher
“It is small firms. They are very open
minded”
Researcher
50
Another
important
factor
seemed to be whether or not the
business school was attached to a
university with a technology faculty.
“They are more closely tied into more
technical research institutions”
Researcher
“They have a culture of having a lot of
doctoral students in the organisation”
Researcher
The rate of organisational
change was a particularly disruptive
influence.
“Traumatic changes occur”
Researcher
“I changed bosses 5 times”
Manager
“They had a big reorganisation and
they took him out”
Researcher
“The original contact person has quit”
Researcher
Managerial
culture
also
seemed to be a problem for some
researchers
“Written documents are not read
Researcher
“I don’t want anything old I want
something new”
Manager
“It’s about quarter by quarter thinking”
Researcher
Others are less convinced.
as did internal politics.
“More difficult to convince them”
Researcher
“Don’t understand what research is
about”
Researcher
“Can’t spell research”
Researcher
I had a feeling that I shouldn’t get too
close to the department”
Researcher
There was really a lot of tension
between the two
departments”
Researcher
The IMP Journal
Volume 4, number 1
“He sent out a letter. I have full access
and everyone has to help me”.
Researcher
Finally there are some obvious
methodological issues about choice of
research sites
“I think that we have a huge bias in the
companies we address.”
Researcher
5.9 Institutions
There are a vast range of
institutions in the 3 countries that
provide both the context within and the
vehicles through which research is
carried out.
Firstly there are projects and
programmes that are essentially time
limited and relatively focused. The
Netlog project at BI is a good example
of the former with 5 senior researchers
and 6 doctoral students working
together on a series of interrelated
research activities within a common
intellectual framework. The Fenix
programme, located in SSE and CUT
as well as the Ecoles des Mines in
Paris, is a well publicised and novel
attempt to allow practising managers
to undertake “industrial PhDs”. Both of
these institutions are funded by
government and by the firms involved.
By contrast the Volvo doctoral
programme is financed by the firm with
supervision
and
certification
outsourced to various universities.
Secondly,
centres
and
institutes can be considered more
permanent entities. Within business
schools there are departmental
centres such as the Centre for
Marketing Communication at CBS that
is largely self financing. It has a paid
membership of Danish advertising and
marketing
organisations,
runs
seminars and produces both academic
and close to market research, drawing
on the members of the marketing
department.
The
Marketing
Technology Centre at SSE was a
cross departmental organisation with a
similar
remit
but
one
which
51
concentrated more on project work to
finance its activities. The Gothenburg
Research Institute, which both raises
research funds and attempts to match
researchers and local firms, is an
illustration of a single business school
centre.
Cross
business
school
examples also exist and seem to be
more salient in Scandinavia. Examples
include
the
Institute
for
the
Management of Innovation and
Technology linking SSE, CUT and
Stockholm University, the SME
consortium in Sweden and LOK, a
logistics research network, run from
CBS but involving most Danish
business schools.
Thirdly there are a number of
more or less informal groups that exist,
mainly comprising researchers and
managers in varying proportions and
for different reasons. These include
departments or business schools
running free or fee based external
seminar series, Executive Clubs,
alumni associations and regional /
local
development
bodies
with
academic membership. In addition
there were one or two small scale
novel institutions such as, at CBS, a
Vice
Dean
of
Knowledge
Dissemination and a half time
academic researcher who uses his
consultancy company as a base for his
research.
5.10 Country culture, nature
and institutions
All three countries are small in
terms
of
population,
highly
industrialised, internationally active
and social democratic in terms of
politics. Most home industries are very
highly concentrated in terms of
domestic production and specialist
education, such as management, is
focused on a small number of
business schools. The term “small
world” was used constantly by
respondents to explain their relatively
close relationships with industry,
commerce and other academics.
However there was a strong belief that
things were changing. Competitive
The IMP Journal
Volume 4, number 1
pressures and moves to more centrist
government policies led some older
academics to hark back to a golden
age when “the doors were always
open”.
6
Reflections
This research project was
essentially exploratory although I
would claim that it has produced some
interesting results. However it should
be remembered that, firstly, the
sample was biased in favour of
researchers who were generally
disposed to carry out empirical
research
involving
managers.
Secondly, many were researchers in
the IMP tradition and they in turn
recruited
other
academics
and
managers who were probably involved
in
research
relationships
with
managers. As a result the picture that
emerges is likely to be one that
overstates the extent of manager –
researcher relationships and which
biases many of their characteristics.
The results were used to
populate an analytical framework
involving a number of categories (e.g.
researcher attitudes) which, on the
whole, worked reasonably well in
terms of presenting the data. Some
categories
showed
a
tendency
towards a common result for some of
the responses. If they could have been
measured in quantitative terms they
might be described as forming a
reasonably tight distribution around a
mean. Conversely other categories
were clearly widely dispersed across a
distribution of the possible “values” of
the “variable” concerned.
Following the analogy with
quantitative data there has been no
systematic attempt to “correlate” the
results between single categories; for
example to compare how PhD
students gained access to managers
as compared to academic staff. This
could be the subject for more
systematic and in depth future
analyses. However an exception was
made in the case of adherents of
different disciplines having rather
52
different views about normativity. In
this case categorisation as a member
of a discipline was clear and
unequivocal, unlike many of the other
categories which were used to analyse
the data.
In what follows the results are
reviewed and an attempt is made to
craft some general observations. The
categories where there seems to be
some uniformity of responses within a
category are presented first.
First of all, not surprisingly,
given the choice of respondents, most
researchers seemed to be enthusiastic
about their profession and gave all the
usual reasons for being so, most
involving trying to understand their
worlds. They were largely interested in
what happened within and between
organisations and the social rewards
of being involved with managers. In
terms of normativity most researchers
seemed to be at least sympathetic
towards the notion of helping
managers to be more effective. What
was even more interesting was some
researchers,
including
historians,
admitted that they were not unhappy
about being of assistance to managers
though that was not their main goal.
By
contrast
managers
revealed, again hardly surprisingly,
that working with researchers was a
relatively minor aspect of their lives.
Managers were generally unsure
about what they had gained from such
involvement.
Access to managers seemed to be
often
serendipitous
and
while
researchers claimed that cold calling
produced good results they rarely
resorted to that mode of contact.
The discussion of the research
process largely centred on the
interview, and most often the semi
structured form. There was a strong
feeling by researchers that the process
should be two sided with the
respondent enjoying the social side of
the process as some sort of reward for
giving up their time. This often led to
long
term
relationships.
The
management of internal and external
politics was clearly important and
The IMP Journal
Volume 4, number 1
some researchers clearly found this
problematic and not something
researchers should necessarily have
to deal with.
A feature of Scandinavian
economies was the relationships of
business schools with industry and
commerce that might help to smooth
the path of researchers. These include
strong regional and national, industry
and function based associations, trade
unions and cross business school
research collaborations.
There were many categories in
which respondents’ answers were
somewhat diverse. For example one of
the main issues that distinguished
among researchers was that of the
ethics of research. Some researchers
clearly had strongly held beliefs about
what was acceptable. Others felt that
the pursuit of knowledge (and
qualifications?) meant that ethics were
less important.
Managers had a wide variety of
different reasons for being involved in
academic research which varied from
near term personal benefits to
accepting research as something all
members of society should support.
Many and various access
networks were used by researchers
from family and friends to industry
bodies.
In the interview situation it was
clear that a wide variety of roles could
be adopted (e.g. the naïve woman)
and that ways of relating to the
respondent in the situation were also
diverse.
While
it
was
generally
accepted that almost all research
represented an intervention in a
situation there was a huge variation in
the extent of such involvement from
the negligible to the considerable.
There was also great variety in the
forms that the intervention took.
Researcher networks included
that most interesting of all researcher
relationships; that between supervisor
and doctoral student. Here there was a
clear dispersion of experiences from
rather
wonderful
to
disastrous.
53
Working with colleagues displayed a
similar heterogeneity of experience
The normativity issue clearly
distinguished
between
different
researchers
with
some
totally
committed to what was, in effect,
consultancy, and others claiming that
they tried not to be involved in this
way. The former mode clearly involved
more contact than a single interview
and this led on to some respondents
discussing the issues and complexities
of long term relationships with partner
organisations
The main theme that emerged
analysing the results from different
disciplines
again
concerned
normativity. Some were clearly totally
at home with the idea that their main
objective was to help organisations in
general and managers in particular.
Other respondents equally clearly
claimed to be members of a discipline
whose only goal was the pursuit of
knowledge for its own sake.
Business
schools
in
Scandinavian countries, even within
their limited numbers, have a variety of
different
forms.
Their
overall
relationships
with
industry
and
commerce appear to be rather strong
although the experience of individual
researchers varied enormously.
7
Further Research
There is a great deal more
work to be done around this research
topic and the argument for doing so is
that any body of researchers should
be reflexive about what they do. This
is necessary and does occur at the
philosophical level in terms what
research methods we should used and
what ontological and epistemological
positions researchers should take.
However there has been relatively little
research on how and why researchers
behave as they do in the field. There
should be more.
In terms of developments stemming
from this piece of research the
following alternatives are offered.
Following directly on from this
research project one option would be
The IMP Journal
Volume 4, number 1
quasi ethnographic study of the
processes which occurred before,
during and after my interviews with
academics and managers, taking into
account the variety of different actors
and settings. An extreme alternative
would be to re-interview one or more
interviewees to get their views about
what was going on in the first interview
process and why. These types of
study might throw light on the
processes
of
semi
structured
interviewing in research using myself
to study a situation I was involved in.
Another alternative with the
same data set would be to attempt to
relate categories to each other. For
example one could check for
consistencies
of
beliefs
within
particular disciplines or the differences
in responses between students and
supervisors. Especially interesting
would be to compare the whole data
set of responses from a researcher
and a manager they had interviewed.
Also, since the interviews were
obviously all recorded it would be
possible to carry out discourse
analyses of the interview situations
with particular emphasis on the
language
differences
between
researchers and managers.
In terms of new research it
would be appealing to carry out a
series of longitudinal case studies of
the focal relationship between a
researcher and a manager. As an
extension of this idea the networks of
both actors could be mapped out and
the extent to which their relationships
are dependent on other relationships,
and the ways in which this occurs,
could be investigated. In both of these
cases this would allow an IMP
researcher to see what, if any, of the
IMP concepts and frameworks would
help us to understand long term
research relationships.
Other important networks also
warrant attention. These might include
within institution networks, cross
institutional networks and cross
disciplinary networks. How any
research project is embedded in a set
of other relationships might suggest
54
how such relationship might best be
developed.
More generally, sociology of
knowledge research rarely attempts to
include in its remit the relationships
between researchers and researched.
The IMP group has a tradition of
crossing all sorts of boundaries.
Perhaps this is an opportunity to use
both IMP theoretical lenses and
methods in an entirely different but
massively relevant field.
References
Axelsson B. & Easton G. (1992)..
(Eds.). Industrial Networks: A
New View of Reality, London:
Routledge.
Caswill, C. & Wensley, R. (2007).
Doors and boundaries: A
recent
history
of
the
relationship between research
and
practice
in
UK
organizational
and
management
research.
Business History 49 (3), 293320.
Das, T. K. (2003). Managerial
perceptions and the essence of
managerial research; What is
the
interloper
business
executive
to
make
of
academicresearcher
perceptions of managers?
British Journal of Management,
14 (1), 23-32.
Fincham, R.
& Clark, T. (2009).
Introduction: Can We Bridge
the Rigour-Relevance Gap?
Journal
of
Management
Studies. 46 (3), 510-515.
Ford, D. (Ed.). (2002). Understanding
Business
Marketing
and
Purchasing, London: Thomson
Learning.
Gibbons, M., Limoges, L., Nowotny,
H., Schwartman, S., Scott, P. &
Trow, M. (1994). The New
Production of Knowledge. The
Dynamics of Science and
Research in Contemporary
Societies. London: Sage.
Grey,
C.
(2001).
Re-imagining
relevance: A Response to
The IMP Journal
Volume 4, number 1
Starkey and Madan, British
Journal of Management, 12,
Special Issue, S27-S32.
Hatchuell, A. (2000). The Two Pillars
of Management Research,
British Journal of Management,
12, Special Issue, S33-S40.
Kilduff, M. & Kelemen, M. (2000). The
Consolations of Organization
Theory, British Journal of
Management,
12,
Special
Issue, S55-S60.
Löwstedt, J. & Sjernberg, T. (2006).
Producing
Management
Knowledge;
Research
as
Practice. London and New
York: Routledge
Starkey, K. & Madan, P. (2001).
Bridging the Relevance Gap:
Aligning Stakeholders in the
Future
of
Management
Research, British Journal of
Management,
12,
Special
Issue, S3-S26.
Stjernberg, T. (2006). Interview as a
Source of Knowledge. In
Löwstedt, J. & Sjernberg, T.
Producing
Management
Knowledge;
Research
as
Practice. London and New
York: Routledge
Wilkinson, I. & Young, L. (2002). On
cooperating - Firms, relations
and networks, Journal of
Business Research, 55 (2)
123-132. .
55
The IMP Journal
Volume 4, number 1
57
Temporal Profiles of Activities and Temporal
Orientations of Actors as Part of Market
Practices in Business Networks
Per Andersson a and Lars-Gunnar Mattsson b
a
Stockholm School of Economics, Sveavägen 65, 113 83 Stockholm, Sweden, e-mail: per.andersson@hhs.se
b
Stockholm School of Economics, Sveavägen 65, 113 83 Stockholm, Sweden, e-mail: lars-gunnar.mattsson@hhs.se
”the element of Time....is the centre of the chief difficulty
of almost every economic problem”
Alfred Marshall (1890; 1961, p.vii)
Acknowledgements: The authors are grateful for useful comments from David Ford and Håkan Håkansson.
Abstract
The purpose of the article is to contribute to concepts and methods for research on network dynamics. We
introduce a business practice oriented perspective on temporality. Based on literature on organizations we
develop a framework identifying temporal profiles of business activities and temporal orientation of business
actors. We apply these temporal dimensions to an STS/ANT - based business practice approach that
distinguishes between three interacting practices: normalizing, representational and exchange practices. We
illustrate the framework with some empirical examples. The paper is concluded with a discussion of how the
approach might be applied to research on temporality in business networks. In an appendix, we comment on how
temporality is treated in marketing and strategy literature.
Keywords: Time, temporality, network dynamics, market practice
1. Introduction: The Construction of
Time in Business Practice
Business managers often bring in
aspects of time in accounts of what they
do and how they perceive certain business
activities.
A former Vice President of Nordea, the
largest banking group in the Nordic
countries, about the merger process that
resulted in the pan-Nordic banking group,
gave the following analysis:
"In October 1997 when the deal was
announced, Merita and Nordbanken at
the same time signalled to other Nordic
banks that they were going for a panNordic
strategy.
Already
then,
Unidanmark in Denmark showed
interest
in
joining
Merita
and
Nordbanken, but it was agreed on both
sides that the time was not ripe for
Unidanmark to join. The idea of
"mergers between equals" was strong
in Merita and Nordbanken, so
Unidanmark needed some time to grow
and expand through acquisitions before
they joined the emerging pan-Nordic
Merita-Nordbanken group. In Norway,
the situation was tricky due to certain
legal restrictions concerning foreign
ownership of banks. Here, Christiania
Bank was interested. The time for them
was also ripe as they feared being
incorporated into the large Norwegian
bank Den Norske Bank. So Merita-
The IMP Journal
Volume 4, number 1
Nordbanken had also to wait as regards
Christiania Bank until the legal barriers
had been lifted. Thus, the timing of
bringing in new partners was delayed,
but for different reasons in Denmark
and Norway.” (From a research
interview by the authors with Karl-Olof
Hammarkvist)
The quote indicates the
central importance attributed to the timing
of certain steps and the sequential order of
business actions. Similar accounts were
given by managers when describing the
intricate processes of a major turnaround
in the marketing organization of a
company (selected from Andersson 1996):
”- It took much longer time than
planned...”
”- The process came to a stop, …which
we had to cope with in some way….”
”- …and then the process took off, and
we were able to work as planned
again…”
”- The process was constantly
interrupted by other things that had to
be prioritized..”
”- We were forced to rethink and
change the order of doing things in the
project.…”
”- Everything suddenly happened much
faster than was originally planned…”
”- The timing was not the best…”
”- The timing was perfect…”
We were able to observe
how different temporal dimensions seem
to be important when accounting for some
business actions, while other actions
seemed to be associated with other
temporal dimensions. The last two lines,
with two opposite accounts of the same
business action (and its timing), also
indicated that different actors, due to their
organizational position, sometimes seem
to interpret the temporal dimensions of
business actions in completely different
ways.
Thus, also derived from our
experiences of management education
programs and case research over the
years, managers’ perceptions of temporal
dimensions and interpretations of business
actions and of market dynamics in
58
general, may differ between business
actors. This might be due to their positions
in the market, their experiences,
cognitions, strategic intentions etc. The
business managers’ representations, or
“constructions”, of time were in many
cases seen as part of the process of
”creating orderliness” in the business
operations. Business managers engage in
activities intended to coordinate temporally
diverse
or
asymmetric
business
operations.
Our observations of how
business managers handle temporality in
their business activities, and how they
choose to account for and represent these
temporal aspects, also indicated a
considerable degree of complexity. First,
as indicated above, the temporality of the
business actions and perceptions seemed
to be associated with several different
dimensions of time. Second, part of the
complexity was associated with the fact
that the dimensions of temporality were
embedded in equally complex intra- and
inter-organizational interactions,
wellknown from IMP research. Two questions
emerged, and have guided us in our work:
How can we account for the observed
temporal dimensions of business actions
and perceptions? And how can we relate
these temporal dimensions to a more
general framework for describing and
analysing business practice?
2.
Purpose, Conceptual Framework
and Disposition
Our purpose is twofold. Our
first aim is to develop a conceptual
framework to analyse the set of temporal
dimensions (such as timing and speed)
that can be associated with business
activities. We will call this the business
activity’s specific temporal profile. We will
also develop a conceptual framework for
how business actors perceive temporality,
their temporal orientation (such as viewing
direction and time horizon).
Our second aim is to relate these temporal
dimensions to business practices, thereby
introducing
a
business
practice
perspective on temporality as a means for
further research on business networks.
This approach, based on actor network
The IMP Journal
Volume 4, number 1
theory (ANT), has recently begun to be
used also by IMP researchers. In
particular, we will use the ideas on market
practices and the shaping of markets
presented in Kjellberg and Helgesson
(2007).
We base our paper on
three conceptual frameworks: a business
network perspective, an ANT-related
market practice approach, and the
literature on temporality and organizations.
We refer to ongoing IMP research, in
particular to the constructivist lines of
research by discussing, for example,
actors’ network pictures, network theories
and network horizons. Our focus on
temporality in networks links to a general
IMP interest in dynamics.
The disposition of the paper
is as follows. We begin by introducing our
conceptual framework and discuss how
temporal aspects are treated in IMP
literature. This is followed by two sections
on temporality concepts: temporal profiles
of business activities and temporal
orientation of business actors. We then
introduce a market practice framework,
apply it to the temporality aspects and
offer some empirical illustrations based on
our earlier research. The paper is
concluded with two sections, one on a set
of business practice phenomena for
research and one on business network
research in general. In the Appendix, we
comment on temporality in marketing and
strategy literature.
3. Some observations on IMP research
and time
From the beginning, aspects
of time have been of major importance in
IMP research. The duration and dynamics
of exchange relationships as well as
investment processes, coordination in
technical development, and sequential
structure in internationalization are
prominent examples of such phenomena
and findings. Easton and Araujo (1994)
present a general discussion of market
exchange, social structures and time.
Methodological aspects have been
addressed in a temporal perspective (e.g.,
Halinen and Törnroos 1998). Papers on
”timing” have appeared more recently
59
(Andersson and Mattsson 1999, 2006;
Hedaa and Törnroos 2002; Medlin 2002).
The time dimension of business interaction
has been part of empirical studies and
conceptual analyses. It has been observed
and explained that exchange relationships
of long duration and characterized by
intensive interaction not only exist, but are
a fundamental feature of industrial
markets. Also the networks in which the
relationships were embedded influenced
and were influenced by such interaction
over time. Change in and stability of
network structures were seen as being
significantly influenced by endogenous
network processes.
Beginning with a focus on
the dynamics of dyadic relations (the
interaction
model),
such
temporal
dimensions as duration, temporal bonds,
investments
in
relationships
and
relationship cycles have been investigated
(Håkansson 1982; Ford 1980). Stability
and change in firms’ supplier structures
over several decades (Gadde and
Mattsson 1987) as well as the long-term
development of one dyadic relationship
(Liljegren 1988) are examples of topics in
early empirical studies of change and
stability covering extensive time periods.
In the business network perspective on
markets, temporal interdependencies were
widened
to
include
connected
relationships and actions of a strategic
nature to change the connectivity pattern,
e.g., by market entry, supply network
strategies, new distribution arrangements,
mergers and acquisitions, strategic
alliances, systems selling (e.g., Johanson
and Mattsson 1992). Network conditions
for such developments change over time,
and therefore the timing of actions is
crucial. How such strategic actions cause
multiple, sequential and interrelated
strategic
reactions
in
others
in
interconnected
networks
has
been
conceptualized as ”domino effects” (Hertz
1998).
What might be included in
the temporality dimensions of actions and
interaction in business markets? In the
network literature, with its emphasis on
interdependence
and
exchange
processes, there are many examples of
temporal
analyses
of
exchange
The IMP Journal
Volume 4, number 1
relationships,
technical
change,
internationalization,
purchasing
and
marketing strategies (e.g., Ford 2002;
Ford et al. 2003). Analyses of
postponement
vs.
speculation
in
distribution provide another example of
network studies on temporality (e.g.,
Gadde 2000). The examples focus on
coordination, sequencing and duration.
Hedaa and Törnroos (2002),
who have a network perspective on
business markets, discuss how timing is
both an aspect of the orderly world of
routines and of the complex world of
unforeseen events. Later these authors
have discussed timing as related to
managerial intuition rather than to
rationalistic decision-making (Törnroos
and Hedaa 2005). Another example is
Andersson and Mattsson (2006), who
analyse timing and sequences of mergers
among wholesalers in an industry.
Using the ARA framework, a
number of dissertations at Uppsala
University
related
to
technology
development and temporal dimensions
have been presented during the past few
years. In his study of demand for electricity
in industry, Wedin (2001) refers to activity
cycles, that is, to the fact that resources
are used at some time, that development
is characterized by co-evolution, that
technology precedes science, and that
specialization becomes more significant
over time. In Bengtsson (2003), a study of
the reintroduction of an old technology,
change is seen as emerging; over time,
interaction among actors results in bonds
and webs and industrial activities have
been developed interactively through time.
In Baraldi (2003), a study on the use of IT
in
product
development,
temporal
considerations include the timeliness of
resources,
speed-up
of
resource
performance, and technology development
understood in a network structural
perspective. In a business network
perspective,
with
its
focus
on
interdependence
and
dynamics
in
networks, temporal issues are inherently
complicated.
In a recent manuscript, coauthored by five core IMP Group members
(Ford et al. 2008), the authors build on the
conceptual and empirical work of the IMP
60
Group as regards business interaction.
They argue that the substantive nature of
the interaction gives it an existence in time
and space and that these two dimensions
“have to be brought to centre-stage” (p. 5)
because “time comprises a major
opportunity/problem for all of those
involved in interaction” (p. 17). In
Håkansson et al. (2009) time is analysed
with reference to sequentiality of
episodes/activities and cumulative effects
of activities. Based on their analyses we
conclude that even if interaction processes
are of central importance to the
performance of firms and markets, they
are, for several reasons, quite difficult for
analysts and managers to understand as
well as for managers to act upon. First,
delimitation of interaction processes is
problematic. Second, such a process is
interdependent with prior, concurrent and
future interaction processes in the dyad.
Third, the dynamic network context also
affects the outcome of dyadic processes.
Fourth, involved actors experience and
interpret the same interaction process
differently.
Still, even if spatial and
temporal dependencies are uncertain, and
open to different, sometimes conflicting
interpretations, interaction processes in
the business world result in products and
services of societal value. The problems
are handled, more or less “well”, in
practice. This causes us to argue for a
“practice
turn”,
involving
special
consideration of temporal aspects. They
can be taken as a starting point for further
analyses of the central role of time in
business interactions. If we take a practice
perspective, the fact that interpretations of
time dimensions in interactions are varied
and often inconsistent between actors is a
driving force in business practice.
Practices are devoted to coping with the
varied interpretations of the business
operations between actors.
4. Starting Point: Business Activities
Have Unique Temporal Profiles
First, we might argue that
time aspects of activities in business
markets, like in social contexts, are
elusive. Time elements often become
The IMP Journal
Volume 4, number 1
internalized, institutionalized, self-evident,
and invisible. Time and temporal
dimensions blend with the general
development and dimensions of business
action and market change processes.
However, the multitude of approaches to
and perspectives with which research has
tried to capture time and temporality in
social and economic research signals an
important
conclusion:
Time
is
a
multifaceted "element" of social change
that cannot be captured easily within the
frames of one or even a few chosen
perspectives, concepts, descriptions or
patterns of thought. The temporality of
business actions, for example, has several
characteristics. We have introduced the
concept of temporal profile to capture the
overall
temporal
characteristics
of
business and market processes. The term
was introduced by Sztompka (1993), who
stated:
“Every social event or social change has
its own proper ’temporal profile’, a
combination
of
four
temporal
characteristics: (1) sequential structure
(the pattern of stages specific, for example
for daily routines, religious rituals,
occupational career, economic growth,
etc.), (2) duration (the length of time it
lasts),
(3)
localization
in
wider
sequences,(when it actually occurs) and
(4) repeatedness or uniqueness.” (p.55)
In organization and strategic
management research (e.g., Bluedorn and
Denhardt 1988; Ancona et al. 2001),
similar ideas have been presented.
Ancona et al. (ibid.) state, for example,
that
”…(one category of temporal
dimensions) maps activities to time.
Examples are rate, duration, allocation,
scheduling, and entrainment… (p. 515)”.
Six universal dimensions (functions) of
time are elaborated by Sztompka:
synchronization, coordination, sequencing,
timing, measuring and differentiating.
It can be assumed that, in all
business actions and activities, some of
these temporal dimensions are underlying
organizers of the actions. We will briefly
describe some of the dimensions, then
select for further discussion a few that are
of particular interest in understanding
61
actions in business markets. Here, we
choose to comment on a set of temporal
dimensions/variables
that
appear
frequently in sociology, organization and
management research: synchronization,
coordination,
sequencing,
timing,
measuring, differentiating, speed, and
duration. Several of them have been
referred to in the section above on IMP
research on time. Therefore we do not
repeat these references here.
Synchronization:
A
large
part
of
organizational life is filled with collective
actions. Things are done together often by
large numbers of people. Actions in
business markets require collective action,
interaction and communication. For such
actions to occur, actors sometimes must
find themselves at the same place at the
same time, or the actions have to be
coordinated in space during a certain,
more extended period of time. A complex
process like a major strategic action
stretching over a period of time requires
certain moments and elements of
synchronization between actors within and
between
firms.
The
greater
the
interdependence of actors, the greater the
necessity for temporal synchronization.
Coordination: Synchronization leads into
the second, more encompassing aspect of
temporality
in
strategic
actions:
coordination. Processes are related and
interdependent
and
need
to
be
coordinated. The division of work in
networks leads to a need for coordination.
The mechanisms by which such
interdependencies
are
handled
in
networks,
whether
or
not
such
coordination involves direct interaction
between actors, constitute the most
important dimension linking actions to
time.
Sequencing: Both synchronization and
coordination are strongly linked to a third
temporality function: sequencing. A
business action most often involves a logic
whereby certain activities or events follow
one another in sequences. These
sequences relate the individual actions
within a time order and to phases or
stages in an overall work process.
The IMP Journal
Volume 4, number 1
Sequencing is dependent on coordination,
because coordination between different
sequences in different processes often
involves more than one actor, and in a
business network setting, coordination
within a sequence is often affected by
unforeseen conditions. Coordination is
thus dependent on sequencing. If
concurrent interdependent sequences are
in conflict with each other, coordination is
different from a case in which such
sequences are complementary. Certain
activities in these processes cannot be
performed earlier or later in such
sequences, and sometimes actors have to
wait (or rush) as a consequence of
changes in such sequences.
Timing: Sequencing leads into the next
temporality dimension: timing. Timing
refers to when an activity is performed, not
in isolation, but in a dynamic context.
‘When’ matters for the outcome of
strategic change processes in business
markets, because conditions change over
time. Timing refers to a number of points
in time when an act could have been
performed and actually was performed.
Timing relates separate acts/activities in a
process to each other, e.g., in terms of a
sequence of acts.
Measuring: Just as in daily life,
organizational and business life is also in
many respects determined by various
ways of measuring and dividing up time
and hence the use of resources. This
applies both to internal business
operations in enterprises and to various
dimensions of exchange processes
between firms, e.g. how firms measure
and value the time spent on certain
activities. The introduction of new market
strategies will have effects on such stable
structures of measuring, also involving
processes of introducing new structures of
measuring time, and using various timerelated measures to organize various work
processes. Measuring as part of
temporality also gives us a link to the fact
that time and temporality in business firms
and networks are also the subject of social
construction.
62
Differentiating: Like in social life, firms
involved in strategic change processes
divide up and demarcate time, allocating
certain resources to certain activities,
while spending other resources during
other periods of time. For example,
strategic change processes will most likely
have effects on established patterns of
time differentiation within and between
firms and their processes.
Speed: Speed, rate, tempo are other
dimensions that relate actions to time. The
speed of business actions can sometimes
be perceived as high (or low), and high (or
low) speed can sometimes also be viewed
as preferred for certain strategic market
actions. Speed can also be assumed to be
strongly connected to social constructions
in processes of business actions.
Duration: Our focus on business actions
does not imply that actions are just events
that take place at a specific point in time.
Instead they are related to episodes of
some duration during which activities may
be carried out, with different and varying
speed.
In the rest of this subsection, we will focus
on five of these temporal dimensions:
timing, sequencing, coordination, speed
and duration. The others will be discussed
later. Bringing these dimensions of
temporal profiles to the forefront, we can
also assume and elaborate on possible
interactions between temporal dimensions
within the temporal profiles of business
activities. Timing, speed and duration are
related to each other and to sequencing
and coordination. Building on a previous
discussion by Andersson and Mattsson
(2006), we assume some potential links
between
timing,
sequencing
and
coordination in connection with activities in
business markets:
Timing, coordination and sequencing:
Timing influences sequencing. Actors
have more or less explicit ideas or plans
concerning the time order of activities.
The sequence relates the individual
activities within a time order and
sometimes also to phases or stages in
change processes. Sequencing also
The IMP Journal
Volume 4, number 1
affects timing because sequencing might
change the pre-planned timing of activities
due to changed conditions. Furthermore,
sequencing is dependent on coordination
because coordination between sequences
involves more than one actor in business
network contexts, and coordination within
a sequence is affected by unforeseen
conditions. Coordination is dependent on
sequencing. If concurrent interdependent
sequences are in conflict with each other,
coordination is different from a case in
which
such
sequences
are
complementary. We can also assume that
the timing of specific activities is
dependent on coordination, because
opportunities for and restrictions on
specific activities are affected by type of
coordination. Coordination is also affected
by the timing of activities; opportunities for
and restrictions on certain activities in a
network change over time.
Speed and duration: On the surface, it
would seem that a business activity should
have a shorter duration, the higher the
speed of the process. This is, however, far
from certain. A change episode is not a
fully predetermined move from condition A
to condition B, independent of other
activities in the network. If the objective,
e.g., is to integrate activities between two
firms, high speed may not be the best
alternative to reach the objective. First,
high speed may cause problems in
coordination within sequences, requiring
more corrections over time than if the
speed had been lower. Second, the speed
also influences the conditions for
coordination between sequences. For
instance, high speed in a business
reorganization
process
may
cause
problems because related sequences with
which a focal sequence is to be
coordinated do not move as quickly. This
may lead to delays or a need to restart the
focal process. We do not mean that high
speed cannot shorten the duration, but we
claim that this is not the case by definition.
Linking speed and duration to timing,
sequencing and coordination: We also
include the idea that business actions can
be aimed at change and consolidation of
new business routines as aspects of the
63
interaction within exchange relationships.
Both positive and negative arguments for
each of the two approaches can be found
and may cause management debates (cf.
the merger case in Andersson 1996). A
high speed/short duration strategy can be
part of a long, planned sequence of
business activities, where the timing of
each activity will depend on how effective
the involved companies are in coordinating
and stabilizing new routines (including new
temporal profiles). The underlying idea is
that the faster the process is carried
through, the less the risk for loss of
resources. Against this we can consider
the idea of a process with lower speed and
longer duration. For instance, radical
market change processes are likely to be
punctuated
by
several,
temporary
reversals in seemingly consistent moves
between system structures, even if in
retrospect, uncertainties and ambiguities
encountered during the process might
become invisible or forgotten. Business
activities related to strategic market
reorientations are not likely to be smooth
evolutions, characterized by a clear
temporal linearity of the reorientation
processes. How companies handle the
oscillations between paths, the reversals
and backtracks, over time becomes
important. It can also be formulated as a
problem of handling the non-linearity of
these activities, and alterations in temporal
profiles during the course of the process.
To sum up, we have
presented a conceptual framework in
which we assume that every business
activity develops its own temporal profile in
interactions with the dynamic market
context.
Many
different
kinds
of
movements occur concurrently in the
context, and the speed of change, the
sequential order, the duration, and the
timing vary between different business
activities.
5.
Temporal
Orientation
Construction of Time
and
One important aspect of
temporal research in organization theory
and sociology concerns the social
construction of time and actors’ ways of
The IMP Journal
Volume 4, number 1
perceiving and relating to time. We call
this the temporal orientation of actors and
see temporal orientation as an important
determinant of the temporal profile of
business activities. Ancona et al. (2001)
differentiate between three categories of
temporal research:
(1) ”Conceptions of Time”, which can
be separated into general types of time
(linear, cyclical, calendar, etc.) and
socially constructed time. The latter is
focused on the ways in which various
social groups in different cultural
settings are involved in the construction
of temporality.
(2) ”Mapping Activities to Time”
describes research that has studied
how events and activities can be
mapped in relation to time, e.g., in
terms of rate, duration and allocation.
(3) ”Actors Relating to Time” focuses on
the way actors engage in temporal
activities, dividing research into two
basic
subcategories:
”Temporal
perceptions”
and
”Temporal
personality”,
focusing
on
”the
characteristic way in which an actor
perceives, interprets, uses, allocates, or
otherwise interacts with time” (p. 519).
The authors present a
number of time studies, with a
constructivist connection, both on a social
group and on a single actor level.
Bluedorn
and
Denhardt
(1988), also using a constructionist
approach, present time-related research.
First, from a macro perspective, they
present
research
describing
the
connection between temporality, culture
and organization. Drawing on well-known
social constructivist ideas (e.g., Berger
and Luckman 1966), they claim that these
macro-oriented texts ”demonstrate most
clearly that time itself is a variable, not a
constant” (p. 300), and ”although time is
so fundamental that people in any culture
regard their conception of it as simply an
immutable part of reality, we will see that
time
is
fundamentally
a
social
construction…”. Second, they observe a
line of time research that has ”a clear
understanding that time is closely related
to organizational productivity and that time
64
can be viewed as a resource to be
managed in the pursuit of organizational
objectives” (p. 303). Here, time is
considered one of several scarce
resources,
to
be
measured
and
manipulated
in
the
interest
of
organizational
efficiency
and
effectiveness. Both measurement and
manipulation refer to social construction.
Third, a line of research focuses on time in
planning and organizational design,
including issues like organizational
perceptions of temporal variability in
planning processes. Finally, they define a
fourth line of research focusing on time
and various types of organizational
behaviour. Here, we find research on the
individual level, looking at similarities and
differences
in individuals’ temporal
orientations, e.g. time horizons and
temporal aspects of decision-making. The
review concludes that more research has
been conducted at the individual and
group level than at the organizational level
and argues for more analysis of “time
management” in organizations (p. 315).
In a third overview, Lee and
Liebenau (1999) sum up organizational
research on time in a matrix. The four
notions of temporality are divided by the
factors clock time vs. social time, and time
as an independent vs. dependent variable.
Social time studies include investigations
of ”varying time”, i.e. how time conceptions
differ between contexts, and of ”changing
time”, i.e. aspects of how temporality can
change in relation to organizational
change.
Pieters
and
Verplanken
(1991) find that actors act, as reflected in
their behaviour, with reference to different
time
horizons
and
different
time
perspectives. They act in an interplay with
the ”moving context”. For example, they
take different temporal vantage points (in
the past, in the present or in the future)
and different temporal viewing directions
(towards the past and/or the future).
Some organizational studies
also address an inter-organizational
perspective on the construction of time.
Bluedorn (1986) describes processes of
integrating
temporally
differentiated
activities
and
behaviours
between
organizations. Others, e.g. Lee (1999),
The IMP Journal
Volume 4, number 1
focus on how changes in organizational
factors such as innovation and technology
affect changes in temporal aspects. Lee
investigated how information systems
affected the temporality of work in trading
companies
using
Electronic
Data
Interchange (EDI).
These research overviews indicate two
important lines of social construction
research: one focusing on macro level
constructions (e.g., temporality in different
cultural contexts), and one on the
individual level (e.g., temporal perception
and temporal personality). In addition, and
as observed by Ancona et al. (2001, p.
518), perceptual variables, although
considered
mainly
individual-level
variables, are often applied directly to
multiple levels of analysis, including
organizational and inter-organizational
levels.
Because we are interested in
temporality in business networks and
markets, we need to especially consider
the multilayer aspects of business
practice.
As argued by Czarniawska
(2004) in a discussion of ANT approaches
to studies of temporal aspects of action in
“modern institutions”, such aspects should
be analysed with recognition of spatially
dispersed actors and a “kairotic” time
perspective, i.e. time is not linear and
objective, but related to the actor’s
subjective perceptions/cognitions and
specific contextual situation. In line with
this, we agree that it is important to
understand
that
perceptions
and
interpretations of the market context and
market dynamics differ between actors,
due to their position in the market, their
experiences,
cognitions,
strategic
intentions, etc.
One ANT based study of
changes in temporality is Kavanagh and
Araujo (1995). They present a multilayered view of time in which different
temporal frames co-exist and draw upon
each other for their existence, illustrating
the processes through which time is
constructed. They show how different
forms of temporality are the product of
heterogeneous
networks
combining
associations of human and non-human
elements. We also acknowledge the role
65
that material resources and other artefacts
play in addition to or as a substitute for
face-to-face interaction. As argued by
Callon and Latour (1981), if the networks
of the social context were simply
associations of people, then the social
order would not rest on particularly stable
or durable foundations.
Thus, we also need to
explicitly introduce spatial aspects of
temporality. Perceptions and cognitions
are influenced by the actors’ network
theories, i.e. their systematic beliefs about
network interdependencies and the effects
of network endogenous and exogenous
forces (Johanson and Mattsson 1992) and
their network orientation (network horizon,
differentiation between types of actors).
To sum up, a business
actor’s temporal orientation concerns its
socially constructed perception and
cognitions on which it partly bases
business activities. The actor’s vantage
point, viewing direction and time horizon
are related to his/her network theory and
network orientation. Temporal orientation
influences the temporal profile of business
activities. One example concerns the
complicated issue of how timing and
sequencing of activities are dependent on
the temporal orientation (e.g., as regards
time horizon and network theory) of
several
connected
actors.
Another
concerns how actors relate the temporal
profile of current activities (e.g., in terms of
timing, speed, duration and sequencing) to
their temporal orientation (with reference
to past, concurrent and future processes).
In networks where temporal orientations
differ
between
actors,
coordinated
activities with regard to their temporal
profiles tend to be less likely.
6.
Introducing a
Business Practice
Framework
for
We ended the previous
section by connecting ideas about the
temporal profiles of business activities with
a constructivist view of actors’ temporal
orientations. We now will relate this
reasoning to a conceptual model of
business practices in markets. Generally,
practice-oriented studies, of which there
are many different types, stress “...the
The IMP Journal
Volume 4, number 1
routine, collective and conventional
nature...” of human action, which is
“...internally differentiated and dynamic”
(Warde 2007). Araujo and Kjellberg (2009)
discuss a practice and performative
approach to markets based on the
intersection between marketing, economic
sociology, and Science and Technology
Studies (STS). Social order emerges as a
consequence of recurrent, interconnected,
and routinized behaviour. This perspective
includes material embeddedness, regards
the constitution of agencies and markets
as outcomes rather than given, and
argues for interaction between ideas and
actions.
Empirical
studies,
often
detailed case studies, analysing business
behaviour in market economies have been
fundamental to the development of the
IMP research tradition. During the past
decade,
IMP-related
research
has
increasingly
taken
on
a
practice
perspective, more or less with an explicit
performative view (e.g., Kjellberg 2001;
Araujo 2007; Finch and Geiger 2008;
Mason 2008; Mattsson 2005; Hagberg and
Kjellberg 2009; Hoholm 2009). Special
sessions on market studies have been
organized at IMP conferences since 2005.
Yet the temporal aspects of practice that
were observed by Kavanagh and Araujo
(1995) have not been in focus.
The
market
practice
framework we are using is the one
proposed by Kjellberg and Helgesson
(2007). In the following section, we will
discuss how temporality is both an
important aspect of practice, and
influences the outcome of such practice.
Three broad interlinked subcategories of
market
practice
are
distinguished:
exchange,
representational
and
normalizing practices.
Exchange practice refers to the activities
necessary to realize exchange between
market actors, as individual transactions
and/or as episodes in exchange
relationships. This includes what is
generally included in selling, buying,
pricing, advertising, distribution, etc. In
exchange practice, we also include
activities that aim to affect the structure of
the market, what may be termed strategic
66
actions, such as market entry, mergers
and acquisitions.
Representational practice refers to
activities that help to describe and analyse
markets and market behaviour, including
the situation of individual actors and
economic calculation of the effects of
market behaviour. An important aspect is
the conceptual and theoretical foundation
for a specific description and analysis, in
our framework an actor’s ”network theory”,
i.e. its set of systematic beliefs about
market
structure,
processes
and
performance and the effects of its own and
others’ strategic actions (Johanson and
Mattsson 1992). Closely related are the
“network pictures” studies within IMP, a
conceptual framework and method used to
analyse network actors’ cognitions about
the network in which they are embedded
(e.g.,
Ford
and
Redwood
2005;
Henneberg et al. 2010). Abrahamsen et al.
(2009) observe that network pictures have
a performative role. However, we argue
that it is important to recognize that
representational practice may include
quite different perceptions of markets,
based on e.g. micro-economic or networkoriented theories, as well as on hybrids
between different theories (Kjellberg and
Helgesson 2006; Mattsson 2005).
Normalizing practice is aimed at shaping
and implementing norms/objectives for
how a market should be shaped and
function at a societal (macro) level and at
a business actor (micro) level. Thus,
normalizing practices include macro-level
activities to develop and implement legal
and voluntary norms for behaviour in
markets and for market reforms in the
public sector: Normalizing also includes, at
the micro-level, activities to develop and
apply individual firms’ objectives and
control systems in the firms’ market
practice. This includes norms for interorganizational interaction.
These practices are linked to
each other through chains of translations
(Callon 1992). Kjellberg and Helgesson
(2007) stress the entangled nature of the
three
practices
by
conceptualizing
“markets as networks of practical
The IMP Journal
Volume 4, number 1
translations” (p. 149). Thus, normalizing
practice produces rules and tools that are
used
in
representational
practice.
Representational practice will produce
both market descriptions that can be
drawn upon in normalizing practice, and
different types of results that feed back
into
on-going
exchange
practice.
Exchange practice, finally, may influence
representational practice through more or
less systematic measurements and
normalizing practice through the interest
that exchange activities create among
actors in keeping or changing norms.
Norms for accounting may translate into
representations of the profitability of
different customers that in turn may
translate into differences in exchange
practice as regards different customers,
e.g.
price
discrimination.
Such
discrimination may then work against the
interests of some actors who refer to
normalizing practice aimed at preventing
price
discrimination.
An
interesting
application of translations in a market
practice perspective is offered by studies
of accounting practices (e.g., Miller 2001;
Ahrens and Chapman 2007).
It is important to consider that the
outcomes of translations are not given,
and may be negotiated. For example,
translations between competition laws,
interaction between competitors and
description of the relevant market may be
negotiated between business actors and
competition authorities.
In the next section, we will discuss how
temporal aspects may enter into the three
practices and the translations between
them.
7. Temporality in Market Practice
IMP studies, as exemplified
above, have resulted in observations of
time, mostly of temporal profiles, less often
concerned with temporal orientation.
Below we make some general remarks on
the three general forms of market
practices, and later we will provide some
examples from two empirical studies of
mergers and acquisitions that we have
analysed with regard to temporality.
Exchange practice and time
67
With a focus on temporal
profiles: Separate interaction processes
need to be synchronized and coordinated.
To achieve this, the sequencing and timing
of processes are crucial. Differentiation
between sub-processes as regards speed
and duration is important. Such temporal
adjustments
between
interaction
processes are partly handled in joint
planning and joint execution processes
between actors. They are also handled
through individual or joint actions as the
processes evolve and are dependent on
how flexible the involved actors are.
orientation: Actors differ with respect to
how they account for experiences (their
own and others’) of the past and how the
past is related to the future, as well as to
what extent their time horizon is short or
long. In an earlier analysis, we discussed
how differences in temporal orientation
affect adjustments to a severe economic
recession (Andersson and Mattsson
2009). The network theories and network
horizons influence exchange practices,
e.g. in terms of development of exchange
relationships, perception of the need to
also include indirect relationships in the
exchange practice, and the opportunities
for structuring changing activities.
Representational practice and time
profiles: Representations include many
standardized reports emanating from the
actor’s own accounting, production and
market information systems, such as
annual and quarterly reports, budgets,
forecasting, monthly sales statistics,
competition analyses, budgets and budget
control, etc. Planning of logistics, of
product development projects, of new
market entries, involves representations of
markets and temporal profiles, such as
synchronization between reports, duration
of processes that are reported, how
temporal aspects are measured. Such
formal,
recurrent
aspects
of
representational
practice
are
likely
important, but informal, idiosyncratic
activities, and interpretations also need to
The IMP Journal
Volume 4, number 1
be considered. We argue that the actors’
network theories and how such theories
are aligned with theories underlying the
formal practices are important. If, e.g., the
formal representation is based on a microtheoretic market view emphasizing
competition, while an actor has a market
theory
emphasizing
exchange
relationships, such an actor will represent
the market with a great deal of information
of an informal character.
orientation: Temporal orientation affects
how learning from the past is represented,
how far in the future the market is
represented, how distant markets are
accounted for. The multilayer aspects of
representational practice need to be
recognized. For example, a structural or
cyclical change at the macro-level may or
may not be represented by a similar
change at a micro-level. Some actors may,
for example, experience growth in times of
a general recession.
Normalizing practice and time
profiles: Societal formal and informal
norms such as market laws, patent laws,
financial market regulations, accounting
standards,
contractual
obligations,
sanctions and rewards related to business
behaviour are related to time. Corporate
strategies are related to time in many
respects. During recent decades, the
development of Swedish competition law
has been synchronized with concurrent
development of the EU competition law.
Patent laws stipulate duration of patent
protection. Accounting standards and
financial market regulations stipulate or
influence all the temporal profile
dimensions mentioned, i.e. repeatedness,
synchronization,
timing,
duration,
measuring, differentiation, coordination
and speed. Corporate strategies also
include varyingly explicit references to all
the dimensions of the temporal profile:
sequencing,
timing,
duration,
synchronization, etc.
orientation: Proposed changes in formal
68
norms such as laws and public and private
regulations are based on investigations
that analyse the past, present and future
attributes and effects of the norms as well
as more or less explicitly on network
theories. Time horizon and network
horizon are both part of the preparation of
the norms and of the norms themselves.
The multilayer aspect of the norms, i.e.
how society as well as individual actors
are affected, also needs to be considered.
As to normalizing practice regarding the
development of corporate strategies, such
norms are less formally determined, but
engage the same temporal orientation
dimensions. The developed strategies that
have a normalizing function in practice
may be challenged within the same actor
organization by informal strategies based
on different network theories and different
time and network horizons. Strategies
employed in a network by different
individual actors may also be incompatible
with each other as a basis for a common
normalizing practice.
Temporal aspects
translation”
of
“networks
of
The three types of practices
are interdependent, as discussed above.
Translations occur over time. Translations
contain temporal attributes. Translations
may affect other attributes of a practices.
Some examples:
a. Competition law is aimed at reducing
synchronization and coordination between
competitors. (A case in the Swedish
Market Court focused on the synchronized
timing of a change in discount structure
among
petrol
retailers.
Marknadsdomstolens
avgöranden,
2005:7)
b. Heterogeneity and asymmetry of
representational practices between actors
as regards the temporal aspects of their
corporate strategies create tensions
between interacting actors, for example
regarding the timing and duration of a
common investment project. It is important
to recognize that calculation, including
temporal aspects, is spatially dispersed
(Czarniawska 2004).
c. Financial market norms require shortterm information from the firms, which
The IMP Journal
Volume 4, number 1
likely affects both how “real” markets and
the situation of individual actors are
represented (quarterly reports, evaluation
of
shortand
long-term
future
developments of individual firms and
“industries”) and how firms change or
stabilize their exchange practice, e.g. in
terms of the timing of an acquisition or the
duration of a contractual agreement.
d. As we have argued above, social
construction of time also involves artefacts
and technology that entail material interdependencies. It is not a matter of “purely
social” construction. Actors are equipped
with resources that may or may not be
available in certain quantities and qualities
for use in the three types of practices at
specific points in time. Investing in a
foreign country may not be possible before
a document allowing this is granted
according to the prevailing norms of that
country. Information technology influences
how rapidly new information representing
certain market conditions affects an actor’s
perceptions of the future. Available
production capacities influence the timing
and speed of deliveries. Production
schedules for a factory affect plans for
deliveries to customers, but the causality
may be reversed depending on exchange
considerations.
Some Empirical Illustrations
To support and illustrate
some of our discussions above, we draw
on three empirical accounts:
1. Information, in the beginning of
2009, from a globally active Finnish
engineering firm, Wärtsilä, a
supplier to power generating and
shipping industries.
2. Reorganization through mergers
and alliances among electronic
component wholesalers
3. Implementing a merger between
two firms in the bio-tech industry
1. Wärtsilä reports during the recession
(Below are excerpts from Wärtsilä
communication on financial year 2008.
After each item we relate to temporal
profiles and temporal orientation.)
69
”The ship power market demand abruptly
stopped during the last quarter due to the
financial crisis. Even if declining demand
had been visible a long time the market
was surprised by the speed of decline.”
(timing, speed, measurement, time
horizon)
”Difficult to foresee how serious and long
the general downturn will be, but effects
for Wärtsilä will be limited because even if
cancellations do occur there is a
substantial order stock for power plant
equipment and services.” (measurement,
sequencing, duration, time horizon)
”Changes in the scheduling of shipyard
activities affect the scheduling in the entire
supply chain.” (coordination, sequencing,
network horizon)
”Demand for services continue to be
positive. Wärtsilä offers products and
services for the entire life cycle of the
installed equipment through a globally
dispersed service organization. Customers
need to upgrade the economic and
environmental
effectiveness
of
technologically more advanced plants and
fulfil
the
stronger
environmental
requirements.” (sequencing, timing related
to normalizing practice, differentiation,
network theory, time and network horizon,)
”Big order stock in the beginning of the
downturn,
flexible
manufacturing
processes, stability of the demand for
services
and
the
global
network
connections give Wärtsilä time to react to
market fluctuations.”(sequencing, duration,
coordination, time horizon, network theory,
network horizon,).
”The fundamental aspects of the power
plant industry are unchanged, but the
financial crisis will likely influence the
timing of the orders.” (network theory,
timing)
”State financed power generating projects
increase as part of economic stimulus
packages, but primarily for infrastructural
projects and not for industrial selfgenerating projects.” (sequencing, timing,
differentiation)
The IMP Journal
Volume 4, number 1
”Difficult to judge when the upturn will
come, and how quickly the balance
between the marine capacity supply and
demand will be achieved, as a result of
increased demand and scrapping of old
ships.” (measurement, speed, sequencing)
Adjustments during economic recession is
shown as a complicated and diverse
process with reference to access to and
interpretation of information, interaction
within and between firms, type of
exchange relationships between suppliers
and customers, interdependencies in
extended supply networks, financial
restrictions and government sponsored
projects. Temporal aspects, especially
temporal profiles, are explicit in the
information from Wärtsilä.
2. Hatteland, a Norwegian wholesaler of
electronic components
The case describes internationalization
processes over a decade, the 1990s
(Andersson 2002). A temporal analysis, of
this case, focusing on timing and
sequencing is reported in Andersson and
Mattsson (2006). The focal firm is
Hatteland (JHE), a Norwegian wholesaler
of electronic components. First, JHE
established a Nordic organization by
acquiring
wholesalers
in
Sweden,
Denmark and Baltic countries and through
a green-field investment in Finland.
Second, JHE joined a strategic alliance led
by a French wholesaler, SEI. This alliance
covered the European region. Third,
towards the end of the decade, one large
US wholesaler acquired SEI. Then another
large US wholesaler acquired JHE. The
driving force for this process was that also
suppliers, other wholesalers, OEM buyers
and contract manufacturers, to which they
outsourced a significant share of their
production
underwent
concurrent,
intensified internationalization during the
1990s, aiming at coordination to increase
the
effectiveness
of
their
global
operations. For wholesalers it was
considered important to be able to control
and coordinate resources for distribution in
increasingly larger regions.
70
Our brief and incomplete interpretation,
related to the framework, is as follows:
Exchange practice included sequences of
mergers,
acquisitions
and
alliance
formation/dissolution in which the focal
firm JHE as well as other wholesalers
participated. Timing and synchronization
were important dimensions of the temporal
profile
because
of
the
on-going,
concurrent structural changes. As to the
operational part of the exchange practices,
temporal aspects of logistics (e.g., timing,
speed, duration, synchronization) and
interaction between the wholesalers and
their supplier and buyer counterparts (e.g.,
coordination,
duration,
measuring,
differentiation) were important reasons for
the structural changes. The formal
changes in ownership were to be followed
by reorganization of operations. The
representational practice likely influenced
the exchange practice a great deal. We
believe that the actors’ temporal
orientation, influenced by their network
theories suggested (were translated to
imply) that a restructuring process in terms
of internationalization was coming and that
the speed of this process was to be quite
high. Also, calculations concerning
temporal aspects of operations were likely
used as a basis for changing exchange
practice. Interaction between exchange
practice and representational practice has
likely widened actors’ network horizons.
Lastly, we have little information about the
normalizing practice, but believe that
efforts to standardize components and to
develop global production strategies
influenced both representational and
exchange practices.
3. Implementation of a merger in the biotech industry.
Pharmacia Biotech (BTG) acquired LKB in
late 1986. The case was reported and
analysed in Andersson (1996; 1996a) and
analysed from a temporal practice
perspective in Andersson and Mattsson
(2007). Stated motives for the merger
were to realize synergies, achieve scale
economies and growth. Below we briefly
analyse the temporal dimensions of two
sub-processes in the overall process
aimed at implementing the merger:
The IMP Journal
Volume 4, number 1
sequencing merger processes and merger
of Production and R&D activities.
Generally, it seems that temporal
orientation related to network theories
differed between the two organizations.
Sequencing merger activities: In order not
to disturb on-going relationships between
LKB sales subsidiaries and their
customers, central management wanted to
delay
the
merging
of
LKB/BTG
subsidiaries. However, on the contrary,
local initiatives speeded up this merger,
leading to an exit of LKB sales personnel
and loss of customers. A rapid
implementation of Pharmacia’s economic
control system, which emphasized central
cost control instead of LKB’s decentralized
profit
centre
concept,
changed
representation practice. This translated
into further problems with continuing LKB’s
strategy to focus on long-term customer
relationships with a strong service
component in its exchange practice. The
normalizing practices in LKB and BTG
differed at the time of the acquisition
(different strategies, different economic
control systems), and the sequencing and
speed of the merger sub-processes
affected the ability to translate the motives
of the merger into changes in exchange
practice.
Representational
practice
residing in the LKB organization was lost
when LKB personnel left.
Merging production and R&D activities:
This sub-process was initiated early but
took longer time than expected. Analysing
overlaps and complementarities between
BTG and LKB as regards customers,
products, techniques and on-going R&D
projects was complicated. An information
system aimed to be a tool in the merged
firm’s development towards an industrial
flow organization, including customer
order-based
production
and
direct
distribution, was not ready when a new
R&D organization was presented. Thus,
temporal
profiles
of
changes
in
representational practices influenced the
ability to change temporal profiles of
exchange practice, e.g. changes in
sequencing for customer order-based
production and the timing of new product
offerings.
8.
Empirical Business
Phenomena for Research
71
Practice
Regarding
organizational
studies in general, Lee and Liebenau
(1999) state that ”…it would be more
appropriate to call them ’time-related
research’ rather than ’research on time’
because they do not deal with time per se
as their main subject…We conclude that
whilst there is much ’time-related
research’, there is little ’research on
time’(p. 1051).
We believe that there are some business
interaction
issues
that
would
be
particularly suitable for a future ’research
on time’ focus (as opposed to ’time-related
research’), issues that would also be
central to business practitioners. It can be
argued that most of the business
phenomena studied in IMP research are
dynamic in nature. Hence, temporality in a
broad sense is addressed in most
empirical IMP studies. However, we have
previously argued that there are probably
certain empirical areas in which a focus on
temporal aspects per se, such as temporal
profiles and temporal orientations, could
enhance our understanding of the
business phenomena in focus. Six such
empirical, business practice areas where
temporality becomes important, and that
we have ourselves some research
background in, are the following.
First, in merger processes,
stabilized temporal structures (as regards
activities and social constructions) of each
of the merging parties tend to come to the
surface. In Ford et al.’s (2008) terms, M&A
processes provide opportunities to study
interactions as ”confrontation” processes
that occur between companies, and where
the different temporal orientations of
actors and temporal profiles of activities
are confronted. Black boxes are opened,
and embedded and taken-for-granted
temporalities, including different temporal
orientations, surface. The bio-tech merger
referred to above shows this.
Second, Lee and Liebenau
also suggest that internationalization
processes are suitable for ”temporally
sensitive research”. ”With the increasing
importance of international business in
The IMP Journal
Volume 4, number 1
cross-cultural environments, the study of
time in different cultures deserves greater
effort from scholars in organizational
studies as well as those in international
business and management” (Lee and
Liebenau a. a. p. 1052). In the wholesale
industry case referred to above, we found
that the timing and sequencing of strategic
actions (like M&A, alliance formation,
market entry) have effects on the shaping
of markets.
Third, business-cycle-related
variation in economic activities, such as
the recent rapid economic downturn, could
provide an example of how temporalities
of different kinds come to the surface.
Fourth, often discussed but
seldom researched, the conflicts between
length of time horizons in the financial
markets vs. “real” markets is an area for
research on the performative role of time.
Fifth, also worth studying is
the impact on business activities of the
implementation
of
information
and
communication systems in business
interaction processes. Lee and Liebenau
(1999) suggest that shifts in time use
could be studied in this research area. We
believe that studies of the development of
exchange practice when new ICT is
introduced would draw attention to
important time-related issues.
Sixth,
obviously,
the
development of practices involved in
logistics provides interesting applications
for research on time, e.g., the network
effects of reduction of lead times, of
postponement
strategies
such
as
customer order-based production, of
implementation of JIT, of Third Party
Logistics, etc.
9. Some Research Issues
Business
practice
and
construction of time are interrelated.
Studies of business practices will
contribute to a better understanding of
dynamics, both as regards processes and
as regards structures. For that purpose,
interdependencies inherent in models,
such as the ARA model in the IMP
research tradition, and the market practice
model in the ANT tradition, on which we
have based our analysis, should be
72
studied with reference to the construction
of temporal orientation and temporal
profiles. As we stated above much of the
research has been “time-related” rather
than “research on time”, i.e. it has primarily
focused on exchange practice and some
dimensions of the temporal profiles of
activities. For a better understanding of
temporality, we argue that knowledge
about the temporal profiles of exchange
practice is not enough. We also need to
include
temporal
orientation,
representational and normalizing practices
and “networks of translation” that link the
three types of market practice.
Business interaction is a
central aspect of business life. Time in all
its shapes is one of its fundamental
attributes. As a consequence, temporal
research should be integrated with the
growing body of business interaction
research. More specifically, we see a
number of potential issues for future
’research on time’ within the IMP tradition
of business interaction research:
First, in contemporary IMP
research, we see a (renewed) focus on its
basic tenets and foundations. In a recent
paper on business interactions, Ford et al.
(2008) draw attention to the importance
and the nature of business interactions,
including their embeddedness in both
temporal and spatial contexts. The authors
argue that interaction can be interpreted
as a ”confrontation” process that occurs
between
companies.
Over
time,
interactions may become structured and
specialized, and interactions such as the
coordination of deliveries or service events
may become standardized or automated.
Other
interactions
may
involve
considerable change, uncertainty and
resource investment for those involved or
be restricted to a specific time period.
We argue that temporality, in
terms of the temporal profiles of business
activities and the actors’ temporal
orientations, may have a central role in
both types of processes, i.e., when
interactions are being stabilized over time,
and when interactions are changed and
re-organized. There are many business
situations where we can anticipate that
interactions are ”confronted”, e.g. merger
processes. It is likely that the temporal
The IMP Journal
Volume 4, number 1
profiles/structures that have been earlier
stabilized in the different actors’ business
contexts before the merger (e.g.,
sequential structures of various production
and distribution activities) will be central to
these confrontations of interactions,
together with the various actors’
interpretations of them. Similarly, the
processes of stabilizing new temporal
profiles (including, e.g., the various actors’
different ways of ”measuring” and
calculating time, and “differentiating” the
time and resources spent on different
activities) should also be a central part of
business interactions, and hence a topic
for more detailed research.
Second, and connected to
the previous point, we suggest that this
renewed IMP interest in the ”micro
dynamics” of business interactions fits
very well with the market practice
perspective previously presented. We
suggest that by linking a narrow
perspective of business interactions with a
focus on market practices, including
exchange practices as well as normalizing
and representational practices, we may
achieve a better understanding of actors’
interpretations and construction of time,
and of the ways in which various
dimensions of temporal profiles become
part of the structuring of business
activities. With a focus on the microprocesses of business interactions, we
may also learn more about the relativity of
actors’ temporal orientations. This also
constitutes a major methodological
challenge for future research. As argued
elsewhere, the reliance on chronology and
sequentiality in our business research
narratives may make our accounts rather
mechanistic in character (Kjellberg and
Andersson 2003). It seems that a rich
account of business activities, besides
accounting for the succession of events
chronologically, also has to account for the
different time perspectives that actors
have as they engage in business action:
What are the future-oriented perspectives
of the actors? How do actors make use of
historical developments when they engage
in business action?
Temporality is multidimensional; activities
also span space; actors can be regarded
as variable. Thus, there exist alternative
73
accounts and analyses of network
processes and structures. There are
therefore
important
methodological
challenges in linking IMP research issues
with a market practice perspective, and an
interest in temporality (Kjellberg and
Andersson 2003). How should we collect
practice data? How can we achieve, in
business narratives, an impression of
presence in business action. We might
construct a succession of events, not
through simple mechanical analogy or
chronology, but by looking for credible
links between business activities and by
making
actors’
different
temporal
perspectives (past, present, future) heard.
This could also be achieved by making our
accounts
of
business
interactions
polyphonic (many-voiced), allowing us to
integrate concordant and discordant
processes and understand the intersection
of different time orientations. We also
need be aware of how “chronotope” in
narratives (time-space integration) can
serve as a means of defining and giving
character to both actors and business
interactions (ibid.). By looking at time and
space and how they interconnect and are
coordinated through the practices in
business interactions, we can cover a
broad range of important research issues.
As mentioned above, the recent network
picture (NP) research stream within IMP
relates to our interest in business practice.
NP research has put much more emphasis
on space than on time. But there are also
some consideration of time. An example is
the model of network pictures developed
by Ramos (2008) that includes a “time
span” dimension that we can interpret as a
temporal orientation dimension. Inherent in
the NP concept is a focus on
representational practice, with more or
less explicit reference to exchange and
normalizing practices.
Third, we suggest that
our focal concepts temporal orientation
and temporal profile require further study.
In the present article, we have made a
distinction between the activity-related
temporal profiles and the more (social)
constructivist
temporal
orientation
variables. However, it can be assumed
that also temporal profile variables such as
”measurement” (of time) includes strong
The IMP Journal
Volume 4, number 1
elements of construction, which might
surface in closer studies of actors’
representational practices. For example, in
their business interactions, firms might put
considerable effort into comparing and
coordinating various ”time schedules” for
production and distribution activities, a
process that involves creating new
temporal structures in joint business
activities, but also some common pictures
or representations of these activities. This
would be an important area for future IMP
studies.
References
Abell, D.F. (1978) ”Strategic Windows”,
Journal of Marketing, Vol. 42,
July, pp. 21-26
Abrahamsen, M.H., Naudé P. And
Henneberg, S.C. (2009),
“Sense making in Networks:
Using Dottograms to analyse
network changes”, Paper
presented at the 25th IMP
Conference , Marseille
Ahrens,T. and Chapman, S.C. (2007),
“Management Accounting as
Practice”,
Accounting,
Organizations and Society,
32, 1-27.
Ancona, D.G, Okhuysen, G. A. and
Perlow, L.A. (2001) ”Taking
Time to Integrate Temporal
Research”,
Academy
of
Management Review, Vol.
28, No 4, pp. 512-529
Andersson, P. (1996) The emergence and
change
of
Pharmacia
Biotech 1959-1995. The
power of slow flow and the
drama
of
big
events.
Doctoral Thesis, Stockholm:
Economic Research Institute
Andersson, P. (1996a) Concurrence,
transition and evolution Perspectives on industrial
marketing
change
processes. Doctoral Thesis,
Stockholm:
Economic
Research Institute
Andersson,
P.
(2002)
”Connected
internationalization
processes: The case of
internationalizing
channel
intermediaries”, International
Andersson,
Andersson,
Andersson,
Andersson,
74
Business Review, 11, 365383
P and Mattsson L.G. (1999)
“Timing of Strategic Actions”,
paper presented at the 1999
IMP Conference, Dublin,
September 1999
P and Mattsson, L.G. (2006)
”Timing and Sequencing of
Strategic
Actions
in
Internationalization
Processes
Involving
Intermediaries-A
Network
Perspective”, In: Solberg, C.
A.
(ed.),
Relationship
between exporters and their
foreign sales and marketing
intermediaries, Advances in
International
Marketing,
Volume 16, Elsevier Science.
297-326
P. and Mattsson, L-G. (2007)
“Market
Practice
and
Construction of Temporality”.
paper presented at the 23rd
IMP Conference, Manchester
P. and Mattsson, L-G. (2009),
“Temporal orientation and
temporal profiles of network
adjustments in times of
economic recession”, paper
Conference,
Marseille,
September 2009
Araujo, L. (2007) “Markets, market making
and marketing”, Marketing
Theory, Vol. 7 (3)
Araujo, L. and Kjellberg, H. (2009)
Shaping
Exchanges,
Performing Markets: The
Study
of
Market-ing
Practices.
The
Sage
Handbook
of
Marketing
Theory 195-218.
Baraldi, E. (2003), When Information
Technology Faces Resource
Interaction. Using IT Tools to
Handle Products at IKEA and
Edsbyn, Doctoral Thesis,
Uppsala: Uppsala University,
Studies
Barius,
B.
(1994)
“Simultaneous
marketing:
a
holistic
The IMP Journal
Volume 4, number 1
marketing
approach
to
shorter
time
market”,
Industrial
Marketing
Management, Vol. 23, pp.
145–154
Bengtson,
A.
(2003),
Framing
Technological Development
in a Concrete Context – the
Use of Wood in the Swedish
Construction
Industry,
Doctoral Thesis, Uppsala:
Uppsala
University,
Studies
Berger, P. and Luckman, T. (1966), The
Social
Construction
of
Reality. New York: Anchor
Books
Bluedorn, A.C., (1986), Primary rhythms,
information processing, and
planning: Directions for a
new temporal technology:
Paper presented at the
meeting of the International
Society for the Study of
Time;
Dartington
Hall,
Devon, England
Bluedorn, A.C. and Denhardt, R.B. (1988),
”Time and Organizations”,
Journal of Management, Vol.
14, No 2, pp. 299-320
Bowman, D. and Gatignon, H. (1996),
“Order of Entry as a
Moderator of the Effect of
the Marketing Mix on Market
Share”, Marketing Science,
15 (3), pp. 222–42
Callon, M. (1992) “The dynamics of
Techno-Economic Networks”
In: R. Combs, P.Saviotti and
V. Walsh (eds.) Tecnology
change
and
company
strategies: economic and
sociological
perspectives
London: Academic Press,
pp. 72-102
Callon, M. and Latour, B. (1981),
“Unscrewing
the
Big
Leviathan:
How
Actors
Macro-Structure Reality and
How Sociologist Help Them
To Do So”, in Knorr-Cetina,
K.; Cicouvel, A.V. (eds.)
Advances in Social Theory
and Methodology: Towards
75
an Integration of Micro and
Macro-Sociology,
London:
Routledge, pp. 277-303
Czarniawska, B. (2004) On Time, Space,
and
Action
Nets,
Organization Vol. 11 (6) 773791
Easton, G. and Araujo, L. (1994) “Market
exchange, social structures
and time”, European Journal
of Marketing, Vol. 28, pp. 7284.
Finch, J.H. and Geiger, S. (2008)
Industrial Sales People as
Market Actors- A Case
Study. Paper presented at
the 24th IMP Conference,
Uppsala
Ford, D. (1980) “The development of
buyer-seller relationships in
industrial
marketing”
European
Journal
of
Marketing Vol. 14, Nos. 5/6
pp. 339-354
Ford, D. (ed.) (2002) Understanding
business
marketing
and
purchasing. An interaction
approach , 3rd ed., London:
Thomson Learning
Ford, D. and Redwood, M. (2005) “Making
Sense of Network Dynamics
through Network Pictures: A
Longitudinal case Study”,
Industrial
Marketing
Management, Vol.34, pp
648-57;
Ford, D., Gadde, L-E., Håkansson, H, and
Snehota, I. (2003) Managing
business relationships 2nd
ed., Chichester: John Wiley
Ford, D. Gadde, L-E., Håkansson, H.,
Snehota,
I.
and
Waluszewski, A. (2008),
”Analysing
Business
Interaction” (IMP working
paper),
http://www.impgroup.org/uplo
ads/papers/6707.pdf
Gadde, L.E. (2000), ‘From Marketing
Channels to Differentiated
Networks-Distribution
Dynamics in a Historical
Perspective’. In Dahyia, S.
(ed.) The Current State of
Business
Disciplines
The IMP Journal
Volume 4, number 1
Spellbound
publications,
2441-2462.
Gadde L-E. and Mattsson, L-G. (1987),
“Stability and change in
network
relationships”,
International
Journal
of
Research in Marketing No 4,
pp. 29–41
Håkansson, H. (ed.) (1982), International
Marketing and purchasing of
Industrial Goods, Chichester:
John Wiley
Håkansson, H., Ford, D., Gadde, L-E.,
Snehota,
I.
and
Waluszewski,
A.
(2009)
Business
in
Networks,
Chichester: John Wiley
Halinen A. and Törnroos J-Å. (1998) “The
role of embeddedness in the
evaluation
of
business
networks”,
Scandinavian
Journal
of
Management
vol.14 no 3, pp. 187-205
Hedaa, L. and Törnroos, J-Å. (2002),
”Towards a theory of timing:
Kairology
in
business
networks”,. In: Whipp, R.,
Adam, B. and Sabelis,
I.(eds.) Making time. Time
and
Management
in
Organizations.
Oxford
University Press
Henneberg,S., Mouzas, S. and Naudé, P.
(2010) Sense-Making and
Management in Business
Networks
Some
Observations,
Considerations
and
a
Research Agenda”, Industrial
Marketing
Management,
(Forthcoming)
Hertz.S. (1998) ”Domino effects in
international
networks”.
Journal of Business to
Business Marketing, 5 (3),
pp. 3-31
Hinings, C. R. and Greenwood, R., (1988)
The Dynamics of Strategic
Change,
Oxford:
Basil
Blackwell
Hoholm, T. (2009) “The contrary forces of
innovation. An ethnography
of innovation processes in
the food industry”, Doctoral
76
thesis, BI Norwegian School
of Management
Johanson, J. and Mattsson, L.G. (1985)
"Marketing Investments and
Market
Investments
in
Industrial
Networks",
International
Journal
of
Research in Marketing, (2),
pp. 185-195
Johanson, J. and Mattsson, L.G. (1992)
“Network
positions
and
strategic
action”,
In:
Axelsson, B. and Easton,
G.(eds.) Industrial Networks.
A new View of Reality
London: Routledge, pp. 205217
Johanson, J. and Vahlne, J-E. (1977),
“The
internationalization
process of the firm – A model
of knowledge development
and
increasing
foreign
market commitment”, Journal
of International Business
Studies, Vol. 8, No 1
(Spring/Summer, pp. 23-32
Kavanagh, D. and Araujo, L. (1995),
"Chronigami: Folding and
Unfolding Time", Accounting,
Management
and
Information
Technologies,
Vol. 5(2), pp. 103-121
Kerin, R. A. Varadarajan, P. R. and
Peterson, R.A. (1992) “Firstmover
advantage:
A
synthesis,
conceptual
framework, and research
propositions.”
Journal of
Marketing, Vol. 56(4), pp. 3352
Kjellberg,
H.
(2001)
Organizing
distribution:
Hakonbolaget
and the efforts to rationalise
food distribution 1940-1960.
Doctoral Thesis, Stockholm:
Kjellberg, H. And Andersson, P (2003),
”Where is the Action? The
Reconstruction of Action in
Business Narratives”, in:
Woodside,
A.
(ed.),
Evaluating Marketing Actions
and Outcomes, Advances in
Business Marketing and
Purchasing, Vol. 12.
The IMP Journal
Volume 4, number 1
Kjellberg, H. and Hagberg, J. (2009)
“Who's
doing
what?
Exploring market practices
and practitioners”, paper
Conference in Marseille,
September 2009
Kjellberg, H and Helgesson, C.-F. (2006)
“Multiple versions of Markets:
Multiplicity and Performativity
in Market Practice”, Industrial
Marketing Management, 35
(7): 839-55.
Kjellberg, H. and Helgesson, C.-F. (2007)
“On the Nature of Markets
and
their
Practices”,
Marketing Theory, 7, 13-162
Lee, H. (1999), “Time in information
technology:
Monochronity,
polychronity and temporal
symmetry”,
European
Journal
of
Information
Systems No 8, pp. 16-26
Lee, H. and Liebenau, J. (1999), “Time in
Organizational
Studies:
Towards a New Research
Direction”,
Organization
Studies, 20/6, 1999, pp.
1035-1058
Liljegren, G. (1988), Interdependens och
dynamik
i
långsiktiga
kundrelationer,
Doctoral
Thesis,
Stockholm:
Madsen, K.T. and Servais, P. (1997), “The
internationalization of born
globals: An evolutionary
process”,
International
Business Review, Vol. 6, no
6, pp.561-583
Marknadsdomstolens avgöranden 2005:7,
Konkurrensverket mot Norsk
Hydro m.fl.
Marshall, A (1890) (1961, p.vii). Quoted in
the introduction of: Winston,
G., The timing of economic
activities
Firms,
households, and markets in
time-specific
analysis,
Cambridge:
Cambridge
University Press, 1982
Mason, K. (2008) “Markets, Business
Models and the Growth of
the Firm”. Paper presented
77
at the 24 IMP Conference,
Uppsala University
Mattsson, L.G. (2005) “Managing in
Markets as Shaping Markets
through
Interacting
with
Market Theories and Market
Practice”. Paper presented at
the
2nd
IMP
Asia
Conference, Phuket
Medlin, C. (2002), “Interaction: a time
perspective”,
paper
presented at the 18th IMPconference
in
Perth,
Australia, 2002
Miller.P (2001) Governing by Numbers:
Why calculative Practices
Matter Social Research, Vol.
68, No.2 ,(Summer) 380-396
Pieters, R.G.M. and Verplanken, B.,
(1991)"Changing our minds
about
behavior",
in:
Antonides,G., Arts, W.,and
van Raij, W.F.(eds.) The
Consumption of Time and
the Timing of ConsumptionToward a New Behavioral
and
Socio-Economics,
Amsterdam: North-Holland,
pp. 63-67
Ramos, C. (2008) , “Developing Network
Pictures as a Research
Tool:Capturing the Output of
Individuals’ Sense-making in
Organizational
Networks,”
PhD Dissertation, University
of Bath
Smallwood, J. E. (1973), “The Product Life
Cycle: A Key to Strategic
Marketing
Planning”,
Michigan State University
Business Topics 21 (Winter)
pp. 29–35
Sztompka, P. (1993), The Sociology of
Social
Change,
Oxford:
Blackwell
Törnroos, J.Å. and Hedaa, L. (2005) “The
role of intuition. time and
timing
in
developing
business
relations:
A
research
note”,
Paper
presented at the 2nd IMP
Asia Conference, Phuket,
December 2005
Warde, A (2005), Consumption and
Theories of Practice Journal
The IMP Journal
Wärtsiläs
Volume 4, number 1
of Consumer Culture Vol. 5
(2), 131- 153
bokslutskommuniké 2008
www.Webfinanser.com
2009-01-30
Wedin, T. (2001). Networks and demand.
The use of electricity in an
industrial process. Doctoral
Thesis,
Department
of
78
Business Studies, Uppsala
University
Vesey, J.T. (1992), ”Time-to-Market: Put
Speed
in
Product
Development”,
Industrial
Marketing Management, Vol.
21, Issue 2, May 1992, pp.
151-158
The IMP Journal
Volume 4, number 1
79
APPENDIX: Temporality in marketing and strategy literature
Temporality appears in marketing and strategy literature in different shapes. Below, we
comment on some such approaches, linking to our chosen temporal dimensions discussed in
the article.
1. ”First mover” and ”Follower strategies”
Research on the advantages and disadvantages of so-called first mover and follower
strategies provides examples of competitor-orientated models. For example, international
marketing research on new market entry strategies often bases its discussions on theories
and the considerable body of previous research on order-of-entry modelling. In a similar
fashion, order of entry assumptions form the basis for much research on new product and
brand positioning in both established and new markets. There are also examples of how both
these research traditions have been combined (Bowman and Gatignon 1996). Kerin et al.
(1992, p. 46) conclude: ”The overall magnitude of positional advantages accruing to the first
mover depends on the comprehensive competitive strategies employed by the pioneer and
followers, in concert with timing.” However, they continue: ”Furthermore, market pioneering is
not a normative strategic behaviour conducive to superior performance for all firms, it can
only provide opportunities for gaining positional advantages. Actual competitive advantages
depend on product-market contingencies and the actions of the first mover and later
entrants” (ibid. p. 48). This model explicitly refers to the timing of strategic actions and
implicitly to speed in a market with interdependent actors. The interdependence, however, is
purely due to competition and the model is silent, or implicit, as regards duration, sequencing
and coordination.
2. ”Sequence models” in Marketing and Market Strategy
A more implicit and taken-for-granted form is when temporal dimensions become part of
sequence models, describing the steps through which different types of marketing activities
are and/or should be taken. Examples of normative models are provided by textbook
treatments of product development, market entry and internationalization. Another type of
sequence model, more descriptive and explanatory than normative, is the ”Uppsala school of
internationalization”, which explains sequentiallity in spatial extension and resource
commitment (Johanson and Vahlne 1977), and the studies of ”Born Globals”, which
describes and explains the high speed of the extension process (e.g., Madsen and Servais
1997). Timing and coordination are at best implicit in the sequence models, while duration
and speed are more in focus. Interdependence implicitly refers more to cooperation than to
competition. An example of sequencing of strategic actions for internationalization is how
production investments and market investment should be timed and coordinated. Production
investments that increase supply capacity need to be matched by market and marketing
investments (Johanson and Mattsson 1985) that make it possible to exchange products
through relationships to distributors and end-users. Resources that have been committed for
distribution and sales in foreign markets cannot be utilized for growth and market penetration
if supply capacity, e.g., depending on delays in production investments, is lacking, and vice
versa.
3. Marketing and ”The Product Life Cycle”
Another example of temporal profiles, especially speed, duration and sequencing, in
marketing analysis is provided by the product life cycle concept. Smallwood (1973) argues
that the product life cycle can be divided into a set of stages or phases with certain
characteristics (introduction, growth, maturity, decline, termination) and that a company may
adapt its sales forecasting, advertising, pricing policies, product planning, and distribution to
these phases. ”The maturation of production technology and product configuration along with
marketing programs proceeds in an orderly, somewhat predictable course over time with the
The IMP Journal
Volume 4, number 1
80
merchandising nature and marketing environment noticeably similar between products that
are in the same stage of their life cycle” (ibid. p. 35). Important for the rather implicit timing
considerations, the cycle may be of long or short duration and develop with high or low
speed. In this model, interdependence in the market is mostly on an aggregate rather than
individual actor level. One implication, that has not been dealt with adequately in the
literature, is that also here, for the individual supplier, supply capacities need to match the
capacities to reach end-users. This is an aspect of coordination. We thus have a link
between the sequence models and the life cycle models that can also be exemplified by
strategies of internationalization. Several temporal aspects are interconnected with the
periodicity of strategic market change processes (Hinings and Greenwood 1992). Changes in
the speed of the change processes, and alterations in the duration (and repeatedness) of
processes are also connected to the linearity of strategic changes. It concerns the directional
consistency, an aspect of coordination, of the change processes over time:
”...one conception of change would anticipate a cumulative momentum or "roll"
from one (strategic) archetype to another. An alternative conception would
perceive the organization as pulled and tugged between competing interests
and, as a result, characterized by disjunctions, oscillations and temporary
reversals or delay in the overall movement towards a different archetype.” (Ibid.
p. 107)
4. ”Time to Market” Ideas
Related to both the ”first mover” and the product cycle analysis, the concept ”time to market”
appeared in managerial texts during the first half the 1990s. The underlying argument was
that by increasing the speed of the product development and market launch processes,
companies could gain a competitive advantage. Increased speed and shorter lead times
(shorter duration) were important temporal dimensions put in focus. Vesey (1992) stated that
”time-to-market is becoming a highly competitive issue for manufacturing companies and in
the 1990s it may be the single most critical factor for success across all markets. A new
group of accelerating competitors is emerging that thinks in terms of ’speed-to-market’ ” (Ibid.
p. 151). Success in managing this is dependent on concurrent engineering, shortened
product life cycles, and technological advances in information processing. We might add that
such concurrent processes involve not only the focal firm, but also suppliers, partners and
end-users, and are therefore an important aspect of coordination. In a similar fashion, Barius
(1994, p. 145) argues that shorter lead times are achieved through parallel and integrated
engineering activities, which increases the speed to market. Hence, concurrence of various
actions (Andersson 1996; 1996a) is an important means to shortening the time to market;
also involved here are speed, duration and timing. Different strategic actions – sequences of
events forming episodes initially connected to a specific group of actors – become each
other’s context. Timing becomes closely connected to aspects of concurrence and
coordination between and within sequences. Interdependencies are made explicit both as
regards cooperation and competition.
5. ”Strategic Windows”
The idea of ”strategic windows” is closely related to temporal profiles, explicitly in relation to
timing and implicitly in relation to speed, duration and sequencing. It is argued that a specific
strategic action can be successfully undertaken only if resources (internal and external) are
available. However, the resources are only available during certain periods of time. Abell’s
original article (Abell 1978) highlights the importance of anticipating and responding to
changes in the marketplace. This is an aspect of coordination. Abell argues that marketing
planning needs to be based on predictions of future patterns of market evolution and
assessments of the firm’s ability to deal with change. ”Short of entry and exit, the allocation
of funds to markets should be timed to coincide with the period when the fit between the firm
and the market is at its optimum” (ibid.). Interdependencies relate to cooperation (availability
of external resources) and less explicitly to competition.
The IMP Journal
Volume 4, number 1
81
Even if the majority of the approaches discussed above (first mover advantage, product life
cycle, time to market and strategic windows) are more or less explicitly based on the
economic theory of imperfect competition and the ”marketing mix” approaches derived from
this theory, there is also some consideration of, or openness to, including temporal profiles of
cooperation between firms. In the network perspectives on markets and market dynamics
referred to in the main text, analyses of temporal profiles of strategic action by firms may be
more explicitly related to a market theory, due to its emphasis on connectivity and on
endogenous structure changing market processes.
The IMP Journal
Volume 4, number 1
Analysing Business Interaction
82
1
David Ford a , Lars-Erik Gadde b , Håkan Håkansson c , Ivan
Snehota d and Alexandra Waluszewski e
a
b
c
d
e
Euromed-Management, Marseille, e-mail : David.Ford@euromed-management.com
Chalmers University, Gothenburg, e-mail: lars-erik.gadde@chalmers.se
Norwegian School of Management BI, Oslo, e-mail: hakan.hakansson@bi.no
University of Lugano, e-mail: ivan.snehota@usi.ch
University of Uppsala, e-mail: alexandra.waluszewski@sts.uu.se
Abstract
This paper develops a conceptualization of the nature of business interaction drawing on the empirical work of the
IMP Group. The paper argues that interaction is the central process within the business landscape. The paper
interprets business interaction as a process that occurs between specific companies and which changes and
transforms aspects of the resources and activities of the companies involved in it and the companies themselves.
Thus the central argument of the paper is that business activities, actors and resources take their form and are
defined by the interactions in which they are involved. The substantive nature of business interaction indicates
that each interaction process will take a unique form in time and network space. This uniqueness has important
consequences both for the structure and processes of the economic landscape and places interaction at the heart
of business development.
1. The Idea Of Business Interaction
The idea that interaction
between individually significant actors is a
primary characteristic of the business
landscape has been a central observation
of IMP studies. (eg Håkansson, ed, 1982,
Håkansson and Snehota, 1995, Ford et al
2003, Håkansson et al 2009).
The
implication of this observation is that it is
not so much what happens within a single
company but what happens between that
company and others that constitutes the
core of business. IMP empirical research
has lead to the conclusion that business
activities, resources and actors take their
form and are defined by interaction. The
idea that interaction is central to economic
life is common throughout the social
sciences. However, there is an important
difference
in
how
interaction
is
approached in theories influenced by
mainstream economic thinking and the
view that has emerged from IMP
research2. Approaches colored by the
traditional assumptions of the market imply
that interaction is a mechanism of
exchange between independent actors
consisting of discrete, frictionless and
generalisable events (Wilk, 1996). Thus,
market theory contributes to the common
understanding that exchange takes place
within “a system that not only regulates
itself but also regulates ourselves, a
process that shapes and forms people
whose relationships with one another are
circumscribed and reduced by the market”
Marglin (2008, p. 2).
IMP
empirical
studies
suggest a view of interaction which is far
away from this simple mechanism. The
basic difference is that the interaction that
has been observed in these studies has a
1
Earlier versions of this paper were presented at the
Annual IMP Conference, Uppsala, September 2008 and as
Chapter 3 in Business in Networks, H Håkansson et al,
John Wiley, 2009.
2
A number of these studies may be found at
www.impgroup.org.
The IMP Journal
Volume 4, number 1
substance.3 The substance of interaction
can be described in several ways, but it
always affects the people and things
involved in it. An important consequence
of substantive interaction is that it always
involves costs for each actor involved in it.
But even more importantly, the benefits of
interaction seem to outweigh these costs
in enough cases to make it impossible for
any company to disregard interaction with
others. For example, interaction may lead
to one of the companies modifying the
product and service offering that it
supplies to a counterpart whilst it may lead
the counterpart to reorganize aspects of its
operations in order to accommodate that
offering. Interaction isn’t just a dyadic
process. All companies simultaneously
interact with several others and interaction
between any two companies may affect
their interactions with these others. Thus
interaction processes are connected
together in a network-like form and these
connections lead to modifications to
activities, resources and to companies
across many organizational borders.
Interaction appears to be the major
means
through
which
companies
systematically relate and combine their
activities and resources to each other. It is
through interaction that the benefits of
widely distributed resources and activities
flow between and into the companies in
the network. As well as being a ubiquitous
process, interaction also forms a working
structure for the network and provides an
element of stability to how different
companies relate to each other in the
network. Interaction provides a means for
companies to address their respective
issues and problems. Interaction may also
generate problems for the companies and
conflict between them. Interaction may
lead to change and dynamism in
companies as well as leading to cooperation and stability.
Interaction is
driven by and produces a world full of
different
and
often
conflicting
interpretations of the meaning of the
3
Similar observations have been made by a number of
other empirical oriented scholars in other fields of
research; see e.g. Hughes, 1983, Latour, 1984, von
Hippel 1988 ,Wilks, 1996, Van de Ven et al 1999,
Marglin, 2008).
83
particular business behavior of different
actors. It is often difficult or impossible for
those involved in the network world to
separate the individual actions, re-actions
and re-reactions of each actor or to trace
their causes, effects and outcomes.
2. Empirical Interaction Processes
IMP
research
has
demonstrated that interaction takes a wide
variety of forms in the business landscape
(For example, Anderson et al 1994; Araujo
et al 2003; Baraldi, 2003; Baraldi and
Strömsten,
2006; Bengtson and
Håkansson, 2007; Blankenburg-Holm,
Eriksson and Johanson, J., 1996). Some
interaction processes have a very long
history
whilst
others
are
more
spontaneous.
Some involve large
volumes of frequently purchased products
or services, others are built around a
single major project. Some interaction
processes include complicated technical
problem-solving,
whilst
others
are
technically simple but time-sensitive. A
significant
category
of
interaction
processes are those that become so
important to those involved in them that
they acquire some quasi-organizational
features. In other words, the interaction
becomes a business relationship (Blois
1972). This type of interaction process
can extend over many years and involve
many resources, activities, individuals and
many different type of problem solving (eg
Bygballe,
2005;
Bocconcelli
and
Håkansson, 2008; Fredriksson and
Gadde, 2005; Gadde and Håkansson
2001; Håkansson, Waluszewski, Prenkert
A typical
and Baraldi
(eds) 2009.).
business relationship may look like this:
The production, delivery and use
activities of the two companies may
have been closely adapted to each
other; Both physical and human
resources may have been adapted
and combined in specific ways;
There may be hundreds of people
from each side that have more or
less frequent contacts with a
similar number of people on the
other side; The two companies
may have accomplished a number
The IMP Journal
Volume 4, number 1
of projects together and there may
be one or more continuing at any
one time; The interaction may have
become quite structured and
specialized involving specifically
designed offerings and procedures
by either or both of the companies;
Some parts of the interaction such
as the coordination of deliveries or
service events may have become
standardized or automated. Other
parts may relate to a particular
problem of one or more of the
participants or be project-related
and involve considerable change,
uncertainty
and
resource
investment for those involved or be
restricted to a specific time period.
Other parts may involve detailed
negotiation and development to
integrate different activities and
resources or even to subsume the
actors into a joint organization or
company.
The
interaction
processes may be sufficiently
critical to one or both of the
companies from a volume, profit or
technological perspective that they
are
closely
monitored
and
systematically evaluated by them.
Each of these relationships is
unique. In each of them the interaction
process is very “heavy” and will have
involved and changed substantial aspects
of the activities and resources of each of
the companies. These changes affect this
particular interaction and also others in
which the companies are directly involved
and others at greater distance across the
network. The total investments of the two
companies in these continuing interactions
are considerable and their effects are
highly significant. A limited number of
these
interaction
processes
often
dominate
a
particular
company’s
operations and can realistically be said to
have formed that company.
In contrast, the business
world also contains many much more
limited interaction processes between
companies that may be either ad-hoc or
short-term. Some of these processes may
be intense and involve important problem
solving or leave significant imprints on the
84
companies concerned. Other interaction
processes may be less significant or
intense but may still be valuable
collectively.
For example, they may
involve a supplier and many of its
customers that only buy occasionally, but
with whom the interaction can be
standardized to reduce costs (Johnsen
and Ford, 2008). Another example of adhoc interaction could be in the case of a
supplier who is able to contribute a
specific technical solution that is crucial for
a particular customer (Bengtson and
Håkansson, 2007).
There are also a large
number of interaction processes, which
involve more costs than benefits for one or
both of the counterparts, despite (or
because) of the efforts of those involved
Håkansson and Ford, 2002. Others may
be in the early stages and may or may not
develop into long-term and important
interaction processes (Ford, 1980; Ford
and Rosson, 1982). It is often difficult to
evaluate these developments and most
companies have to engage in a large
number of such interaction processes just
in order to find the few that are worth
developing further (Buttle and Naude,
2000).
Because interaction is a
process over time, it is likely that
connections will develop between different
interaction processes in which the two
companies are involved.
These
connections may or may not be systematic
or conscious. But their outcome is that
participation in a single interaction process
with a single counterpart relates a
company to a set of many others about
which they may know little or nothing. In
this way, business interaction is a process
in which ideas, solutions, technologies and
problems and interdependencies are
transferred across a network of companies
(Mattson 1989, Håkansson and Johanson
1987 and 1993; Håkansson and Snehota
1989).
Interaction
between
companies enables each to take
advantage of an economic world
characterized by continuous change, but
with many potentially cooperative or at
least mutually beneficial counterparts.
Continuing interaction with others provides
The IMP Journal
Volume 4, number 1
some kind of stability in a world of
unpredictable outcomes and unknowable
influencing factors. In this way, interaction
is both a dynamic and a stabilizing force.
3.
An Initial Conceptualisation Of
Business Interaction
We will use three simple
diagrams
to
develop
an
initial
conceptualization of business interaction
(Figures 1, 2 and 3). A first step will be to
make a distinction between the ideas of
interaction and of exchange. The idea of
exchange is based on the transfer
between actors of unchanging entities;
products, services or money. Exchange
can take place without there being any
significant intervening process between
the counterparts. A simple example of
exchange occurs when someone buys a
newspaper from a street vendor, the
exchange of coin for paper involves no
alteration to either of the exchanged items
and the only interaction is a polite “please”
and “thank you”. This idea of exchange is
A
Figure 1
Exchange
85
represented in Figure 1. Thus we can
interpret exchange as a mechanism that
connects the actors for the time of the
exchange, but which does not have any
content of its own. In general terms this
mechanism is discussed as the “market
mechanism”.
The functioning of this
mechanism without any independence of
its own makes it a very powerful
theoretical construct. It assumes that the
parties to the exchange have all necessary
knowledge and that the objects exchanged
are unchanged in the process.
However, the typical process
that we have observed in the business
landscape and illustrated above is rather
more complex than that of exchange. This
is because there seems to be some sort of
change process that occurs between
business actors. This process and its
content may from an analytical point of
view be separated from the two actors
themselves. This separated interaction
process is pictured by the diagram in
Figure 2.
B
The IMP Journal
Volume 4, number 1
A
86
B
Figure 2
Interaction
This way of conceptualizing
interaction assumes that it is a process
that occurs between actors over time. The
interaction process derives its unique
content from the two involved actors but
develops in a way that is not fully
controlled by either of them. In this way
and over time, interaction changes what
each actor contributes to and receives
from the other and also changes the
actors themselves.
Figure 2 may be
interpreted in the following way:
•
The spiral at the centre of the
figure is a representation of the
process of interaction. The raw
material for this process is the
respective inputs of the two actors.
But it is from each unique process
of
interaction that products,
services, deliveries, adaptations,
developments and payments all
emerge, each with their particular
characteristics and timing.
•
The arrows to A and B from the
spiral represent A and B’s
interpretation and assessment of
what has emerged from the
interaction and what has been their
counterpart’s
intentions
and
approach
to
it.
These
interpretations relate to each
actor’s assessment of its own
approach, to their problems and
aspirations, to their resources and
activities and to their other
interactions and their positions in
the wider network.
These interpretations and
assessments of their interaction form the
basis for the actors’ approaches to further
interaction. The arrows from A and B to
the curve represent these approaches.
These approaches may take many forms,
such as in a change to the quality of a
service delivery; the effort (or lack of it)
that is devoted to a product adaptation; the
stance taken in a negotiation; the timing of
a payment or the commitment to a joint
development. Some of these approaches
to interaction may be in line with a clear
intent or strategy by one or both of the
companies. But some or all may be
unconsidered, inconsistent, or be the
result of inertia and simply continue the
status quo. These approaches may be
oriented towards a single episode of
interaction. They may be unique to a
particular counterpart or be part of an
attempted common approach to a number
of counterparts by either of the companies.
It is likely that there will be inconsistency in
the approach to interaction with a single
counterpart, both between different
individuals in a company and by that
company over time (Ford et al 1988).
The spiral indicates that
interaction is an evolving process. It has
no single identifiable outcome or end-point
because each output is an input into the
continuing process and will be interpreted
differently
by
each
counterpart
The IMP Journal
Volume 4, number 1
A
87
B
C
Figure 3
Interaction Between Three Counterparts
involved in the interaction and by others.
Interaction has both immediate and long
term effects and current interaction is
affected by what has taken place
previously and by the perceptions and
expectations of future interaction held by
the actors. The content of interaction is
always produced by more than one party.
The two arrows that separate A and B in
Figure 2 are intended to show that the
connection between the approach of each
actor to their interaction and to its
outcomes is beyond their individual
intentions or control.
Instead, the
interaction between business actors is
influenced by their approach or intentions
and by the process of interaction itself.
This creates an outcome for each actor
which in turn will be interpreted by both
counterparts.
Each actor probably has a
view of the activities and resources they
wish to contribute and the approach they
wish to take to the interaction. Each
probably also has a view of what they
want to gain from the interaction. But
there is no reason to assume that the
wishes of each will be the same in either
respect.
Each actor has some initial
control over their own activities and
resources.
But the form that these
activities and resources subsequently take
and how they are delivered to and
received from the counterpart is affected
by the way that they interact with those of
that counterpart.
Interaction is an
intervening variable between the activities
and resources of the two companies as
they come together.
Successive interaction over
time can lead to outcomes that mean that
the activities and resources of the actors
and
the
actors
themselves
are
transformed through interaction.
The process of interaction
may occur as a routine or without
conscious effort or planning by any of the
actors involved. In contrast, it may involve
The IMP Journal
Volume 4, number 1
extensive
planning,
development,
negotiation, bargaining or conflict. But
irrespective of how the process develops,
the interaction of resources, activities and
actors means that no single actor is or
could ever be in control of what emerges
from its interactions or be independent in
the world of business.
No actor ever interacts with
just one counterpart. The typical situation
is that interaction is a more or less
continuous problem solving process in
which more than two actors are involved
as portrayed in Figure 3. Each actor will
be taking part in this extended process in
order to address their individual problems
and each dyadic interaction will be
affected to a greater or lesser extent by
those with which it is connected.
This leads us to the following
initial conceptualization of interaction:
Interaction is the substantive
process that occurs between
business actors through which all
of the aspects of business:
material, financial and human and
all of the elements of business:
actors, activities and resources
take their form, are changed and
are transformed.
One important consequence
of this conceptualization is that business
interaction should never be seen simply as
communication or negotiation, even if
these may be important aspects of it. The
greater the involvement of a company in a
particular interaction, the greater will be
the effects on its own activities, on its
resources and on the company itself.
Interaction is a cumulative process over
time. Hence, the characteristics of actors
themselves and of their activities and
resources are as much an outcome of
interaction as they are an input into it. The
actors, activities and resources of
business are defined by interaction. This
view of business interaction has been
refined in the Actor-Activities-Resources
Model
(ARA-model,
Håkansson
&
Johanson, 1992)
4.
The Process And Outcomes Of
Interaction: The A-R-A Model.
88
The ARA Model provides a
conceptual structure for the process and
outcomes of interaction, based on
empirical studies in the IMP research
stream4. The model suggests that the
outcomes of an interaction process (or the
content of a business relationship) can be
described in terms of the three layers:
Actor Bonds, Activity Links and Resource
Ties
between
the
counterparts
(Håkansson and Snehota, 1995). The
model also suggests that each of these
three layers are inter-connected and each
affects and is affected by the wider
constellations of resources, patterns of
activities and webs of actors of which they
form part.
The Actor Layer: This layer
relates to the interpersonal bonds that
develop between individuals through their
interaction. This layer reflects the degree
to which the actors see, know and feel
close to each other; how they trust,
appreciate and influence each other and
become mutually committed (Wilson &
Jantrania 1994, Wilkinson & Young 1994,
Huemer 1998). Bonds that arise between
actors may be more or less strong and will
influence to varying extent what the
individuals involved in a process perceive
as possible and feasible directions for that
interaction. Actor bonds are important for
the
“learning”
and
“teaching”
of
counterparts about opportunities and
solutions, as pointed out in some of the
studies of learning in relationships
(Dahlquist 1998; Håkansson & Johanson
2001; Håkansson and Johanson, 1987;
Håkansson, Havila and Pedersen, 1999;
Håkansson, Huysman and von Raesfeld
Meijer,2001; Håkansson and Johanson
2001.
The Activity Layer: This
layer relates to the integration and coordination of activities that may develop
between actors. Various activities such as
production,
logistics,
administration,
deliveries, information handling may
become integrated and linked together. In
this way, the two companies’ activity
structures can become more or less
4
Håkansson and Snehota (1995) provides case
illustrations of the various elements within the ARA Model.
The IMP Journal
Volume 4, number 1
systematically and tightly linked. The
relative strength of specific activity links, or
their absence, in a business relationship
has been shown to have substantial
economic effects on the actors involved
(Richardson 1972, Dubois 1998, Torvatn
1996).
The Resource Layer: This
final layer relates to how the two actors’
resources may become adapted and more
or less mutually tied together as their
interaction develops.
Specific mutual
adaptations
may
concern
tangible
resources such as physical items of plant
or equipment, but may also include
intangible resources such as knowledge.
Resource ties arise as the two parties in a
relationship confront and mutually adapt
their resources over time (Hallen et al
1991, Waluszewski 1990).
Resource
adaptations can make resource usage
more efficient. But more importantly, the
systematic confrontation of resources also
underlies the development of new joint
resource combinations in the process of
innovation (Håkansson 1987, 1989,
Biemans 1992, Lundgren 1995, LaageHelman 1997, Holmen 2001, Håkansson &
Waluszewski 2002, 2007).
The three layers of content of
buyer-seller
relationships
are
not
independent and there is important
interplay between them: Activity links may
limit or facilitate resource adaptations;
resource ties may limit or favor the
possibility of activity co-ordination and
actor bonds may open up the possibility of
developing activity links and resource ties.
The ARA-model also takes
into account another aspect of business
relationships, namely that actor bonds,
resource ties and activity links do have
consequences that go beyond the
particular relationship in which they arise.
They result from and have effects not only
on what is happening between the actors
but also within the actors themselves and
within their other relationships.
The
content of a particular relationship can be
used by the counterparts to affect their
organization, their use of resources and
the structuring
of their
activities.
Conversely, the content of the relationship
also reflects the characteristics of the two
actors
in
the
same
dimensions.
89
Additionally,
third
parties
to
the
relationship may also take advantage of
developments within the relationship. For
example,
other
actors
that
have
relationships with the two actors involved
in a relationship and the broader network
of businesses can affect and be affected
by the ways in which the content of a
relationship develops (Easton & Lundgren
1992, Blankenburg-Holm et al 1996,
Pedersen et al 2008). Every relationship
is a more or less important connection in a
number of webs of actors, constellations of
resources and patterns of activities that
stretch across many other businesses
(Håkansson and Snehota 1995).
Interaction and development
The existence of interaction
means that it is not enough to look inside a
business company for explanatory factors
in the development of that company. If we
want to understand the development of
business actors, or their activities, or their
resources, or the economic logic between
these elements then we have to
understand the interactions in which those
actors, activities and resources are
currently and have previously been
involved. This means that if we wish to
examine business in an interactive world
then our unit of analysis must be the
specific process of interaction, how it
occurs between particular combinations of
companies and its outcomes in terms of
Activity Links, Resource Ties and Actor
Bonds.
Similarly an analysis of the
development of business in an interactive
world must centre on the development of
specific interaction processes rather than
on the apparent changes that occur in any
single company. These company changes
are likely to be more the outcome of those
processes than the determining factor in
them.
5. Parameters Of Interaction: Time And
Space5
We have emphasised in this
paper that business interaction is a
continuing process of evolution rather than
5
This section is based on Håkansson and Ford (2002).
The IMP Journal
Volume 4, number 1
a series of discrete or independent events.
Each interaction process affects and is
affected by others, but each process is
unique in network space. We can use the
two variables of time and space to
examine the characteristics of interaction
processes in more detail, as follows:
Interaction and Time
Time largely defines the
nature of interaction as a process in which
sequential events are related to each
other. But interaction is difficult to delimit
in time. Interaction can have no easily
identifiable beginning or end. No matter
when or where we look at interaction, what
we see is the continuation of things from
before. This applies just as much to the
interactions surrounding the start-up of an
apparently new company as it does to the
interactions involved in the latest delivery
of a continuously purchased component to
a long-established customer.
A consequence of the
importance of time when analysing
business interaction is that there is no
such a thing as a new network. If we
recognise the existence of a particular
network for the first time, then we are
simply isolating part of a pre-existing and
wider network. Similarly, neither a new
actor nor a newly developed relationship
creates a new network. Instead, new
actors and new relationships always
emerge from something that pre-exists
them and there is always a history behind
them. Each new actor or relationship is
always related to others that already exist.
A new actor will have some, but probably
a rather limited effect on the existing
network. The new actor’s interactions with
others will be affected by and will affect
the continuing and future interactions of
those around it. A useful analogy here is
of the introduction of a new product into a
supermarket. The supermarket is not
constructed around the new product, nor is
there empty shelf-space waiting for it.
Other existing products will have to move
sideways to accommodate it, although the
effect on each one may be small. Some of
these existing products may benefit and
others may lose because of the entry of
the new product. In the same way a new
90
actor’s interactions will affect those around
it, but those interactions will be built upon
the previous experiences of those involved
in the interaction and those of the others
around it. Each actor brings its own
baggage from the past. This phenomenon
is familiar from technological studies
where
path-dependence
has
been
identified as a key issue, but here that
path-dependence is within a wider context.
Path dependence means that the analysis
of interaction must always look behind
current patterns of interaction to what has
preceded them and framed their evolution.
In the same way, it is difficult
to identify the final completion of any
interaction. Each interaction will affect
subsequent interactions between the
participants and others. There are no endresults in business!
The problem of
predicting future directions is multiplied
because the characteristics of each
interaction can affect that subsequent
interaction in multiple directions.
Another problem in the
analysis of interaction is that it is not
evenly distributed over time. Interaction is
likely to be “lumpy”, so that there are
periods of more intense episodes of
interaction than others. It is also difficult to
characterize what defines a single episode
of interaction or to find a neat way to
identify its boundaries or when it starts or
finishes. There is likely to be an important
element of interdependence over time
between single episodes, but these
episodes may also be very important to
understand as units in themselves. Many
of the preoccupations of managers are in
trying to manage effectively within a
particular episode, whether that episode is
defined in the form of a meeting, an order,
a delivery or a financial transfer. One way
for analysis to cope with “lumpy”
interaction is to identify “significant events”
or “critical incidents”.
This approach
clearly provides historical information, but
has similar boundary problems to those of
“episodes”. More importantly, the idea of
critical incidents may also involve
assumptions about the causality of
outcomes that are likely to be unwarranted
in a situation of complex, multi-party
interaction.
The IMP Journal
Volume 4, number 1
The interaction between two
actors will evolve over time through
experience and learning. But interaction at
any one point in time is not pre-determined
by what has happened before. Interaction
will also be affected by the concerns or
problems of the actors as they arise and
by influences on them from their parallel
interactions or those from elsewhere in the
network. However, an interaction episode
is not just an island of significance in a sea
of ordinariness, if for no other reason than
that its significance will be impossible to
assess at the time. Most interaction
episodes are each relatively insignificant
among many others, such as deliveries,
payments, communications etc. These
episodes may be interrelated in an
obvious or in a confusing way, but taken
together they comprise the relationship
between the participants.
A single
episode will affect each of those involved
in it differently. A single episode is also
likely to be interpreted differently by each
of them and by others around them.
A series of episodes will in
many cases simply be continuous or
“normal”, such as a normal flow of orders,
normal terms of payment, normal products
etc. These episodes are part of everyday
life for those involved in them and existing
business relationships and routines play
an important role in providing a basic
structure to business activity.
Most
interaction episodes are not critical
incidents. Many are not significant in
themselves at all, although each may
include some new element. But these
interaction episodes, together with various
actors’ interpretations of them, define the
life of the individual or corporate actor.
Each single element of
newness in an interaction episode
simultaneously restricts and expands the
opportunities for future interactions for
both of the participants in it and for others.
In fact each single element of newness
may have multiple sequential effects in
many directions.
These multiple
processes of restriction and expansion
produce at least two problems for both
actors and analysts:
The first problem is that it is
difficult to make sense of the alternative
possible outcomes of interaction. Thus,
91
actors may not be aware of how their
options may have been broadened or
narrowed by some outcomes: The
multiplicity of simultaneous interactions
both inside and outside of any dyad
means that it is effectively impossible to
construct distinct causal links between
particular episodes and outcomes in
interaction.
This problem makes it
impossible to predict the direction of future
interaction and its effects on individual
companies or to attribute causality for a
current business situation – success or
failure - to a specific action in the past.
The second problem for the
actor is how to anticipate and cope with
the chain of events in interaction. It is
difficult for both actors and researchers to
understand interaction because things
happen in a causal, but unknown
sequence. Also, each actor will have a
view of a preferred or probable sequence
and will interact today with an eye on
subsequent interaction in the future.
These subjectively preferred or predicted
sequences mean that a researcher
seeking to explain interaction over time will
have to be interested both in the evolving
views and pictures of the actors as well as
how activities and resources are actually
evolving.
We will close this section by
relating our treatment of time to four ways
in which researchers have conceptualised
the
problematic
characteristics
of
interaction over time and of the ways that
subsequent interaction episodes are
related to each other:
1. The easiest way to cope with the
issue of time is to ignore any effect
between episodes by assuming
that each episode or exchange is
independent of all other episodes,
as is done within transaction-costs
economics (Williamson and Ouchi,
1981). On this basis, each
interaction
episode
may
be
analysed and managed separately.
The assumption of independent
exchange situations means that
the best total result for the actor
will appear if each situation in itself
is handled in the best way.
The IMP Journal
Volume 4, number 1
2. An alternative approach is to
consider that episodes are related
together over time in a process of
development that comprises a life
cycle consisting of a number of
different stages. This approach
sees episodes as part of a process
of
learning,
adaptation,
commitment
and
distancereduction over time. However, this
and similar stage models tend to
infer that the development of
relationships
is
a
rather
deterministic, unidirectional and
linear process.
This does not
relate well to the much more
complex empirical reality of
change, decay or re-development
(Ford et al 2003).
3 A third way is to assume that the
process of interaction over time
has a cumulative effect. One way
of doing this is to consider the
economics of interaction over time
as an investment process. In this
interpretation, the companies are
considered to be investing in each
other through their relationship.
One
consequence
of
an
investment view of interaction over
time is that these experiences and
processes must be taken into
account when the value of a
company is assessed.
An
investment view also indicates that
a company’s activities should be
steered in ways that develop and
capitalize on these investments
(Johanson & Wootz 1986). The
investment logic puts an emphasis
on long –term relationships since
initial costs can only be balanced
by revenues generated over time.
Such a long-term view of business
processes is advocated also by
researchers in related fields.6
6
A long-term view has been taken by many other
researchers in related areas. Examples include those in
the history of technology (i.e Hughes 1983, Lindqvist
1984), history of science (i.e. Galison 1997), science in
action (Latour 1984, Law 1992) and economic history
(Rosenberg 1994 and David 1985). Others are those
where change, growth or evolution have been central
questions such as Nelson & Winter (1982), Pasinetti
(1981), Penrose (1959), Nonaka (1991), and Kauffman
92
Interaction and Space
The unique substance of
each interaction process positions the
process and the actors, activities and
resources within it in network space. The
relative position of an interaction process
has a number of dimensions of which its
geographical location is just the most
obvious. Other dimensions of the position
of an interaction process include the
particular knowledge that is activated and
produced within the process; the specific
resources that are mobilized and affected
and the activities that are performed; the
form and intensity of their interaction and
the benefits and costs that each accrues
through their interaction (Johanson &
Mattsson, 1988, Henders 1992).
A consequence of their
relative positions in space is that we
cannot explain what happens in a single
interaction process in isolation from those
others with which it is connected. Nor can
we realistically describe an interaction
process except as relative to the other
interactions that may exist in parallel or in
sequence with it. Connections in space
will lead a particular interaction process to
become more or less close to other
interaction processes, in at least some
their aspects.
For example, joint
technological development between two
companies may lead to their increasing
interaction with others that can provide
support technologies or that may be
potential applications for the technology.
In this way an actor becomes related to
other actors of which it may know very
little.
Interaction with a specific
counterpart indirectly but systematically
relates an actor to a whole set of other
actors. Interaction is a way for ideas,
solutions and technologies to travel across
several actor boundaries. The facilitation
of these connections is the classic role of
distributors, such as wholesalers and
export/import
agents
and
financial
intermediaries such as brokers.
The
(1995). A third type is organizational studies dealing, for
example, with organizational learning (such as March 1988
and Powell et al 1996).
The IMP Journal
Volume 4, number 1
interactive
business
landscape
is
characterised by a large and increasing
number of companies with few but highly
specialized internal activities or resources
that operate almost solely on the basis of
their ability to access the activities and
resources of others.
Hence these
companies operate on the basis of their
ability to interact on behalf of counterparts.
This role is illustrated in the earlier
presented Figure 3. Actor A is positioned
so that it interacts with both B and C and
this gives it the opportunity to influence
two adjacent interaction processes and to
mediate between two or more adjacent
actors that do not interact with each other.
This mediating effect may extend to other
more distant interactions across the
network and is commonly seen in the case
of search engines, import houses, trade
organisations
and financial service
providers.
Interaction provides a way
for companies to take advantage of an
economic world that is characterised by
diverse, distant and often unknown but
potentially
co-operative
counterparts.
Interaction
creates
stability
in
a
continuously changing landscape. This
stability is necessary in a world that is full
of influences that are unknowable by any
individual actor.
The structure of
interaction relates a single company to
particular others that in turn are also
related to others. In this way, every
interaction process, every involved actor,
activity and resource has a specific
position that is determined by the
processes in which it is involved. In the
short term these positions provide the
multiple and relative contexts within which
interaction takes place. In the long term,
continuing
interactions
successively
change their positions and the structure.
Business actors employ their
resources differently in interaction with
different counterparts and develop their
interdependencies differently with each of
them. The value of an actor’s resources
and the usefulness of its activities vary
between the particular processes in which
they are employed and depend on the
connections between each process and
others. This variation means that we
cannot analyse a single interaction
93
process in isolation, but only in relation to
others that exist in parallel or in sequence
with it.
The importance of space for
interaction raises the issue of what
approach
to
interaction
may
be
appropriate for an actor in a particular
setting. A consequence of the importance
of space is that there are no general rules
to enable us to determine what interaction
is appropriate. What is good in one
situation may not work in another and
what is right for one company given its
place may be wrong for others. But even
more problematic is that what is right in
the short run may be wrong in the long run
and what is perceived in a positive way by
one counterpart may later be viewed
negatively by the same counterpart.
The ability to analyse and
cope with changes in relation to space
dimensions becomes a key issue for
actors. These changes will involve relative
movement
between
one
particular
interaction and others. This is part of the
critical question for those involved in
analysing interaction, “who should a
particular actor prioritize in its interactions
and who should it not?”
The
close
connection
between space and interaction creates a
dynamic structure in which each process
(and specific elements within it) is related
to particular others that in turn are also
related to particular others. For example,
business interaction may lead a particular
company to systematically adapt towards
a particular counterpart.
This will be
manifested in changes in the company’s
resources,
activities
and
relative
interdependence. But at the same time,
the particular counterpart may be moving
toward some other counterpart and that
counterpart may also be moving in relation
to others and so on and on. Companies
evolve in relation to each other: It is a
case of movements within a moving world!
Space and Routines in Interaction
A considerable proportion of
an actor’s interactions are likely to become
routinised within its different continuing
relationships, encompassing such things
as “normal” deliveries, services, payments
The IMP Journal
Volume 4, number 1
etc.
This normal interaction may be
contrasted with that which consciously or
unconsciously changes the characteristics
of
particular
relationships
or
the
connections between them.
Routine
interaction may lead to the development of
formal or informal rules between specific
companies and across the wider network,
so that each knows what should be done,
or what each can get away with. It is
common for the effects of these rules, both
formal and informal, legal and illegal, to
extend over many participants in a
network and to produce a contrast
between the relative interactions of
“insiders” and “outsiders” (Kriesberg1955,
Palamountain, 1955). Common examples
of these rules include professional ethics,
trade association rules, contract law and
dispute resolution, common terms of trade,
market sharing and price fixing (Mouzas
and Ford 2009).
94
Routines have an important
two-fold effect on interaction. On one
hand, routines create predictability and
trust between the counterparts and can
increase the efficiency of a relationship as
they can reduce many of the costs of
handling the relationship, such as making
deliveries, payments and other day-to-day
activities. However routine interaction may
also become “institutionalised” so that
ways of working are unquestioned and
inefficiency and other problems can
develop.
6. A Model Of The Interaction Process
We can now continue this
conceptualisation in a model that seeks to
systematically relate interaction to time
and space. The model is shown in Figure
4.
TIME
SPACE
Resource Constellations
PATH
HETEROGENEITY
CO-EVOLUTION
Actor Webs
JOINTNESS
SPECIALISATION
Activity Patterns
INTERDEPENDENCY
Figure 4
A Model Of Business Interaction
The IMP Journal
Volume 4, number 1
The model is based on the
idea that business interaction is a process
that takes place within and between the
three layers of the A-R-A Model: the
activities and resources of each of the
counterparts in the process and the
counterpart actors themselves. Each of
these three layers is modified and shaped
by the particular interaction process. But
the model also shows that each interaction
process is part of a network that involves
many others and each activity, resource
and actor forms part of a wider pattern of
activities, constellation of resources and
web of actors across the network. The
form of each actor, activity and resource at
any particular is defined by its position
within these wider webs, patterns and
constellations across the network. The
form of each layer at any one time is the
outcome of previous interactions and part
of a continuing evolution.
We can now use the model
to examine the connections between each
interaction layer and time and space, as
follows:
Interaction, Activities and Time:
The model refers to the
evolution of activities over time as a
process of specialization.
Business
actors7 build specialisation into their
activities relative to counterparts and
others as interaction develops. Interaction
constantly relates individual specialization
processes to each other.
The
specialization of activities by actors is an
important factor in the development of
long-term business relationships and
activity patterns. Specialisation involves a
willingness by actors to forgo short-term
gain for long-term reward and a
commitment to a particular counterpart at
the expense of others. Actors have to be
able to assess and re-assess the costs
and benefits of these specialisations as
their
relationships
develop,
their
interdependence grows and the problems
on which they are based evolve. Many
7
Throughout this paper we use the term business actor to
refer without distinction to companies, subgroups or
individuals. For a detailed discussion of the concept of the
interactive business actor see Hakansson et al (2009)
95
specialisations involve significant costs for
both actors, but many are critical for the
development of particular relationships.
But specialisation is neither a simple nor
an
uncontroversial
process:
the
specialization of a single activity may
affect many other activities either
positively or negatively across different
activity patterns. Specialisation towards
one interaction process frequently involves
specialization away from another.
The specialization process is
closely related to the development of
interdependencies between activities in
the space dimension as will be further
developed below.
Interaction, Activities and Space:
Activities distributed in the
space dimension are interdependent: They
may be in different geographical locations;
they may arise from different problems; be
for specific or wide application or involve
different types of costs and benefits.
Some of these activities may appear to be
independent, but they are more likely to be
connected to others in a variety of ways.
They are more or less interdependent.
Interdependence of activities is both a preexisting structure of interaction and is an
outcome
of
interaction
and
the
development of business relationships.
Interdependence both affects and is
affected by interaction.
The
interdependence of
activities is an unavoidable consequence
of the distribution of activities across the
business landscape.
This distribution
develops over time in order to gain the
benefits of specialisation.
These
interdependencies are both important and
complex and interaction will be strongly
influenced by the possibilities and
problems that arise from them. But the
complexity of interdependencies may
mean that actors are not aware of the
existence of all of them, nor may they
appreciate their implications. Hence a key
aspect of business interaction is the
building, managing and exploitation of
interdependencies.
Dependence on the activities
of others is not a negative, but an
The IMP Journal
Volume 4, number 1
essential
aspect
of
interaction.
Companies can exploit the specialization
of activities in a more extensive way
through
seeking
and
accepting
dependence on others. Dependence also
enables them to develop and exploit other
activities and achieve efficiencies. But
interdependencies also increase the
commitment of companies to specific
counterparts over time as well as the
relative importance of their interaction with
them. Companies also seek to build the
dependence of others on themselves in
order to achieve stability in their
interactions over time with consequent
gains in efficiencies.
The interdependencies in an
actor’s
existing
relationships
simultaneously empower and constrain its
ability to achieve change and growth
(Håkansson and Ford 2002). Thus for a
company relying on the activities of others
increases its freedom to invest its own
resources in more productive areas within
that relationship or elsewhere and provide
the basis for it to develop in new
directions. But at the same time, an actor’s
dependence in its existing relationships
restricts its freedom to act in the directions
of its own choice and require it to invest in
interaction within its existing relationships.
Interaction, Resources and Time:
The development of a single
resource or a combination of resources;
physical, human or financial into particular
technologies or abilities often follows an
identifiable path over time. These paths
have been observed by a number of
studies of technological development that
have explored the existence of pathdependency.
The development of a
resource along a path is closely connected
to the interaction which can be observed in
the use of that resource and in its
combination with others in a resource
constellation. This path can often be
observed in the use of particular resources
in different applications in sequence. For
example, the technological resource of
electronic control was sequentially applied
in different applications such as petrol
pumps, taximeters, domestic appliances
and vehicle engines by combining it in
96
different resource constellations with other
technological as well as physical, financial
and human resources.
The development of single or
constellations of resources over time,
whether depicted as path-dependence,
technology-trajectories
or
life-cycles
seems to be based on two basic features
of resources: The first is concerned with
the resource’s potential to be developed,
which is always unknown; the second is
concerned with how this potential is
related to other resources. For both of
these reasons, the path followed by a
particular resource is difficult to forecast
owing to the multiplicity of problems to
which it may be addressed in
constellations
with
other
multiple
resources accessed by multiple actors.
This difficulty is reinforced because the
evolving path of a resource may depend
on decisions on specialising activities that
are difficult to reverse. For example, the
decision to commit a resource in a
relationship with others will take that
resource down the path of a particular
constellation that may involve significant
investment
by
the
company
in
development activities.
Conversely, a
decision not to invest in a particular
resource may well be difficult or
impossible to reverse at a later time after
others have made similar investments.
Resource decisions may effectively be
once and for all.
Interaction, Resources and Space:
Resources in the business
network are heterogeneous. This means
that the usefulness and value of a single
resource depends on the other resources
with which it is combined, or in other
words, where it is located in network
space. Resource heterogeneity means
that interaction is a means for value
creation across company boundaries.
Conversely, a company can increase the
value of single heterogeneous resource
and of its total resources through
interaction. Interaction positions different
resources more or less close to each
other. Changes in interaction can move
the location of resources towards or away
from each other along a number of
The IMP Journal
Volume 4, number 1
aspects of the space dimension:
geographic,
problem
orientation,
technology. Thus interaction effectively
“moves” resources relative each other and
consequently affects their value and how
embedded they are in each other.
Interaction, Actors and Space:
The existence of the space
dimension has similar implications for the
actor layer as for resources.
These
implications centre on the differences
between
actors
in
their
various
interactions.
Each actor acquires its
identity through its interactions with
particular counterparts. A business actor
cannot exist in isolation, it is always
contiguous with some others and this
leads to jointness (Håkansson and
Snehota 1989). Jointness is a way of
characterizing the specific relationship
between any two actors in relation to all
others. Jointness is a central feature of an
economic world where interaction is a key
attribute and it has a number of aspects.
Jointness
implies
that
interaction is never simply dyadic. Even if
interaction appears to take place between
only two parties, the intentions of those
parties, the content of their interaction and
its outcomes will not be limited just to
them. Any actor interacting with a specific
counterpart depends on the intentions,
resources and activities of all those others
with which it also interacts. Thus, any
company that supplies another does so by
using its own activities and resources. But
it also uses the activities and resources of
its own suppliers as well as those of the
customer. It also uses the activities of its
other customers. In this way, all business
interaction has an important “joint” content.
Thus a business company cannot be
adequately described in terms of its own
internal activities and resources (Ford and
Håkansson 2010). A company is probably
more accurately described as a “node”, or
the point at which the activities and
resources of others come together with its
own, through its interactions with those
others.
Jointness can also take
organizational forms such as when actors
take part in directed, collective or joint
97
interaction with specific others. Examples
include joint technological development,
joint logistics or the development of joint
sales or procurement organizations. Thus
we can identify jointness in the design of
resources, or in the performance of
activities as well as in the holding of
similar ideas about the context of
interaction.
An important effect of
jointness is to reduce the importance of an
actor’s own intentions in determining the
direction of its development and increases
the importance of the combined intentions
of interacting parties in the development of
them all.
The concept of jointness also
covers two related aspects that describe
the orientation of companies towards each
other: Mutuality and Reciprocity (Ford et al
1986).
Mutuality exists when the
interacting
parties
explicitly
pursue
common aims. Mutuality is a measure of
how much a company is prepared to give
up its own individual goals in order to
improve the relative outcomes of specific
others and through this to increase its own
ultimate well-being. Hence mutuality is
closely related to time and the trade-off
that actors have to make between shortterm opportunism and longer-term gain.
Reciprocity, exists when parties feel
obliged to interact on the basis of the
previous actions of a counterpart.
Reciprocity can involve both positive
rewards and negative pay-backs and is
closely related to time: Interactions may be
based both on assurances of long-term
future pay-back and on revenge for
grudges from long ago.
The underlying logic of
jointness is that the space dimension in
interaction makes it both possible and
necessary to create jointly positive results
with others. In the long run any company
in an interactive world is dependent on the
success of its counterparts. The
interaction between two actors is related to
interactions with others and it may be
influenced, mediated or facilitated by these
others. This is seen clearly in the context
of a so-called distribution channel or
supply-chain. In both of these cases it is
the interdependencies and interaction
between the actors, rather than the plans
or control of any one of them that jointly
The IMP Journal
Volume 4, number 1
allow goods and services to flow between
them.
The existence of jointness
fundamentally
questions
the
meaningfulness of analyzing a single
business alone or a single action in itself.
We cannot separate and isolate any action
from the corresponding reactions of
counterparts. All are part of the interaction
between multiple actors. An actor exists in
the context of its network and is defined by
its
relationships
and
through
its
interactions in that network. An actor’s
interactions effectively determine its
characteristics, its capabilities, its scope,
its freedoms, its obligations and its
restrictions. Each actor and each
interaction will depend on and be based
on the actor’s own resources and those of
others who stand with it, behind it and
against it.
Interaction, Actors and Time:
Actors
evolve
in
an
interactive
landscape.
Business
companies successively change both in
terms of the activities they perform, the
resources they control and with whom they
interact. But the evolution of each single
actor is not an individual process, but one
that takes place interactively with others.
Actors co-evolve. Co-evolution means
that if an actor seeks to cope with its own
problems or opportunities it has to do so
by also coping with those of its
counterparts. Co-evolution does infer that
any two business companies necessarily
evolve by becoming closer to each other
or that relationships have a deterministic
life leading to ever greater mutuality.
Instead,
co-evolution
is
a
multidimensional process that takes place
within two or more actors in parallel as
each seeks to relate its own problems,
resources and activities with those of
others. In this way, co-evolution can
actually lead actors to become more
diverse.
The importance of working
together and different suggestions to how
it can develop has been discussed in other
marketing studies such as Achrol 1991
and Achrol and Kotler 1999, in general
network based studies such as Castells
98
2000, Jarillo 1988, and Freeman 1991, in
strategic alliance studies such as Gulati
1998, Gulati et al 2000 and Spekman et al
1998, interorganizational studies (Powell
et al 1996) and can also be related to
research in political science and game
theory such as Axelrod (1984), but also to
research based on social network studies
(Nohria & Eccles 1991, Podolny 1994).
7. Conclusions
The aim of this paper has
been to develop the empirically derived
idea that interaction is the core process of
the business landscape. The idea of
business interaction involves a radical
departure from the view of the business
landscape as a market in which
transactions can be considered as discrete
entities and in which the activities of
marketing and purchasing can be
understood as generalities or in isolation
from each other. The idea of business
interaction suggests that the process
between each set of counterparts is
unique and may involve multiple episodes
over time. This infers a structure for the
business
landscape
composed
of
individually significant and interconnected
interaction processes of which products or
services are simply the most immediately
apparent outcomes. The idea of business
interaction shifts analysis to the activities
and resources of companies and the ways
that these are adapted through interaction.
This analysis shows that business actors,
activities, resources are not simply the
result of individual company strategy but
are equally the outcome of specific
interaction processes. An interactive view
of business opens up a wealth of issues
and opportunities for managers:
Firstly, an interactive view of
business de-emphasises the independent
status and the internal resources of a
company and envisages management as
a process of working within a structure of
specific interdependencies with others.
These interdependencies exist within each
layer of interaction; activities, resources
and actors and emphasise that the
knowledge on which operations are based
is not bound by corporate borders, but
effectively exists between companies.
The IMP Journal
Volume 4, number 1
Secondly, this
way of
conceptualizing
business
interaction
suggests that it plays a key role in
constructing and modifying the value of
physical and intellectual assets in specific
situations. Thus, interaction can be the
means through which business actors coevolve and co-create the physical as well
as the economic context. Of course, this
development might lead to negative as
well as positive outcomes for companies
by leading them into uneconomic activities
or to unproductive investments.
Thirdly,
this
way
of
conceptualizing
interaction
enables
managers to create a measure of stability
and predictability in their operations. The
development of interaction processes with
others enables each company to
rationalise investment in its own physical
and intellectual assets.
However, the
conscious or unconscious development of
interdependencies with others enhances
the importance of managerial skills in
interaction and in the development of
productive relationships.
This paper has presented a
model of the underlying process of
business that is central for the supposedly
separate activities of purchasing and
marketing.
Analysis of business
interaction not only requires a change in
language. It requires us to go beyond the
ideas of purchase and sale, beyond
generalities of supply or customer market,
beyond products or services and the
boundaries of single companies.
An
interactive view of business also changes
the orientation of business management
from that which takes place within an
organisation’s boundaries to its interaction
and co-evolution with others.
References
Achrol, R. S. (1991). Evolution of the
Marketing Organisation: New
Forms
for
Turbulent
Environments, Journal of
Marketing. 55 (Oct) 77-93.
Achrol, R. S. & Kotler, P. (1999).
Marketing in the network
economy.
Journal
of
Marketing,
63
(Special
issue), 146-163
Anderson,
99
J. C., Håkansson, H. &
Johanson, J. (1994). Dyadic
business relationships within
a business network context.
Journal of Marketing, 58, 1–
15.
Araujo, L., Dubois, A. & Gadde, L.-E.
(2003).
The
Multiple
Boundaries of the Firm.
Journal
of
Management
Studies. 40 (3): 1255-1277.
Axelrod, R. M. (1984). The Evolution of
Cooperation. New York:
Basic Books.
Baraldi, E. (2003). When Information
Interaction. Doctorial thesis
no 105, Department of
University.
Baraldi, E., & Strömsten, T.
(2006).
Embedding and utilizing low
weight: Value creation and
resource configuration in the
networks around IKEA’s
Lack table and Holmen’s
newsprint. The IMP Journal
1:1, 39-70.
Bengtson, A. & Håkansson, H. (2007).
Introducing “old” knowledge
in an established user
context: how to use wood in
the construction industry. In
Håkansson,
H.
&
Walusewski,
A.
(Eds.).
Knowledge and Innovation in
Business and Industry. The
importance of using others.
Abingdon: Routledge, 54-78.
Biemans, W. G. (1992). Managing
Innovation within Networks,
London, Routledge
Blankenburg-Holm, D., Eriksson, K. &
Johanson,
J.
(1996).
Business
networks
and
cooperation in international
business
relationships.
Journal
of
International
Business Studies. 27 (5)
Blois K. J. (1972). Vertical QuasiIntegration,
Journal
of
Industrial Economics, 20 (3),
253-272.
Bocconcelli, R. & Håkansson, H. (2008).
External interaction as a
The IMP Journal
Volume 4, number 1
means of making changes in
a company: The role of
purchasing in a major
turnaround for Ducati. The
IMP Journal 2 (2), 25-37.
Buttle, F. & Naude, P. (2000). Assessing
relationship quality, Industrial
(4), 351-361.
Bygballe, L. (2005). Learning Across Firm
Boundaries The Role of
Organisational
Routines,
Ph.D Norwegian School of
Management, BI, Oslo.
Castells, M. (1996). The Rise of the
Network
Society.
The
Information Age: Economy,
Society, and Culture. 1.
Oxford: Blackwell Publishers.
Dahlquist, J. (1998) Knowledge Use in
Business Exchange, Acting
and
Thinking
Business
actors, Doctoral thesis no 74,
Studies, Uppsala University.
David, P.A. (1985). Clio and the
Economics of QWERTY. The
American Economic Review,
75 (2), 332-337.
Dubois, A. (1998). Organising Industrial
Activities
Across
Firm
Boundaries,
London:
Routledge.
Easton, G. & Lundgren, A. (1992).
Changes
in
industrial
networks as flow through
nodes. In Axelsson, B. &
Easton, G. (Eds.). Industrial
Networks – A New View of
Reality, London: Routledge.
Ford, D. (1980). The development of
buyer-seller relationships in
industrial markets, European
Journal of Marketing, 14
(5/6), 339-353.
Ford, D. & Håkansson, H. (2010).
Accounting
and
Interorganizational issues. In
Håkansson, H., Lind, J. &
Kraus, K. (Eds.). Accounting
in
networks,
London:
Routledge.
Ford, D., Håkansson, H. & Johanson, J.,
(1986). How do Companies
100
Interact? Industrial Marketing
and Purchasing, 1 (1), 26-41.
Ford, D., Gadde, L-E., Håkansson, H. &
Snehota, I. (2003). Managing
Business Relationships, 2nd
edition, Chichester: John
Wiley.
Ford, D. & Rosson, P. (1982).
Manufacturers-overseas
distributor
relations
and
export performance, Journal
of International Business
Studies, Fall, 57-72.
Fredriksson, P. & Gadde, L.-E. (2005).
Flexibility and rigidity in
customization and build-toorder production, Industrial
(7), 695-705.
Freeman, C. (1991). Networks of
innovators: a synthesis of
research issues. Research
Policy, 20 (5), 499-514.
Gadde, L-E. & Håkansson, H. (2001).
Supply network Strategies,
Chichester: Wiley & Sons
Galison, P. (1997). Image and Logic: A
Material
Culture
of
Microphysics,
Chicago:
University of Chicago Press
Gulati, R. (1998). Alliances and Networks.
Strategic
Management
Journal, 19, 293-317.
Gulati, R., Nohria, N. & Zaheer, A. (2000).
Strategic Networks. Strategic
Management Journal, 21,
203-215.
Håkansson, H. (Ed.). (1982). International
Marketing and purchasing of
industrial
Goods:
An
Interaction Approach, New
York: John Wiley & Sons
Håkansson,
H.
(1987).
Industrial
Technological Development:
A
Network
Approach.
London: Croom Helm.
Håkansson,
H,
(1989).
Corporate
Technological
Behaviour.
Co-operation and Networks.
London: Routledge
Håkansson, H. & Johanson, J. (1987).
Formal
and
Informal
Cooperation Strategies in
International
Industrial
Networks. In Contractor, F. J.
The IMP Journal
Volume 4, number 1
&
Lorange,
P.
(Eds.).
Cooperative Strategies in
International
Business.
Lexington, Mass.: Lexington
Books.
Håkansson, H. & Johanson, J. (1992). A
Model of Industrial Networks.
In Axelsson, B. & Easton, G.
(Eds.). Industrial Networks: A
New
View
of
Reality,
London: Routledge, 28-34.
Håkansson, H. & Johanson, J. (2001).
Business Network Learning,
Amsterdam: Pergamon.
Håkansson, H. & Snehota, I. (1989). No
Business is an Island,
Scandinavian Journal of
Management, 5 (3), 187200.
Håkansson, H. & Snehota, I, (1995).
Developing Relationships in
Business Networks, London:
International Thomson.
Håkansson, H. & Ford, D. (2002). How
Should Companies Interact?
Journal
of
Business
Research, 55, 133-39.
Håkansson, H. & Waluszewski, A. (2002).
”Managing
Technological
Development. IKEA, the
environment
and
technology”.
London:
Routledge,
Håkansson, H., Havila, V. & Pedersen,
A-C. (1999). Learning in
Networks.
Industrial
Marketing Management, 28,
443-52.
Håkansson, H., Huysman, M. & von
Raesfeld Meijer, A. (2001).
Inter-Organizational
Teaching. In Håkansson, H.
& Johanson, J. (Eds.).
Business Network Learning,
Amsterdam: Pergamon.
Håkansson, H. & Johanson, J. (Eds.).
2001,
Business Network
Learning,
Amsterdam:
Pergamon.
Håkansson, H. & Waluszewski, A. (Eds.).
(2007). “Knowledge and
Innovation in Business and
Industry. The importance of
using
others.”
London:
Routledge.
Håkansson,
101
H.,
Waluszewski,
A.,
Prenkert, F. & Baraldi, E.
(Eds.). (2009). “Use of
Science and Technology in
Business:
Exploring
the
Impact of Using Activity for
Systems, Organizations and
People.
Bingley: Emerald
Group.
Snehota, I. & Waluszewski,
A. (2009). Business in
Networks, Chichester: John
Wiley.
Hallen, L., Johanson, J. & SeyedMohammed,
N.
(1991).
Interfirm
adaptation
in
business
relationships.
Journal of Marketing, 55 (2),
29-37.
Henders, B., (1992). Position in Industrial
Networks.
Marketing
Newsprint in the UK, PhD
Dissertation, Department of
Business
Administration,
University of Uppsala.
Holmen, E. (2001). Notes on a
Conceptualisation
of
Resource-Related
Embeddedness
of
Interorganizational Product
Development.
Ph.Ddissertation, Institute for
Marketing,
University
of
Southern Denmark
Hughes, T. P. (1983). Networks of Power:
Electrification in Western
Society
(1880-1930).
Baltimore: John Hopkins
University Press.
Huemer, L. (1998). Trust in Business
Relations. Economic Logic or
Social Interaction, Umeå,
Borea
Jarillo, C. J. (1988). On Strategic
Networks,
Strategic
Management Journal, 9, 3141.
Johanson, J. & Mattsson, L. G. (1988).
Network
Position
and
Strategic
Action
– An
Analytical
Framework,
Working Paper, University of
Uppsala,
Department
of
Business Studies.
The IMP Journal
Volume 4, number 1
Johanson, J. & Wootz, B. (1986). The
German approach to Europe,
In Turnbull, P. W. & Valla, JP. (Eds.). Strategies for
International
Industrial
Marketing. London: Croom
Helm.
Johnsen, R. & Ford, D. (2008). Exploring
the concept of asymetry: A
typology for the analysis of
customer-supplier
relationships,
Industrial
(4), 471-483.
Kauffman, S. (1995). At Home in the
Universe, Oxford: Oxford
University Press.
Kriesberg, L. (1955). Occupational Control
Amomg Steel Distributors.
American
Journal
of
Sociology, 61 (3), 203-212.
Laage-Hellman, J. (1997). Business
Networks in Japan, SupplierCustomer
Interaction
in
Product
Development,
London: Routledge.
Latour, B. (1984). Science in Action. Milton
Keynes and Cambridge,
Mass.:
Open
University
Press and Harvard University
Press.
Law, J. (1992). Notes on the Theory of the
Actor-Network:
Ordering,
Strategy and Heterogeneity,
Systems Practice 5 (4), 379393.
Lindqvist, S. (1984). Technology on Trial.
The Introduction of Steam
Power
Technology
Into
Sweden,
1715-1736.
Uppsala Studies in History of
Science
1.
Stockholm:
Almqvist
&
Wiksell
International.
Lundgren, A. (1995). Technological
Innovation
and
Network
Evolution,
London:
Routledge
March, J. G. (1988). Decisions and
Organizations.
Oxford:
Blackwell Publishing.
Marglin, S. (2008). The Dismal Science:
How
thinking
Like
an
Economist
Undermines
Community.
Cambridge
102
(MA): Harvard University
Press.
Mattsson, L-G. (1989). Development of
firms in networks: Positions
and investments. In Hallen,
L., Johanson, J., (Eds.).
Advances in International
Marketing, Vol 3 Networks of
Relationships in Internaional
Industrial
Marketing,
Greenwhich, CT: JAI Press,
3,121-139 .
Mouzas, S. & Ford, D. (2009). The
Constitution of Networks,
Industrial
Marketing
Management, 38, 495-503.
Nelson, R. R. & Winter, S. G. (1982). An
Evolutionary
Theory
of
Economic
Change.
Cambridge, Mass.: Belknap
Press
of
the
Harvard
University Press.
Nonaka, I. (1991). The knowledge-creating
company, Harvard Business
Review, 69 (3), 27-38.
Nohira, N. & Eccles, R. G. (Eds.). (1991).
Networks and Organizations:
Structure, Form, and Action.,
Boston,
Mass.:
Harvard
Business School Press.
Palamountain, J. (1955). The Politics of
Distribution,
Harvard
University Press, Cambridge
Pasinetti, L., 1981, Structural Change and
Economic
Growth.
A
Theoretical Essay on the
Dynamics of the Wealth of
nations.
Cambridge:
Cambridge University Press.
Pedersen, A-C., Torvatn, T. & Holmen, E.
(2008). Towards a model for
analysing
supplier
relationships
when
developing a supply network.
IMP Journal 2 (2), 38-58.
Penrose, E. (1959). The Theory of the
Growth of the Firm. Oxford:
Oxford University Press.
Podolny, J. M. (1994). Market uncertainty
and the social character of
economic
exchange,
Administrative
Science
Quarterly, 39, 458-83.
Powell, W. W., Koput, K. W. & SmithDoerr,
L.
(1996).
The IMP Journal
Volume 4, number 1
Interorganizational
Collaboration and the Locus
of Innovation: Networks of
Learning in Biotechnology,
Administrative
Science
Quarterly, March 1996, 41
(1), 116-145.
Richardson,
G.
B.
(1972).
The
Organisation of Industry,
Economic Journal, 82, 883896.
Rosenberg, N. (1994). Exploring the Black
Box:
Technology,
Economics,
History.
Cambridge:
Cambridge
University Press.
Spekman, R., Kamauff, J. Myhr, N. (1998).
An empirical investigation
into
supply
chain
management: a perspective
on
partnerships.
Supply
Chain Management, 3 (2),
53-67.
Torvatn, T. (1996). Productivity in
Industrial Networks – A Case
Study of the Purchasing
Function. Ph.D dissertation,
Department of Industrial
Economics and Technology
Management,
Norwegian
University of Science and
Technology.
Van der Ven, A. H., Polley, D. E., Garud,
R. & Venkatarman, S.
(1999).
The
Innovation
Journay, New York: Oxford
University Press
Von Hippel, E. (1988). The Sources of
Innovation,
New
York:
Waluszewski, A. (1989). Framväxten av
en ny massateknik - en
utvecklingshistoria.
(The
development of a new
mechanical
pulping
technique)(dissertation). Acta
Universitatis
Upsaliensis,
Studiae
Oeconomia
Negotiorum, 31. Almqvist
&Wiksell, Uppsala.
Williamson, O. E. & Ouchi, W. G. (1981).
The
Networks
and
Hierachies
Program
of
Research:
Origins,
Implications and Prospects,
103
in van den Ven AH and
Joyce,
WF
(eds),
Perspectives
on
Organisation Design and
Behavior, New York: John
Wiley.
Wilson, D. T. & Jantrania, S. (1994).
Understanding the value of a
relationship, Asia-Australia
Marketing Journal, 2 (1), 5566
Wilk, R. R. (1996). Economics & Cultures.
Foundation of Economic
Anthropology.
Oxford:
Westview Press.
Wilkinson, I. F. & Young, L. C. (1994).
Business dancing – the
nature and role of interfirm
relations
in
business
strategy.
Asia-Australia
Marketing Journal, 2 (1), 6779.
The IMP Journal
Volume 4, number 2
104
Letter from the editor
In this issue of the IMP journal two articles, both based on large case studies of
technological development, are presented. One describes the development process
taking part between two companies and the other, when the same appears on a
country level, within a whole set of companies. In the first case written by Patrick
Lynch and Thomas O’Toole, we can follow the ups and downs in a process including
two companies fighting for solving severe technical production problems. It is the
buying company that identifies a problem as it is getting a high amount of waste
when using the seller’s product in its production process. This is the starting point for
a complex process where a number of technical issues have to be identified and
solved. In the process there are several critical moments where the conflict intensity
between the two companies is very high and where the process could have ended.
The case indicates that single persons and their personal relationships play an
important role to handle at least some of these critical situations. The case also
illustrates the opposite – that single managers can not make decisions regarding the
relationship in isolation. As an example, the heaviness of the existing relationship in
terms of functional aspects; the importance of the product and how it functions in the
buyer’s production process is more important than what a new purchaser thinks. The
case is an excellent example of describing all those complexities that appear within
business interaction processes going on between single companies and how the
solving of these more or less continuous problems are a condition for the way the
companies can function technically and commercially.
In summary the single relationship has both to function and to produce value in itself
but it has at the same time to fit into the larger picture – the two counterparts have to
fit into each other’s network. It is a long and often cumbersome process when the
relationships are forged together into business networks. But when successful this
process also gives them a much higher long term value which is exemplified in
several ways in the article.
The case in the second article written by Tommy Shih is a description and analysis of
the development of the Taiwanese semiconductor industry during thirty years. It is
giving a very similar but also very complementary picture to the first article in
picturing the development of a whole industry. In the same way as for the company
case above there are also here single persons who play an important part. We have
in this way crucial micro processes. At the same time there are important macro
processes where the large international companies and the Taiwanese government
together are combing resources into something useful and therefore also a powerful
constellation. The single episodes and the combining of this large resource
constellation slowly and stepwise are forming a total network structure. The role of
the industrial policy by the Taiwanese government is never to control the
development. That is impossible! Instead it has to mirror the development and
become an integrated part of this development which includes a much more
complicated combination of resources than any single actor can handle. The
development in Taiwan has to be complementary to the development of some big
actors and also complementary to the total development in some other regions. In
this development a key issue is to take advantage of the resources already existing.
The IMP Journal
Volume 4, number 2
105
The two articles together give an interesting picture of how micro and more macro
processes are related within networks. Single episodes within interaction processes
between single companies are crucial for the development at the same time that the
forming of the total network structure give a frame within which these single
interaction processes have to develop. It gives a picture both of what can be
achieved within relationships as well as how they become a frame within which the
company has to develop.
Enjoy the reading
Håkan Håkansson
The IMP Journal
Volume 4, number 2
106
A Critical Episode Analysis of the Dynamics of
the Interaction Atmosphere in a New Product
Development Relationship
Dr. Patrick Lynch a and Dr. Thomas O‟Toole b
a
Lecturer in Marketing, Department of Management and Organisation, Waterford Institute of Technology, Waterford, Ireland,
e-mail: plynch@wit.ie
b
Head of School of Business, Waterford Institute of Technology, Waterford, Ireland, e-mail: totoole@wit.ie
Abstract
This paper deals with collaborative user involvement in the new product development (NPD) process. The
purpose of this paper is to describe the dynamics of the interaction atmosphere that occurred in a long-term
relationship between a packaging technology supplier and its food manufacturer partner in the development of a
plastic film during the NPD process. Utilizing an interpretative case study approach, the empirical evidence is
based upon interviews, reflective practices, observation and documents. The paper will detail seven critical
interaction episodes and provides valuable insight into the dynamics of the interaction atmosphere that occur in
close collaborative relationships. The findings showed that the interaction atmosphere between the two
companies went through alternating cycles of divergence and convergence to maintain and re-negotiate an
already established belief structure of expected and accepted behaviour.
Keywords: Critical episodes, dynamics of interaction atmosphere, user involvement in new product
development, case study
1. Introduction
Within the new product
development (NPD) literature there has
been an emerging consensus that while
certain success factors pertain to the
development
and
commercialisation
stages, the majority are determined much
earlier in the project‟s life, explicitly in the
early or pre-development stages (Stevens
et al. 1999; Cooper and Kleinschmidt,
1996; Cooper, 1993; Booz, Allen and
Hamilton, 1982). Developing a product
that
delivers
superior
benefits
presupposes an understanding of user
needs and wants, a process that should
ideally be undertaken prior to the
commencement
of
any
actual
development (Stevens et al. 1999;
Cooper, 1988). Without this up-front user
knowledge, significant problems in later
stages of the development process can be
expected including development activities
taking longer than expected, increased
costs, delayed time to market and even
product failure (Cooper and Kleinschmidt,
2000; NICB, 1964). However, user need
information can be costly, complex and
often sticky (von Hippel and Katz, 2002;
von Hippel, 2001). Moreover, in business
markets, conventional research tools are
often of limited utility and since these
markets have relatively small numbers of
users, leading edge companies such as
3M, HILTI and Johnson & Johnson are
increasingly involving their users in the
early stages of new product development
(Lilien et al. 2002). This is done to
enhance a firm‟s competitive advantage
through the provision of innovative and
appealing new product concepts (Stevens
et al. 1999; Cooper and Kleinschmidt,
1996; Madique and Zirger, 1984; Cooper,
The IMP Journal
Volume 4, number 2
1979a; b; NICB, 1964). Others (von Hippel
and Katz, 2002; Tidd et al. 2001; Cooper
and Kleinschmidt, 2000; Voss, 1985)
suggest that user involvement in front-end
activities can also reduce need and
market
uncertainty
by
supplying
manufacturers with a more accurate
assessment of user requirements and
consequently reduce the potential risks of
misfitting buyer needs to a deficient or
poor product idea (Johnsen and Ford,
2000).
However,
despite
the
importance the literature assigns to user
involvement in these predevelopment
activities, there is also ample evidence to
suggest that many firms do not bring their
industrial users into the NPD process (O‟
Toole and Lynch, 2004: Adams et al.
1998; Cooper, 1996) and in most
instances, projects enter the development
phase lacking any clear definition, often as
a result of superficial user involvement in
these early stages (Lynch and O‟ Toole,
2006; Cooper, 1999; Biemans, 1992;
Mahajan and Wind, 1992; Cooper and
Kleinschmidt, 1986). Other evidence
suggests that while most firms consider
user involvement in predevelopment
activities
to
be
beneficial,
they
nevertheless felt that it complicated the
development process and made it more
difficult to control and manage (Olson and
Blake, 2001). An alternative approach to
activating user involvement in industrial
new product development happens when
partners collaborate to solve ongoing
product and process problems in some
part of their dyadic interaction, or indeed
network (Håkansson, 1987; Ford and
Saren, 1996). When this occurs it has the
additional advantage of being a codeveloped solution and thus gets over
some of the user involvement challenges
outlined. Nonetheless the reasons cited
as to why user involvement in the early
stages of NPD remains, for many,
aspirational and difficult also apply to codeveloped new product solutions. Hence
there is still much to understand.
Describing the dynamics of interaction of a
NPD project will depend on the nature and
urgency of the change required of the
parties (Halinen, Salmi and Havila, 1999).
107
In this paper, the problem is amplified in
the atmosphere of the interaction.
The purpose of this paper is
to describe the dynamics of the interaction
atmosphere of a NPD project in a long
term collaborative relationship. This goal
was met in two ways: the authors first
classified the data into a series of critical
episodes which showed the development
of the new product project in decisive
stages. The episodes were labelled on the
basis of major shifts in the relationship
atmosphere
in
either
a
positive
(convergence) or negative (divergence)
direction.
Secondly, each episode is
described by interaction processes that
are either pulling the relationship together
or apart. Describing both the „movement‟
of the project through critical stages and
the related processes achieved the
purpose of the paper. The bulk of the
paper is given over to the case, the
partners‟ story. The case description is in
the processual research tradition and, as
such, is contextual and situational
(Pettigrew, 1997; Sminia, 2009).
Its
richness is evident in the insights into the
dynamics of relationship atmosphere
which was the dominant interaction
process in the case. This case gives full
exposure to the dynamics of interaction
atmosphere which is relatively rare and
unique in the new product development
literature. The interaction atmosphere is
often classified as an intervening variable
moderating the relationship, or empirically
measured by the interdependence of the
parties at a point in time whereas in this
paper it is the core dynamic during the
period under investigation. Few studies
outside the industrial markets-as-networks
approach study underlying interaction
process dynamics (Ring and Van De Ven,
1994; Doz, 1996; Welsh and Wilkinson,
2002) and much less that of interaction
atmosphere especially in the context of a
NPD project.
Given the lack of prior
research into the interaction dynamics of a
new product development project, the
authors committed themselves to an
inductive data gathering approach.
However,
related
empirical
and
conceptual research literature was used to
ensure that no obvious area was missed
The IMP Journal
Volume 4, number 2
in data collection or analysis.
The
relationship atmosphere aspect of the
interaction model provided the contextual
setting for the data (Håkansson, 1982).
The paper addresses the dynamics of the
interaction
atmosphere
in
the
development of a new product.
A
multidisciplinary research literature on
relationship and alliance development was
used to identify processes that impact on
this dynamic.
From the industrial
marketing
and
industrial
network
approach,
relationship
development
models such as those by Ford (1980),
Dwyer, Schurr and Oh (1987) and Ring
and Van De Ven (1994) were used to
identify processes that might be germane
to the case setting. From the alliance
formation and development literature work
that focused on development dynamics
also proved instructive, for example, Zajac
and Olsen (1993), Büchel (2000) and
Ariňo and de la Torre (1998). The insights
from these works kept the data analysis
on track and were used as a comparison
base to sift through the data and
eventually arrive at the final set of
descriptors of each critical episode
mentioned in the case narrative. Little of
the prior literature addressed the
atmosphere dynamics of an established
relationship
with
the
specific
characteristics of the case companies but
key processes such as expectations
(Dwyer, Schurr and Oh, 1987), learning
(Doz, 1996), problem solving (Powell,
Koput and Smith-Doerr, 1994), negotiation
(Dwyer, Schurr and Oh, 1987; Ring and
Van De Ven, 1994), commitment (Ring
and Van De Ven, 1994), personal and role
relationships (Ring and Van De Ven,
1994; Welsh and Wilkinson, 2002) were
central to framing the interaction dynamics
specific to each episode.
The case is organised on the
basis of critical episodes at a dyad level
(Halinen, Salmi and Havila, 1999;
Edvardsson and Strandvik, 2000; Schurr,
2007; Schurr, Hedaa and Geersbro,
2008). For an episode to be classified as
critical in this case, it had to be a major
shift in the relationship atmosphere. This
shift had a cycle from positive to negative
(Daft and Weick, 1984; Büchel, 2000,
Schurr, 2007); convergence where a
108
direction and mutual expectation was
perceived to exist then the atmosphere
became positive and reinforcing, to
divergence where the dynamics of the
interaction were negative, crises ridden
and shared systems of understanding
including personal relationships were
under severe strain. The authors attempt
to describe the dynamics of the interaction
atmosphere during the NPD process
across the cycle of convergence and
divergence and in doing so narrate unique
insights into the process.
The rest of the paper is
organised as follows. In the next section,
the methodology employed in this
research and the data analysis is
described. The research site and context
of the new product development is then
provided. The main section of the paper
presents the seven critical interaction
episodes between the collaborating
partners.
Case analysis is then
addressed before a conclusion to the
paper is drawn.
2. Research Design
2.1 Sample and Method
This paper reports on an indepth case study of a long-term
relationship
between
a
packaging
technology company and its food
manufacturer partner in the development
of a plastic film during the early stages of
the NPD project. Data was gathered in
real time, over a six-month period,
between October 2005 and April 2006.
Since this research was occurring in real
time, care had to be taken not to influence
the ongoing interaction process between
the two companies. Indeed, as observed
by Doz (1996), with “real-time process
studies…it is extremely difficult to be sure
not to influence ongoing processes and
still maintain a legitimate presence in the
field insofar as mangers would quickly be
tempted to seek advice from the
researcher and ask the researcher to
intervene in the process, as a quid pro
quo for allowing access” (58). To ensure
the prevention of researcher bias on the
case, from the outset, a boundary was
firmly established and reasons explicitly
stated to informants. In fairness to the
The IMP Journal
Volume 4, number 2
case companies that boundary was not
breached
Data was collected from four
main sources: interviews, reflective
practices, documents and observation.
The variety of data collection techniques
allowed for greater possibility of
discrepancies or anomalies to be noted in
research data, and should compensate for
any limitations in individual collection
techniques (Eisenhardt, 1989). Multiple
sources also counteract potential validity
concerns
in
relation
to
theory
development, because multiple lines of
enquiry converge towards a particular
proposition or conclusion (Yin, 2003).
All in all, 15 interviews were
conducted with the key members involved
in the development project. Of these 9
were personal in-depth interviews and 6
were telephone interviews. The personal
interviews ranged in length from 1 hour to
3 hours each. The telephone interviews
lasted about 10 to 30 minutes, with the
shortest of these aimed at collaborating
existing
information
obtained
via
interviews, documents etc or at alleviating
confusion over some point. The telephone
interviews were mainly made after
personal interviews had been carried out.
An interview guide was made before each
interview. Nevertheless, the interviews
took an unstructured format. The
individuals that were being interviewed
were
highly
educated,
competent
executives
and
understood
the
cooperation process between their
company and their partner. Thus, they
talked freely, only to be interrupted by the
researcher on some follow-up issue. The
role the interviewer played was only that
of a guide through the interviewees‟
stories. For instance when it was felt that
a topic was exhausted, the researchers
would introduce a new topic, based on the
interview guide, or some issue that may
have materialised in the interview. Thus,
the interviews had a very relaxed feel to
them, even conversational, and rich
insightful data about the interaction
processes involved in their cooperation
emerged.
The
reflective
practice
involved the researchers analysing data
gathered from in-depth interviews and
109
documents, and putting it into a story, and
then presenting that story to the
respondents. Gaps of understanding
about what was going on, where evident
in the narrative, were subsequently filled
in by the respondent. In one way, the
participants became, in part, the analyst of
the data and in the act of writing the
narrative and listening to the respondent,
the researchers were able to immerse
themselves in their experiences and get at
what is going on (Denzin, 2001). The
routine that emerged in conducting
reflective interviews was that at the start of
a scheduled in-depth interview the
researchers would present the narrative
about the data collected to that point and
discussion would ensue, errors would be
highlighted and understanding would
emerge on issues. When the discussion
was exhausted, the interview would then
revert back to gathering and uncovering
the next episode of the story. At the next
meeting, the same process would occur
as just outlined, the researchers would
show the updated narrative, including new
data, and the respondent would be asked
to fill in the blanks. Once understanding
was arrived at, the next episode would be
investigated.
Documentary data, printed
as well as electronic, was collected from
various sources such as annual reports,
academic
databases,
commissioned
company reports and product design
specifications. In total 25 documents were
used. In most instances, documents were
studied in preparation for interviews.
Finally, secondary data such as
government reports were assessed to
understand the specific nature of the
industry context and policy during the
period of the study.
Observations influenced and
contributed to casework to the extent that
a picture paints a thousand words. It
facilitated the researchers in observing the
products and production process “in use”,
and so, provided a greater contextual
understanding of the product concepts
and development issues under research.
For instance, the researchers were
brought onto the factory floor and were
able to see how the packaging was made,
what a packaging sealing issue was and
The IMP Journal
Volume 4, number 2
110
what slippage means. The observation
allowed the researchers to gain insight
into knowledge that was hard to
communicate
without
actually
experiencing it, and in this fashion,
fertilised the researchers‟ understanding
of concepts that the participants were
talking about.
coding process allowed the researcher to
retrieve information when needed, thus
facilitating the refining and regrouping of
knowledge units through the linking of
ideas
and
sources,
identifying
contradictions
in
arguments
and
comparing dissimilarities (di Gregorio,
2000).
2.2 Data Analysis
In order to categorise and
identify the processes that occurred in the
interaction atmosphere between the
manufacturer and user in the early stages
of new product development, this research
builds upon the analytical ideas of
Pettigrew (1997), Fox-Wolfgramm (1997),
and Lincoln and Cuba (1985) which
incorporates a constantly iterating cycle of
deduction and inductive category coding
and
pattern
recognition
across
categorised
phenomenon.
In
this
analytical model, social phenomena were
continuously being compared across
categories so that new dimensions could
be discovered. The identification of the
dynamics of interaction atmosphere began
with the analysis of initial observations.
However, these initial observations
underwent
continuous
refinement
throughout the data collection and
analysis process, because data collection,
analysis and interpretation were occurring
simultaneously, and so continuously fed
back into the process of category coding.
Category codes and their content were
continuously compared with previous
events and so new insights were
discovered (Lincoln and Cuba, 1985).
Finally, it was possible to group the data
into critical episodes and describe the
process dynamics of the interaction
atmosphere at each of these points.
Nvivo, qualitative analysis
software, was utilised to manage the
process of coding, retrieving, memoing
and data linking. In the domain of coding,
text chunks of transcribed data were
unitised and categorised by giving them a
code. The code that was assigned
resulted from keywords or labels that were
used by the research subjects to describe
the phenomenon under discussion. When
unitising, each unit was coded according
to the source, site, date, label etc. This
2.3 Research Site and Context
Starting
in
1985
and
continuing today, Packfex and Farmfresh
collaborated
closely
in
developing
packaging film for Farmfresh‟s food
products. The relationship began in 1985,
when both companies co-developed an
innovative and revolutionary film (Chubb
X1) that actually became the cooking
vessel, in that the raw meat entered the
packaging and is cooked. Indeed, the
innovativeness of the packaging is still
evident today as only a few companies
have been able to produce a similar film.
Presently Chubb X1 is Packfex‟s biggest
selling product and as Packfex is currently
becoming more global, new applications
for the plastic are emerging such as
plastic covering for army artillery and
sales of the film are expected to increase
exponentially. Over the past twenty-one
years, the relationship between Packfex
and Farmfresh has become close and can
be characterised as being highly
integrative, with high levels of trust,
commitment,
and
cooperation.
Communication
between
the
two
companies is high with a considerable
amount of information sharing. In addition,
there are a number of close professional
and interpersonal relationships between
individuals in both companies.
However, in late October
2005, in two of Farmfresh‟s production
facilities,
Chubb
X1
was
uncharacteristically causing significant
wastage, approximately 33 per cent. To
put this into perspective, in production
terms, the golden rule is that production
efficiency running in to double figures is
totally unacceptable. The problem was
communicated when one of the Farmfresh
plant directors contacted Packfex to
request that someone come down to their
plant and rectify the situation as it was
costing a significant amount of money.
The IMP Journal
Volume 4, number 2
Assurances were given that the problem
would be dealt with. However this did not
materialise and Farmfresh got extremely
upset at the lack of response from
Packfex in solving the wastage issue.
Tension between the two companies was
evident when the Engineering Manager
(EM) contacted Packfex and very
aggressively demanded that they come
down to the plant at 6am the following
morning to witness the wastage first hand
for themselves and a meeting with
Farmfresh‟s management board was to be
scheduled after the production run to
discuss how this quality issue was going
to be solved.
I wasn’t happy overall…there
were a lot of questions being
asked [about Packfex] like
what’s happening now. There
could have been a quicker
reaction time (Farmfresh: EM).
You have got to show quite a
lot of commitment at that
particular point to handle the
situation. When Farmfresh
rang up and said that we are
going to run at 6 o’clock
tomorrow morning and as
unreasonable as it sounds and
in relation to what other
appointment or meeting you
have got, I had to jump in and
bite the bullet, because they
would not be interested in
talking to us if we didn’t turn
up. If we didn’t turn up,
Farmfresh would have thought
that there was no commitment
from us to solve the issue we
were having (Packfex: TD).
When Packfex‟s Technical
Director (TD) and Account Manager went
to the site the next morning at the agreed
time, the first thing that both of them
noticed was that none of Farmfresh‟s
management team were present. In fact,
they did not arrive until 9am, three hours
later. For Packfex, that was quite an
amazing statement. They felt that
Farmfresh were trying to show them that
they were the dominant actor in this
111
relationship. However, the Packfex
representatives did not mind being left on
the factory floor with Farmfresh‟s
operatives because it allowed them to
observe the production process in real
time. It also allowed them to interact with
the operatives and to get first hand
illustrations and explanations of the
problems that were occurring in the
production process.
When
Farmfresh‟s
management came in at 9am a conflict
immediately ensued between EM and TD,
in that the former was blaming the film for
the high wastage and the latter was
blaming a catalogue of mechanical errors
with the packaging machine.
We stated to them that the
machine had a catalogue of
mechanical problems. At this
particular point EM is angry
because here am I going
through the issues, the
problems with his machine. He
had worked on this machine
for the last 7 years trying to
get wastage down and here
am I jumping all over his toes.
And the conflict at this
particular point is intense. The
anger
and
the
facial
expressions in his face as I
was going through this is really
visible… The tension at that
moment was very high. We
were standing on his territory,
discussing his machines in
such a manner that he is going
to find it extremely aggressive.
The only way to describe that
interaction is that the guy took
a pasting from me (Packfex:
TD).
At the formal meeting
between the two companies Packfex
reiterated the technical faults that they had
discovered with the machine and stated
that no alterations to the film would be
made without these mechanical issues
being fixed. Within Farmfresh, the
production and engineering departments
were
not
happy
with
Packfex‟s
The IMP Journal
Volume 4, number 2
assessment of their machines because it
meant that production would have to be
stopped and there was now an internal
conflict
from
other
departments
demanding to know why these technical
issues were not resolved before.
However, EM was claming that it had to
be Packfex‟s film since other packaging
ran on the machine with no faults.
However, despite Farmfresh‟s arguments
that the film was to blame for the high
wastage, Packfex was adamant that the
mechanical issues had to be dealt with
before any product alterations were
contemplated. The meeting ended with an
agreement that the technical faults would
be rectified.
When we went into the
meeting, we stated that before
any changes are to be made
to the product, they had to fix
the machine and we gave
them 5 faults that needed to
be corrected before we would
get involved. EM did not agree
with us and had stated that the
machine manufacturer had
been here last week and that
the machine was in pristine
condition. They were telling us
that our film had faults and
that when they put on our roll
they got 33% wastage,
however, whenever they put
on other rolls they were getting
none. First of all alarm bells
were ringing. We knew we
were not been told the truth.
We knew the wastage was
high on the other rolls as well.
So we told them that there
were 5 faults with the machine
and that they needed to be
corrected before we would
even contemplate adjustments
to the product. In fairness, to
EM, he agreed and the
production line was shut down
until the faults were corrected
(Packfex: TD).
To resolve some of the
tension between the companies TD and
EM worked out a methodology to fix the
112
technical faults in the fastest time possible
so as not to delay production.
TD stayed with us when we
were trying to figure out how to
solve the problem and gave us
ideas on how to fix the
machine. I found him very
good…We actually had to do a
full diagnostic on the machine
and it meant that we had to
spend a lot of money changing
parts and there was the time,
the labour and production had
to be stopped (Farmfresh:
EM).
Also at this time, Farmfresh‟s
corporate buyer with whom Packfex had a
very close relationship with for a number
of years retired and the new corporate
buyer began rationalising his supplier
base. For the new buyer Packfex was a
small packaging supplier with a small
spend (number 10) and was informed that
due to the significant ongoing quality
issues with their film it was been de-listed.
Packfex did not argue with the new buyer
and told him that it understood his
rationale and that its services were always
available to him if he so needed.
He was showing his metal, it is
all about pecking order and we
knew that if he was going to
upset the apple cart with us,
he was probably going to do it
with all the rest of the
suppliers. What he was doing
was very controversial and
most companies would adopt
a stance that I will teach him
and so a conflict would ensue.
We didn’t see it like that at all.
We saw it as an opportunity to
move up the supplier list, even
though we were technically
de-listed. We knew that our
product was very unique to
Farmfresh and that they would
have a hard job replacing
us…we had to be patient. He
couldn’t de-list us, but he had
not realised that yet (Packfex:
TD).
The IMP Journal
Volume 4, number 2
The
buyer
found that
because Chubb X1 is such a unique and
complicated product that he was unable to
locate an alternative supplier to Packfex.
Indeed, he found that he was dependent
on Packfex. Moreover, the buyer found
significant
resistance
from
other
departments such as marketing within
Farmfresh to the de-listing decision. Thus,
while Packfex was technically de-listed
they were still supplying Farmfresh with
the Chubb X1. This also meant that the
quality issues at Farmfresh had to be
resolved. As a result of the corrections to
the machine the wastage level on
Packfex‟s film dropped from 33% to 15%.
Packfex was also informed that wastage
on competitor‟s films had also significantly
been reduced and was now running at
approximately 9%. Despite the ongoing
wastage issue Farmfresh were ecstatic.
They had a significant drop in wastage,
not only on Packfex‟s film, but also on
other suppliers. At this point, Packfex
knew there was a problem with the film.
The waste figure was very high at 15%
and the fault was not with Farmfresh‟s
machine.
In the beginning there was a
machine issue and a film
issue. When the machine
issue got sorted, the film was
still causing unacceptable
levels of wastage. So for a
while, there was a grey area
there when there was not a
definitive problem with the
film…however
when
the
machine was fixed we knew
there was something wrong
with the film (Farmfresh: EM).
Packfex
travelled
to
Farmfresh‟s factory floor to witness the
wastage. What they found was that the
packaging film, filled, it sealed, it clipped, it
cooked and it was during the chilling
stage, which is at the very end of the
process, the fault appeared. The film
actually split. Since the film was in
Farmfresh‟s cooking process for four days
when the packaging split, it meant that
besides the high wastage and the extra
113
production time needed to compensate, it
also meant serious time delays in
Farmfresh‟s delivery of its product to
customers.
So the second time we went
back. EM has now fixed the
machine and got less than 9%
wastage with other films on
the machine and we were still
getting 15-20%. EM put the
roll on and we saw it run. It
was quite clear now, black and
white that it was now a film
issue. Now the situation
changed the other way around
and now EM was saying that
there is something definitely
wrong with the film and that he
had accepted everything I said
about his machine and so
what was going to be done
about the film.
Now this
wasn’t a conflictual meeting at
all, because the EM is now a
hero in Farmfresh, because
following the conversation with
me over the issues with the
machine, after 10 years of
waste, he had got the waste
level down on his machines to
between 8-10%. If you are
turning several million euros,
you are talking about serious
cost savings for the company.
EM got an awful lot of respect
from his colleagues (Packfex:
TD).
Although a professional and
personal relationship had existed prior to
the problems with the film, the
consequential effect of both individuals
working together to fix the machine was
that the personal bond between the EM
and TD had deepened. Since both had
similar
engineering
backgrounds,
interacting with each other was relatively
fluid and a relational trust had formed
between the two. Both TD and EM agreed
that the film had to be altered and a codevelopment meeting was scheduled.
Also at this time the buyer, having realised
that it would be difficult to de-list Packfex,
offered them a substantial increase in
The IMP Journal
Volume 4, number 2
orders and Packfex moved up the supply
chain to the number five supplier.
3. Critical Interaction Episodes
The
following
project
narrative traces the dynamics of the
interaction atmosphere inherent in the
product
development
collaboration
between Packfex and Farmfresh in the
development of a new packaging film. As
is clearly evident from the above
description,
an
established
close
relationship already exists prior to project
initiation. Table 1 presents a chronological
map of the interaction episodes between
the interactants, detailing both actions and
perceptions.
3.1 Episode 1. Convergence: Shared
Expectations
Both companies entered the
first development meeting with the
expectation that the adjustment to Chubb
X1 would be rectified within a relatively
short period of time, approximately one
week. The urgency to rectify the problem
stemmed from two main rationale. First,
Packfex faced a potential consequential
loss claim for the wastage which can
amount to ten times the cost of
manufacturing the film and second,
Farmfresh were in a situation where they
had to use the faulty film until the problem
could be rectified and the film up-lifted and
replaced. For Farmfresh, this meant a
period of high wastage, delayed
production, and unsatisfied customers and
they were willing to accept this while the
problem was been corrected.
Part of the problem was the
film had to be fixed, but we still
had to produce. It meant that
we had to work more to reach
our quotas. We really had to
have great patience. The
wastage was still going on
while the film was been
fixed…but the ultimate goal at
the end of the day is that the
114
problem is fixed (Farmfresh:
EM).
The meeting evolved into a brainstorming
session in which both actors discussed, in
detail, the problem and potential solutions.
As both parties were familiar with each
other, the level of engagement from both
sides to solving the issue was said to be
very high and intense.
Both parties
openly communicate and shared ideas. As
a show of commitment, Packfex agreed to
take corrective action and sent a team to
Farmfresh to go through their stock of
Chubb X1 film and any product that was
deemed not fit for process was removed.
3.2 Episode 2. Divergence: Problem
Ambiguity and
Expectations Suffer
Expectations suffered. The
initial goal of a quick product improvement
was diminished as the problem with the
film was not easily identifiable.
We thought perhaps a week…
What we did was that we
broke down the film and did a
technical analysis on it. The
problem is that there are about
5 million parts to a film and
locating what is specifically
wrong is very difficult…but
when we did the analysis on
the film we knew that this was
not a simple problem and we
had to confirm to Farmfresh
that to get a resolution would
take a minimum of 4 weeks
(Packfex: TD).
The tension between the two
companies was very high. Farmfresh had
an expectation that the fault with the film
would be resolved in a quick period of
time and from their perspective Packfex
had let them down. Farmfresh were
extremely frustrated as it meant that the
high wastage and the costs associated
with it would continue into the near future.
The IMP Journal
Volume 4, number 2
115
Table 1. Interaction Episodes
= Divergence
Legend
Packfex
= Convergence
Episode 1. Convergence: Expectations of Cooperative
Development
Assumption of a relatively easy task
Short time frame expected
Potential consequential loss claim for the wastage
See Farmfresh as partner, close ties
Level of engagement high
Intensive interaction
Bilateral communication of ideas
= Exit threat
Episode 2. Divergence: Problem Ambiguity and Expectations
Suffer
Expectations suffer
Realises that interaction is going to be more difficult than
previously expected
Concern over consequential loss
Reassurances given that problem will be resolved
Episode 3. Convergence: Problem Identified
Expectation high that the problem is resolved
Engage in intensive interaction and communication
Joint decision making is undertaken
Likelihood
of conflict
Threshold of
evaluation
Problem not easily identified
Adjustments
to old film
made
Agree to codevelop – film
uplifted
Joint
Analysis
Problem identified
Farmfresh
Nov 2005
Dec 2005
Jan 2006
Assumption of a relatively easy task
Short time frame expected
Nervous over high wastage – patience
See Packfex as partner, close relationship
Level of engagement is high
Meaningful communication
Perceived up-lift as an illustration of commitment
Expectations suffer
Perceive Packfex has let them down
Frustrated at lack of progress
Expectation high that the problem will be resolved
Engage in intensive interaction and communication
Joint decision making on all activities
Kept up-to-date of all development
The IMP Journal
Volume 4, number 2
116
Table 1. Interaction Episodes (Cont.)
= Divergence
Legend
Packfex
= Convergence
Episode 4. Divergence: Expectations Suffer over Problem
Realises the difficulty involved
Empathizes with Farmfresh’s situation
Concern over lack of progress
Concern over the threat of consequential loss
Time pressure is critical
= Exit threat
Episode 5. Convergence: Solution found
Interaction is extensive, brainstorming sessions
Intensive communication
Past history facilitates efficiency
No relational ambiguity
Sense of relief over threat of consequential loss being
removed
Sense of closeness with Farmfresh
Likelihood
of conflict
Threshold of
evaluation
Problem larger than
expected: new supplier
and ingredient needed
Supplier product
analysed
Brainstorming
session
Supplier refuses to
help
Farmfresh
Solution found:
concept developed
Jan 2006
Feb 2006
Frustrated at the lack of progress
Becomes worried at the degree of difficulty
Concern at the mounting cost
Questioned whether Packfex appreciated the cost endured
Felt that Packfex’s reaction times could be quicker
Is appreciative of Packfex’s dilemma
Discontentment communicated
Engaged in brainstorming sessions
Intensive interaction and communication
Sense of relief and euphoria
Expect cost and wastage to decrease
Sense of relational closeness with Packfex
Concept developed
March 2006
The IMP Journal
Volume 4, number 2
117
Table 1. Interaction Episodes (Cont.)
= Divergence
Legend
Packfex
= Convergence
Episode 6. Divergence: Expectations Suffer. Tensions reach a
new high
Expectations suffer
Concerned over tensions
Communication is perceived as aggressive
Concerned about the future of the relationship
Realised confidence most be restored in EM and PM
Reliance on interpersonal and professional relationship
Concerns over interpersonal relationships
Likelihood
of conflict
= Exit threat
Episode 7. Convergence: Solution Found
Conflict must be resolved
Reliance on interpersonal relationship with EM
Communication perceived as intense
Perceive high commitment from Farmfresh to resolve concept problem
Concerns over interpersonal relationship resolved
Expected continuity of the individual and inter-organisational relationship
Threat of exit
Threshold of
evaluation
Engineering and
production attempt to
de-list
Meeting
scheduled
Development team
reconvening
Chubb X2 fails.
Wastage high
Solution found
Farmfresh
March 2006
April 2006
Expectations suffer
Perceived that Packfex and TD let them down
PM felt that Packfex should be de-listed
Psychological contract for cooperation decreases
Production and engineering stop communication with Packfex
Internal conflict emerges over de-listing
Shared commitment
Quality of communication increases
Limited conflict at both inter-organisational and interpersonal level
Resolution of conflict is seen as part of doing business
Compatibility exists
Expected continuity of the individual and inter-organisational relationship
The IMP Journal
Volume 4, number 2
Farmfresh were saying to
us that had they known
that the problem could not
be solved quickly; that they
would have went to
another supplier. If the
person’s expectations are
not met, conflict will occur
and we had to try and
reduce the conflict by
reassuring them that their
expectations will be met to
a certain degree but that
they might not be what
they initially thought - at
that point we had to
change their perception of
what
was
feasible
(Packfex: TD).
To alleviate some of the
conflict Packfex did make some
adjustments to the film that would
reduce the wastage slightly, however
the film could not be up-lifted as there
was nothing to replace it with. This
meant that Farmfresh had to keep
using the existing film while the
solution
was
being
developed.
Nevertheless, pressure was been
exerted
on
Packfex
from
all
departments in Farmfresh to fix the
problem. At this stage, the tension and
the conflict between both companies
was extremely dense.
There appeared to be no
progress, we were still
getting the same film, the
same wastage. There just
wasn’t
any
progress.
Whatever way you look at
it. I don’t think Packfex
came across this problem
before and so there was a
good bit of confusion. The
process for figuring out
what was wrong with the
film was a bit all over the
place. I don’t think that
they were following any
real
guidelines
in
identifying the problem
(Farmfresh: EM).
Moreover,
from
Packfex‟s perspective the threat of
consequential loss was ever present.
3.3 Episode 3. Convergence:
Problem Identified
Despite
the
tension
between the two companies, both
needed the problem with the film to be
rectified. The conflictual tension was
reduced
through
continuous
engagement and a joint analysis of the
problem
was
undertaken.
The
dialogue between the two companies
and the level of intertwinement among
various departments was very regular
and frequent, to the point where they
were jointly making decisions in
relation to the course of action been
taken. They uncovered that a supplier
was guaranteeing its raw material
would
meet
particular
sealing
requirements but analysis showed this
not to be the case.
It worked out that one of
our suppliers changed an
ingredient in the formula
without actually informing
us. Once we got into it we
identified
the
problem
within 24 hours (Packfex:
TD).
Having identified the
problem, the supplier was contacted
and insurance was given that
requirements would be met. Both
Packfex and Farmfresh felt that within
a relatively short period of time, the
film would be rectified and the
production issues at Farmfresh
eliminated.
3.4 Episode 4. Divergence:
Expectations suffer over Problem
However, it was soon
realised that the problem was far
greater than was anticipated. First, the
supplier refused to admit that it made
any alterations to the sealing layer and
refused to engage in any discussion
on the matter, mainly due to the
possibility of consequential loss. This
118
The IMP Journal
Volume 4, number 2
dramatically delayed the progress as it
meant that Packfex had to analyse
their supplier‟s product to identify
which ingredient was altered.
If they had to turn around
to us and say the fault was
actually theirs we could
have been able to work
with them and Farmfresh
to ensure that the film
worked. But instead they
opted not to inform us of
any change. Which ended
in the fact that we had to
make a new product, when
probably
slight
adjustments would have
only been needed…and
we had to de-list them
(Packfex: TD).
Having identified the
ingredient, it was then discovered that
new
regulations
prohibited
the
ingredient from being utilised in a
sealing layer and hence, the reason
why the supplier had removed the
ingredient from its product in the first
place. However, this finding caused a
significant problem in that an
alternative supplier and alternative
ingredient to the formula had to be
located. What is more, tensions began
to mount significantly between the two
companies due to the ever increasing
catalogue of setbacks that were been
experienced on the project. Farmfresh
were getting frustrated at what
appeared to be a lack of progress.
They felt that the film should have
been rectified by now and that perhaps
Packfex did not fully appreciate the
cost to Farmfresh, which was clearly
expressed to them through several
communication mediums at that point.
While consequential loss was not
explicitly
threatened,
it
was
nevertheless
implicitly
implied.
Farmfresh were increasingly becoming
worried that the problem could not be
solved.
At this particular point, we
had to solve the problem;
we had to solve it today.
There just wasn’t any
question of that at all, we
had to solve it. Time was
now
the
constraint.
Farmfresh could not stop
producing and so our back
was up against the wall.
The clock is ticking and the
sense of urgency becomes
absolutely critical from the
point of view that the
tensions between the two
companies is intensifying
and about to boil into a full
conflict (Packfex: TD).
It has to be stressed that
Farmfresh understood and empathized
with the difficulties Packfex were
experiencing, they still needed the
problem to be solved as fast as
possible. For Packfex, the threat of
consequential loss was real.
There could have been
quicker reaction time on
some issues. This was a
big headache for us,
especially production…but
how do you put a time limit
on something that has
never happened before. It
was new to both of us
(Farmfresh: EM).
Solution Found
A concept solution was
developed through a series of formal
and informal brainstorming sessions.
At the formal sessions, all suggestions
were documented, discussed, and
eliminated if not appropriate. Informal
sessions occurred over the phone or
on Farmfresh‟s factory floor. Three
main individuals involved in this
process were TD, EM and Framfresh‟s
production manager (PM). The level of
interaction between these people was
extensive and intensive.
To get to the point of
developing
a
concept
solution, there was a lot of
119
The IMP Journal
Volume 4, number 2
brainstorming
between
Farmfresh and ourselves. I
had
so
many
conversations one to one
and over the telephone.
Most of the time it was
between myself and the
production manager or the
engineering manager…In
relation
to
Farmfresh,
brainstorming
didn’t
necessarily have to take
place
around
a
table…often we were on
the
factory
floor
hammering out ideas…that
is
brainstorming
and
sometimes
the
most
valuable ideas comes from
that informality. These
brainstorming sessions are
invaluable because you
are filtering out what
concepts or ideas you are
going to use (Packfex:
TD).
Because there was an
existing working relationship and
personal relationships between these
individuals, they understood how the
other
operated
and
so
the
communication of ideas, what they felt
did not work, or did, was clearly
expressed, and taken on board as a
valid input. In essence, because of
their relational history, there were no
relational ambiguities surrounding the
issue of trying to solve the problem. All
ideas in relation to changing the
sealing
formula
were
clearly
communicated and decisions were
jointly made. Indeed, it was stated by
both companies that their interactions
were characterized by a sense of
honesty and openness.
Our
relationship
with
Packfex
is
good.
Interacting with TD is
good. There is honesty
there. We know them and
we work well together
(Farmfresh: EM).
As a result, an innovative
solution to the problem was devised
and tested for requirements. It was felt
by both parties that the problem had
been solved. Indeed, there was a
sense of euphoria amongst the people
involved and the relationship between
the two companies was said to get
even tighter. Moreover, there was a
sense of relief for both Packfex and
Farmfresh. The threat of consequential
loss was removed for Packfex and for
Farmfresh, the high cost and wastage
would be reduced.
What we did in Farmfresh
was very controversial. We
changed the sealing layer
completely away from
traditional practice (xx) to
what it is now - yy. What
happened was that we
created a brand new
product. So what we did
was that we took that old
product apart and rebuilt it
as if it were the ideal
utopian
product
for
Farmfresh. Rather than
using xx as a sealing layer
we used yy. The idea for
using yy instead of xx
came as a result of new
product that we were
developing
with
a
European
manufacturer
and in the course of testing
suitable ingredients we
discovered that one of the
properties of yy was that it
was a good sealing layer
(Packfex: TD).
At this stage it would
appear the NPD project was
concluded. However, in this particular
instance, this is not the case. The film
solution, Chubb X2, devised by both
companies was developed but when
the film ran in Farmfresh‟s cooking
process the wastage was higher than
ever before, approximately 50%.
Although Chubb X2 had resolved the
sealing issue, another problem
materialised as a result of the changes
120
The IMP Journal
Volume 4, number 2
to the sealing formula. In essence, the
new film was thicker and not as
smooth as the original and as a
consequence, the film moved slower
through the packaging machine
causing it to be misplaced.
What happened was that
when we sorted out the
sealing issue, because we
changed the formula, we
found
a
completely
different
problem
materialised.
What
happened was we sorted
out the sealing issue but
now there was a problem
with the slip … they [the
film] were out of place –the
product was misplaced- it
was not where it should
have been. We did alter
the feel of their product
(Packfex: TD).
Thus the narrative continues of the
development process of Chubb X3.
3.6 Episode 6. Divergence:
Expectations
suffered. Tensions
reach a new high
When Chubb X2 failed,
the tension between the two
companies and the key individuals TD,
EM and PM, escalated to an extreme
intensity. The communication between
the two companies was said to be very
aggressive and threatening. As far as
EM and PM were concerned, Packfex
and TD had let them down, and worse,
had made them look bad within
Farmfresh. Indeed, both EM and PM
were under fierce internal pressure for
the failure of Chubb X2. The animosity
between the key individuals was
clearly evident when EM and PM
would not even communicate to TD in
relation to rectifying the new problem
and as far as they was concerned
Packfex should be de-listed.
It was like we had made no
progress at all. There was
a lot of patience on our
part when the problem with
the
film
was
been
sorted…because of the
fault we had a lot of
waste…there just wasn’t
any progress and that is
the way we looked at it. It
really wasn’t what we
expected (Farmfresh: EM).
However, internal conflict
broke out between other departments
in Farmfresh, when engineering and
production tried to de-list Packfex. As
was mentioned earlier, Packfex had
multiple
working
and
personal
relationships in different areas of
Farmfresh and now they relied upon
those multiple relationships to stem
the conflict from engineering. It has to
be
stressed
that
although
communication broke down between
production
and
Packfex,
communication was still ongoing
between
Packfex
and
other
departments
within
Farmfresh.
However, TD knew that if he did not
get EM and PM involved, the project
would fail. They were key individuals.
Sometimes you have a
situation where one person
in their company is your
provocateur, he is the one
with the issue. So you
have to surround him with
people within Farmfresh
that like your product and
your company (Packfex:
TD).
In relation to resolving
the conflict TD eventually contacted
EM with whom he had a very close
bond. EM reluctantly took the call and
stated that while he would like to keep
working with TD he was under
pressure from other factions in
Farmfresh not to have dealings with
Packfex anymore.
EM was telling me that
there was no reason for
me to be phoning him, we
had our chance, we would
121
The IMP Journal
Volume 4, number 2
be de-listed and that I was
wasting his time. I told him
that innovation is an
iterative process and that
this was an engineering
situation. I was trying to
relate the whole situation
back to his experiences so
I asked him how many
times
had
he
been
convinced that welding a
bracket to a machine
would fix a fault, you weld
the bracket and the
machine is still faulty and
all the time it was only a
loose screw and you didn’t
see it. I asked him how
many times had that
occurred and if it hadn’t I
will put down the phone
but if it had, then he would
understand where I was
coming from (Packfex:
TD).
I understood where he was
coming from. I understood
that it is an iterative
process, that it is a stepby-step
process.
I
understood
that
(Farmfresh: EM).
Having convinced EM to
provide assistance a meeting was
scheduled
between
the
two
companies.
The
meeting
was
extremely confrontational, yet the level
of communication was also high.
Farmfresh were annoyed with the lack
of progress and placed the majority of
the blame on Packfex. Since both
parties knew each other very well
frustrations and accusations were
openly communicated and dealt with.
In essence both companies managed
to argue their way to an understanding
that the project should continue.
In relation to the failure of
Chubb X2, it was really
hard because we had to go
back to Farmfresh and pick
them up of the floor and
there was also a loss of
faith in us. At this particular
point we are trying to
positively reinforce the
relationship, no negativity
at all (Packfex: TD).
We had to be positive the
whole time in that meeting.
It was not a matter of
saying how are we going
to come up with a solution;
it is a matter of when we fix
this. We were constantly
reminding them of the
benefits that they were
going to get…All the time
we are trying to give them
positive reassurance that
we are going to get to the
goal, it’s just taking a little
bit longer than anticipated
and yet this is perfectly
normal because innovation
is an iterative process. Is it
painful? Yes. Is it going to
get an easier? No. It is
probably going to get
harder (Packfex: TD).
The meeting ended with
both sides in agreement that the
development team reconvene to solve
the issue. Although at an interorganisational level the conflict had
been resolved, at an individual level
there was still considerable animosity
between TD, and EM and PM and that
tension had to be resolved to ensure
project success. After the meeting, all
three individuals had a private
meeting. At first, there was little or no
communication from either EM or the
PM and it really was TD who did all the
talking. Relying on their long-standing
interpersonal
and
professional
relationship, TD tried to convince them
that he understood their frustration and
that the issue would be resolved. In
essence,
the individuals talked
themselves into a resolution and
although tension between them had
been abated, it had not been fully
resolved at that particular time.
122
The IMP Journal
Volume 4, number 2
Solution Found
Like before, the level of
interaction
and
communication
between the two companies was
intense, especially between TD, PM
and EM. Within a relatively short
period of time the problem with the film
was located and a simple alteration to
the sealing formula was devised that
would rectify the current problem. Both
parties were confident that the Chubb
X3 film would work. In addition to
solving the misplaced product problem
the development team also came up
with an additional innovative alteration
to the film that would not only reduce
waste but speed up the rate at which
the film went through Farmfresh‟s
production
process
significantly
increasing efficiency. The alteration
also dramatically sped up Packfex‟s
actual production of the film and stock
turn
around
thus
eliminating
Farmfresh‟s need to hold vast
quantities of stock.
3.8 Epilogue
Chubb X3 film was
developed and far exceeded the
expectation of both companies running
at efficiency levels between 2 and 4
per cent, far surpassing other films at
an industry average of approximately 8
per cent. Due to the success of the
product and the strong relationship
between both companies, Farmfresh‟s
corporate buyer (with whom many
interpersonal relationships have now
been formed) asked Packfex to
become their joint number one
supplier on all packaging. For Packfex
it meant that their sales would
increase by at least €7-9 million in
2006. Both companies are currently in
the planning stages of collaborating
again on Farmfresh‟s other packaging
films.
4. Analysis
The purpose of the paper
was to describe the dynamics of the
interaction
atmosphere
in
a
collaborative NPD relationship. The
process was categorized into seven
episodes based on major shifts in the
interaction atmosphere in a positive or
negative direction.
The interaction
dynamic at these convergent and
divergent points was presented.
Relationship new product development
doesn‟t require any of the classic
structures for working out processes of
collaboration where parties do not trust
each other or are developing an
agreed framework for interaction for
the first time. Given the relationship‟s
long term nature, 21 years, it had an
inbuilt structure of close collaboration
for
new
product
development.
Therefore, the product was developed
in response to solving a problem
rather than by negotiation and
agreement around how this might be
done, the roles of the parties, the
resources required, and how benefits
would be shared. The case context
appeared to have no specific structure
for the new product being developed –
there were no clear boundaries
between
the
firms,
continuous
interaction between the parties just
happened, there was no official new
product development team as people
got involved according to the demands
of the problem, the process appeared
fluid.
This type of NPD is only
possible in a collaborative relationship
and facilitates a shortening of the time
it takes to set-up mechanisms and to
define the character of the NPD
process. The relationship approach to
NPD with existing users in a problem
setting is about the process of
interaction as the parameters of the
new product development, so much a
feature of the NPD-specific literature,
are set-up for better or worse. The
case can be contrasted with other
forms of user involvement in NPD
where structures and processes of
cooperation have to be negotiated in
advance by the parties to the project,
or where projects are driven and
controlled by one actor who aims to
keep ownership of the entire process.
Therefore, this case, as structures and
systems for collaboration were set,
provided insights into the interaction
123
The IMP Journal
Volume 4, number 2
process dynamics which are rare in
the findings of previous NPD studies.
Not all the dynamics of
interaction atmosphere from the
relationship and alliance development
literature were dominant in the case.
As the relationship was already long
established, processes such as
negotiation
(the
structures
for
collaboration were already there),
learning (the parties already knew
each
other),
commitment
(interdependence in the relationship
was high), were not the critical
dynamics of interaction atmosphere in
the case.
However, expectations,
problem
solving,
personal
relationships
and
communication
appeared to be more critical process
dynamics across most of the episodes.
Mindsets around expectations at a
personal and organisational level were
major triggers of episode change and
challenge.
Parties were felt, at
individual and organisation level, not to
be meeting what was expected and
promised thus letting the relationship
down. Problem solving followed a
parallel cycle. It was almost a natural
feature of the relationships and thus
the focus for the NPD – the solution to
the film problem. The ebb and flow of
the problem was part of the dynamic of
each episode. The role and personal
relationships also were core aspects of
the cycle of atmospheric high and low.
The density of ties often held what was
at times near breakdown in personal
relationships. Goodwill trust was often
challenged where a “loss of faith” was
evident. The critical relationship in the
case and the one that held the process
together, that between the engineering
manager in Farmfresh and the
technical director of Packfex, further
exhibits the importance of personal
relationships in a time of relationship
crisis. Many of the episodes and
quotes in the case illustrate personal
frustrations and empathy trailing the
highs and lows of the interaction
atmosphere.
The communication
dynamic, both its absence and
presence, were revealing of the
atmosphere of the NPD relationship.
At times there was no communication
and at other times high quality
communication systems occurred
without huge thought – open access
across both organisations, idea
sharing,
and meetings labelled
“brainstorming” to focus on maximising
ideas and solutions. The descriptors
of these dynamics of the interaction
atmosphere are illustrated in table 1 of
the case and provide a comparator for
other such relationships especially
where parties cooperate intensely.
The dynamics were heightened in the
seven critical episodes.
The
cycles
of
convergence and divergence in the
case were evident in the seven
episodes and in the interaction
dynamics in each one.
The
descriptors of the dynamics, as in
table 1, illustrated the convergence
and divergence, for example, the
euphoria in meeting expectations
when they converged and the near
breakdown in expectations with
suspicion and blame in another. It
was where several of the dynamic
processes of interaction atmosphere
reach a high or low that convergent or
divergent episodes were grouped.
The severity of conflict across the
relationship atmosphere led to an
episode of divergence and likewise the
highs of the parties coming together
on a solution or agreement in an
episode
of
convergence.
Convergence and divergence were not
discrete processes. Indeed, without
them would rich solutions emerge?
They
appeared
to
make
the
interactions richer and the parties to
the relationship work harder. The
conflict in the divergent cycles was
related to the task at hand and forced
more creativity and innovation. The
relationship appeared well able to
handle vast amounts of task related
conflict in the interaction atmosphere
dynamics. This was especially so
given the lack of prescribed outcomes,
except for solving the problem with the
film, in the case.
The range of
outcomes finally reached was much
higher for both parties and would not
124
The IMP Journal
Volume 4, number 2
have happened without the cycle of
convergence and divergence. Only
once in the case, illustrated in table 1,
did the divergence become too great
(episode 6). This episode saw many
of the interaction dynamics at breaking
point but also conflict became
affective, it became interpersonal and
this affected the social fabric of the
relationship (Ariňo and de la Torre,
1998). This was particularly evident in
the near collapse of the personal
relationship between two key actors –
the engineering manager in Farmfresh
and the technical manager in Packfex.
Even in this episode the actors were
capable of managing the conflict within
the relationship.
The long term nature of
the partnership and the security and
stability this provides sustained the
parties throughout the progressive
turmoil the uncertainty threw up. The
substance (Ford and Håkansson,
2006; Håkansson and Ford, 2002;
Ford, Håkansson and Johanson, 1986;
Håkansson and Snehota, 1995) of the
interaction between the parties was
really only threatened once in what
was a highly charged engagement.
The actor bonds in the case were very
strong on many levels. The resolve of
the actors to solve the problem was
sustained throughout the case and
acted as a bridge to resolution in times
of deep crisis. The interdependencies
between Packfex and Farmfresh were
visible in the case on many levels.
The relationship was buyer-supplier
with the supplier‟s product being part
of the final consumer packaging. The
nature of the supplier‟s product
brought additional interdependencies
into the relationship. The two firms are
specialised and their interdependent
activities reinforced through the case.
Farmfresh‟s buyer changed and tried
to de-list Packfex – an action resisted
by other actors within Framfresh and
by the nature of the product
specialisation achieved by Packfex
over time. Packfex is much smaller
than its buyer and could be
disproportionately affected by its
decisions. In this way, it can be
argued that Farmfresh has a dominant
power position in the relationships if it
tried to use it. However, the joint
activity built up over time with
Farmfresh has embedded a culture of
mutuality and reciprocity in the
relationship.
Both parties have
committed much resource to the
relationship. They seemed to do it to
solve the current crisis without
question of cost or clear solution
patterns.
The resources of both
parties were different but it was the
resources embedded in people that
appeared to ultimately solve the
problem in the relationship and smooth
over, potentially deleterious, conflict.
The ability to commit people, spend
large amounts of time, do analysis
work, test solutions without assessing
risks, make clear the interdependency
of the parties. The learning from each
other is a well understood dialogue
from deeply connected patterns of
activity and resource sharing readily
switched on. The pattern of what is
there is almost not emphasized in the
interaction as it is taken for granted.
Routines
and
processes
long
established are part of a unique
context-specific structure.
Joint
problem solving was a major feature of
the case and a major relationshipspecific resource. As outlined at the
beginning of our paper solving
innovation
problems
is
usually
complex and not linear with few clear
solution pathways (Powell, Kogut,
Smith-Doerr, 1996; Håkansson and
Waluszewski, 2007). The solution in
this case was far from certain and
evolved in a messy unstructured way.
Problem solving was facilitated by the
nature of the relationship and open
sharing
of
information
which
deemphasized the extent of the almost
natural joint problem solving behaviour
which took place. While we cannot,
due to the specific nature of the
collaboration, generalise, the case
shows how actors co-create and solve
new product problems.
The NPD problem in the
case created the need for change in
what was a stable relationship. The
125
The IMP Journal
Volume 4, number 2
critical event, unacceptable levels of
wastage, started the cycles of
episodes in which the parties
ultimately resolved the problem and
further strengthened their relationship.
The critical episode classification
provided the structure for organising
the case data and for revealing the
interaction atmosphere, as a whole at
heightened points of convergence or
divergence.
Critical
episode
classification seeks out major points of
inflexion which gives an overall picture
of how the underlying processes of
interaction work in cycles, reinforcing
each other, compensating for each
other and leading to enhanced
relationship value as perceived by the
parties. The episodic classification
demonstrates how solutions focused
the companies were and the jumps to
solutions and the trails and errors
involved rather than any planned
approach. There were no clear stages
or development patterns to the NPD
process. Potential solutions at the
manufacture‟s site were followed until
a major solution (Chubb X3) was
found.
Many of the classic
cooperative mechanisms for NPD
were embedded in the relationship but
more importantly in the dynamic
processes that connected each
episode such as communication,
problem
solving,
and
personal
relationships.
Critical
episode
classification facilitated the authors‟
analysis of the dynamics of interaction
atmosphere which, in turn, will enable
firms further integrate and increase the
level of user involvement in NPD and
enhance its success.
5. Conclusion
This paper describes
the dynamics of the interaction
atmosphere in a case study of a
dyadic NPD project. The description
of the dynamics, whilst context and
situation specific, is novel. Two main
aspects of dynamics were considered.
The first was movement or change in
the interaction atmosphere. These
changes were classified using critical
episodes on the basis of major shifts in
the relationship atmosphere across a
number of processes. The second
level of analysis was the identification
of the process dynamics that most
impact in the case. The understanding
of these processes can be used in
other settings to handle the NPD
interaction
atmosphere
by
collaborating parties. The case clearly
underlies the strength of relationship
NPD as an approach, even in a severe
test of the relationship where, among
the
process
failures,
personal
relationship also broke down, the
parties found a way back to an
ultimately successful outcome.
References
Adams, M., Day, G. & Dougherty, D.
(1998). “Enhancing New
Product
Development
Performance:
An
Organisational Learning
Perspective”, Journal of
Product
Innovation
Management, 15, 403422.
Ariňo, A. & de la Torre, J. (1998).
“Learning from Failure:
Towards an Evolutionary
Mode of Collaborative
Ventures”,
Organizational Science,
9 (3), 306-325.
Biemans, W. (1992). Managing
Innovation
Within
Networks,
London:
Routledge.
Büchel, B. (2000). “Framework of Joint
Venture
Development:
Theory-Building Through
Qualitative
Research”,
Journal of Management
Studies, 37 (5), 637-661.
Booz, Allen and Hamilton, (1982). New
Products
Management
for the 1980s, Booz,
Allen and Hamilton Inc.,
New York, NY.
Cooper, R. (1979a). "Identifying
Industrial New Product
Success: Project New
Prod",
Industrial
126
The IMP Journal
Volume 4, number 2
Marketing Management,
43, 93-103.
Cooper, R. (1979b). "The Dimensions
of Industrial New Product
Success and Failure”,
Journal of Marketing, 8,
124-135.
Cooper, R. (1988). “Predevelopment
Activities Determine New
Product
Success”,
Industrial
Marketing
Management, 17, 237247.
Cooper, R. (1993). Winning at New
Products:
Accelerating
the Process from Idea to
Launch,
3rd
edition.
Cambridge, MA: Perseus
Publishing.
Cooper, R. (1996). “Overhauling the
New Product Process”,
Industrial
Marketing
Cooper, R. (1999). "From Experience:
The Invisible Success
Factors
in
Product
Innovation", Journal of
Product
Innovation
Management, 15 (2),
115-133.
Cooper, R. & Kleinschmidt, E. (1986).
“An Investigation into the
New Product Process:
Steps, Deficiencies, and
Impact”,
Journal
of
Product
Innovation
“Winning Business in
product
development:
The
critical
success
factors”,
Research
Technology
Management, 39 (4), 1830.
“New
Product
Performance:
What
Distinguishes the Star
Products”,
Australian
Journal of Management,
25 (1), 17-45.
Daft, R. & Weick, K. (1984). “Towards
a
Model
of
Organizational
as
Interpretation Systems”,
Academy
of
Management Review, 9,
284-295.
Denzin, N. (2001). “The Reflexive
Interview
and
A
Performative
Social
Science”,
Qualitative
Research, 1 (1), 23-46.
di Gregorio, S. (2000). “Using NVivo
for
your
Literature
Review”, Conference for
Strategies in Qualitative
Research: Issues and
Results from Analysis
using
QSR
NVivo
andNUD*IST, Sept 2930, Institute of Education,
London.
Doz, Y. (1996). “The Evolution of
Cooperation in Strategic
Alliances:
Initial
Conditions or learning
Processes?”,
Strategic
Management
Journal,
17, Special Issue, 55-83.
Dwyer, F. R., Schurr, P. & Oh, S.
(1987).
“Developing
Buyer-Seller
Relationships”, Journal of
Marketing, 51, 11-27.
Edvardsson, B. & Strandvik, T. (2000).
“Is a Critical Incident
Critical for a Customer
Relationship?”,
Managing
Service
Quality, 10 (2), 82-91.
Eisenhardt, K. (1989). “Building
Theories
from
Case
Study
Research”,
Academy
of
Management Review, 14
(4), 532-550.
Ford, D. (1980). “The Development of
Buyer-Seller
Relationships
in
Industrial
Markets”,
European Journal of
Marketing, 1, (5/6), 339354.
Ford, D. & Håkansson, H. (2006). “The
Idea of Interaction”, The
IMP Journal, 1 (1), 4-27.
Ford, D., Håkansson, H. & Johanson,
J. (1986). “How do
127
The IMP Journal
Volume 4, number 2
Companies
Interact?”,
Industrial Marketing and
Purchasing, 1 (1), 26-41.
Ford, D.
& Saren, M. (1996).
Technology Strategy for
Business,
London:
International
Thomson
Business Press.
Fox-Wolfgramm, S. (1997). “Towards
Developing
a
Methodology for Doing
Qualitative
Research:
The
DynamicComparative Case Study
Method”, Scandinavian
Journal of Management,
13 (4), 439-455.
Håkansson,
H.
(1982).
Ed.,
International Marketing
and
Purchasing
of
Industrial Goods, New
York: Wiley.
Håkansson, H. (1987). Ed., Industrial
Technological
Development, A Network
Approach, Kent, UK:
Croom Helm.
Håkansson, H. & Ford, D. (2002).
“How Should Companies
Interact
in
Business
Networks”, Journal of
Business Research, 55,
133-139.
“Analysing
Business
Relationships”.
In
Håkansson,
H.
&
Snehota,
I.
(Eds.).
Developing Relationships
in Business Networks,
24-49,
London:
Routledge.
Håkansson, H. & Waluszewski, A.
(2007).
(Eds.).
Knowledge
and
Innovation in Business
and Industry – the
importance
of
using
others,
London:
Routledge.
Halinen, A., Salmi, A. & Havila, V.
(1999). “From Dyadic
Changing
Business
Networks: An Analytical
Framework”, Journal of
Management Studies, 36
(6), 779-794.
Johnsen, T. & Ford, D. (2000).
“Managing Collaborative
Innovation in Complex
Networks: Findings from
Exploratory Interviews”,
16th
Annual
IMP
Conference,
Bath,
England.
Lilien, G., Morrison, P., Searls, K.,
Sonnack, M. & von
Hippel,
E.
(2002).
“Performance
Assessment of the Lead
User Idea Generation
Process for New Product
Development,
Management
Science,
48 (8), 1042-1059.
Lincoln, Y. & Cuba, E. (1985).
Naturalistic
Enquiry,
London:
Sage
Publications.
Lynch, P. & O‟Toole, T. (2006).
“Involving External Users
and Third Parties in the
New
Product
Development Process”,
Irish Marketing Review,
18 (1&2), 29-37.
Madique, M. A. & Zirger, B. J. (1984).
"A Study of Success and
Failure
in
Product
Innovation: The Case of
the
U.S.
Electronics
Industry",
IEEE
Transactions
on
Engineering
Management, 31 (4),
192-203.
Mahajan, V. & Wind, J. (1992). “New
Product
Models
–
Practice, Shortcomings
and
Desired
Improvements”, Journal
of Product Innovation
Management, 9 (2), 128139.
National Industrial Conference Board
(1964).
“Why
New
Products
Fail”,
The
Conference
Board
Record,
NCIB,
New
York.
128
The IMP Journal
Olson,
E.
Volume 4, number 2
& Blake, G. (2001).
“Implementing the Lead
User Method in a High
Technology
Firm:
A
Longitudinal Study of
Intentions
versus
Actions”,
Journal
of
Product
Innovation
Management,18 (6), 388395.
O‟Toole, T. & Lynch, P. (2004). “A
Study of the Practice of
User Involvement in the
Early Stages of New
Product Development”,
11h International Product
Development
Conference,
Dublin,
Ireland.
Pettigrew, A. (1997). “What is
Processual Analysis?”,
Management, 13 (4),
337-348.
Ring, P. & Van De Ven, A. (1994).
“Developmental
Processes
of
Cooperative
Interorganisational
Relationships”, Academy
of Management Review,
19 (1), 131-145.
Powell, W. W., Koput, K. W. & SmithDoerr,
L.
(1996).
“Interorganizational
Collaboration and the
Locus of Innovation:
Networks of Learning in
Biotechnology”,
Administrative Science
Quarterly, 41 (1), 116145.
Schurr, P. H. (2007). “Buyer-Seller
Relationship
Development Episodes:
Theories and Methods”,
Journal of Business and
Industrial Marketing, 22
(3), 161-170.
Schurr, P. H., Hedaa, L. & Geersbro,
J. (2008). “Interaction
Episodes as Engines of
Relationship
Change”,
Journal
of
Business
Research, 61, 877-884.
Sminia, H. (2009). “Process Research
in Strategy Formulation:
Theory,
Methodology
and
Relevance”,
International Journal of
Management Reviews,
11 (1), 97-125.
Stevens, G., Burley, J. & Divine, R.
(1999).
“creativity
+
Business Discipline =
Higher Profits Faster
from
New
Product
Development”, Journal of
Product
Innovation
Management, 16 (5),
455-468.
Tidd, J., Bessant, J. & Pavitt, K.
(2001).
Managing
Innovation,
London:
Wiley.
von Hippel, E. (2001). “Perspective:
User
Toolkits
for
Innovation”, The Journal
of Product Innovation
Management, 18 (4),
247-257.
von Hippel, E. & Katz, Ralph (2002).
“Shifting Innovation to
Users
via
Toolkits”,
Management
Science,
48 (7), 821-833
Voss, C. (1985). “The Role of the user
in the Development of
Applications Software”,
Journal
of
Product
Innovation Management,
2 (June), 113-121.
Welsh, C. & Wilkinson, I. (2002). “Idea
Logics
and
Network
Theory in Business”,
Journal of Business-toBusiness Marketing, 9
(3), 27-48.
Yin, R. (2003). Case Study Research:
Design and Methods,
Thousand Oaks, CA:
Sage Publications.
Zajac, E. & Olsen, C. (1993). “From
Transaction
Cost
to
Transaction
Value
Analysis: Implications for
the
Study
of
Interorganizational
Strategies”, Journal of
129
The IMP Journal
Management
30, 131-145.
Volume 4, number 2
Studies,
130
The IMP Journal
Volume 4, number 2
131
The emergence of a successful business
network – What was the role of public policy?
Tommy Shih
Department of Business Studies, Uppsala University, Box 513, 75120 Uppsala, Sweden, e-mail: tommy.shih@fek.uu.se
Abstract
To promote industrial development and economic growth is a vital issue for governments all over
the world. The ideals guiding policymakers in their endeavours, strongly influenced by traditional economics and
the innovation system approach, are that innovations based on new and advanced knowledge are central for
industrial and economic development. As is exemplified through the quote below policymakers have no problem
with finding inspiration from regions such as Silicon Valley.
The idea that so much could grow in so short time within such small geographical area sent
planning bodies from Albuquerque to Zimbabwe scrambling to grow the next Silicon Valley on their
own backyard. Sturgeon (2000: p.15)
But although the identified “generic” features have been copied, there are few examples of how
ambitions to “artificially” create policy supported high-tech based business networks and industries have
succeeded. One of the few successful examples of policy created high-tech industries often mentioned is the
Taiwanese semiconductor industry. The story of the Taiwanese semiconductor industry is impressive as the one
of Silicon Valley; in just a few decades a booming industry developed from scratch. One of the most common
explanations to the transformation addresses the governing role of the state in coordinating industrial
development and creating a successful semiconductor business network. Some of the major factors mentioned
were for example the creation of public research institutes, the public provision of R&D, and the subsequent
transfer of technologies to a downstream sector created by Taiwanese policy. This envisioned development
scenario has been strongly supported in Taiwanese policy circles and forms a foundation of contemporary
Taiwanese industrial development policy. However this model of business creation applied to other industrial
areas has been widely criticized for not fulfilling it promises.
To investigate this issue, this paper takes a different and complementary view of the emergence
of a Taiwanese semiconductor business network. Based on a resource interaction perspective the study aims to
increase the understanding of forced network creations. The findings argue that the understanding that a network
was created by policy is clearly an over-simplification which omits several important factors in the emergence of
the semiconductor business network.
1. Introduction
The promotion of industrial
development and economic growth is a
vital issue for governments all over the
world. The ideal that guides policymakers
in their endeavours, strongly influenced by
traditional economics and the innovation
system approach, is that innovations
based on new and advanced knowledge
are central for industrial and economic
development (OECD, 1996; Eklund,
2007). This observation is explained by
the OECD (2007: p5):
Today,
performance
determinant
is
innovation
a crucial
of
competitiveness and national
progress.
Moreover,
innovation is important to help
address global challenges,
such as climate change and
sustainable development. But
despite the importance of
innovation,
many
OECD
countries face difficulties in
strengthening performance in
this area. […] Governments
can also play a more direct
role in fostering innovation.
Public investment in science
and basic research can play
an important role in developing
ICT and other general-purpose
technologies and, hence, in
The IMP Journal
Volume 4, number 2
enabling further innovation.
This highlights the importance
of reforming the management
and
funding
of
public
investment in science and
research, as well as public
support to innovative activity in
the private sector.
To support development of
advanced knowledge and to create a
system that facilitates the transfer of the
results from research to industry has
consequently been a main concern in
contemporary policymaking. However,
empirical
evidence
suggests
that
commercializing
knowledge
is
a
cumbersome task with few traces of
linearity. That it is not that easy to support
artificially the development of new hightech solutions which will lead to
knowledge-based industries and business
clusters has been experienced by many
governments. An editorial in The
Economist (2007: p4) gave the following
opinion on this experience:
EU officials, like government
bureaucrats everywhere, are
obsessed
with
creating
geographic clusters like Silicon
Valley. The French have
poured billions into pôles de
compétitivité; and Singapore,
Dubai and others are doing
much the same. There are
dozens of aspiring clusters
worldwide, nicknamed Silicon
Fen, Silicon Fjord, Silicon Alley
and Silicon Bog. Typically
governments pick a promising
part of their country, ideally
one that has a big university
nearby, and provide a pot of
money that is meant to kickstart entrepreneurship under
the guiding hand of benevolent
bureaucrats. It has been an
abysmal failure.
Despite these disappointing
results there are examples of high-tech
business networks and industries that are
presented as successful creations of
policy. A salient example is the Taiwanese
132
semiconductor industry based in Hsinchu.
The development of this industry is
intimately linked with Taiwan’s economic
success. In just a few decades, the
Taiwanese economy transformed itself
from being dependent on agriculture to
become one of Asia’s high-tech centres. In
short the story commonly told is that in the
early 1970s Taiwan was a backwater
economy. The country was dependent on
agricultural
production
and
labourintensive manufacturing of textiles,
electronic components and plastics. At
that time, Taiwanese policymakers
decided that it was time to direct industrial
production towards more knowledgeintensive sectors and move up a step on
the economic development ladder. A field
that was identified by the government as a
future industry which would allow Taiwan
to take this development leap was
semiconductors. Public policies were
implemented to speed up development in
a hitherto non-existent semiconductor
industry. The focus on semiconductors
turned out to be beneficial for the
Taiwanese economy. Since the 1980s the
economic growth of Taiwan has been
closely associated with the development of
the semiconductor industry located in
Hsinchu, also known as the Silicon Valley
of Taiwan. Two decades after the
emergence of the first few semiconductor
businesses in the early 1980s, the
Taiwanese semiconductor industry was
ranked the fourth largest in the world1 and
consisted of nearly 400 companies2. At the
end of 2005 the Taiwan Semiconductor
Industry Association (TSIA) estimated that
60 per cent of worldwide semiconductor
foundry, package and testing revenue, 25
per cent of worldwide semiconductor
design revenue and 25 per cent of
worldwide DRAM revenue were generated
by Taiwanese companies. The total
economic value generated by the
Taiwanese semiconductor industry totalled
1118 billion New Taiwan Dollars (roughly
1
Defined in terms of production value, surpassed only by
the USA, Japan and Korea.
2
The companies can be classified as: 268 IC design
houses, 6 wafer suppliers, 4 mask makers, 13 fabrication
companies (fabs), 33 packaging houses, 35 testing
houses, 15 substrate suppliers and 19 chemical suppliers
(TSIA, 2006).
The IMP Journal
Volume 4, number 2
33 billion USD) at the end of 2005 (TSIA,
2007).
Regardless
from
what
vantage point the development of the
semiconductor industry and the Hsinchu
region is viewed, it appears impressive.
Within a few decades, a new industry and
a flourishing business network resting on
high-tech and innovation emerged in a
country which had previously relied on
traditional industries and small and
medium-sized companies with weak R&D
capacity. The most common interpretation
of
the
Taiwanese
semiconductor
development is that it was a result of
public policy engagement in coordinating
industrial development (see, e.g., Liu,
1993; Mathews & Cho, 2000), a view
which is also heavily stressed by
Taiwanese public policy (MOEA, 2003).
Hence the picture of a dynamic
semiconductor business network created
in the hands of foreseeing government
bureaucrats has come to serve as an
important role-model for how to create
new industries in Taiwan (Liu, 1993). As
has been interpreted by e.g. Liu (1993) or
Chang et al., (1994) the main policy
measures undertaken were aimed at the
foundation of research institutes, a public
provision of R&D, the subsequent diffusion
of the research results to the private
sector, and the establishment of sciencebased parks. This template has in the last
decade aggressively been applied by the
Taiwanese
government
to
other
technological fields. The application of the
“semiconductor development template” on
other
fields
have
however
been
considered disappointments, for instance
in the case of biotechnology (see e.g.
Swirbanks & Cyranoski, 2000; Hsu et al.,
2005; Shih, 2009).
But could the perceived
failure of these attempts rather be
symptoms of unrealistic expectations on
how industries and business networks
actually emerge. In this case what is then
the role of policy? To investigate these
issues the emergence of a Taiwanese
semiconductor business network will be
studied from a resource interaction
perspective (Håkansson & Waluszewski,
2002). The rationale of using a resource
based perspective is that it can catch
133
interdependencies beyond spatial and
organizational borders even when they are
not
represented
through
direct
relationships.
1.1 Literature review
In
contemporary
policymaking system approaches in
general have provided a rationale for the
public support of industrial development
and business networks. In particular
research on clusters and innovation
systems have been highly influential to
policymakers. In the innovation system
perspective the innovative performance of
a country depends to a large extent on
how actors relate to each other as
elements of a collective system of
knowledge creation and use. Commonly
understood is that the analytical focus is
on the elements and relationships which
interact in the production, diffusion and
use of new, and economically useful,
knowledge (Lundvall, 1988; Edquist,
1997). Thus from the innovation system
perspective it is the ecology of actors
within a system, which through interaction
create and diffuse innovations, that is of
interest. The cluster framework proposed
by Porter (1990) is a comparable concept.
Cluster analysis has been applied and
used extensively in policy circles (Teigland
et al., 2004). Similar to innovation systems
it does not have a universally accepted
definition. A definition which is used by
Porter (1998: p78) is:
Clusters
are
geographic
concentrations
of
interconnected companies and
institutions in a particular field.
Clusters encompass an array
of linked industries and other
entities
important
to
competition.
Although
the
conceptualization of innovation systems
and industrial clusters often differs from
policymakers’ use of these notions
(Eklund, 2007; Lundvall, 2007) the policy
implications are straightforward. That is,
system approaches suggest the possibility
of a certain level of control and implies that
The IMP Journal
Volume 4, number 2
industries and business networks can be
created, and the components necessary
for their development identified. However
empirical evidence of how industries or
networks emerge often shows a messy
and non-linear picture (Håkansson &
Waluszewski, 2002; Shih, 2009). As
suggested by Wedin & Waluszewski
(2003: p4) the development of a
technology or industry is a myriad of
“expected
outcomes
as
well
as
unexpected effects, where new and old
solutions are tried and retried”. With this
understanding, the creation of networks is
logically also an intricate issue which
offers few certain pre-destined outcomes.
The reason, as described by
Håkansson and Snehota (1989) is that
companies
interact
with
suppliers,
customers, competitors, authorities and
non-governmental organizations in order
to create value. There are always
anticipated and unanticipated effects from
interaction, as neither the motives of
people nor the content of the resource
combinations and activity links they
represent can be fully known in advance
(Håkansson & Waluszewski, 2002; Ford et
al., 2002). Thus interaction between
various actors creates effects with both
positive and negative consequences
(Håkansson et al., 2009). The effects of
prolonged interaction are that relationships
between actors are established and
interdependencies created between actors
and resources. A result of the formation of
relationships over time is that companies
and organizations become increasingly
dependent on each other, on their
customers,
suppliers
and
other
counterparts. Thus actors, material and
immaterial resources, and activities are
systematically related to each other (Ford
et.al., 2003). As relationships are
systematically developed, Håkansson et
al. (2009: p2) conclude that “they do not
only connect dyads, but they do also
connect indirect related companies in
network-like structures”.
From the IMP perspective
the dictum holds that networks cannot be
managed, however firms manage within
networks (Ford et al., 2002; Öberg. et al.,
2007). How firms manage in networks has
been described in the IMP literature (e.g.
134
Håkansson
&
Waluszewski,
2002;
Harrison & Waluszewski, 2008; Baraldi &
Strömsten, 2009; Öberg & Brege, 2009).
Few studies within the IMP tradition,
however, have followed actual attempts of
forced network creations and the
involvement of policy. In this study we will
follow the case of a business network in
the semiconductor field, which has been
argued to have been artificially created
and managed by Taiwanese policy. The
case will provide an opportunity to
investigate an example of a forced
network and the role of policy in creating
and supporting a network. The paper is
structured as follows in the section 2 the
research design is presented. This is
followed by the empirical case description
in section 3 and an analysis in section 4.
In section 5 the conclusions are
discussed.
1.2 Research Design
The
Taiwanese
semiconductor business network has often
been portrayed as one of few examples of
how policy has managed to create and
control the emergence of an industry. In
this context the example provides an
opportunity to study forced network
creations and to investigate the role of
policy. The methodology used in this
paper is a single case study (Yin, 1994).
The material has been based on a number
of secondary sources. Initially a number of
interviews were made that gave me a
structure of the semiconductor story.
However as most of the events occurred in
the 1970s and 1980s, the respondents
had little first-hand information of what
happened. Hence, this paper is principally
based on a broad secondary literature
survey synthesising different areas of
literature such as economic geography
and history, business studies, strategic
management, organizational studies, and
political science; and it draws on analysis
of policy, industry and academic
documents. The sources I have used
include official policy documents, white
papers, articles from academic journals,
media coverage, industry reports, and
statistical material. These sources have
helped me to identify events, establish
The IMP Journal
Volume 4, number 2
timelines, and have given me varied
perspectives
of
the
development
processes.
For
example
historical
accounts
have
provided
extensive
development descriptions and analyses of
the Taiwanese economy and, industrial
and technological policies from the 1970s
onwards. There are also a large number of
publications, in academic journals, aimed
at describing the Taiwanese economic
miracle and the emergence of the
semiconductor industry. The analyses
made in the articles mostly discuss the
role of the government and policy related
research institutes in the development
processes.
By
using
the
above
mentioned
sources
a
case
was
constructed of how the Taiwanese
semiconductor industry and a business
network emerged. To structure the case
and the analysis this study will apply the
resource interaction model developed in
Håkansson & Waluszewski (2002). It will
be used to capture the processes in which
heterogeneous resources3 have been
combined in the Taiwanese semiconductor
industry and investigate the role of policy
in the creation of the network. The
resource interaction model investigates
direct and indirect interaction between
resources, on the basis that it is possible
to catch interdependencies even when
they are not represented through direct
relationships (Håkansson & Waluszewski,
2002). The model has been applied to
areas such as product, technological,
logistics, and industrial development (see,
e.g., Wedin, 2001; Håkansson &
Waluszewski,
2002; Baraldi,
2003;
Gressetvold, 2004; Jahre et al., 2006,
Håkansson
&
Waluszewski,
2007,
Waluszewski et al., 2009, Shih, 2009).
In the resource interaction
model, resources are separated into four
3 An important assumption of resources is that they are
heterogeneous. The notion of resource heterogeneity was
early on suggested by Penrose (1959) who argued that a
resource is “a bundle of possible resources per se but the
services they create that make them valuable. In the IMP
setting these ideas were adopted by Hägg and Johanson
(1982) who proposed that the value of a resource depends
on how it is combined with other resources. Hence
resources alone are not productive and have no value
unless they have a use or a function to fulfil in combination
with other resources, i.e., forming a network-like
structure.services”. In other words, it is not the
135
categories where two are mainly tangible
or physical: (a) products and (b) facilities
or equipment. The other two types of
resources are mainly intangible or
organizational: (c) organizational units and
(d) organizational relationships. Below is
an overview of the four types of resources
(for a more detailed description see
Håkansson & Waluszewski, 2002; 2007).
In the context of resource
interaction, the identification of the
resources that are relevant to study is a
matter of choosing a focal resource
through which ties to other resources can
be identified. How the resources affect
each other are investigated through the
interfaces that are created between
resources which are defined by Strömsten
and Håkansson (2007: p29) as follows:
No resource is used in
isolation. Every resource has
interfaces to both physical and
organisational resources. […]
“interface” is defined as “a
place or area where different
things meet and have an effect
on each other”.
As can be seen in Figure 1, both
material (physical) and immaterial
(organizational)
resources
are
combined into a larger resource
structure,
connected
through
interfaces. The attention is directed
to
the
interaction
between
resources, how they are combined
and, developed over and beyond
time, organizational and spatial
boundaries. In the search for
resource interfaces there are no exante distinctions made concerning
technological sectors, spatial or
organizational borders. Instead the
focus is on the search for related
resource interfaces that occur
across various technological, spatial
and organizational fields (Strömsten
& Håkansson, 2007). In this study
the
focal
resource
is
an
organizational
unit,
namely
Taiwanese policy. By originating
from the focal resource the
interaction with other relevant
resources
creates
a
context
The IMP Journal
Volume 4, number 2
136
Resource interface
Resource interface
Resource interface
Products
Facilities
Org. Units
Relationships
Source: Modified from Waluszewski et al. (2009)
Figure 1
An illustration of the resource interaction model and interfaces
(which is discussed in section 4). We can
also
investigate
if
the
resource
combinations occurred consciously or
unconsciously. What follows hereafter is
the empirical case describing the
emergence of a Taiwanese semiconductor
business network.
1.3 The emergence of the Taiwanese
semiconductor business network
Taiwan has been considered
one of the economic miracles of the
twentieth century (World Bank, 1993;
MOEA, 2005). The annual growth rate
from 1952 to 1993 was 8.7 percent, an
impressive number which few other
countries have surpassed over such a
long period of time (Chuang, 1999). In just
a few decades, the Taiwanese economy
went from being dependent on low-tech
agricultural production to become a
technological powerhouse and one of the
leading semiconductor manufacturers in
the world. How this was achieved has
been studied extensively and often it is
attributed to the government’s active role
in economic planning and coordination
(Wade, 1990).
Today Taiwan is the twentyfourth largest economy and has the fourth
largest semiconductor industry in the world
(TSIA, 2007; IMF, 2008). Based on the
economic success of the industry, the
government has lately been vigorously
promoting a knowledge-based economy
and aimed to transform Taiwan into a
green silicon island. With the hopes of
creating a second economic miracle, the
semiconductor industry has played an
important role as an inspiration and model
to follow (MOEA, 2003). What now follows
is an empirical account of how a
Taiwanese
semiconductor
business
network emerged. In the chronology
below, some major events leading to the
development
of
a
Taiwanese
semiconductor industry are outlined.
The IMP Journal
Volume 4, number 2
137
Table 1
Chronology: major events in the Taiwanese semiconductor industry
1961
1964
1966
1973
The first foreign electronics companies, such as Philips and IBM, establish a
presence in Taiwan.
National Chiao Tung University establish the first semiconductor laboratory in
Taiwan
Texas Instruments establish the first semiconductor assembly operation in Taiwan.
The Taiwanese government decides to develop a semiconductor industry.
The first public research institute, Industrial Technology Research Institute (ITRI) is
founded through the merger of three government laboratories in Hsinchu.
1974
1976
1978
1980
1986
1988
2004
The Electronics Research Service Organisation (ERSO), a sub-department of ITRI,
aimed at developing semiconductor technology is founded.
A technology transfer of a mature technology to ITRI from US semiconductor
producer RCA Semiconductor and Materials.
A special government expert committee created, known as the Science and
Technology Advisory Group (STAG).
Taiwan’s first semiconductor company, United Microelectronics Company (UMC), a
spinoff from ITRI, is founded.
The Hsinchu Science Based Park is established.
UMC is the first company to locate in the Hsinchu Science Based Park.
The second spinoff from ITRI, Taiwan Semiconductor Manufacturing Company is
founded.
Semiconductor foundry as a business model is established with the emergence of
TSMC.
The Taiwanese semiconductor industry starts to grow rapidly.
The Taiwanese semiconductor industry the fourth largest in the world.
2. The Taiwanese electronics industry
– paving a way into semiconductors
The
ambition
of
the
Taiwanese government to make the
transition into technology-intensive sectors
formally appeared in the 1970s. The move
was believed to be needs driven for
reasons such as industrial development
and international recognition (Chang et al.,
1994). Prior to that, public policies had
been aimed at measures which would
build up a military capacity in Taiwan in
order to launch an attack to retake
mainland China. Initially, production on the
island had been directed towards
agriculture, but after the Kuomingtang4
4
The Kuomintang was the first political party of the
Republic of China. During the Second World War, the KMT
was the ruling party in China, but after the war internal
conflicts and the growing strength of Communist party led
(KMT) assumed control over the former
Japanese colony5 in 1949 an import
substitution policy was adopted. This
stimulated the growth of new industrial
sectors, such as plastics and textiles. In
the 1960s, the Taiwanese leaders started
to promote the export industry in order to
increase national income and earn foreign
currency as a result of reduced US
financial aid.6 The government policies
encouraged the development of labourintensive light industries (Wade, 1990;
Chen, 1999).
By the 1970s the import
substitution and export subsidy policies
to the defeat of the KMT, which had to flee into exile. The
KMT leader Chiang Kai Shek brought over to Taiwan a
whole administration and an army in 1948; a total of 2
million people moved.
5
Taiwan was a Japanese colony between 1895 and 1945
6
In the 1950s it was financial aid from the United States
that helped Chiang Kai Shek to maintain a large military
force without overheating the weak economy.
The IMP Journal
Volume 4, number 2
had turned the trade deficits into regular
trade surpluses. The momentum was
however temporarily brought to halt due to
competitive pressure from emerging
neighbouring economies and political
crisis as a result of China taking over
Taiwan’s mandate at the United Nations in
1971. The global oil crisis in 1973 also
brought an economic downturn. These
events forced the Taiwanese government
to search for new avenues through which
sustainable
economic
and
political
development could be created. To realize
these goals it was believed that the focus
had to shift from the labour-intensive
consumer goods industry to technology
intensive manufacturing industry. The
industries that were targeted for export
promotion to attract foreign currency and
investments had been identified with the
help of Stanford Research Institute (SRI)
in the early 1960s (Wade, 1990; Chen,
1999; Hsu & Cheng, 2002). As noted by
Ernst (1997: p7): “SRI chose those
product
groups
where
American
companies had strong interests: certain
petrochemical
intermediates,
plastic
resins, synthetic fibres, transistor radios,
electronic components, watches and
clocks”. To motivate foreign investments,
an export processing zone was also
established in Kaohsiung in Southern
Taiwan in 19657. As a result, increased
amounts of investment by US, Japanese
and European electronics companies
started to flow in. The operations, taking
advantage of the low-cost labour, were
concentrated towards the manufacturing of
electronics and electronic components
(Mathews & Cho, 2000).
As mentioned, an active
export promotion policy was implemented
in the 1960s. A reason for this was the
reduced financial aid from the US, which
prompted the Taiwanese government to
seek income and foreign currency through
other means. Generous incentives were
given to foreign companies willing to invest
in Taiwan. The foreign direct investments
came in the field of consumer electronics
and the pioneers were IBM and Philips.
IBM had set up operations in Taiwan in the
7
The first in the world; various tax incentives were given to
local as well as foreign companies interested in investing
in the zone.
138
late 1950s, and also established an
affiliate producing core wires by the early
1960s. The business model was geared
towards moving labour-intensive stages of
final assembly to low cost countries.
Similarly, Philips took advantage of low
cost manufacturing by establishing a
subsidiary
in
Taiwan
in
1961,
manufacturing TV sets, audio equipment
and related components. Soon an inflow of
Japanese direct investments came, the
first was Matsushita that set up a majority
owned joint venture in 1962. Up to the
mid-1980s this venture was one of
Matsushita’s major production facilities in
South East Asia. Sanyo followed in 1963,
Hitachi in 1965, and Sony in 1967. By the
1970s, most of the leading Japanese
electronic producers had established a
presence in Taiwan or were engaged in
labour intensive assembly with a growing
share of output going to Japan or
Japanese affiliates in Asia. American
companies had also realized the benefits
of being in Taiwan. For instance, in 1964
General Instruments directed production of
transistor radios to Taiwan (Ernst, 1997;
Mathews & Cho, 2000).
While several companies
had set up subsidiaries, others acquired a
direct stake in existing local companies.
The latter strategy was for example used
by Toshiba, which had in the 1950s
acquired a 5 percent equity-share in
Tatung Co. Taiwan’s only integrated
electronics company at the time. Initially,
Tatung was only a distributor, selling
various electronic products produced by
Toshiba. In the 1960s the cooperation
deepened and Tatung also received
technology licenses from its Japanese
partner, allowing the company to become
a supplier of key components, such as
high-end compressors, picture tubes and
LCDs. Other Japanese companies such
as Fujitsu followed with a similar approach
when it in 1973 established a joint venture
with Tatung. The deal gave Tatung the
rights to both sell and service Fujitsu
computer systems and peripherals. These
events eventually led to a number of joint
ventures and OEM (Original Equipment
Manufacturing) contracts with Taiwanese
companies. Thus the investments made
by foreign manufacturers of consumer
The IMP Journal
Volume 4, number 2
electronics gave rise to a rapid growth in
demand for electronic components
produced in Taiwan. Although most of the
high value-added key components were
imported, both local production and
capacity were increased (Ernst, 1997; Tu,
2001).
The
foreign
direct
investments played an important catalytic
role for the emergence of a Taiwanese
electronics industry. For example, the
Japanese companies offered intensive onthe-job training as well as developing
close links with local suppliers that
focused especially on the domestic
market. A significant scale of local
linkages was created by the foreign
investments. Furthermore, the companies
that invested provided the local employees
and suppliers with education, knowledge
and technology, although not advanced.
Some of the employees also started new
local companies. For instance General
Instruments’ Taiwanese affiliate itself gave
rise, through former employees, to the
founding of 11 local companies. In addition
to being an incubator for local suppliers,
foreign companies also established other
facilities. Matsushita for instance created
the Matsushita Electric Institute of
Technology in 1981 with a work force of
around 40 researchers (Ernst, 1997; Lin,
2003).
The events mentioned above
preceded the growth of a domestic
semiconductor industry. The first company
to introduce semiconductor related
business to Taiwan was General
Instruments,
who
established
a
semiconductor assembly plant in Taiwan
in 1967. Between 1969 and 1973, other
multinational companies such as Philips,
RCA and Texas Instruments followed suit
and established their semiconductor
assembly operations in Taiwan (Mathews
& Cho, 2000). By contrast, the first
semiconductor related research activities
in Taiwan had local roots as discussed by
Chang & Tsai (2000: p186): “The theory
and technology of semiconductors was
first systematically introduced in Taiwan
when National Chiao Tung University
started a course in 1960. The university
built a semiconductor laboratory in 1964
that succeeded in manufacturing its first
139
integrated circuit in 1965. National Chiao
Tung University then chose semiconductor
technology as the main focus of its
curriculum, with the aim of training more
high-tech manpower”. According to Chang
& Tsai (2000), National Chiao Tung
University later also cooperated with
governmental units and provided a
foundation for the semiconductor industry
in terms of basic research and human
resource development.
3. A government initiative to create a
new industry
Foreign direct investment
was a factor that contributed to the
emergence of a Taiwanese electronics
manufacturing industry. Although the
manufacturing of electronics products
brought income to the export sector, those
activities were believed by Taiwanese
policymakers to be isolated from the rest
of the economy and to have little value in
terms of industrial development. The
reason expressed by Lin (2003) was that
the foreign companies saw Taiwan only as
a low cost manufacturing resource.
Furthermore, there were no local
companies conducting any technologically
advanced R&D. However, the fast growth
and the volume of applications possible, in
for
example
consumer
electronics,
telecommunications
and
industrial
electronics, made the electronics industry
an attractive sector for Taiwanese
policymakers to promote. With this
ambition, the main issue became to find a
key technology that would help the
Taiwanese electronics industry to develop
in the direction of technology-intensive
products.
Hence,
expert
advisors
suggested that Taiwan should develop
semiconductors, specifically integrated
circuit
design
and
manufacturing
technology in order to stimulate innovation
throughout
the
island’s
electronics
industry. Chang et al. (1994: p163) provide
the following explanation for why the
Taiwanese government
decided to
concentrate on semiconductors:
Since the integrated circuit
was introduced in 1958, its
The IMP Journal
Volume 4, number 2
small
size,
low
power
consumption, rapid operating
speed, reliability, and low cost
per electronic function have
led to significant changes in all
electronics products, including
consumer electronics. If the IC
industry were developed in
Taiwan, a spillover effect
would be generated for
industries which use ICs. The
IC was thus selected as the
key
technology
to
be
developed.
Taiwanese companies had
however no experience in making
semiconductors. Beside the foreign
manufactures there were no local
companies with experience or knowledge
concerning semiconductor design or
manufacturing. A task force, The
Technology Advisory Committee (TAC),
funded by the Ministry of Economic Affairs
(MOEA) was therefore set up with the
mission of investigating how to carry out a
development
strategy
for
the
semiconductor industry. The TAC was
formed by Y.S. Sun8 (at the time the
Minister of Economic Affairs) and P.W.
Yuan, an engineer at RCA, in Princeton.
The formation of TAC had been preceded
by the belief that the key to a successful
technological upgrading was to leverage
the experience and knowledge of
overseas Chinese engineers working in
the US (Mathews, 1997). It was this group
of highly skilled Chinese engineers, and
academic scholars, working at various
semiconductor companies and universities
in the US that became the recruiting base
for the TAC. Eventually the TAC also
provided the guidelines concerning how to
develop a semiconductor industry (Tung,
2000). The main areas of the strategy are
highlighted below (Chang et al., 1994:
p163):
1. TAC became responsible for
140
the
planning
of
the
development.
This
was
decided because there was
no local experience in
integrated circuit design and
manufacturing available.
2. Since the gap between
advanced
semiconductor
producing countries and
Taiwan was very large, the
main strategy to quickly
develop an industrial base
was through technology
transfer.
3. The purpose of introducing
semiconductor
technology
was to create an industrial
base and to establish this
kind of technology in Taiwan.
The technology would have
to be assimilated and
developed. For this purpose
a new research institute,
ITRI was formed to reach the
initial goals.
4. Over a period of 4 years,
410 million NTD (13 million
USD) was to be invested by
the government to purchase
the
manufacturing
technology, product design
and training personnel
4. The creation of a public research
institute and technology acquisition
Who would take the lead in
developing a new industry? The private
sector companies, the majority of them
being small or medium sized9, were not
technologically sophisticated enough.
Neither did those companies place much
emphasis on increasing R&D activities and
investments (Liu, 2002). The few large
companies, all involved in traditional
industries, were reluctant to invest in new
unproven industries (Mathews & Cho,
2000). Consequently it was believed by
policymakers and experts that “no existing
8
Sun was responsible for laying the foundations for
Taiwan’s technological upgrading. Both he and Yuan
agreed that the electronics industry would be the key to
Taiwan’s transformation, and that semiconductors should
be a key technology. Furthermore, they believed that the
required knowledge needed to be leveraged from abroad.
9
According to Saxenian (2000), SMEs make up 95
percent of all companies in Taiwan. MOEA states that 90
percent of all Taiwanese companies in the 1950s were
enterprises with 10 or fewer employees. In the 1960s the
proportion of SMEs was 95 percent.
The IMP Journal
Volume 4, number 2
industry in Taiwan could lead the way in
developing future high-tech industries for
more than ten years” (Chang & Hsu, 1998:
p350). In addition, the Taiwanese capital
market was underdeveloped and financial
institutions were conservative in lending
out capital for risky ventures (Saxenian,
2000). Due to these circumstances, there
was no other choice than for the
government to assume the responsibility
of being in the frontline in building up a
semiconductor industry. In order to
commence
semiconductor
related
activities, the Ministry of Economic Affairs
merged three government laboratories
located in Hsinchu to form ITRI in 1973.10
The government commissioned the newly
founded research institute to carry out the
introduction
and
assimilation
of
semiconductor technology. ITRI was thus
the sole institution in Taiwan chartered to
develop a semiconductor industry. With
that purpose, ITRI established in 1974 the
Electronic
Research
and
Service
Organization (ERSO)11, a unit specifically
concentrating
on
semiconductor
technology. The responsibility for planning
and coordination was however still in the
hands of the TAC. Since no domestic
proprietary technology existed. TAC
decided to acquire it from abroad (An,
2001). What technology would be suitable
to license?
The first integrated circuits
had already been developed in 195912,
and by the 1970s a large number of
integrated circuits with various features
and technology platforms existed. In the
mid 1970s the most advanced integrated
circuit designs had a 3.0 micron
bandwidth. After some initial enquiries
however, no companies were interested in
transferring cutting-edge technology to
ERSO. The only technologies available for
licensing were 7.0 micron chips. After
lengthy discussions concerning the
opportunities, the conclusion reached by
the TAC was to obtain low power, high
density technology that would provide
10
Union Industrial Research Laboratories, Mining
Research and Service Organization, and Metal Industrial
Research Institute were donated to ITRI by the Ministry of
Economic Affairs
11
At the time the lab was known as Electronics Industrial
Research Center, in 1979 the name ERSO was adopted.
12
By Kirby, at Fairchild Semiconductor
141
submicron development potential. The
main points in the discussions of the TAC
were according to Chang et al., (1994:
p164) as follows:
1. It would be very difficult to
license
an
advanced
technology.
Either
the
companies that possessed
that technology would not
agree to a transfer or the
price would be very high. It
was believed by the TAC
that it would be more
feasible to license a mature
technology
with
lower
competitive advantage.
2. The 7.0 micron technology
was mature, and thus also
held several advantages for
a country which had no prior
experience in semiconductor
manufacturing
and
development,
including
higher consistency, complete
technical documents, many
skilled
technicians,
and
effectiveness
in
the
operation of the equipment.
3. Products manufactured with
7.0 micron technology were
already out on the market
and feedback was available
concerning
process
technologies,
product
development,
design
technology and marketing
channels. Acquiring the 7.0
micron technology would
therefore allow Taiwan to
learn about all aspects of
integrated circuit technology
from
R&D
to
commercialization.
The search for partners was
conducted by ITRI which believed that
American semiconductor companies were
the ones most suitable to license
technology from. Hence, over twenty
requests to companies in the US were
sent out and a handful of companies
returned a proposal for a technology
transfer. After the Taiwanese selection
committee had visited the prospective
The IMP Journal
Volume 4, number 2
companies, two were selected as potential
partners,
RCA
Semiconductor
and
Materials (hereafter RCA) and company X.
The cost for RCA’s deal was twice as high
as the one given by company X, but the
terms of the company’s proposal were
better.
RCA’s
proposal
included
technology,
process
design
and
manufacturing management skills for
integrated circuit fabrication, whereas
company X’s proposal consisted of
process design and design technology.
However, another dimension that came
into play was also that RCA could provide
a year-long training for 35-40 ITRI
engineers at its laboratories in the US. In
contrast company X only suggested
training for 3 months for 3-4 persons.
Since it was believed by TAC that the
success of the project would be reliant on
the extensive training of human resources,
the difference in the suggested training of
Taiwanese engineers came to be the
critical factor in the decision-making.
RCA’s proposal was considered as the
better choice. Although the guiding
principle had been to select the deal with
the lowest price, the technology content
and personnel training proposed by
company X was believed not good enough
to achieve the goal of introducing
semiconductor technology in Taiwan. The
technology that was licensed to Taiwan
was the so called Complementary Metal
Oxide Semiconductors (CMOS), which
originally was developed by RCA13. The
technology corresponded to the goals of
TAC to acquire a “low power, high density
technology that would provide submicron
development potential”. Targeting this
technology meant also that ITRI would not
be competing directly with established
manufacturers (Chang et al., 1994; Chen
& Sewell, 1996; Hung et al., 2005).
Responsible for coordinating
the technology transfer was ERSO. While
the agreement with RCA was being
negotiated talented young Taiwanese
engineers were recruited and trained at
13
CMOS technology was developed at Fairchild
Semiconductor in 1963. In 1968 the first CMOS based ICs
were developed at RCA. At the time it was a low power but
slow alternative to the standard NMOS (another
technology which ITRI wanted to license but proved to be
too expensive) and TTL technologies.
142
ERSO for a period of time while waiting for
the pending transfer. After the agreement
with RCA had been finalized in 1976, 37
engineers
were
sent
to
different
laboratories and plants in the US operated
by the company for one year of technical
training. Many of these engineers would
later become the corporate leaders of
Taiwanese semiconductor companies
(Chang & Hsu, 1998). The agreement with
RCA included the transfer of a 7.0 micron
CMOS process technology, product
specifications,
design
and
testing
technology for a digital electronic watch.
Assistance in building a semiconductor
plant and training of personnel were also
included in the licensing agreement. While
the engineers were sent to the US for
training, ERSO were setting up a 4 inch
wafer pilot plant for semiconductor
manufacturing back in Hsinchu, Taiwan.
When the engineers returned in 1977 the
plant was already operational for test runs.
The same year the first integrated circuits
were produced by the pilot plant. The
standard of the product complied with
what had been agreed in the licensing
contract (Chang et al., 1994).
ITRI had accomplished the
introduction of semiconductor technology
to Taiwan. Of course Taiwan was still far
from catching up with advanced nations
but the main goal was to learn more about
semiconductor
technology,
and
to
accumulate knowledge. For this goal, a
pilot plant, with design, manufacturing and
testing capabilities had been built, and
was geared towards producing simple
semiconductors. As noted, RCA had
developed the first CMOS integrated
circuits, but CMOS was at the time not a
widespread technology. RCA was actually
about to withdraw from the semiconductor
industry and the licensing deal with ITRI
was an opportunity to squeeze some last
income from a mature technology. The 7.0
micron CMOS was mature, and far behind
the worlds leading LSI 2.0 micron circuit
designs, nonetheless for ERSO this was a
way of gaining access to the world of
semiconductors.
In
retrospect,
the
licensing of CMOS technology proved to
be a wise choice. First of all ITRI did not
have to directly compete with established
producers on a global market. Second the
The IMP Journal
Volume 4, number 2
market share of CMOS was relatively
small at the end of the 1970s, but started
to expand rapidly afterwards to become
the most used technology in IC design
today.14 After RCA withdrew from the
semiconductor industry in the early 1980s,
ITRI also inherited the intellectual property
portfolio from RCA that had been related
to CMOS technology (Mathews and Cho,
2000).
ERSO’s
semiconductor
fabrication plant had been built under the
guidance of RCA. After being able to
produce integrated circuits in 1977, used
in electronic watches, ERSO soon also
started
to
produce
experimental
semiconductors by using its own designs.
By 1979 ERSO were getting better yields
from these integrated circuits than what
the licensed technology had given. In the
early 1980s ERSO could provide CMOS of
4.5 micron and in the mid 1980s of 1.0
micron (Chang et al, 1994). In 1979 ERSO
also established a customer relationship
with a Honk Kong electronic watch
producer that bought integrated circuits
from the pilot plant. The order of 10000
integrated circuits was small and the
owner of the Hong Kong firm was a former
college classmate of
the person
responsible for running the pilot plant, Shih
Ching Tay. This deal provided ERSO with
an opportunity to interact with a user. (Tu
et al., 2006) The total amount of capital
invested from 1975 to 1979 was 410
million NTD dollars (roughly 12 million
USD)15. After the introduction of CMOS
technology the government’s commitment
also increased. Between 1979 and 1983,
670 million NTD was to be invested. The
goal that had been set up by ITRI was to
upgrade the technology from 7.0 to 3.0
micron (Chang et al., 1994).
14
Originally a low-power but slow alternative to TTL,
CMOS found early adopters in the watch industry and in
other fields where battery life was more important than
speed. Some twenty-five years later, CMOS has become
the predominant technology in digital integrated circuits.
This is essentially because area occupation, operating
speed, energy efficiency and manufacturing costs have
benefited and continue to benefit from the geometric
downsizing that comes with every new generation of
semiconductor manufacturing processes.
15 It was a substantial sum to be invested by the
Taiwanese government in a single technology, but
compared to the research budgets of large semiconductor
companies it was not a considerably large R&D budget.
143
In 1979 the Taiwanese
semiconductor sector still only consisted of
ERSO’s plant and a handful of foreign
assembly plants. Local companies were
not interested in semiconductors, as it was
considered a risky and unproven business.
The technology which ERSO had acquired
and continued to develop was still far
behind the global standards, which were
getting to below 2.0 micron bandwidth. In
addition, there was no real infrastructure to
support high-tech development in Taiwan
and the investments required to resolve
this issue would have to be quite large
(Chang & Tsai, 2000).
5. Hsinchu Science Park and the first
ITRI spinoff - UMC
The ambition of creating
high-tech industries in Taiwan had strong
support in policy circles. A person who
came to play an important role for the
high-tech development was former
Minister of Economic Affairs, Li Kwoh
Ting. He had taken an initiative for the
creation of a permanent advisory body to
the government in science and technology
issues. The group that was established in
1978, headed by Li, was named the
Science and Technology Advisory Group16
(STAG) consisted mainly of overseas
Chinese with technical backgrounds. Many
of the advisers in STAG had worked in the
US and experienced the growth of hightech regions such as Route 128 and
Silicon Valley.
Based upon their
experiences, STAG suggested that
Taiwan needed a specialist infrastructure
to support advanced industries such as
semiconductors (Saxenian & Hsu, 2001;
Yu, 2007). The ambition to set up a
specialized
infrastructure
gained
adherence in the Executive Yuan, and
under the sponsorship of the National
Science Council (NSC) a science park
was to be established. The decision was
however not well received in all political
camps. The efforts to set up a science
park were met with considerable
opposition and scepticism in the
16
STAG remains the main science and technology
advisory group to the government up to the present. Since
1979, together with the NSC it has also served as the main
organ for science and technology policy.
The IMP Journal
Volume 4, number 2
Taiwanese Cabinet. The NSC was
nevertheless successful in securing land
near Hsinchu, where both the ITRI
campus and National Chiao Tung
University were located. In 1978 210
hectares of land had been expropriated by
the Hsinchu county government to create
the new park and in 1980 the Hsinchu
Science
Park
Administration
was
established (Mathews & Cho, 2000).
The
establishment
of
Hsinchu Science Park was to facilitate the
creation of a high-tech industry, but there
where no local companies that could
locate in the park. What existed were a
few foreign subsidiaries that were involved
in the downstream stage of packaging and
testing semiconductor products. There
was also ERSO which had set up a pilot
plant manufacturing semiconductors, but
other than that there were no local
companies
specifically
involved
in
semiconductor
development
and
production. Since no private Taiwanese
companies were involved in Large Scale
Integration (LSI) and semiconductor
related R&D activities, the ERSO
management decided to create a company
(Chang & Hsu 1998; Mathews 2000). It
was believed by ERSO that the prospects
of a Taiwanese semiconductor industry
would be threatened if foreign companies
would first establish subsidiaries (Liu et al.,
2005). Hence, the pilot plant at ERSO was
to be spun off, and form the foundation of
a
new
company
named
United
Microelectronics
Company
(hereafter
UMC). The spin-off would mark an
important milestone in the development of
a semiconductor industry. ERSO was now
ready to exploit commercial opportunities
with the technology that had been
acquired 3 years earlier (Chen & Sewell,
1996). The idea of a spin-off from ITRI
was however novel, and there were
difficulties with raising capital for a project
of this kind. ITRI sought funding from both
private and public sources, and in the end
the majority of the capital was provided by
the government (mainly by securing funds
from state owned banks). A large stake
was also taken by five large private
companies (Saxenian, 2002).
ERSO not only spun off the
pilot plant, there was also an extensive
144
technical personnel transfer in which
around 180 persons were transferred to
the new company. In addition, the process
technologies which had been modified and
developed at ERSO were given to UMC,
mainly a 5.0 micron CMOS process
technology. Furthermore UMC received
ten Application Specification Integrated
Circuit (ASIC) products as well, including
integrated
circuits
for
calculators,
melodies, timers and telephones. By 1982
the transfer process had been completed
and the operations of the new company
began the same year (Chang et al., 1994).
UMC pursued a niche strategy by focusing
on ASIC products that had been
transferred from ERSO. The first
customers had been inherited from ERSO,
but the company also started to attract low
end electronics manufacturers from
Taiwan and Southeast Asia as customers.
The strategy to concentrate on these
customers meant that UMC was able to
avoid direct confrontation with the large
Japanese
semiconductor
companies
which were concentrating on standard
products such as memory. Focusing on a
niche market turned out to work well, and
in November 1982 UMC had reached
break-even point (Liu et al., 2005).
Even with the ongoing spin-off of UMC
and the ensuing reorganization, ERSO
remained
active
with
continued
development of the licensed technology.
By 1980 ERSO engineers had reduced the
bandwidth of the process technology from
7.0 micron to 5.0 micron. This was further
improved to 4.5 micron the year after.
Although this could be seen as an
achievement in itself for a new
organization with little experience in
semiconductor R&D, ERSO was not
getting closer to catching up with the
leading standards. At the time the world’s
top semiconductor manufacturers were
producing products using Very Large
Scale Integration (VLSI) technology of 2.0
micron bandwidth. It was clear that Taiwan
was still far behind the top countries such
as the US and Japan in terms of
technological levels (Mathews & Cho,
2000).
Thus, with the current rate of
progress would it be possible for Taiwan to
catch up with the advanced nations? The
The IMP Journal
Volume 4, number 2
advisers at STAG believed that although
ERSO was successful in introducing and
assimilating the CMOS technology,
Taiwan was still far behind the advanced
semiconductor nations, and some argued
that the gap was actually increasing. The
STAG advisers strongly advised that
Taiwan should set its target at achieving
VLSI capacity of 1.0 micron standard or
higher. This would bring the technology
competence in Taiwan on par with the top
companies in the world. ERSO strongly
objected to STAG’s advice and argued
that Taiwan should be patient in its efforts
to develop an industry, and not take on
that much risk by directly try to challenge
the large semiconductor companies. This
could quickly jeopardize what had already
been built up. Officials at the state
departments such as the Ministry of
Finance and the Economic Council for
Planning and Development were also
opposing the suggestion from STAG.
These departments were more concerned
with issues related to macro-economic
stability and were not interested in
promoting a single technology. STAG’s
suggestion to achieve VLSI capability was
however supported by some high
government officials, such as the president
and the premier. Hence, in 1983 it was
decided that the government would invest
2.9 billion NTD (roughly 85 million USD) to
pursue the plan to achieve 1.0 micron
VLSI capability by 1988 (Chang & Tsai,
2000; Mathews & Cho, 2000).
This was a very ambitious goal
considering the then current state of the
Taiwanese semiconductor sector and the
small involvement of the private sector. As
earlier, the government entrusted the VLSI
project to ERSO. UMC had also tried to
convince the government that it was
capable of handling the task, but it was
considered too risky to hand over such a
mission to a newly started company. Thus
the plan was that a VLSI plant would be
set up at ERSO. But where would the
VLSI technology come from? Instead of
turning to another large company as
before, ERSO signed an agreement with
two Silicon Valley start-ups, Mosel and
Vitelic, to develop VLSI semiconductor
chips. Already by 1985 a bandwidth of
1.25 micron had been achieved at ERSO
145
for the CMOS technology, and in 1986
CMOS memory chips of 1.0 micron were
available. Taiwan now had the capability
of designing 1.0 micron chips. There were
however no fabrication facilities in the
country to produce these semiconductor
chips (Mathews, 1997; Mathews & Cho,
2000).
6. The growth of design capabilities
and the emergence of TSMC
The results of the VLSI
project were advanced design capabilities
and “state of the art” technology in one of
ERSO’s special laboratories. Where was
Taiwan heading from here, should the
designs be licensed to third parties for
fabrication? The problem with lack of
fabrication capacity became more obvious
with the growing number of semiconductor
design companies in Taiwan. As
mentioned earlier ITRI had started to
transfer the capabilities and resources
which had been built up, the first one
being UMC. In 1982 ERSO had also spun
off the first two independent Taiwanese
semiconductor design houses, first Syntek
and shortly thereafter Holtek. But if no
private sector companies would willingly
get involved in the semiconductor industry,
ITRI would have to create an industry
through spin-off companies (Mathews &
Cho, 2000).
In addition to the VLSI laboratory,
ERSO had in 1985 set up a Common
Design Center for chip design companies
to develop application products, which was
mainly aimed at start-up companies (Liu et
al., 2005). This encouraged several
overseas Taiwanese from Silicon Valley to
return to Taiwan and start their own
companies or expand their business with
the support of the Common Design Center
(Chiang & Hsu, 1998; Mathews & Cho,
2000; Liu et al., 2005). The semiconductor
design industry in Taiwan did not really
take off, however, even though the
technological levels had been raised and
were approaching those of the advanced
companies. A reason was not only that
there were no customers, but also a lack
of fabrication capabilities in Taiwan
contributed to the situation (Chen &
Sewell, 1996). UMC was the only
The IMP Journal
Volume 4, number 2
semiconductor company in Taiwan with a
fabrication plant prior to 1987 (An, 2001).
ERSO also had some fabrication capacity
since it had retained a part of the plant for
continued research after the UMC spin-off.
Nevertheless, none of these plants were
intended for VLSI manufacturing, and as
noted earlier the development was moving
towards VLSI technology. This capability
was believed to be necessary in order to
catch up with the advanced semiconductor
nations. So how would the products
developed with VLSI technology be
manufactured?
In 1985 Morris Chang had become
the new president of ITRI. Chang, an
overseas Chinese with a Ph.D. from
Stanford University in engineering, had
three decades of working experience in
the semiconductor industry and prior to
joining ITRI was head of the global
operations
department
at
Texas
Instruments. In Chang’s first week at ITRI
he proposed a new spin-off from ERSO.
He suggested that this spin-off should be
focusing strictly on manufacturing chips,
i.e. semiconductor foundry, for local and
international customers based on VLSI
technology. The rationale for this was
mainly based on two reasons. First, most
of the top 20 semiconductor companies in
the world did not have financial capital to
quickly upgrade their fabrication facilities
to VLSI-standard. Second, the growing
Taiwanese semiconductor design sector
needed fabrication plants to meet their
production needs (Liu et al., 2005).
The idea was quite novel,
since up to now the semiconductor
companies had been vertically integrated,
involved
in
both
design
and
manufacturing. Although these two
activities are separable, the companies
with fabrication capabilities were also
designing their own semiconductors in
order to reduce the risk of having
semiconductor designs copied. The new
spin-off from ITRI would be the first
company focusing strictly of foundry. The
proposal to create a pure foundry
company
was
accepted
by
the
government, but it was not to be fully
funded by the state; instead it was to have
both public and industry support. This
would be a way to push the private sector
146
to participate in the semiconductor
industry. The government gave ITRI the
task to find a multinational company as a
sponsor. The ambitions for the new
company was to become a global
semiconductor company, and in order to
receive credibility, technology and a cross
licensing portfolio it was believed that a
venture with a leading semiconductor
company would be best (Chang & Hsu,
1998; Mathews & Cho, 2000).
The possibility of creating a
large scale VLSI semiconductor business
through financial support from the
government
and
combined
with
engagement
from
an
international
semiconductor
user,
appeared
an
attractive solution. Interest was shown
from four multinational companies: Texas
Instruments, Intel, Philips, and Matsushita
(Mathews & Cho, 2000). All of these
companies, with the exception of Intel,
already had prior production activities in
Taiwan. Philips was the pioneer, starting
production in Taiwan when the company
established its production of TV sets,
audio equipment and related components
in 1961. In 1962 Matsushita established a
production facility in Taiwan, to be
followed by among others RCA and Texas
Instruments.
After
the
Taiwanese
government had negotiated with all four
companies, Philips proved to be the only
serious candidate. In 1986 it was
announced
that
the
Taiwanese
government and Philips would be the
largest shareholders of the new company,
Taiwan Semiconductor Manufacturing
Company (TSMC)17, and the company
was established the year after (Chen &
Sewell, 1996). ITRI provided the technical
personnel, around 150 persons, of which
most had been involved in the VLSI
project. ERSO also spun off its 6 inch
VLSI manufacturing plant that became
TSMC’s first fabrication facility. With this
TSMC became the first dedicated foundry
in the world, and pioneered a concept
which became a central element of the
semiconductor supply chain. Since Philips
17
According to Saghafi & Davidson (1989) 10 billion NTD
was raised. Philips became the largest private shareholder
with 27.5 percent of the equity. The largest shareholder
was the Taiwanese government, with 48.3 percent of the
equity.
The IMP Journal
Volume 4, number 2
147
production activities in Taiwan already
included
semiconductor
assembly
operations, the step to an engagement in
semiconductor foundry was already locally
established (Saghafi & Davidson, 1989).
Through
Philips’
engagement, TSMC not only received a
financier but also a large, skilled and
demanding customer. In the technology
area, Philips agreed to transfer 2.0 and 1.5
micron process technology to produce
VLSI devices. For more advanced
technologies Philips, would be paid royalty
fees. The condition for the deal was that
the new company would not become a
competitor to Philips own products in
Taiwan. The initial technology inputs
supplied by Philips accounted for 80
percent of TSMC’s original capability.
Philips transferred its portfolio of cross
licenses to TSMC to avoid the company
being accused of infringing intellectual
property rights of other semiconductor
companies,
something
which
had
happened to several upcoming Korean
semiconductor companies. In addition,
Philips also supported TSMC with product
and process know-how, but more
importantly what was gained was
legitimacy for the new company. As a
result of the extensive support, TSMC
experienced strong growth and was
successful in upgrading its technology to
world standards in a short period of time.
Until the end of the 1980s TSMC had to
rely on the support from Philips in order to
be able to produce advanced integrated
circuits. However, at the end of the 1980s
both the customer base and the
knowledge
of
making
advanced
semiconductors had grown so much that
TSMC was able to design 0.8 micron
semiconductors without any technical
support from Philips. In the early 1990s, a
decade after the operation started,
TSMC’s annual sales surpassed 1 billion
USD, and the production activities
included design and manufacturing of
semiconductor chips (Mathews & Cho,
2000).
semiconductors many foreign electronics
companies had a steady presence in
Taiwan. Philips had already been involved
in Taiwan since the early 1960s when the
company had set up a transistor and
television tube factory, which today is the
largest of its kind in the world and the main
supplier of tubes to the Philips group. The
company’s commitment came to grow
stronger over the years. Hence, when the
Taiwanese government searched for a
partner to form TSMC, Philips was a
potential sponsor. The reasons that Philips
turned out to be the only serious candidate
was not only because the company had
the financial and technical resources but
equally important was its long term
dedication to Taiwan. It must be taken into
consideration that TSMC was an unproven
business idea and the burden of proof was
on ITRI. The other companies, Texas
Instruments and Intel were just not
convinced of TSMC’s potential, but for
Philips the incentive to invest was the
opportunity to gain a stronger foothold in
the emerging Taiwanese market (Chang &
Tsai, 2000; Mathews & Cho, 2000).
The
development
of
TSMC
functioned as a catalyst for the continued
start-up
of
new
semiconductor
18
companies
in the Hsinchu region.
Around TSMC and its interaction with
customers such as Intel and Texas
Instruments, a structure of related
companies started to emerge. ITRI had
also continued to run its R&D operations,
and fuelled by its proven spin-off strategy,
projects became companies as soon as
technologies were considered ready for
commercialization. The research institute
maintained a liberal view on employees’
ambitions to create new companies, direct
as well as indirectly19, and this benefited
the enlargement of the semiconductor
industry. With the growing opportunities,
Taiwanese private capital was starting to
flow into the semiconductor industry in
larger amounts (Chang & Hsu, 1998).
UMC,
the
only
other
company in Taiwan at the end of the
7. A growing semiconductor industry
18
As discussed earlier, by the
time the government decided to promote
E.g. Destiny Technology Corp., Realtek, Weltrend,
Sunplus, ICSI, Eltron et cetera.
19
This high mobility of labour was also a major contribution
to the successful development, according to Saxenian
(2001).
The IMP Journal
Volume 4, number 2
1980s with fabrication capabilities, had
already been listed on the Taiwan Stock
Exchange in 1985. But although the
company was profitable it was lagging
behind
TSMC
in
technological
sophistication. For example, in 1987 when
TSMC’s technological capabilities were
almost similar to the world leading
producers, i.e. close to 1.0 micron, UMC
only had 3.5 micron process technology.
Furthermore, while TSMC was attracting
large
multinational
companies
as
customers, UMC was serving mostly
“small” customers (Chang & Hsu, 1998).
However this did not mean that UMC was
unsuccessful, the CMOS technology which
the company inherited from ITRI was also
becoming a standard technology used in
producing integrated circuits. Initially ITRI
had chosen to license a mature CMOS
technology from RCA because the more
advanced technological solution could not
be afforded. Although CMOS based
integrated circuits were a somewhat
slower alternative to some more advanced
solutions, it was also less power
consuming. This meant that CMOS
became an attractive solution for products
where low power consumption was of
greater importance than speed, for
example in the watch industry. Since the
CMOS technology was considered by the
dominating
US
and
Japanese
semiconductor companies as obsolescent,
it became a niche product which UMC
later became one of the few to supply.
About two decades after ERSO started the
development and production of CMOS
technology it had emerged to become one
of the predominant standards in integrated
circuits20 (Mathews & Cho, 2000). TSMC
and UMC had proven to be triumphant
cases which encouraged private sector
and non-public investors to participate in
an industry which had earlier been
dominated by government organizations.
The development progressed quickly, and
by the early 1990s Taiwanese companies
had similar technology levels to those of
20
It was a combination of CMOS features, for example the
geometric downsizing, the development of operating
speed together with energy efficiency, and the low
manufacturing costs that made CMOS a dominant
standard in semiconductors.
148
the advanced global semiconductor
manufacturers (Chang & Hsu, 1998; Hsu
& Cheng, 2002).
Today the semiconductor
industry is considered an icon of success
in Taiwan. At the end of 2005, Taiwan
Semiconductor
Industry
Association
(TSIA) estimated that 60 percent of
worldwide
semiconductor
foundry,
package and testing revenues were
generated by Taiwanese semiconductor
companies, with a majority of them located
in Hsinchu. For worldwide revenue in
semiconductor design as well as dynamic
random access memory, Taiwanese
companies held around 25 percent. The
total economic value generated by
Taiwan’s semiconductor industry totalled
1118 billion NTD (roughly 33 billion USD)
at the end of 2005 (TSIA, 2006).
7.1
Key interfaces in
semiconductor business network
the
The story of the emergence
of the Taiwanese semiconductor business
network is both an interesting and
impressive
example
of
industrial
development. Some of the major actors
that contributed to the development were
policy actors, foreign manufacturers,
public research institutes and local
industry. Figure 2 below is a network map
of some of the key resources in the
emergence of the semiconductor business
network. The following analysis will take a
closer look at Taiwanese policy’s
interfaces to: 1) ITRI & TAC; 2) RCA and
the CMOS technology; 3) Foreign
electronics
producers;
4)
Spin-off
companies
and;
5)
Advanced
semiconductor users. These interfaces will
illustrate the role of policy in the
emergence
of
the
Taiwanese
semiconductor business network.
1. Interface with ITRI & TAC
In 1973 the public research
institute ITRI was established and
commissioned
by
the
Taiwanese
government to develop semiconductor
technology. Accordingly this was the
formal start of a policy created Taiwanese
semiconductor business network. The
The IMP Journal
Volume 4, number 2
149
ITRI
TAC
ERSO
R.I. 1
UMC
TSMC
RCA
R.I. 2
CMOS
Taiwanese
Policy
R.I. 4
R.I. 5
R.I. 3
Products
Facilities
Foreign
electronics
manufacturers
Philips
Advanced
semicoductor
users
Org. Units
Relationships
Figure 2
Network map of key resources involved in the semiconductor case
capabilities and knowledge within ITRI
emerged over an extended period of time.
For example national Chiao
Tung University had already established
Taiwan’s first semiconductor laboratory in
the early 1960s and played a significant
role in training and educating personnel at
ITRI. Another important factor in the
establishment of the local research
institutes was existing sources of
semiconductor knowledge outside of
Taiwan. Much of the knowledge of
semiconductors stemmed from Taiwanese
professionals working within existing
companies and research environments.
Almost all the experts engaged in the
expert committees that the Taiwanese
government created, were US-based
Chinese and Taiwanese engineers
involved in semiconductor research or
business.
Members of the Technology
Advisory
Committee
(TAC)
were
especially influential in establishing a local
structure. It was these experienced
semiconductor experts who helped set up
ITRI, and ERSO in 1974. Hence, already
from the start of attempting to develop a
domestic
base
of
semiconductor
technology, Taiwanese policymakers and
engineers were interacting with individuals
working in world leading semiconductor
companies and research units. This
created a large number of organizational
interfaces not only to other developing
structures but also to existing producing
and using structures mainly in the US. It
was through these interfaces which ITRI
were able to access technologies and
knowledge.
2.
Interface with
CMOS technology and RCA
A local structure was built up
through using experienced semiconductor
professionals and organizational units
outside of ITRI. In addition to the creation
of such organizational interfaces, through
policy support, there were also important
physical interfaces that shaped ITRI and
the Taiwanese semiconductor business
network. One important physical interface
was the mature CMOS technology that
ITRI licensed from RCA in 1976. Although
The IMP Journal
Volume 4, number 2
the government provided the funding to
develop semiconductor technology it was
not
going
to
be
developed
instantaneously. ITRI needed first an
existing technology to experiment on and
learn from. However no advanced
producers were interested in licensing any
cutting edge technology to Taiwan. The
only viable option was to license mature
and less advanced technologies. The
decision to license an “obsolete”
technology from US producer RCA served
as a way to educate ITRI and its personnel
on how to manufacture semiconductors.
The fact that it was mature had several
advantages; one was that the cost of the
technology was within the proposed
budget. Another advantage was that it was
already thoroughly tested in existing
producing and using structures. In other
words the technology already had
established user and producer interfaces.
Although the technology lacked novelty,
which was considered as a weakness by
the multinational companies, it made it
possible for ITRI to work on solutions
suitable for both production and use
already in the 1970s. Consequently it was
not only the new institutions that
policymakers created that had an imprint
on the emergence of a business network.
ITRI’s ability to take advantage of already
embedded resources, such as a mature
technology was decisive.
About two decades after ITRI
started to engage in the CMOS technology
it had emerged to become a dominant
standard in integrated circuits. From a
using perspective the features of the
mature technology (for example, low
power consumption) were much more
important than novelty and untried
solutions. The CMOS technology later
became a niche product which ITRI’s spinoff UMC was one of the few manufacturers
to supply. Of course this could not have
been known by Taiwanese policymakers
at the time of the technology transfer.
However, an important aspect to point out
is that it was enabled due to the fact that it
could increase the value of the users’
existing resource structures, and thus
providing opportunities for Taiwanese
companies to become suppliers of
semiconductors.
Furthermore,
an
150
important reason for ITRI choosing RCA
was the extensive support program the
company offered. RCA provided a
complete production technology and
training for 37 Taiwanese engineers at
RCA’s US laboratories for a year. Since
the technology transfer also entailed
extensive personnel training, ITRI had a
large number of trained engineers by the
mid-1970s. In addition RCA helped ITRI
set up a fully operational semiconductor
production facility. RCA was willing to help
ITRI set up a production facility since the
technology was considered obsolete and
the company was about to withdraw from
the semiconductor field. The production
facility that was established was fully
functional within a year after the signing of
the contract with RCA. As the technology
transfer was accompanied by interaction
related to other complementary resources,
capabilities
and
knowledge
of
semiconductor development could be built
up. However, as mentioned in the case,
this did not lead to any major
achievements commercially. But the
explicit goal of ITRI was not primarily to
make
economic
returns
on
the
investments made, at least not at that
stage. The main aim was to learn how to
develop and produce semiconductors.
Until the 1980s the capabilities and
technologies of the local structure had
already been built up through a large
number of resource combinations with
existing developing and producer-user
structures, where the interface with RCA
was quite substantial and important.
3.
Interface with
established producers
Besides RCA, other foreign
semiconductor companies were not
actively collaborating with the emerging
Taiwanese
semiconductor
network,
although relationships to multinational
electronics companies had already existed
since the 1960s. The view of Taiwanese
policymakers was that the business units
of the foreign companies had positive
effects on Taiwan’s economic growth but
had added little value to the emergence of
a home-grown industry. It was not until the
mid-1980s that they were considered to
have an important role by the Taiwanese
The IMP Journal
Volume 4, number 2
government. Nonetheless, it is difficult to
separate and neglect the role the foreign
companies played before the Taiwanese
industry started to grow rapidly in the late
1980s. Although the presence of the
foreign companies in Taiwan in the 1960s
had no immediate impact on the
development of advanced semiconductor
technology in Taiwan, it served as an
important platform whereby important
relationships and commitments came to
be established. By the time Taiwanese
policy decided to start promote the
semiconductor field, foreign electronics
companies had been present in Taiwan for
over a decade. The relationships which
were established between Taiwanese and
foreign companies provided knowledge to
Taiwanese employees, gave rise to new
companies, and set the foundation for the
electronics industry (which later became a
source
of
users
of
Taiwanese
semiconductors).
Furthermore,
established Taiwanese companies also
received a share of technologies and
business as they were seen as important
business
partners to
the
foreign
companies. Thus an intricate network of
interfaces to producer-user settings had
emerged. In this period the local structure
had already had extensive interaction with
established business structures, which
was also inherited by the spin-offs.
What the case indicates is
that the organizational interfaces that were
created were often not consciously part of
an ambition to build up a semiconductor
industry. For instance the relationships,
developed between foreign electronic
companies
and
the
Taiwanese
government were built up over decades,
starting with the establishment of a
foreign-owned electronics industry in
Taiwan in the 1960s. The activities to
develop semiconductor technology as well
as business in the 1970–1980s were thus
undertaken in an environment where
major global suppliers of semiconductors
were already active in the Taiwanese
economy, as producers of electronic
appliances. As relationships between
Taiwanese companies, policymakers and
the foreign companies were established
there was continuity in their interaction.
However, it was after many years of
151
infrastructure build-up and commitment
from the Taiwanese government that the
foreign companies present in Taiwan
eventually became interested in the
Taiwanese semiconductor industry.
4.
Interface with
UMC and TSMC
Following the establishment
of ITRI the local resource structure
became incrementally more advanced and
was on par with world standards in the
mid-1990s. This was in part due to the
structure of local companies emerging in
the 1980s. ERSO had improved the
CMOS technology in the production
facilities created with the help of RCA.
Eventually a part of ERSO’s production
facility was spun off into a new company,
UMC. Later ERSO’s VLSI production
facility was spun off laying the foundation
to TSMC. Not only were these two spin-off
companies the first two Taiwanese
producers of semiconductor technologies,
today they are also the two largest
semiconductor foundry companies in the
world. When UMC was spun off from ITRI
it inherited both a production facility and its
first customer. The first Taiwanese
semiconductor company became a
producer of reliable but non-advanced
semiconductors, catering mainly for small
South-east Asian electronic companies.
This business idea changed when Philips
became interested in a joint venture with
ITRI. The creation of TSMC had a
profound effect on the Taiwanese
semiconductor industry, and also the
global semiconductor business. The birth
of the semiconductor foundry and
Taiwan’s flagship company TSMC was the
result of the interaction between ITRI,
Philips and the Taiwanese government.
These organizational units had at the time
goals which were commensurable. The
Taiwanese government wanted to create
an industry and ITRI had reached a stage
where it could spin-off a part of its
facilities. For Philips there were clear
business opportunities from outsourcing its
production. Hence, Philips transferred
technology, know-how, a cross-licensing
portfolio, as well as legitimacy to the new
start-up (each resource being instrumental
to the development of the TSMC). But
The IMP Journal
Volume 4, number 2
perhaps more important was the fact that
TSMC had one of the largest electronics
companies in the world as its customer
from the start. This allowed TSMC to
upgrade its manufacturing technology and
skills in a short amount of time. Becoming
a supplier to a large and advanced user
not only proved beneficial in upgrading the
technology of TSMC but it also drew the
attention of other large electronics
companies such as Intel and Texas
Instruments to mention a few that later
also became customers of TSMC.
5.
Interface
Philips and other advanced
users
A
major
reason
why
advanced semiconductor companies were
not initially customers of Taiwanese
semiconductor products is quite simple.
The Taiwanese companies did not offer
any complementary resources which they
needed. Most of these advanced
companies were fully vertically integrated
in terms of design and production, and had
no interest in what was being developed at
ITRI. The only part of the production which
was outsourced was the testing which did
not require any advanced capabilities.
Thus in the beginning ITRI’s production
catered to a largely “low-tech” segment of
the semiconductor market. ITRI and later
UMC was not regarded as a threat by the
top semiconductor manufacturers, neither
did they produce anything of economic
value for them. This changed with the idea
of semiconductor foundry and the
opportunity to create an external supplier.
The established semiconductor companies
were not interested in another company
that could develop advanced technologies.
Instead the solution that was created, and
which provided a complementary resource
base to these advanced users’ existing
resource structures, was a Taiwanese
company TSMC. The business idea was
to supply advanced semiconductors based
on users’ specifications. This business
was not a result of ITRI creating a hightech production plant and then finding
customers. The demand was created
through interaction between the ITRI and
an established business structure. For
example, the business relationship
152
between TSMC and Philips had a long
history. Philips had been present in
Taiwan since 1961 and over the years the
commitment had come to grow stronger.
When
the
Taiwanese
government
searched for a partner to form TSMC,
Philips was a potential candidate. Other
companies that were approached were
Intel and Texas Instruments, all advanced
semiconductor companies, but in the end
Philips turned out to be the only serious
candidate. The reason was not only
because it had the resources, of
importance was also the company’s longterm dedication to Taiwan. It must be
taken into consideration that TSMC was
an unproven business idea and the burden
of proof was on ITRI. The other companies
were just not convinced of TSMC’s
potential. However, for Philips the
incentive to invest was to increase the
value of its already made investments
abroad and in Taiwan. The company also
wanted a supplier for a set of VLSI
technologies, the leading standards at the
time. The idea was something which
quickly became embedded into the
existing structure of related producer-user
interfaces. Later on TSMC also became a
major supplier to other semiconductor
companies such as Intel and Texas
instruments among others.
8. Discussion and conclusions
In this article the purpose
has been to increase the understanding of
forced network creation in the context of
resource interaction, and in relation to the
role of public policy. The understanding
reached from this study will be discussed
through three propositions which have
been identified in earlier studies of
resource interaction (Håkansson &
Waluszweski, 2002, 2007; Waluszewski
et. al., 2009; Shih, 2009 among others).
The propositions are as follows: 1)
Resources are combined over multiple
spaces and times; 2) New resources are
always combined with existing resource
structures and; 3) The economic value of
any new resource is closely associated
with how it can embedded in a using
structure.
The IMP Journal
Volume 4, number 2
153
Table 2
Summary of resource interfaces
1
ITRI & TAC
ITRI created from a directive from Taiwanese policy with mandate to
establish an industry and undertake research in the field of
semiconductors. Established with the help of TAC which consisted of
semiconductor professionals from abroad provided ITRI with direction,
knowledge and contacts.
2
RCA
& RCA not willing to license an advanced technology but saw a chance
Technology
to make some revenue by licensing out an obsolete technology.
Provided ITRI with knowledge, support, facility, training and a mature
technology which already had established user-producer interface.
Policy wanted advanced technology, acquiring the mature technology
within the realms of the government budget.
3
Electronics
Multinational producers already in Taiwan during the 1960s, wanted to
producers
take advantage of low cost production. Government policies of import
substitution and export promotion provided incentives for these
companies to set up shop on Taiwan. Important relationships and
commitment from the established producers created, especially with
Philips. Also helped establish a local electronics business network,
although not a direct goal of Taiwanese government.
4
Spinoffs
Government supported the spinoffs, UMC and TSMC. TSMC was
created through the combination of spinning off ITRI’s VLSI production
facility and the technological know-how from Philips. This resource
combination brought forward a new production process,
semiconductor foundry, a novel idea which was enabled partly due to
the support of the Taiwanese government. It later in turned out to
become a money-earning business model for both TSMC and other
Taiwanese businesses.
5
Advanced
Advanced users not interested in Taiwanese semiconductor
users
technology or products. Taiwanese policy not able to control or
influence these actors. After a complementary resource was
established base (TSMC) the advanced users were willing to
cooperate with Taiwanese semiconductor companies.
1) Resources are combined over
multiple spaces and times
As has been highlighted in
the analysis it is not single events at a
certain time or place which triggers the
formation of new industries. What this
study illustrates is that the emergence of
network was the result of both planned
and unplanned combinations of various
resources over an extended period of time
and in different places. The emergence of
a semiconductor business network
happened without following a linear path
with first R&D, then production and finally
use, at consecutive separate stage.
Rather
use-production-development
happened concurrently, where developing
structures emerged in relation to already
existing using and producing structures.
With the assistance of established
knowledge sources, it took more than a
decade to establish research and
development capabilities. A producing
structure was built up over an even longer
period with close contact to users. These
users had an established presence in
Taiwan already in the 1960s and, although
they were not active at the time, resource
synergies were created. The emergence
of a Taiwanese semiconductor business
network was thus a result of combinatory
efforts stretching over at least three
decades and shows the importance of the
close ties between developing, producing
and using structures.
The IMP Journal
Volume 4, number 2
2) New resources are always
combined with existing resource
structures
As discussed above a factor
for value creation was the ability of
different actors to take advantage of what
had already been created in other
resource structures both locally and
internationally.
The
Taiwanese
government’s effort to create space for
Taiwanese organizations and companies
in an international network, covering
development, production and use of
semiconductors, was a key factor for the
emergence of a business network. In
particular the connections to established
producer-user settings were imperative in
the development. For instance the
subsidiaries of the foreign companies
provided local employees and suppliers
with
education,
knowledge
and
technology. Already from the 1960s new
local companies were started in the wake
of the foreign investments. What was in
creation was the emergence of a
producer-supplier network which continues
until today, where semiconductors became
an extended business activity due to
already
established
business
relationships. These interfaces brought
forward knowledge and also various
solutions which could benefit the
Taiwanese
semiconductor
business
network. Thus the notion that there is
always something to build on is imperative
when formulating viable public policies to
develop business networks.
3) The economic value of any new
resource is closely associated
with how it can embedded in a
using structure
From the empirical account it
is also evident that the value of any new
resource is strongly related to how it can
be embedded in using structures.
Although there were interfaces to
advanced using structures already before
network
emerged,
multinational
semiconductor companies did not become
customers
of
Taiwanese
made
semiconductors until the early 1990s. Why
the advanced multinational companies
154
eventually
became
customers
of
Taiwanese semiconductors products was
to a large extent dependent on the
possibility to create additional value of
their already made investments. What
happened in the Taiwanese case were
Taiwanese companies becoming part of a
global semiconductor supply chain. By
concentrating on only a part of the
production chain and becoming suppliers
of semiconductors to advanced using
structures, the Taiwanese companies
were not competing with their customers
and did not risk eroding the value of their
investments. Instead they complemented
the resource structures of the advanced
using structures.
9. Policy implications
What
can
the
above
discussion tell us about forced network
creation and what implications are there
for policy? What the analysis of the
resource interfaces in the Taiwanese
semiconductor business network has
portrayed is that the emergence of the
network came about through interaction
between both established and new
resources over several decades and
geographical borders. The industry was
not an instant economic success nor did it
just surface in a setting with tabula rasa
antecedents.
Development
of
semiconductor technology occurred in
close relation to already existing produceruser structures and the closeness
functioned as a catalyst to the emergence
of a semiconductor business network in
Taiwan.
As is evident from the
empirical account the different actors in
the network that emerged had various
goals which often were not compatible.
But Taiwanese policy was an important
and skilful network actor. For example
Taiwanese policymakers assisted in
establishing and creating interfaces to
various organizational and physical
resources, in particular to existing
producing-using structures. Policy was
also
innovative
in
creating
new
organizations, empowering professionals
with experience to lead the way in
The IMP Journal
Volume 4, number 2
development and allowed the policysupported institutions to take new
directions. What seemed to be an
essential factor is the flexibility through
which policy acted, a trait which does not
always seem to go well with pro-active
government intervention and guidance.
Thus the notion of Taiwanese policy
creating and controlling the business
network omits several important empirical
conditions in the emergence of the
network. An important factor seems to be
the reliance on an existing network of
resources, locally and internationally,
within and beyond organizational borders.
Taiwanese policy was far away from
controlling this larger network, especially
the users. Furthermore the ability of other
network actors to take advantage and
create value of already embedded
resources,
independent
of
policy’s
ambition should not be downplayed.
What can policy then actually
do? Governments can support business
networks
through
funding
and
infrastructure build-up. Nonetheless, this
does not equal control of the network, or
controlling the goals of a larger number of
actors within the network. As this study
suggests policy can also facilitate the
processes through which relationships are
established and interfaces are created.
Especially how to create persisting
interfaces between developing, producing
and using structures should be of higher
importance.
Furthermore
promoting
flexibility and providing institutional stability
are measures which policy should actively
pursue.
References
An, C. T. (2001). Taiwan’s semiconductor
industry: What the state did
and did not. Review of
Development Economics. 5
(2), 266-288.
Alchian, A. A. & Demetz, H. (1972).
Production, information costs
and economic organization.
American Economic Review.
62, 777-795.
Baraldi, E. (2003). When Information
Interaction Using IT tools to
155
Handle Products at IKEA and
Edsbyn. Doctoral Thesis,
Studies, Uppsala University
Chang, P. L., Shih, C. T. & Hsu, C. W.
(1994).
The
formation
process of Taiwan’s IC
industry
method
of
technology
transfer.
Technovation. 14 (3), 161171.
Chang, P. L. & Hsu, C. W. (1998). The
development strategies for
Taiwan’s
semiconductor
industry. IEEE Transaction
on
Engineering
Management. 45 (4), 349356.
Chang, P. L. & Tsai, C. T. (2000).
Evolution
of
technology
development strategies for
Taiwan’s
semiconductor
industry:
Formation
of
research consortia. Industry
and Innovation. 7 (2), 185197.
Chen, C. F. & Sewell, G. (1996).
Strategies for technological
development in South Korea
and Taiwan: The case of
semiconductors. Research
Policy. 25 (5), 759-783.
Chen, P. C. (1999). The role of industrial
policy
in
Taiwan’s
development. In Thorbecke,
E. & Wan, H. Taiwan’s
Development
Experience;
Lessons
on
Roles
of
Government and Market.
Norwell: Kluwer Academic
Publishers.
Chiang, J. T. (1990). Management of
national
technology
programs
in
a
newly
industrializing
country
Taiwan. Technovation. 10
(8), 531-554.
Dubois, A. & Araujo, L. (2004). Research
methods
in
industrial
marketing
studies.
In
Håkansson, H., Harrison, D.
& Waluszewski, A. (Eds.).
Rethinking
Marketing:
Developing
a
New
Understanding of Markets.
The IMP Journal
Volume 4, number 2
Chichester: John Wiley &
Sons..
Easton, G. (1998). Case research as a
methodology for industrial
networks: A realist apologia.
In Naudé, P. & Turnbull, P.
(Eds.). Network Dynamics in
International
Marketing.
Oxford: Pergamon Press.
Edquist, C. (Ed.). (1997). Systems of
innovation;
Technologies,
Institutions,
and
Organizations. Pinter.
Eklund, M. (2007). Adoption of the
Innovation System Concept
in Sweden. Doctoral Thesis,
Department of Economic
History, Uppsala University
Ernst, D. (1997). What permits David to
defeat
Goliath?
The
Taiwanese model in the
computer
industry.
International
Business
Economics Research Paper
Series
Ford, D., Gadde, L-E., Håkansson, H. &
Snehota,
S.
(2003).
Managing
Business
Relationships.
Chichester:
Wiley & Sons..
Snehota, S. & Waluszewski,
A. (2009). Business in
Networks. Chichester: Wiley
& Sons.
Gadde, L.E. & Håkansson, H. (2001).
Supply Network Strategies. London:John
Wiley.
Gressetvold,
E.
(2004).
Product
Development – Effects on a
Company’s
Network
of
Relationships.
Doctoral
Thesis. Norwegian University
of Science and Technology
Harrison, D. & Waluszewski, A. (2008).
The development of a user
network as a way to relaunch an unwanted product.
Research Policy. 37 (1), 115130.
Hsu, J. Y. & Cheng L. L. (2002). Revisiting
Economic Development in
Post War Taiwan: The
Dynamic
Process
of
Geographical
156
Industrialization.
Regional
Studies. 36 (8), 897-908.
Hughes, T. P. (1987). The evolution of
large technical systems. In
Bijker, W., Hughes, T. P. &
Pinch, T. J. (Eds.). The
Social Construction of Large
Technological
Systems.
Cambridge, MA: MIT Press.
Hung, S. C. (1999). Policy system in
Taiwan’s industrial context.
Asia Pacific Journal of
Management, 16 (3), 411428.
Developing Relationships in
Business Networks, London:
Routledge.
Håkansson, H. & Waluszewski, A. (2002).
Managing
Technological
Development. IKEA, the
Environment
and
Technology.
London:
Routledge.
Håkansson, H. & Waluszewski, A. (Eds.).
(2007).
Knowledge
and
Innovation in Business and
Industry. The importance of
using
others.
London:
Routledge.
Hägg, I. & Johanson, J. (Eds.). (1982).
Företag
i
Nätverk.
Studieförbundet
Näringsliv
och Samhälle. Stockholm.
IMF (2008). Nominal GDP list of countries.
World Economic Outlook
Database.
International
Monetary Fund.
Jahre, M., Gadde, L-E., Håkansson, H.,
Harrison, D. & Persson, G.
(Eds.). (2006). Resourcing in
Business Logistics – The Art
of Systematic Combining.
Sweden: Liber AB.
Lin, M. K. & Trappey C. V. (1997). The
development of Taiwan’s
integrated circuit industry.
IEEE
Transactions
on
Components, Packaging and
Manufacturing Technology Part C. 20 (4), 235-242.
Lin, Y. (2003). Industrial Structure,
Technical Change, and the
Role of the Government in
Development
of
the
The IMP Journal
Volume 4, number 2
Electronics and Information
Industry in Taipei, China.
Asian Development Bank.
ERD Working paper no. 41.
Liu, C. Y. (1993). Government’s role in
developing a high tech
industry:
The
case
of
Taiwan’s
Semiconductor
industry. Technovation. 13
(5), 299-309.
Liu, T. H., Chu,Y. Y., Hung, S. C. & Wu S.
Y.
(2005).
Technology
Entrepreneurial styles: a
comparison of UMC and
TSMC. International Journal
of Technology Management.
29 (1/2), 92-114.
Lo, C. L. & Tung L. R. (1997). Integrated
packaging
industry
in
Taiwan. IEEE Transactions
on Components, Packaging
and
Manufacturing
Technology - Part C. 20 (4),
243-255.
Lundvall, B-Å. (1988). Innovation as an
interactive process: from
user-producer interaction to
the
national
innovation
systems,
In
Dosi,
G.,
Freeman, C., Nelson, R.R.,
Silverberg, G. & Soete, L.
(Eds.). Technical Change
and
Economic
Theory.
London: Pinter.
Lundvall, B-Å. (2007). National Innovation
System: Analytical Focusing
Device and Policy Learning
Tool.
Working
paper
R2007:004. ITPS.
Mathews, J. (1997). A Silicon Valley of the
east:
Creating
Taiwan’s
semiconductor
industry.
California
Management
Review. 39 (4), 26-54.
Mathews, J. & Cho, D. S. (2000). Tiger
Technology - The creation of
a Semiconductor Industry in
East
Asia.
Cambridge:
MOEA (2003). Introduction to Investments
in the Biotechnology and
Pharmaceutical Industries in
Taiwan, ROC. Ministry of
Economic Affairs, BPIPO.
Taiwan, ROC.
MOEA
157
(2005).
Introduction
to
Biotechnology
and
Pharmaceutical Industies in
Taiwan, Republic of China.
Ministry of Economic Affairs,
BPIPO. Taiwan, ROC.
OECD (1996). The Knowledge Based
Economy. Organisation for
Economic Co-operation and
Development. Paris.
OECD (2007). Innovation and Growth
Rationale for an Innovation
Strategy. Organisation for
Economic Co-operation and
Development. Paris.
Penrose, E. (1959). The Theory of the
Growth of the Firm. New
York: John Wiley.
Peng, B. W., Chen, H. G. & Lin, B. W.
(2006). A Taiwan research
institute as a technology
business incubator – ITRI
and
its
spin-offs.
Comparative
Technology
Transfer and Society, 4 (1),
1-18.
Porter, M. E. (1990). The Competitive
Advantage
of
Nations.
London: Macmillan.
Porter, M. E. (1998). Clusters and the New
Economics of Competition.
Harvard Business Review.
76 (6), 77-90.
Rosenberg, N. (1982). Learning by using,
In Rosenberg, N. Inside the
Black Box: Technology and
Economics.
Cambridge:
Saghafi, M. M. & Davidson, C. S. (1989).
The new age of global
competition
in
the
semiconductor
industry:
enter the dragon. Columbia
Journal of World Business.
24 (4), 60-70.
Saxenian, A. L. & Hsu J. Y. (2001). The
Silicon Valley – Hsinchu
connection:
Technical
communities and industrial
upgrading. Industrial and
Corporate Change. 10 (4),
193-202.
Saxenian, A. L. (2002). Taiwan Hsinchu
region: imitator and partner
The IMP Journal
Volume 4, number 2
for
Silicon
Valley.
Unpublished Saxenian.
Sher P. J. & Yang P. Y. (2005). The
effects
of
innovative
capabilities
and
R&D
clustering
on
firm
performance: the evidence of
Taiwan’s
semiconductor
industry. Technovation, 25,
33-43.
Shih, T. (2009). Scrutinizing a Policy
Ambition to Make Business
Out of Science – Lessons
from
Taiwan.
Doctoral
Thesis,
Department
of
University.
Strömsten, T., & Håkansson, H. (2007).
Resources
in
use:
Embedded
electricty.
In
Håkansson,
H.,
&
Waluszewski,
A.
eds.
Knowledge and Innovation in
Business and Industry. The
importance of using others.
London: Routledge.
Sturgeon, T. J. (2000). How Silicon Valley
Came to Be, In: Kenney, M.
(Ed.). Understanding Silicon
Valley. The Anatomy of an
Entrepreneurial
Region.
Standford:
Stanford
University Press.
Swirbanks, D. & Cyranoski, D. (2000).
Taiwan backs experience in
quest for biotech success.
Nature. 407, 417.
Teigland, R., Lindqvist, G., Malmberg, A. &
Waxell,
A.
(2004).
Investigating the Uppsala
Biotech
Cluster.
Cind
Research Paper 2004:1.
The Economist (2007). Special report on
innovation: The fading lustre
of clusters.
October 13,
2007.
TSIA (2007). Overview on Taiwan
semiconductor
industry.
Taiwan
Semiconductor
Industry Association. Taiwan,
ROC.
Tidd, J., Bessant, J. & Pavitt, K. (2005).
Managing
Innovation:
Integrating
Technological,
Market and Organizational
158
Change (3rd edition). Oxford:
John Wiley and Son.
Tu, Y. Y. (2001). The electronics and
information
technology
industries. In Hsueh, L. M.,
Chen, K. H. & Perkins, D.
Industrialization
and
the
State – The Changing Role
of the Taiwan Government in
the
Economy.
Boston:
Harvard University Press.
Van de Ven, A., Polley, D., Garud, R. &
Venkataraman, S. (1999).
The Innovation Journey. New
York:
Oxford
University
Press.
Von Hippel, E. (1988). The Sources of
Innovation.
New
York:
Wade, R. (1990)0 Governing the Market:
Economic Theory and the
Role of Government in East
Asian
Industrialization.
Princeton:
Princeton
University Press.
Waluszewski, A. (2004a). A Competing or
co-operating cluster or seven
decades of
combinatory
resources? What’s behind a
prospering biotech valley?
Management. 20, 125-150.
Waluszewski, A., Baraldi, E., Linné, Å. &
Shih, T. (2009). Resource
interfaces telling other stories
about the commercial use of
new
technology:
The
embedding
of
biotech
solutions in US, China and
Taiwan. IMP journal. 3 (2),
86-123.
Wedin, T. (2001). Networks and Demand The Use of Electricty in an
Industrial Process. Doctoral
Thesis,
Department
of
University.
World Bank. (1993). The East Asian
Miracle: Economic Growth
and Public Policy. Oxford:
Yang, W. T. & Lee, W. H. (2000). The
cradle of Taiwan high tech
development
Hsinchu
The IMP Journal
Volume 4, number 2
science park. Technovation.
20 (1), 55-59.
Yin, R. (1994). Case study research:
Design and methods (2nd
edition). Beverly Hills, CA:
Sage Publishing.
Yu, M. F., Hung, S. C., Lin, B. W. & Shih
C. T. (2006). ITRI in
Taiwan’s
Semiconductors:
Fueling
New
Industry
Formation. Paper presented
at IAMOT.
Öberg, C. & Brege, S. (2009). What
Happened
with
the
Grandiose Plans? Strategic
Plans and Network Realities
in B2B Interaction. Paper
presented
at
IMP
Conference in Marseille,
2009.
159
The IMP Journal
Volume 4. Issue 3
A letter from the guest editor
At the 26th IMP Conference in Marseille, The IMP Journal announced a Special Issue on
Distribution Networks. Potential authors were requested to provide extended abstracts in the
late fall 2009 and the first full paper versions in the beginning of 2010. Papers were then
discussed at the IMP Journal Seminar in Lugano, April 9-10. After further review procedures
the following papers were accepted for publication in this Special Issue:
Kajsa Hulthén and Lars-Gunnar Mattsson:
Distribution network dynamics: Evolution in the PC distribution network
Leif-Magnus Jensen:
Opportunities and constraints for intermediaries in distribution: The challenge of variety
Chiara Cantù, Francesca Montagnini and Roberta Sebastiani:
Organizing a network within the network: The case of MC Elettrici
In addition, the editor-in-chief of the IMP Journal, Håkan Håkansson, asked for an extended
editorial showing how these publications are positioned in a distribution network context and
in which ways they contribute to enhanced understanding of this business reality. For this
reason the Special Issue is introduced by the following paper:
Lars-Erik Gadde:
Distribution network evolution – challenges for practice and theory
Enjoy the reading
Gothenburg, December 2010
Lars-Erik Gadde
The IMP Journal
Volume 4. Issue 3
160
Distribution network evolution
– challenges for practice and theory
Lars-Erik Gadde
Industrial Marketing, Chalmers University of Technology, Gothenburg. Email: largad@chalmers.se
Abstract
The aim of this editorial is to provide a context for the three following papers dealing with distribution networks
from an IMP perspective. All three papers are concerned with distribution dynamics and reorientation of
established arrangements. In this editorial these current challenges are explained in terms of long-term evolution
of distribution leading to increased specialisation and fragmentation in both practice and theory. The conclusion
of the paper is that industrial network theory is a relevant framework for regaining the holistic perspectives that
once characterised both practice and theory in distribution.
Keywords: Distribution networks, distribution dynamics, marketing-logistics integration.
1. Introduction
This special issue of the IMP Journal focuses on
distribution networks and in particular on the
dynamics of these arrangements. The industrial
network approach (INA) has been used in a huge
number of studies of technological development
and business relationships in customer and supplier
networks. This framework is also considered of
relevance for the scientific and practical field identified as ‘distribution’ (see, for example, Gripsrud,
2004; Gadde and Ford, 2008). Despite this potential
usefulness only scant attention has been devoted to
this particular area where the channel concept is
totally dominant.
The overall purpose of the special issue is to illustrate the applicability of the industrial network
approach (Håkansson and Snehota, 1995) and its
three layers: activities, resources and actors (the
ARA model), for the analysis of distribution. For
this special issue the editor of the IMP Journal
wanted an extended editorial and asked me to
review the distribution literature with the aim of
showing the relevance of this framework for studies
of distribution.
The editorial begins with an overview of some of
the major challenges that are currently facing
distribution in both practice and theory. A multitude
of papers indicate that most firms are reconsidering
their distribution arrangements. In order to illustrate
the usefulness of the network approach for analysis,
interpretation and handling of these challenges, we
continue with an historical exposé of the evolution
of distribution. This exploration shows that distribution was originally considered from a holistic
perspective on society, involving what is identified
today as marketing and logistics as two subfields
within this broader context. For reasons explained
below this holistic view dissolved in practice, and
marketing and logistics became separate company
departments. Concerning theoretical issues, several
schools of thought focusing on isolated aspects of
the distribution totality replaced the holistic
frameworks. In the final section we explain how a
network approach would be a useful framework in
today’s distribution landscape and show how the
papers in this special issue contribute to an
enhanced understanding of this business reality.
2. Current distribution challenges
The ongoing reorientation of distribution was
initiated some 25 years ago. One article from 1987
in the Harvard Business Review marks this change,
claiming that distribution is a neglected aspect of
strategy and that many companies “are reaching
their markets in similarly outmoded ways” (Stern
and Sturdivant, 1987: 34). However, another
finding in their study was that “in contrast a number
The IMP Journal
Volume 4. Issue 3
of companies have outperformed their competition
with imaginative strategies for getting products to
their customers.” This reorientation was considered
complicated because “of all marketing decisions the
ones regarding distribution are the most farreaching and difficult to change.”
A decade later an article in the same journal
argued that “forward-looking companies are experimenting with their channels to make them more
flexible and responsive” (Narus and Anderson,
1996: 112). The authors concluded that “tremensdous innovation in distribution” had begun and that
a number of companies “have outstripped their
competitors with imaginative strategies.” One of
the main challenges confronting firms at the time
was the growth in global competition, which put
virtually all industries under “increasing pressure to
cut costs, while at the same time providing even
greater customer service” (Oswald and Boulton,
1995: 138). For these reasons suppliers in general
were being forced to “reconsider fundamental
assumptions about how they reach their markets”
(Anderson et al., 1997: 59).
These fundamental assumptions concern several
issues, each one imposing its particular challenge
since it calls for rethinking of established formulas.
First, one of the most significant changes in
distribution is the increasing attention to postponement at the expense of speculation (van Hoek,
2001). Mass production, mass distribution and huge
inventories of standardised offerings have long
ruled the distribution arena. Over time, however,
individualised solutions in relation to specific
customers have become increasingly important and
suppliers are therefore involved in designing
differentiated distribution arrangements (Gadde,
2000; Cao et al., 2006).
The second challenge is that these changes imply
a shift from the established view of distribution as
starting with the selling firm, seen “through the
eyes of marketing management in production
firms” and mainly dealing with the route taken by a
product as it moves from the producer to the user
(Rosenbloom, 1996:5). Instead, when customisation
and differentiation become critical, the analysis of
these arrangements must ‘start with the customer’
(Holweg and Pil, 2001). Of particular importance
for customised distribution are just-in-time deliveries (Kaneko and Nojiri, 2008) and build-to-order
production (Gunasekaran and Ngai, 2005). The
central features of these arrangements lead to the
third challenge, since they require reorganisation of
established structures. For these reasons distribution
activities have become increasingly interdependent
because buffering inventories are being replaced by
more or less seamless flows.
Such conditions call for enhanced coordination
both within and between firms. Concerning internal
coordination, the performance of distribution
services is dependent on well-developed interfaces
161
between the firm’s marketing and logistics
functions. The fourth challenge is that close crossfunctional collaboration between marketing and
logistics is not always at hand. The absence of these
conditions “may result in promises made by the
firm's sales force that have not been coordinated
with logistics, promotions that are not synchronised
with delivery schedules, and failure to deliver the
product in a specific, requested format because it is
not the most efficient way to do so” (Ellinger, 2000:
86). Moreover, owing to enhanced outsourcing,
these interdependences tend to cross corporate
boundaries, making distribution coordination an
inter-organisational issue. These conditions imply a
fifth challenge since efficiency in operations calls
for coordinated action and partner-specific adjustments among companies. Achieving this state of
affairs requires considerable rethinking since relationships in distribution have normally featured
transactional processes based on arm’s length
arrangements (Hoyt and Huq, 2000). A movement
towards cooperative alliances is suggested as a
means of addressing “the never-ending battle to
reduce costs while maintaining and improving
quality and service” (Wagner et al., 2002: 254).
3. The roots of the challenges
In order to further analyse these challenges we
need to understand their underlying causes, so an
exploration of the evolution of distribution is
needed. Historically, distribution was not perceived
as something starting in the factory of a manufacturer of finished products. Business history tells
another story and claims that the development of
society has been driven by the distributive trades
and their specialisation over time (Heilbroner,
1962; Hicks, 1969). Before the industrial revolution, the scene was long dominated by the ‘grand
distributor’ who sold all types of products and
performed all the basic commercial functions. The
grand distributor was “an exporter, wholesaler,
importer, retailer, ship-owner, banker and insurer”
(Chandler 1977: 15).
During the nineteenth century, distribution
structures changed significantly. Various types of
enterprises specialising in some of these activities
appeared; banks, insurance companies, common
carriers, etc. Merchants began to specialise in one
or a few lines of goods such as cotton, wheat, dry
goods and others. Furthermore, they tended to
“concentrate on a single function: retailing, wholesaling, importing, or exporting” (ibid.). These firms
jointly linked what was identified as the technology
of production with the technology of use. This
bridging concerned actions necessary to secure that
“an end-product is placed in the hands of the
ultimate consumer” which required the undertaking
of all activities “back to conglomerate resources”
The IMP Journal
Volume 4. Issue 3
(Alderson, 1965: 92). These conditions provided a
holistic perspective on distribution where ‘massmarketers’ played the most significant role
(Chandler, 1977).
In other industries “the mass producer rather
than the mass marketer took over the role of
coordinating the flow of goods” (Chandler 1977:
239). Complex high-volume processes of production required manufacturers’ control of these
operations. These conditions were the origin of the
in-house sales organisations of manufacturing companies, which seem to have been established for the
first time in the mid-1800s (Bucklin, 1972). This
was a necessary step for mass production firms
because existing intermediaries were unable to sell
and distribute the manufactured products in the
volumes in which they were produced. Other
manufacturers of industrial goods had their own
reasons for engaging in sales and distribution.
Producers of complex, advanced machinery needed
specialised marketing services such as demonstration, installation, etc., which existing middlemen
had neither the interest in providing nor the
competence to supply (Chandler, 1977).
According to Chandler (1962: 28), enterprises
developing technologically advanced durable
products “became dissatisfied with their existing
sales channels.” The marketing of electric lighting,
power machinery and traction equipment demanded
highly trained salesmen who understood the needs
of their customers. Through a direct sales organisation, covering major accounts, manufacturers
gained access to “direct technical interchange,
direct price negotiations and market feedback”
(Corey et al., 1989: 223). However, these direct
sales organisations could cover only large accounts
in geographically concentrated markets. Therefore,
manufacturers had problems serving two types of
customers – the small users and the geographically
dispersed ones. This paved the way for the
industrial distributor, who entered the arena in the
second half of the 1800s.
Thus, in the early twentieth century, distribution
involved several types of organisations for the
division of labour of activities “performed by
middlemen, but often to a greater extent by
producers themselves [and] it should be noted that
the final consumer performs part of the functions”
(Weld 1917: 306). These functions were categorised in different ways. For further exploration of
distribution evolution we rely on Clark (1922), who
distinguished between three major functions:
- exchange functions (buying and selling),
- physical distribution functions (storing and
transportation),
- facilitating functions (financing, risk-taking,
standardisation and market information).
The features of that which is exchanged in
buying and selling determine “the most appropriate
162
and economical method for distributing it” (Aspinwall 1956: 1). According to Aspinwall, effective
strategies in these respects are contingent on the
interplay between “the manner of physical distribution and the manner of promotion.” This interplay
is critical and problematic since “the movement of
goods and the movement of information are obviously quite different processes.” These conditions
lead to a separation of the two processes which
were originally closely integrated within the selling
firm. In the mid-twentieth century, efficiency in the
movement of goods became increasingly significant
and resulted in physical distribution being recognised as a separate organisational function (Heskett
et al., 1964).
Over time this function became increasingly
strategic and evolved towards a management discipline in a broader sense under the new label of
logistics (Ballou, 1978). The outcome of these
changes was that “marketing and logistics which
were initially closely linked drifted apart” (Juttner
et al., 2007: 379). This drifting apart had a considerable impact on distribution issues at the selling
company as is discussed below, concerning the
separation of marketing and logistics into the ‘two
halves of distribution’ (Converse, 1958). This
disconnecting affected the interplay between the
flows of goods and information, examined below in
terms of the separation of the movement of goods
and the movement of information.
4. The separation of marketing and logistics
Converse (1958) claimed that this separation into
‘two halves’ was counterproductive. In a similar
vein Bartels (1987: 25) argued that the segregation
of the negotiating function from the physical
distribution function was problematic since
customer satisfaction is “as dependent upon
availability of physical products, parts, and services
as it is upon psychological satisfactions.” The
dissolving of these activities into two separate
departments (marketing and logistics) over time
proved to become a severe drawback. One example
is the claim by Murphy and Poist (1996: 16) that
“poor coordination between marketing and logistics
can be quite dysfunctional to corporate operations
and also have negative effects on corporate
performance.”
These effects tend to appear owing to the lack of
understanding about how the coordinated efforts of
marketers and logisticians impact on the ability to
deliver appropriate service levels in relation to
customers (Rinehart et al., 1989). The problems
related to this lack of understanding can be
illustrated from both sides. On the one hand “the
functions grouped under marketing management
had little, if anything, to do with physical distribution processes and costs” (Bartels, 1976: 24). On
The IMP Journal
Volume 4. Issue 3
the other hand “marketing issues, such as customer
service [...] have not appeared to be the principal
concern of physical distribution managers“ (Schary
and Becker, 1973: 249). On the whole, these conditions did not foster cooperative relationships
between the two corporate departments and “a
number of studies indicate that this interface is
often fraught with conflict and disagreement”
(Murphy and Poist, 1992: 21).
Enhanced influence from the marketing
management approach (Alexander at al., 1940) and
particularly the marketing-mix-model (McCarthy,
1960), reduced the previously broad perspective on
distribution and the linkages between technology of
production and technology of use. This shift is
described as a movement “from a system-wide
perspective to a focus on how the channel captain
should behave to secure an efficient distribution of
his products” (Gripsrud, 2004: 195). Bartels (1976)
concluded that this concentration to the marketing
concept neglected the prosaic responsibilities
comprising transfer, storage and inventory costs.
From a logistic perspective, this reorientation was
perceived as determined by the emphasis put by
marketing on sociological and psychological
disciplines, important to product promotion and
development (Bowersox et al., 1968). On this basis
“the physical handling of goods seem to be pretty
much overlooked by executives” (Schary and
Becker, 1973: 247).
A common view in the mid-twentieth century
was that “firms had to carry out logistics just to be
in business [and] all too often the activities were
treated as cost absorbing” (Ballou 1992:10). Similar
thoughts were expressed by Drucker (1962: 103) in
the claim that “physical distribution is a cost area
and purely a cost area”. This view of logistics was
even prevalent mid-1980s when it was concluded
that “for much of the U.S. industry logistics
remains a secondary class citizen and a logistics
strategy at best an afterthought” (Shapiro 1984:
126). However, the author concluded that owing to
changing conditions this view of logistics as “a
necessary but uninspiring collection of apparatus
[...] can no longer be tolerable.” Similar arguments
were raised by a logistics advocate, who argued that
in prevailing business situations “mastery of
logistics has become an essential ingredient of
competitive success” (Sharman, 1984: 71).
Once logistics became a strategic function the
logistics perspective changed “from a relatively
compartmentalised orientation toward a relatively
integrated one” (McGinnis et al., 1994: 299). This
shift emphasised the interfaces to other disciplines,
including marketing, and the need for linkages.
Kent and Flint (1997: 25) argued that logistics was
becoming considered “a critical component in the
strategy of the firm” and the problem in focus was
how to link together “the whole supply chain and
163
create value for the consumer and be competitive in
world markets.” These claims were reinforced over
time by, for example, Martin and Grbac (2003) and
Svensson (2002). Despite these proclamations,
integration between logistics and marketing
remains an unresolved issue since ”marketing
academics have been slow to rise to the occasion of
combining logistics research into their studies of
channel systems” (Alvarado and Kotzab, 2001:
184), and the fact that “many companies have failed
to realise that supply chain coordination is not
possible without an adequate understanding of
demand” (Mentzer and Moon, 2004: 45). The call
for enhanced coordination between those handling
the demand side and those handling the supply side
in the company has therefore been repeated by
Juttner et al. (2007) and others. Recently, a special
issue dealing with the connections between marketing and supply chain management concluded
that despite the progress made “the nature and
implications of the interrelationship of marketing
and SCM have not been explored at great length”
(Mentzer and Gundlach, 2010: 1).
5. The separation of the movement of goods and
the movement of information
The industrial distributors who became
important around 1900 were more specialised than
traditional intermediaries, thus emphasising the
development towards more specialised actors, as
identified by Chandler (1977). Corey et al. (1989:
253) described how, in the early 1900s, distributors
emerged “whose businesses were defined in terms
of user industries”. Examples of such intermediaries were found in steel, mining, textiles and
shipbuilding. Other firms specialised in certain
product categories, for example welding supplies,
cutting tools and industrial chemicals, while yet
others specialised by addressing the service needs
of small industry in local trading areas. These
distributors tended to be involved in all the main
functions defined by Clark (1922): exchange,
physical distribution and facilitating functions. The
characteristics of these firms long remained largely
unchanged. They continued to be involved in broad
ranges of activities. By the end of the twentieth
century this full service intermediary “contacts
customers and makes the product available by
providing necessary supporting services such as
delivery, credit, technical service, repair service,
assembly, and promotion” (Herbig and O’Hara,
1994: 99).
Over time, however, the full-service distributor
has been supplemented and sometimes replaced by
increasingly specialised intermediaries. These new
actors appeared owing to technological developments in logistics and enhanced opportunities for
The IMP Journal
Volume 4. Issue 3
exchange of information. Firms focusing on specific distribution functions are thus improving their
positions in distribution today. Functional specialisation occurs, for example, in warehousing (see
Faber et al., 2002), and logistics and transportation
(Carbone and Stone, 2005). Furthermore, firms
specialising in roles as ‘information brokers’
(Clarke and Flaherty, 2003) and ‘electronic intermediaries’ (Tamilia et al., 2002) have become
significant distribution actors.
These new actors have emerged because they are
able to improve the efficiency in the undertaking of
their specialised activities. By concentrating their
efforts they can invest in sophisticated equipment to
a greater extent than companies involved in the
whole bundle of distribution functions. This
enhanced specialisation and separation of the flows
of information and goods is no surprise in itself.
Aspinwall (1956) concluded that the two are
obviously quite different processes and may
therefore be tricky to handle within one and the
same firm. However, the separation of the two can
be expected to lead to problems since, as he also
claimed, effective strategic action depends on the
interplay between the two. Efficient and effective
industrial arrangements thus require specialised
activities to be integrated and form a coherent
totality (Håkansson et al., 2009). The relationship
between specialisation and integration is quite
straightforward, implying that the greater the
specialisation, the greater the need for integration.
One consequence of the increasing specialisation is therefore that new types of actors have
emerged with specific competence in coordination
of the operations of specialised firms. For example,
in logistics and transportation, large-scale logistics
service providers connect the various physical
distribution activities of a huge cadre of small firms
(Marasco, 2008) while manufacturing companies
like Flextronics coordinate the operations of many
specialised production subcontractors (Bitran et al.,
2007). Another significant example is the Li &
Fung Group, which evolved from being a textile
trade broker into a coordinator of whole networks
of private label apparel firms in Europe and North
America (Bitran et al., 2007). Li & Fung integrate
the processes in the supply chain from raw material
sourcing to allocation of production to factories,
control of manufacturing operations, shipping
consolidation, customs clearance, and also local
forwarding logistics.
These conditions imply that separation in order
to promote efficiency in one part of an activity
structure (or in one flow) must be accompanied by
integration in another, since specialised activities
have to be “re-assimilated into a cognitive form
where they have economic meaning” (Piore, 1992:
442).
164
6. Theoretical approaches to distribution
Similar to the evolution of distribution in
practice, the development of the theoretical
conceptualisations of distribution are characterised
by an increasingly narrow focus. Early frameworks
employed a holistic perspective of distribution
illustrated by the conclusion in an overview of
schools of thought in marketing: “Functions were
concerned with ‘what’ marketing activities were
performed. Institutions with ‘who’ performed them.
Commodities dealt with ‘how’ products (and
occasionally services) were distributed” (Shaw and
Jones, 2003: 39). The frameworks within these
three research fields were mainly descriptive and
are considered as “largely a-theoretical enumerations and classifications” (Wilkinson, 2001: 27).
These conditions continued for a long time with a
few notable exceptions discussed by Wilkinson. For
example, Breyer (1934) introduced the idea of
marketing flows and tried to apply a holistic view
of marketing by providing a systems oriented
approach to the conceptualisation of channels.
These ideas did not receive widespread attention,
and not until the 1950s did more systematic
theoretical framing of distribution occur.
The main contributions in this respect were two
books by Wroe Alderson (Alderson, 1957, 1965),
still considered “unquestionably the most fully
developed theoretical exposition of marketing up to
that time” (Bartels, 1988: 238). Furthermore, the
functional approach refined by Alderson has been
perceived as having contributed most to the
development of marketing as a science (Hunt and
Goolsby, 1988). Alderson developed a distribution
framework based on the evolving system theory
and stated that his approach “stresses the whole
system and undertakes to interpret the parts in terms
of how they service the system” (Alderson 1957:
16). The main task of the system is to “bring
together heterogeneous supply on the one hand and
heterogeneous demand on the other” (Alderson
1965: 20). This was achieved by overcoming the
discrepancies of assortments that separated the
conglomerations of resources found in nature from
the meaningful assortments of goods and services
demanded by society (Wilkinson, 2001), and by
closing the time, place and form gaps between
provider and user.
Alderson’s perspective was thus a system of
action of which individual firms were elements and
where the activities of one firm supplemented the
activities of others. These arrangements required
sophisticated means of coordinated action since
such utility-creating processes “cannot function
without sustained cooperation in which each party
knows what to expect from its opposite number”
(Alderson 1965: 239).
The IMP Journal
Volume 4. Issue 3
However, this organised system was not
designed and managed by a ‘channel captain’, but
was instead the outcome of the joint actions of
many firms. Space does not allow here for a
comprehensive description of Alderson’s contributions, but it should be clear that Alderson’s work
functioned as a source of inspiration for the
development of the industrial network model, as
explained in greater detail in Gadde and Ford
(2008).
Although Alderson’s theoretical contributions
are recognised as highly significant, his view of
distribution never gained the attention and impact it
deserves. In the same way as critical elements of
distribution reality ‘drifted away’ from each other,
Alderson’s holistic approach was outperformed by
theories covering bits and pieces of this totality. As
mentioned above, marketing management models
emphasised the channel captain concept that
directed the attention to the efficient flow of goods
out of the factory of the manufacturing firm.
Various forms of middlemen are involved in these
channels and these firms have both common and
contradictory objectives in relation to manufacturers. In order to explore and explain these
conditions, theories from the behavioural sciences
were introduced to the study of distribution. The
first major contribution in this respect was that of
Stern (1969), applying a social system perspective
to relations between firms in distribution, in terms
of role theory, power, conflict and communication.
The studies in this area were originally overly
focused on power and conflict (Wilkinson, 2001).
Over time they expanded to cover non-confrontational dimensions such as commitment and trust
(e.g. Kumar, 1996) as well as recognising the
cooperative aspects of power and conflict (for an
overview see Gadde, 2004). When it comes to the
unit of analysis these studies tend to focus on
dyadic conditions, implying a further loss of the
holistic perspective.
Another strong influence emerged from
transaction cost economics (TCE), originating in
the work of Coase (1939) and later refined by
Williamson (1975). The issue of the efficient
organisation of the transaction between separate
businesses is obviously a significant one, appearing
at several levels in a marketing channel. Therefore,
the TCE approach received widespread attention in
distribution research and gradually came to
dominate channel studies (Rindfleish and Heide,
1997) through its focus on a very specific issue
within the wider context of distribution.
The ‘political economy’ approach developed by
Stern and Reve (1980) was an attempt to integrate
the behavioural approach with the economic
analysis important in previous distribution research
(e.g. Bucklin, 1960, 1965). The framework focuses
on the interplay between the internal economy and
165
internal socio-political factors for structures and
processes in the channel in relation to the external
economy and external socio-political factors in the
environment. Despite its merits in terms of applying
a broader perspective on distribution, the interest in
this approach declined in the shadow of the
emerging TCE perspective.
In the late 1970s, a review of the evolution of
distribution research concluded that “every conceptual approach reviewed has added something to
our cumulative knowledge” (Gattorna, 1978: 505).
The same conclusion is valid for the years after
1978 since the behavioural and transaction cost
approaches have added considerably to the understanding of the specific aspects of distribution
networks on which they focus. Another conclusion
concerning distribution research from the same
review was that “one major detraction from the
work so far is the piecemeal way in which it has
been approached” (Gattorna, 1978: 506). As shown
in the examination above, also this statement
remains valid today. Even Gattorna’s recommendation for future research remains in force since he
claimed that “there is an urgent need to explain
entirely new avenues in an effort to achieve a
meaningful conceptualisation of channel structure
and mechanics”. This editorial suggests the
industrial network model as an alternative meaningful conceptualisation, in order to handle the fragmentation characterising distribution research.
7. An industrial network view of distribution
The analysis above shows that alternative
perspectives on distribution are needed for both
practical and theoretical reasons. Concerning
business reality, this paper shows that today’s
companies are engaged in reorientation of their
distribution arrangements. Some conclusions from
studies describing the changes of reality are that
innovative companies can be characterised as
“webs of capabilities embedded in an extended
enterprise” (Narus and Anderson, 1996: 112) and as
“networks of value-adding partnerships like
confederations of specialists” (Anderson et al.,
1997: 59). In both these papers access to the
resources of business partners is highlighted as the
main driving force behind such constellations. The
same factor is emphasised in another study where it
is concluded that these collaborative arrangements
make firms able to “combine resources in new
ways, gain additional resources and dispose of
superfluous resources” (Wilson and Daniel, 2007:
10). An illustration of resource sharing in such
constellations is provided by Stern and Sturdivant
(1987:41) in their example of the ways in which a
specialty grocery products manufacturer distributed
its products onto supermarket shelves:
The IMP Journal
Volume 4. Issue 3
“It was using an array of third-party players,
including food brokers, grocery wholesalers,
and health food distributors, some of whom
carried out a remarkable range of functions
between the manufacturer and the retail level
of the distribution chain. When the company
drew a structured diagram it looked like a
bowl of spaghetti.”
Obviously the ARA model used for analysis of
industrial networks is a highly relevant framework
for investigation of such conditions. Not only is it
helpful for managers in their attempts to handle the
current challenges in practice, since it provides
opportunities for analysis of both marketinglogistics interfaces and the interplay between the
movement of goods and the movement of information. The ARA model also represents a powerful
tool for theoretical recapturing of the holistic
perspective of distribution articulated through the
claim of Gripsrud (2004: 189) that it is “time to
regain lost territory”. It should also be noted that
the argument for more holistic views, exemplified
by industrial network theory, is raised by several
advocates of research approaches that have deliberately narrowed the perspective on distribution.
Such claims have been put forward by representatives of transaction cost economies (Heide,
1994; Wathne and Heide, 2004), political economy
(Acrol et al., 1983), supply chain management
(Christopher, 2005), and agency theory (Rokkan
and Haugland, 2002), among others.
The papers in this special issue are excellent
illustrations of the usefulness of the INA and ARA
frameworks for providing alternative conceptualisations of distribution issues. All three deal with
two of the most significant aspects of distribution.
The first is the dynamics and continual evolution of
distribution arrangements called for by changing
conditions. The second is the variety and
differentiation featuring distribution today in all
three network layers. In the first paper, Kajsa
Hulthén and Lars-Gunnar Mattsson analyse the
evolution of the distribution network for personal
computers from the 1970s until today. Their
contribution is highly unusual in that it describes
the development of a distribution network from the
very outset. However, the paper shows that nothing
is entirely new since the first arrangements utilised
distribution resources developed for other products
and services. Hulthén and Mattsson contrast
mainstream distribution literature by analysing the
evolution of distribution in terms of connected
processes in the PC network, while the common
approach in the channel literature is to consider
distribution dynamics as driven by a ‘channel
captain’ in response to external driving forces. The
investigation of the evolution over four decades
emphasises the variety in the actor layer, the inter-
166
dependence among activities and the aspirations for
control in the resource dimension.
Various types of intermediaries are significant
actors in the PC distribution network, and in the
evolution over time their roles have changed. This
issue is the main theme in the second paper, by
Leif-Magnus Jensen, dealing with a logistics
service provider. These firms have been able to
strengthen their positions in contemporary distribution networks through their specialisation in the
flow of materials. The focal company serves car
importers in Norway with transportation and
logistics services. Although the business conditions
of the various customers are similar, their service
demands are different, which calls for variety in the
service offerings of the provider. Jensen shows that
this role variety expands the business opportunities
for providers at the same time as problems appear
in terms of combining the various roles. An
interesting conclusion, in contrast with established
views, is that “the role of an intermediary is
determined less by the resources it controls or the
activities it is capable of undertaking and more by
the resources that others have and the activities that
others are capable or incapable of undertaking.”
Furthermore, in traditional conceptualisations, the
service provider would not be considered a member
of the channel since this firm is not involved in title
holding.
The changing role of intermediaries is also the
theme of the third paper by Chiara Cantù, Francesca
Montagnini, and Roberta Sebastiani. Their study
focuses on a wholesaler in the Italian electrical
equipment industry involved in reorganis-ation of
distribution arrangements owing to changing
conditions. The focal firm is actually in the process
of developing “a network within the network” by
establishing closer connections to some suppliers
and wholesalers/retailers within the larger network.
The expansion in the number of network members
is accompanied by modifications of the services
provided by the central actor, thus illustrating the
variety of roles. In the first phase of the network
evolution, efficient physical distribu-tion was
emphasised, followed in the second phase by
enhanced focus on dissemination of knowledge
from the central actor to network members. In the
third phase the attention is being directed toward
promoting the development of relationships among
network members and establishing a common
community. The consequences of these efforts of
the central actor are illustrated from the perspective
of one of the members of the community. Cantú,
Montagnini and Sebastiani show that the entrance
of an additional intermediate level can improve
distribution performance. This stands in contrast to
the mainstream view of efficient distribution, where
the key seems to be the elimination of middlemen.
The conclusion drawn by the authors is that a new
The IMP Journal
Volume 4. Issue 3
intermediary “will always be able to get a foothold
as long as it can contribute to value generation for
other firms”.
The three papers contribute substantially to our
understanding of distribution from an industrial
network perspective. Specifically, they call for
reinterpretation of the role of intermediation in
distribution networks. Traditionally, intermediation
has been concerned with efficiency in the
transactions between actors in terms of “the effort
to reduce the number of transactions between
producers and final users” (Rosenbloom, 1995: 45).
The papers in this special issue show that an
industrial network view broadens the perspective on
intermediation. The main benefits of intermediation concern the intra- and inter-organisational
linking of activities, tying of resources and bonding
among actors. PC distributors, the logistics service
provider and the network organiser in the electric
industry all illustrate these roles of intermediation.
Having said that we must acknowledge that in the
current business landscape intermediation is no
longer an issue reserved for firms identified as
‘intermediaries’. For example, manufacturers of
PCs may also be seen as involved in intermediation
since they connect activities and resources of
component manufacturers with those of distributors
and end users. In other words, from an industrial
network perspective, all firms are involved in
intermediation (Ford et al., 2003), since every
company is connected to many others. Intermediation in the resource and activity layers is becoming
increasingly important owing to enhanced reliance
on outsourcing and escalating interdependence
among activities. The papers in this special issue
provide interesting illustrations of the enhanced
significance of intermediation.
References
Achrol, R., Reve, T. and Stern, L. (2003). The
environment of marketing channel dyads: A
framework for marketing analysis. Journal of
Marketing, 47 (4): 55-67.
Alderson, W. (1957). Marketing Behaviour and
Executive Action. Homewood, Ill.: Richard, D.
Irwin.
Alderson, W. (1965). Dynamic Marketing
Behaviour. A Functionalist Theory of Marketing.
Homewood, Ill.: Richard D. Irwin.
Alexander, R., Surface, F. and Alderson, W.
(1940). Marketing. Boston: Ginn and Company.
Alvarado, U. and Kotzab, H. (2001). Supply chain
management. The integration of logistics in
marketing. Industrial Marketing Management,
30 (2): 183-198.
167
Anderson, E., Day, G. and Rangan, K. (1997).
Strategic channel design. Sloan Management
Review, 38 (4): 59-69.
Aspinwall, L. (1956). Parallel systems of promotion
and distribution. Cost and Profit Outlook, 9 (10):
1-4.
Ballou, R. (1978). Basic Business Logistics. Englewood Cliffs: Prentice Hall.
Ballou, R. (1992). Business Logistics Management.
Englewood Cliffs: Prentice Hall.
Bartels, R. (1976). Marketing and distribution are
not separate. International Journal of Physical
Distribution, 7 (1): 22-29.
Bitran, G., Gurumurthi, S. and Sam, S. (2007). The
need for third-party coordination in supply chain
governance. MIT Sloan Management Review, 48
(3): 30-37.
Bowersox D., Smykay, E. and LaLonde, B. (1968).
Physical Distribution Management: Logistics
problems of the firm. New York: MacMillan.
Breyer, R. (1934). The Marketing Institution. New
York: McGraw-Hill.
Bucklin, L. (1960). The economic structure of
channels of distribution. In: Bell, M. (ed.)
Marketing: A maturing discipline, 379-385.
Chicago: The American Marketing Association
Bucklin, L. (1965). Postponement, speculation and
the structure of distribution channels. Journal of
Marketing Research, 2 (1): 26-31
Bucklin, L. (1972). Competition and evolution in
the distributive trades. Englewood Cliffs:
Prentice Hall.
Cao, J., Wang, J., Law, K., Zhang, S. and Li, M.
(2006). An interactive service customization
model. Information and Software Technology, 48
(4): 280-296.
Carbone, V. and Stone, M. (2005). Growth and
relational strategies used by the European
logistics providers: Rationales and outcomes.
Transportation Research Part E, 41 (6): 495510.
Chandler, A. (1962). Strategy and Structure.
Chapters in the History of the Industrial Enterprise. Cambridge: The MIT Press.
Chandler, A. (1977). The Visible Hand. The
Managerial Revolution in American Business.
Cambridge: Harvard University Press.
Christopher, M. (2005). Logistics and Supply Chain
Management - Creating Value Adding Networks.
London: Prentice Hall, Financial Times Pearson
Education.
Clark, F. (1922). Principles of Marketing. New
York: McMillan.
Clarke, I. and Flaherty, T. (2003). Web-based B2Bportals. Industrial Marketing Management, 32
(1): 15-23.
Coase, R. (1937). The nature of the firm.
Economica, (4): 386-405.
The IMP Journal
Volume 4. Issue 3
Converse, P. (1958). The other half of marketing.
In: Seyle, A. (ed.) Marketing in Transition. New
York: Harper & Row.
Corey, R., Cespedes, F. and Rangan, K. (1989).
Going to Market. Distribution Systems for Industrial Products. Boston: Harvard Business School
Press.
Drucker, P. (1962). The economy’s dark continent.
Fortune, 72 (4): 103-104.
Ellinger, A. (2000). Improving marketing/logistics
cross-functional collaboration in the supply
chain. Industrial Marketing Management, 29 (1):
85-96.
Faber, N., de Koster, R. and van de Welde, S.
(2002). Linking warehouse complexity to warehouse planning and control structure.
International Journal of Physical Distribution &
Logistics Management, 32 (5): 381-395.
Ford, D., Gadde, L-E., Håkansson, H. and Snehota,
I. (2003). Managing Business Relationships.
Chichester: Wiley.
Gadde, L-E. (2000). From marketing channels to
differentiated networks – distribution dynamics
in a historical perspective. In: Dahiya, S. (ed.)
The Current State of Business Disciplines, 26412662. Rohtak: Spellbound Publications.
Gadde, L-E. (2004). Activity coordination and
resource combining - implications for relationship involvement and the relationship atmosphere. Journal of Marketing Management, 20 (12): 157-184.
Gadde, L-E. and Ford, D. (2008). Distribution
research and the industrial network approach.
IMP Journal, 2 (3): 36-52.
Gattorna, J (1978). Channels of distribution
conceptualizations: A state-of-the-art review.
European Journal of Marketing, 12 (7): 470509.
Gripsrud, G. (2004). The marketing discipline and
distribution research: time to regain lost territory. In: Håkansson, H., Harrison, D. and
Waluszewski, A. (eds.) Rethinking Marketing.
Developing a new understanding of markets.
Chichester: Wiley.
Gunasekaran, A. and Ngai, E. (2005). Build-toorder supply chain management: A literature
review and framework for development. Journal
of Operations Management, 23 (5): 423-51.
Håkansson, H. and Snehota, I. (1995). Developing
Relationships in Business Networks. London:
Routledge.
Håkansson, H., Ford, D., Gadde, L-E., Snehota, I.
and Waluszewski, A. (2009). Business in
Networks. Chichester: Wiley.
Heide, J. (1994). Interorganizational governance in
marketing channels. Journal of Marketing, 58
(1): 71-85.
Heilbroner, R. (1962). The Making of Economic
Society. Englewood Cliffs: Prentice Hall.
168
Herbig, P. and O´Hara, S. (1994). Industrial
distributors in the twenty-first century. Industrial
Marketing Management, 23 (3): 199-203.
Heskett, J., Ivie, R. and Glaskowsky, N. (1964).
Business Logistics. Management of physical
supply and distribution. New York: The Ronald
Press Company.
Hicks, J. (1969). A Theory of Economic History.
Oxford: Oxford University Press.
Holweg, M. and Pil, K. (2001). Successful build-toorder strategies start with the customer. MIT
Sloan Management Review, 43 (1): 74-83.
Hoyt, J. and Huq, F. (2000). From arm’s length to
collaborative relationships in the supply chain.
International Journal of Physical Distribution &
Hunt, S. and Golsby, J. (1988). The rise and fall of
the functional approach to marketing. In: Nevitt,
T. and Fullerton, R. (eds.) Historical Perspectives in Marketing, 35-51. Lexington: Lexington
Books.
Juttner, U., Christopher, M. and Baker, S. (2007).
Demand chain management – integrating marketing and supply chain management. Industrial
Kaneko, J. and Nojiri, W. (2008). The logistics of
just-in-time between parts suppliers and car
assemblers in Japan. Journal of Transport
Geography, 16 (3): 155-173.
Kent, J. and Flint, D. (1997). Perspectives on the
evolution of logistics thought. Journal of
Business Logistics, 18 (2): 15-29.
Kumar, N. (1996). The power of trust in
manufacturer-retailer relationships. Harvard
Business Review, 74 (6): 85-96.
Marasco, A. (2008). Third-party logistics: A
literature review. International Journal of
Production Economics, 113 (1): 127-147.
Martin, J. and Grbac, B. (2003). Using supply chain
management to leverage a firm’s market
orientation. Industrial Marketing Management,
32 (1): 25-38.
McCarthy, E. (1960). Basic Marketing - A
Managerial Approach. Homewood, Ill.: Richard
D. Irwin.
McGinnis, M., Boltic, S. and Kochunny, C. (1994).
Trends in logistics thought: An empirical study.
Journal of Business Logistics, 15 (2): 273-303.
Mentzer, J. and Gundlach, G. (2010). Exploring the
relationship between marketing and supply chain
management: introduction to the special issue.
Journal of the Academy of Marketing Science, 38
(1): 1-4.
Mentzer, J. and Moon, M. (2004). Understanding
demand. Supply Chain Management Review, 8
(4): 38-45.
Murphy, P. and Poist, R. (1992). The logisticsmarketing interface: Techniques for enhancing
cooperation. Transportation Journal, 32 (2): 1423.
The IMP Journal
Volume 4. Issue 3
Murphy, P. and Poist, R. (1996). Comparative
views of logistics and marketing practitioners
regarding interfunctional coordination. International Journal of Physical Distribution &
Narus, J. and Anderson, J. (1996). Rethinking
distribution – adaptive channels. Harvard
Oswald, S. and Boulton, W. (1995) Obtaining
industry control. The case of the pharmaceutical
industry. California Management Review, 38 (2):
138-162.
Piore, M. (1992). Fragments of a cognitive theory
of technological change and organisational
structure. In Nohria, N. and Eccles, R. (eds.)
Networks and Organisations. Structure, form
and action. Boston: Harvard Business School.
Rindfleisch, A. and Heide, J. (1997). Transaction
cost analysis: Past, present, and future applications. Journal of Marketing, 61 (4): 30-54.
Rinehart, L., Cooper, B. and Wagenheim, G.
(1989). Furthering the integration of marketing
and logistics through customer service in the
channel. Journal of the Academy of Marketing
Science, 17 (1) 63-71.
Rokkan, A. and Haugland, S. (2002). Developing
relational exchange – effectiveness and power.
European Journal of Marketing, 36 (1/2): 211230.
Rosenbloom, B. (1995). Marketing Channels. Fort
Worth: The Dryden Press.
Schary, P. and Becker, B. (1973). The marketinglogistics interface. International Journal of
Physical Distribution, Monograph Series, 3 (4):
246-288.
Shapiro, R. (1984). Get leverage from logistics.
Harvard Business Review, 62 (3): 119-126.
Sharman, G. (1984). The rediscovery of logistics.
Harvard Business Review, 62 (5): 71-79.
Shaw, E. and Jones, B. (2003). The history of
schools of marketing thought. Proceedings of
CHARM, volume 11, 38-52. East Lansing:
Michigan State University.
Stern, L. (ed.) (1969). Distribution Channels:
Behavioral Dimensions. Boston: Houghton
Mifflin Company.
169
Stern, L. and Reve, T. (1980). Distribution channels
as political economies: A framework for
comparative analysis. Journal of Marketing, 44
(3): 52-64.
Stern, L. and Sturdivant, F. (1987). Customer
driven distribution systems. Harvard Business
Review, 65 (4): 34-41.
Svensson, G. (2002). Supply chain management:
The re-integration of marketing issues and
logistics theory and practice. European Business
Review, 14 (6): 426-436.
Tamilia, R., Senecal, S. and Corrivecu, G. (2002).
Conventional channels of distribution and
electronic intermediaries: A functional analysis.
Journal of Marketing Channels, 9 (3/4): 27-48.
van Hoek, R. (2001). The rediscovery of
postponement – a literature review and directions for research. Journal of Operations Management, 19 (2): 161-184.
Vorhees, R., Teas, K., Allen B. and Dinkler, E.
(1988). Changes in the marketing logistics
relationship. Journal of Business Logistics, 9 (1):
34-50.
Wagner, B., Macbeth, D. and Boddy, D. (2002).
Improving supply chains relations: an empirical
case study. Supply Chain Management: An International Journal, 7 (4): 253-264.
Wathne, K. and Heide, J. (2004). Relationship
governance in a supply chain network. Journal
of Marketing, 68 (1): 73-89.
Weld, L. (1917). Marketing functions and
mercantile organization. American Economic
Review, 7 (2): 316-318.
Wilkinson, I. (2001). A history of network and
channels thinking in marketing in the 20th
century. Australasian Marketing Journal, 9 (2):
23-52.
Williamson, O. (1975). Markets and Hierarchies:
Analysis and antitrust implications. New York:
The Free Press.
Wilson, H. and Daniel, E. (2007). The multichannel challenge: A dynamic capability
approach. Industrial Marketing Management, 36
(1): 1-20.
The IMP Journal
Volume 4. Issue 3
170
Distribution Network Dynamics:
Evolution in the PC Distribution Network
Kajsa Hulthén and Lars-Gunnar Mattsson
Industrial Marketing, Department of Technology Management and Economics, Chalmers University of Technology,
Gothenburg, Sweden. E-mail: kajsa.hulthen@chalmers.se
Stockholm School of Economics, Stockholm, Sweden. E-mail: lars-gunnar.mattsson@hhs.se
Abstract
The distribution literature has often treated distribution dynamics as either initiated by a producer in the form
of a channel captain or by external forces. In this paper we take a network perspective on distribution dynamics.
In order to illustrate the evolution of a distribution network, we have chosen the distribution network for the
product category personal computers (PCs). The paper shows how changes are initiated from all kinds of actors
in the network and in all three network layers: actors, resources and activities. The paper illustrates how activity
interdependencies, resource control, and variety in the web of actors interplay with distribution dynamics.
Key words: distribution, networks, dynamics, evolution, PC.
1. Introduction
This paper deals with distribution dynamics. It
illustrates how distribution arrangements for a
specific product category evolve over time. We
simply define distribution as the closing of the
space, time and form gaps between production on
the one hand and consumption on the other. In
traditional marketing management literature distribution arrangements are analyzed as distribution
channel issues. In the channel literature business
actors select the preferred distribution arrangement
for their products (e.g. Stern and El-Ansary, 1992).
Dynamics are either driven by a focal channel
member or by conditions external to the channel. In
this paper, however, we take a network perspective.
We believe that it is more fruitful to see dynamics
as a result of interconnected processes in networks.
Networks have wider scope than distribution
channels. In a network perspective, aspects of ‘use’
and ‘production’ as well as ‘distribution’ are considered interdependent. Interconnected network
processes involve a number of different kinds of
firms in the distribution network. Actors that are
not traditionally identified as parts of the
distribution channel, such as third party logistics
providers and firms specializing in information
handling, are also central actors in such a network.
Gadde and Ford (2008: 46) argue that the
traditional channel concept is inadequate for under-
standing what “actually takes place in current
distribution arrangements”. A number of authors
have discussed how modern distribution reality
seems to be more network-like than channel-like.
For example, Andersson (1992: 48) discusses
“distribution systems and processes as embedded in
networks of exchange relationships”. Similarly,
Cox (1999: 211) states that “the process in which
raw materials are turned to end-products and
services is rarely a simple linear process chain, and
much more like a spaghetti web of complex interconnected relationships.” Distribution structures
can be regarded as “networks of value-adding
partnerships like confederations of specialists. They
are flexible, specialized, and emphasize interfirm
relationships, with a pooling of complementary
skills and resources” (Anderson et al., 1997: 61).
We support the belief that a network perspective
facilitates understanding of distribution and its
evolution. A channel perspective on distribution
dynamics implies that by regarding distribution
arrangements as separate channels, virtually as
entities in themselves with common goals,
dynamics is discussed in terms of channels reacting
to outside changes. Along these lines, McCammon
(1964: 75) argues that institutions and marketing
channels “must adapt continuously to the environment in order to avoid economic obsolescence.”
However, every actor involved in channel arrangements has its own goals, strategies, and network
The IMP Journal
Volume 4. Issue 3
contexts, not only linked to one specific channel but
to many types of arrangements. Hence, as argued
by Gadde and Håkansson (1992: 171) “some actors
are interested in stabilizing the present structure
while others will be interested in changing it.” Each
actor always tries to modify its way of working in
order to better fit into the context or, more
radically, change a certain network position. This
means that distribution dynamics can involve both
radical and minor changes (Gadde and Håkansson,
1992).
The evolution of a distribution network of a
certain product category will most likely involve
both radical changes and minor adaptations.
Furthermore, both “kinds of changes might be
found inside as well as outside the present
distribution channel” (ibid: 171). Confronted with a
change in technological development, two firms
might react very differently, owing both to their
present position in the network and their history.
With such a view, a change in distribution is not a
reaction to external forces but can instead be traced
back to very diverse sources related to different
actors, their activities and resources. Furthermore,
as will be shown in this paper, these changes can
also be related to production, distribution, and use
aspects.
In line with the discussion above, the aim of the
paper is to explore how a distribution network
evolves over time. The paper aims to provide an
illustration of long-term development of a network
in contrast to more frequent studies of the developments and changes in single ‘channels’ from a
single manufacturer’s perspective.
2. Distribution dynamics from a network
perspective
As pointed out above, we take a network
perspective on distribution. More specifically we
apply the ARA conceptual framework (Håkansson
and Snehota, 1995). This model has been argued to
be suitable for the study of distribution as a
complex phenomenon and also for the study of
dynamics. With reference to the ARA model, a
distribution network consists of three overlapping
and interrelated layers: a web of actors, a constellation of resources and a pattern of activities.
Activities can be defined on various levels and
include aspects related to sales, transportation,
assembly, materials handling, production, and the
search for offerings, etc. Resources can be products,
facilities such as lorries, inventories, assembly and
production equipment, logistics hubs, etc. The
various actors involved in a distribution network
include firms such as distributors, brand owners,
producers, logistics firms, information specialists,
resellers, retailers, and distributors, etc. There can
also be other types of organizations and users. The
171
actors in a distribution network deploy resources
and carry out activities in order to handle gaps in
space, time and form in the process by which
heterogeneous resources in production are
transformed into heterogeneous products (and
services) for use (Alderson, 1965).
At a specific point in time, the distribution
network has a certain structure with regard to the
organization of activities, resources and actors.
However, a network is never static, nor complete,
and it is never in equilibrium. According to
Håkansson (1992) network evolution is systematic
in the way that actors, resources and activities are
continuously related to each other in novel ways.
Based on this Gadde (1993) distinguishes between
two evolutionary processes in distribution networks. The first concerns how activities are
reallocated among actors, i.e. it concerns changes in
the division of work among actors in the network.
The second deals with the actors’ control over
resources in the network. Furthermore, these two
processes are tightly interlinked, i.e. changes in the
activity pattern will imply changes in the resource
constellation and the other way around.
Evolution signifies a gradual process of change
of an innovative nature with many, interwoven
influencing forces. The network structure at a
certain point in time is a temporary outcome of this
evolution that is of a long term innovative nature,
engaging the three, interconnected layers in the
distribution network. We consider the evolution as
predominantly endogenous network processes and
agree with Alderson when he states that: “in the
transforming society, the marketing system transforms itself.” (Alderson, 1965: 47).
We identify three, in principle different, changes
in a distribution network. First, networks change as
new entities appear and established ones disappear.
Some actors enter the network and others exit. New
resources are introduced and others are removed.
New activities are performed while others are taken
away. Second, the existing entities in the network
might change for some reason. Established actors
change direction of business, resources are changed
radically or adapted slightly and activities are
performed in new ways. Third, the connections
among the entities in the network can change. This
can be discussed as changes in the connection
within each layer in terms of actor bonds, resource
ties and activity links (Håkansson and Snehota,
1995). Furthermore, the connections among entities
of different layers can also change, e.g. how
resources and activities are connected or how actors
and activities are connected. (See Figure 1 for the
analytical model).
The outline of the paper is as follows: First we
provide an empirical illustration of the personal
computer (PC) distribution network and its
evolution in three decades. The case shows how the
network evolves over time as actors in the network
The IMP Journal
Volume 4. Issue 3
Change type 1
Change type 2
- New actors, resources,
activities enter
- Actors, resources, and
activities exit
Established actors, resources
and activities change
- Radical changes
- Minor adaptations
Distribution network time 1
Web of
actors (t1)
Resource
collection (t1)
172
Distribution network time 2
Distribution network dynamics
Activity pattern
(t1)
Resource
collection (t2)
Web of
actors (t2)
Activity
pattern (t2)
Change in the connections
- Within one of the layers
- Among the layers
Change type 3
Figure 1: The analytical model of the paper
try to change their positions. By doing so the way
resources are exploited and combined as well as
how activities are coordinated are affected. Second,
the case is analyzed with regard to dynamics in the
three network layers; the pattern of activities, the
resource constellation, and the web of actors. This
is followed by a discussion on how the interplay
among these layers impacts on dynamics. From this
discussion three themes to discuss further are
identified, each relating to one of the network
layers: variety in the web of actors, interdependencies in the activity pattern, and control in
the resource constellation.
3. Evolution in the ‘PC distribution network’
In order to illustrate the evolution of a distribution network, we have chosen the distribution
network for the product category personal
computers (PCs). We take our point of departure in
the introduction of the microcomputer in the mid
1970s and analyze how the PC distribution network
evolved from that time until 2008, over three
decades. This specific case is interesting since the
distribution network had to handle a new product
category, the personal computer. We recognize that
the distribution network in focus is embedded in the
wider network. However, we have delimited our
network boundary for both analytical and practical
reasons.
The empirical material is collected from
secondary data in terms of articles on PC history.
Fortunately, the PC is a subject of great interest to
academicians, practitioners and other ‘enthusiasts’.
This means that information about the ‘history of
the PC’ is well documented in articles and on
company websites. By studying a large number of
such sources of information, a somewhat unified
picture has evolved. Thus, the history provided in
the following sections reflects this picture as
interpreted by the authors of this paper. Obviously,
we have only been able to examine a limited part of
the overall distribution network and the processes
involved. However, we believe that the short story
presented in the article can be used to analyze some
general aspects of dynamics in distribution
networks.
The vocabulary used to describe the different
actors involved in the network is that used in the PC
industry, inspired by the terminology of Gabrielsson et al. (2002). Distributors by definition do not
sell directly to end-users, but to either resellers or
retailers. Resellers sell to businesses and organizations and can be further divided into sub-groups
such as dealer chains, local dealers, mail-order or
Internet dealers, value added dealers, system
integrators, etc. Retailers sell to consumers and can,
The IMP Journal
Volume 4. Issue 3
for example, be divided into PC superstores,
electronics stores, department stores, catalogue,
mail-order and Internet stores as well as general
merchandisers. The customers are most often
divided into private consumers and business
customers. Sometimes we refer to a ‘dealer’ in a
general sense, as an actor acting in between the
producer and the end-customer.
3.1. Introduction of new products
The introduction of what we know as the
‘original PC’ was preceded by earlier developments
in computers. Between the 1950s and 1970s the
main type of computer was the mainframe
computer, used by large corporations and organizations. The suppliers of mainframe computers at
this time were often referred to as ‘IBM and the
seven dwarfs’ indicating one dominant and seven
smaller suppliers (NetworkWorld, 2010). They sold
their products through their own subsidiaries, with a
high degree of technical competence.
A major event in the history of computers was
the development of the microprocessor in the early
1970s. This new processor could be manufactured
and then programmed to meet various demands.
Earlier, each processor had been designed and
manufactured to fit specific purposes. This development meant that computers were no longer reserved
only for large organizations and governments. In
the mid-1970s microcomputers entered the arena,
giving small companies and households access to
the new technology. In 1975, the first microcomputer was launched by a company named MITS,
with a history in programmable calculators (Langlois, 1992). The calculators had paved the way for
the use of such electronics in households, and
courses in programming had created a customer
base for computers (Ceruzzi, 2003). The (MITS)
Altair 8800 was shipped as kits through mail order
and advertisements or sold at trade fairs.
Later in 1975 the consultancy firm IMS
associates (IMSAI) ordered a number of Altair
8800s in order to set up a computer system for one
of their customers. Since MITS could not deliver
for at least 90 days, IMSAI decided to make their
own computer based on the design of the Altair
8800. This resulted in the IMSAI 8080, partly
compatible with the Altair 8800. In 1975, the first
store specializing in computers also opened, The
Computer Store. They sold Altair 8800 kits but also
peripheral equipment, accessories, expansions,
books, magazines, and repairs. The Computer Store
was a one brand store. Soon other stores appeared,
focusing on selling different brands, such as
Computer Mart. It was soon followed by Byte
Shop, which opened in late 1975. The owner of this
store was soon approached by people who wanted
to open their own Byte Shop stores. He signed
franchise dealership agreements in exchange for a
173
percentage of their sales, and many Byte Shop
stores soon emerged. The first true retail chain
specialized in computers was born. The founder of
IMSAI soon realized that IMSAI was better
equipped to sell computers than to manufacture
them. They started a franchising business called the
Computer Shack. However they had to change their
name to ComputerLand because of the similarity to
the large radio retail chain RadioShack. Computer
Land focused on selling IMSAI computers but also
added other computers to its assortment. Computer
Land was the first full service computer store,
focused on selling everything a customer would
need to support its computer under one roof.
(www.pc-history.org). Soon, there were computer
stores ‘all over’. However, very few early retailers
survived the first decade (Creative Computing,
1984).
In the mid and late 1970s buying a computer
meant entering a retail store for a demonstration by
the seller who spoke ‘computerese’. Hence, the
buyer needed to be very savvy about the technical
aspects of a computer. If the customer decided to
buy a product, 1/3 of the price was left as a deposit
and when the retailer had orders for about five to
ten computers he would place an order with the
manufacturer who bought parts and put the product
together. The first microcomputers came in kits for
the buyer to assemble. According to Langlois
(1990:96) to “accomplish anything, one needed not
just the box itself but also the know-how, add-on
boards, and software provided by a large network
of external sources.” Furthermore, the handling
required very high technical skills since the instructions were very basic and the parts needed to be
partly soldered together. The microcomputers were
more or less only for hobbyists at the time. For
many retailers, the sales of computer magazines
became an important income. (Creative Computing,
1984).
In 1975, IBM, the leading actor for mainframe
computers, had launched its first desktop microcomputer, the IBM 5100. It was a complete system
with keyboard and built-in monitor, aimed at
professionals and scientific problem-solvers, and
not at businesses or consumers. Furthermore, it was
very expensive with a price of up to $20,000. It was
sold by IBM’s established subsidiaries who also
served the large clients with mainframe computers.
These subsidiaries were technologically advanced
and focused on providing total solutions for large
customers with a high degree of technical support.
In 1977, Commodore launched the Commodore
PET 2001, later labeled ‘the world’s first complete
computer’. The price was $795, much less than the
earlier IBM computer. It had an integrated monitor
and keyboard wrapped up in a good looking white
chassis. It was aimed primarily at the business
segment and at early adopters. It was distributed
through authorized office dealers which gave
The IMP Journal
Volume 4. Issue 3
Subsidiaries
174
Fairs
Mail-order
IBM
IMS/Imsai
MITS/Altair
The Computer
Store
Computer Shack/
ComputerLand
Byte Shop
Apple
Commodore
Department
stores
Toy stores
Office dealers
Computer Mart
Tandy
Radio Schack
Producer
End-customer interface
End-customers
Figure 2: Part of the distribution network at the time of the introduction of the microcomputers 1975-1977
Commodore PET a professional image, and offered
qualified technicians and services. Commodore had
a history in typewriters and calculators and had
great advantage of the already established network
developed for distribution of calculators. To
become authorized the dealer needed to have a
service engineer that could provide services, e.g.
repairs, as well as a retail outlet. Soon Commodore
expanded its dealership network to include larger
retail chains supplying a variety of brands,
something that upset the authorized dealers.
Commodore followed up with the launch of the
VIC-20 in 1980, directed to the consumer market.
This product was ‘sold everywhere’ from
authorized dealers providing high levels of services
to department stores and toy stores. (This section is
mainly based on information from www.
commodore.ca)
Apple Computer, founded in 1977, introduced
Apple II, a tremendous success sold through
independent dealers. The first dealer to sell them
was the Byte Shop retailer chain followed by
others, for example ComputerLand. In 1977, Tandy
Corporation introduced the TRS-80, a personal
computer that also became a commercial success.
Tandy had acquired RadioShack in the 1960’s
(Langlois, 1992). They could now use this retail
chain for mass distribution of personal computers.
At this time there were thousands of RadioShack
stores in the United States (Dwyer, 1982). Part of
the Byte Shop business later became MicroAge
Computers and developed into a major national
distributor with its own chain of stores. In the early
1980’s the personal computer market was
dominated by three actors: Apple, Commodore, and
Tandy. With these new models the customer base
expanded from a hobbyist market to a larger and
more diverse customer base (Langlois, 1990).
Thus, from the launch of the microcomputers
sold in kits through advertisements, mail-order, and
trade fairs, the distribution of these products soon
came to take other routes to their customers. These
routes consisted of newly opened ‘independent’
specialized computer stores and retail chains,
already established radio retailers, and office
retailers selling, for example, calculators, depart-
The IMP Journal
Volume 4. Issue 3
ment stores, toy stores, etc. Thus, a previously
established network was partly supplemented by
new actors, specializing in this new kind of product.
Other actors brought in these new products as part
of a broader assortment (see Figure 2: most important connections in bold).
3.2. Towards increasing standardization
Until the 1980’s, manufacturers had more or less
relied on internally designed configurations and
components. Apple was an exception in this case,
relying on outside suppliers for all the hardware,
and enabling other suppliers to develop accessories.
This in turn helped to develop and disseminate the
product further (Dwyer, 1982). However, at this
time no suitable available operating system existed
on the market. Thus Apple had to rely on an
internally developed operating system, not open for
use by other firms.
In 1980, IBM decided to get into the business
and consumer market, wanting to develop a new
personal computer to be available for shipments
within a year. At first glance this seemed an almost
impossible mission, since at this time IBM had a
tradition of keeping everything in-house, with very
long development times. Dwyer (1982: 142)
explains that until then: “IBM wasn’t even using
standard components, choosing instead to guard
their proprietary technology through self-manufactured components.” However, a team was appointed
and their decision was to work in a totally new way.
This meant using already tested ‘off the shelf’ components provided by external suppliers (Langlois,
1992). Intel supplied the processor and Microsoft
supplied the operating system. Microsoft had
bought the operating system, MS-DOS, from a
local Seattle software firm, and modified it slightly.
However, the agreement also allowed Microsoft
to sell this software to other companies. Some components, such as the monitor and printer, had been
designed by IBM for an earlier product range and
were ready to use. Even the actual fabrication of the
PC was outsourced to suppliers through competitive
bids, some of which were actually won by IBM
departments. In order for other suppliers to be able
to make peripheral equipment and compatible
software for the PC, an open architecture was used.
The open architecture and the standardized
interfaces of components made it quite easy to
combine components into a functioning PC. This
also meant that other suppliers could produce IBM
compatible clones. The PC was launched in 1981;
one year after the decision was taken. The name
‘PC’ as we know it today has come to be equal to a
computer that is IBM compatible using the
operating system MS-DOS (later upgraded to
Windows), compared with a Macintosh computer
from Apple, using the operating system MAC OS.
175
When the PC was introduced in the USA in
1981, IBM already had established sales
subsidiaries with direct sales to customers for its
existing computer systems and mainframe
computers. However, “IBM recognized the need to
approach the consumer market much differently
than its industrial market.” (Samli and Wills, 1986:
27). In order to get quick market response the IBM
team decided to use external partners. Computer
Land and the Sears Roebuck department stores
became involved with the IBM team for the sales of
the new product from an early stage (Langlois,
1992). IBM relied heavily on their knowledge about
customers. They became the main outlets for the
new IBM PC. Sears Roebuck set up five business
systems centers. Most importantly, more than 190
ComputerLand stores already existed. This meant
that immediate widespread distribution across the
USA could be accomplished. Sears Roebuck,
however, had problems with sales since the new PC
turned out to sell to the office market rather than to
consumers, who were the main target of Sears
Roebuck. Hence, IBM took advantage of the
growing distribution network that had been
established for the earlier generations of microcomputers. The large business accounts were still
handled directly by the IBM subsidiaries.
3.3. New actors enter the distribution network
The effects of the open architecture became
immediately apparent. As early as six weeks after
the launch of the IBM PC, the IT equipment
supplier Tecmar had 20 IBM compatible PC related
products available for sale. Furthermore, according
to Dwyer (1982), some 40 manufacturers or
potential manufacturers applied to Microsoft for
licenses to use MS-DOS during the first year. In
1982 the first ‘IBM clone’ was launched by a new
company, Compaq. This meant that this PC was
based on the same architecture as, and 100%
compatible with, the IBM standard. As a newly
established firm, Compaq did not have any distribution arrangements of their own. Instead, they
approached the established dealers and invited them
to also become authorized Compaq dealers. By
promising not to use direct sales, Compaq could
establish a loyal dealership network with trained
and motivated salesmen and technically educated
service personnel who worked closely with
Compaq’s sales force. Like IBM, Compaq used
Sears business systems centers and specialized
computer stores such as ComputerLand to reach out
to customers. Alongside the ‘brand-clones’, such as
Compaq, a large number of assemblers of ‘nonbrand-clones’ appeared. These firms bought standard components, assembled them into PCs, and
sold them by mail order or storefront dealers
(Langlois, 1992). Hewlett Packard (HP) launched
The IMP Journal
Volume 4. Issue 3
its first IBM compatible PC in 1985 and, like
Compaq, worked only with dealers.
The decision by IBM to source components from
outside suppliers, thereby defining the product and
its technical characteristics as a standard, provided
an important prerequisite for the forthcoming
development of PCs. The standardized bundle of
resources comprising a PC, and the standardized
resource interfaces, meant that many actors could
engage in the assembly of these products. This PC
did not require high-tech facilities, and could be
made almost anywhere, ‘with nothing but a
screwdriver’.
Owing to the modularity and open architecture
of the product, “as the industry grew, more
specialized companies developed to produce
specific components” (Magretta, 1998: 74). Hence,
standardization of interfaces paved the way for
specialization in production of compatible
components. Eventually manufacturers of PCs
almost entirely outsourced the production of
specialized components and most PC ‘manufacturers’ became assemblers of third party components rather than manufacturers who designed
developed and manufactured components in-house.
Related to this specialization is another feature of
the electronics industry: the fast pace of technological development driven by these component
suppliers. This is most clearly expressed in Moore’s
Law, that the power of the CPU (processor) is
doubled every 18 months. This, even today, has
implications for assembly and distribution activities
since it creates short lifecycles of both these
components and final products.
Until the mid 1980’s the PCs had mainly been
produced to stock using intermediary actors for
distribution. The three large suppliers often referred
to as ‘the big three’: IBM, Compaq and HP, all
relied on this structure for production and
distribution. This meant that the actors involved in
production and distribution estimated future
demand and organized production and distribution
activities based on forecasts. Inventories at
component suppliers, PC producers and distributors
and resellers served as buffers in these systems to
provide availability of components and finished
products. Owing to the fast pace of technological
development, this system was vulnerable. When
new product launches were announced, the actors in
the system tried to get rid of the ‘old’ products,
resulting in serious price reductions and sell-outs.
This became an increasing problem as the number
of product variants grew. The cost of holding total
stocks of variants in the distribution network was
very high.
Dell Computer (Dell) was founded in 1984.
Dell’s primary idea was to sell PCs directly to
customers, without using title-holding dealers.
Another difference was that whereas other PC
producers built standardized PCs to estimated
176
demand, Dell built and configured to individual
customer orders. Dell relied on outside suppliers
but kept assembly in-house. It contracted
component suppliers to hold inventories close to
Dell’s assembly site and used transportation firms
to physically distribute the PCs to the customers.
This meant a system based on a different logic than
that of the established producers. Dell used
telemarketing and advertising to reach out to
customers. In these ads they clearly took a distance
from the ‘traditional channel structure’, which they
argued was inefficient and costly.
3.4. Changing customer needs and numbers
In the early 1990s the number of both business
and private customers increased, which also meant
an increase in the number of dealers. In 1992,
Compaq had 4,200 distribution partners in Europe
alone (Gabrielsson, 1999). In order to manage the
large number of dealers in 1992, both IBM and
Compaq began to use distributors. The distributors
took over the responsibility for the relationships
with resellers and retailers who, in turn, kept up the
relationships with end-customers. This added one
more ‘level’ in the distribution arrangements and
also added another level of inventory, held by
distributors. Furthermore, components and their
technical interfaces became more important when
PCs were integrated into larger networks. It was
then crucial that the PCs could readily interact, and
this could not be accomplished with so many
different configurations. Following the introduction
of ‘PC server products’, Compaq introduced system
resellers in 1991. They specialized in providing
total IT solutions to business customers and had
knowledge of how to configure PCs in order for
them to interact in IT networks. The system
resellers were authorized and had trained personnel
who could handle customers’ more complex needs
and environments (Gabrielsson, 1999).
At the beginning of the PC era, many small
specialized dealers had been established and the
margins on hardware were fairly high. In the mid1980s the conditions for these small dealers
changed. With the PC becoming more and more of
a commodity the dealers found their margins
dropping. This started a price war among the
established dealers and many ended up in trouble,
having expensive salespeople working with all
kinds of customers. A period of differentiation
began implying that dealers were forced to focus
more on some specific customer needs. A large
segment of PC sales also shifted from small local
dealers to computer superstores or electronics
superstores. For example, Tandy Corporation
opened the Computer City chain in 1991. Computer
City was the first superstore to offer IBM, HP,
Apple, Compaq, Tandy computers, accessories, and
software under one roof. The fact was that because
The IMP Journal
Volume 4. Issue 3
dealers were profiting very little from PC sales,
they needed to make money on other tasks such as
service, repair, software sales, or other peripheral
equipment.
In the early 1990’s the distribution network was
huge, involving hundreds of thousands of dealers,
with different focuses and directed to different
customer needs. The network was composed of a
mix of, for example, value adding resellers (VARs),
system integrators, distributors, superstores, mail
order firms, department stores, local vendors
specializing in IBM compatible clones with local
assembly and sales, and direct vendors such as Dell.
Furthermore, transportation and logistics firms had
an important role to play, especially in the direct
sales of Dell and other direct vendors.
At about this time, Dell’s business model based
on direct sales to customers was facing a challenge
with the increasing number of customers. This was
due to the fact that it became very costly to
approach the increasing number of private
consumers directly with telemarketing. In 1990
Dell, too, approached dealers. The first step was to
use superstores such as CompUSA. This step was
followed by the use of value adding resellers and
system integrators working more with customization and solutions for business customers.
(Hayes et al., 1996). However, this was not an easy
task since the dealers were very well aware of
Dell’s main business model and vision of direct
sales. Therefore, very few of them wanted to
include Dell PCs in their assortments.
Thus, in the early 1990’s the use context of the
PC had changed significantly. In the early years,
PCs were mainly restricted to business firms of
different sizes and to administrative activities in
individual work stations and for implementation of
EDI processes. Public agencies became more
computerized and individual households began to
use PCs for a variety of activities related to
administration, communication, entertainment, etc.
Design of integrated systems became more
important in firms, as did compatibility of
equipment and processes between organizations.
All these changes in the use of PCs also changed
the buying behavior of customers in different
directions. Some needed less services, some more
but of different types. Others needed installation
services and repairs on site while others wanted to
buy through mail order.
At this time all PC producers were more or less
using dealers for their distribution. This was a
necessity in order to handle the diverse endcustomer needs and the increasing number of endcustomers. However, Dell still had direct sales as
their main business logic and, as pointed out above,
were having problems with credibility in the eyes of
dealers.
177
3.5. Reorganizing for improved performance
In 1994, a new phenomenon entered the arena;
the Internet. The Internet was very important to
Dell, providing them with a new tool to reach out to
many customers in a cost-efficient way. As a
consequence, Dell withdrew from the established
relationships with dealers and went back to the
original idea of direct sales. The same year, Dell
launched www.dell.com and began to interact with
customers through this communication tool. But it
took until 1996 for customers to be able to purchase
from the website. With the Internet as their new
tool, Dell could focus more on private consumers.
In 1997, Dell created a special group to handle
private consumers, introducing a product line
especially for this kind of customer (Thompson and
Strickland, 1999). It became apparent that the most
interested customers were those who already had a
high degree of computer competence and were
buying their second or third PC. These customers
did not need a lot of support and could order
exactly what they wanted and have it delivered to
the door. This was made possible through close
cooperation between Dell and different transporttation and logistics firms. Hence, the modularized
design of the PC, together with technological developments in production equipment and information
technology enabled efficient mass-customization,
as applied by Dell. The information technology,
together with developments in logistics facilities
and material handling equipment also helped to
enable customized distribution solutions, as used by
Dell in their customer-order based system. About
the same time, in 1995 IBM decided to withdraw
totally from direct sales through subsidiaries and
rely only on indirect sales for its PC business. The
main reason was that the dealers viewed IBM as a
competitor, which led to conflicts (Gabrielsson,
1999). The same year, Hewlett Packard (HP)
introduced their first consumer-oriented PC for
private consumers and home offices, the Pavilion.
HP also chose to use only dealers to reach out to
these new customers. Thus, in 1995 there were two
main types of PC distribution arrangements. One,
as applied by Dell, was based on ‘direct sales’ to
customers and the other, as applied by for example
IBM, Compaq and HP, was based on the use of title
holding dealers. (see Figure 3).
One of the most important effects of Dell’s
distribution set-up is that it implies a very limited
level of inventory. This, in turn, puts pressure on
the companies working with production to stock,
and two-tier channels, including a distributor level
and a reseller/retailer level, both of whom hold
inventories. To create more efficient distribution
set-ups IBM, Compaq and HP all launched ‘channel
assembly programs’.
The IMP Journal
Volume 4. Issue 3
178
Component suppliers
Component suppliers
IBM, HP, Compaq
Dell
Distributors
Logistics and
transportation firms
Resellers
Retailers
Customers
Inventory
Business
Customers
Private
Customers
Information exchange
Physical flow
Figure 3: The two dominant distribution arrangements in 1995
Essentially this meant “shifting some final
assembly operations to distributors, thereby
bringing the final product closer to the customer”
(Curry and Kenney, 1999:23).
Thus, companies working with these two-tier
channels started to reorganize their distribution
arrangements to make them more efficient. The
new strategies were partly a response to directselling PC supplier Dell. These initiatives had both
similarities and differences. These new arrangements illustrate an increase in the variety of ways in
which a PC was assembled and distributed.
In 1995, IBM introduced the Authorized
Assembly Program (AAP), which allowed some
authorized partners to configure and complete final
assembly of PCs. The partners were supplied with
partially assembled PCs which they then configured
in accordance with customer specifications prior to
delivery. In 1997, IBM launched the Advanced
Fulfillment Initiative (AFI), which is a further
development of the AAP. IBM gave some of their
partners increased responsibility for component
inventory management and final assembly. They
received components directly from component
suppliers and IBM and assembled the PCs to
customer specifications. The activities were
coordinated by an advanced EDI-system, which
linked IBM, the suppliers and the assembly
partners. The objective of the program was to
improve all steps in the supply chain and to reduce
inventories at both IBM and resellers. In order to
facilitate final assembly at the partners’ locations,
IBM redesigned the chassis of some PC models so
components could easily be ‘snapped on’ rather
than ‘screwed in’. It also became important to be
able to replace one component without having to
remove the others. For example in one model, the
motherboard, which contains all the core electronics, could be slid out of a socket without
removing cables first. In the same PC, the hard disk
drive and CD drive were placed in a cage that could
be flipped out of the way of other components.
They also worked with co-location, meaning that
the partners performed assembly activities at IBM’s
manufacturing facility. From there, they shipped
assembled PCs directly to the end-customers
without going through the partner’s location.
In 1997, Compaq reorganized its distribution
strategy in a way similar to that of IBM. The
objective was to reduce inventories, internally as
well as for the approximately 40,000 channel
partners (Gabrielsson, 1999). The new strategy, the
Optimized Distribution Model (ODM), also
included a reduction of the number of distributors
The IMP Journal
Volume 4. Issue 3
to be able to work more closely with a few selected
partners. In the USA this meant a reduction of
distributors from 39 to four. The ODM program
incorporated three different levels of customization.
The first, Built-to Order (BTO), meant that part of
the assortment could be built and shipped on receipt
of a specific order. This was followed by (CTO)
(Configure-to-Order), which meant that a variety of
commonly requested PCs could be configured by
Compaq according to end-customer requirements.
Then the dealer could do the final customization by
installing the specific components specified by endusers. The last step, the Channel Configuration
Program (CCP), enabled dealers to build PCs to
individual customers’ specifications from components provided by Compaq. For a detailed
description and analysis of Compaq’s channel
assembly program see Kaplan (2002).
HP introduced a revised distribution program in
1997 called the Extended Solutions Partnership
Program. The program comprised four distribution
models: the traditional push-model through
distributors and resellers, Channel Assembly, Economy Program, and Vendor Express. The channel
assembly program was similar to Compaq’s CTO
program and IBM’s AAP program, meaning that
HP provided the partners with semi-assembled PCs
which were then finally assembled and customized
by HP’s partners at their locations with authorized
components supplied by HP. The partners had
access to local inventories, which were owned and
managed by HP.
These inventories were replenished by HP and
contained the ‘bare-bones’ as well as the components needed for customization and final
assembly. In order to coordinate these activities, the
partners were connected to, and had access to, HP’s
inventory system. The Economy Program involved
the production of certain top-selling models,
labeled Top-Value PCs. These were produced on
the basis of forecasts made jointly by the partners
and HP. In Sweden, HP introduced six Top-Value
models. They were priced very competitively and
offered to four distributors. By connecting to the
distributors’ inventories HP could place orders online according to certain inventory levels. The goal
was to reduce inventories so that distributors would
not get stuck with unsold products. If a distributor
could not sell the PCs kept in stock, there were two
possibilities: either to send the PCs back to HP or to
lower the prices. Both these alternatives were bad
from HP’s perspective, which is why they were
very anxious to reduce inventories at distributors’
facilities.
The fourth distribution model is called Vendor
Express and provided large business customers with
dedicated web pages, called Hewlett Packard
Enterprise Web Page, and from them the customers
could configure and order products, track deliveries
and submit service requests. This new way of
179
working was initiated in response to customers who
wanted more direct contact with HP rather than
with the intermediaries. They also wanted a wider
variety of ways to receive HP’s different products
and services. In this case the products are shipped
directly to end-customers and the reseller becomes
more of a service provider. For a summary of these
initiatives see Table 1.
These channel assembly programs illustrate how
firms tried to experiment with distribution
arrangements in order to reduce inventories, and at
the same time keep relationships with dealers. In
some cases logistics service providers were also
appointed as partners to keep assembly close to
end-customers.
Thus, for firms working with distribution
arrangements based on production to stock, the
modularized design of the PC and developments in
information technology were what made channel
assembly programs possible. This led to some
assembly activities being able to be postponed and
performed closer to the end-customers by
distributors, resellers or logistics providers, giving
these actors some new roles. These arrangements
also required efficient information exchange among
the actors involved and called for more integration
of systems among them. The reorganization of
distribution arrangements such as channel assembly
led to decreased lead times and a reduction in
inventory as well as providing greater variety in the
ways the PCs were assembled. It enabled customization, even if limited, in these systems, too, which
in turn implied greater variety in both products and
distribution alternatives for end-customers.
Dealers were also challenged by the pressure to
become more cost efficient and at the same time to
provide appropriate service levels. Many
distributors and resellers made joint efforts in this
respect, resulting in establishment of partnerships.
For example, in 1999 Swedish IT supplier IMS
teamed up with US IT distributor Ingram Micro to
work more closely together, making their ‘total’
arrangement more efficient. According to the
agreement, Ingram Micro took over all physical
handling of the products, order handling and
invoicing from IMS. IMS, in turn, specialized in
sales activities and services such as installation and
support as well as the design of solutions for
customers. IMS also focused on customer
relationships by maintaining a local presence and
contact with customers. IMS also agreed to use
Ingram Micro as its exclusive distributor. This is
only one example of many such arrangements
appearing, creating new relationships in the
network and changing established ones.
3.6. Changing end-customer interfaces
In 2000, IBM announced that they would begin
both direct sales via the Internet and telesales, as a
The IMP Journal
Aim of the initiative
Main levels and steps
Volume 4. Issue 3
IBM
To improve all steps in
the supply chain and
reduce inventories at
IBM and dealers.
Authorized Assembly
Program
-Final configurations to
customer specifications
by distributors.
Advanced Fulfillment
Initiative
-Inventory management
and final assembly by
partners.
-Advanced EDI system.
-Co-location at IBM’s
facilities.
Compaq
To reduce inventories,
internally and for
distribution partners.
Optimized Distribution
Model
-Reduction of the
number of distributors.
-Build-to-order: PCs are
built and shipped based
on customer orders.
-Configure-to-order:
Final configurations of
pre-assembled modules
by partners to endcustomer specifications.
-Channel configuration
program: Dealers
assemble customized
PCs from components
provided by Compaq.
180
HP
To approach customers
in new ways and to
reduce inventories
internally and at dealers.
Extended Solutions
Partnerships Program
-Traditional push
-Channel assembly: Final
configurations of preassembled modules by
partners to customer
specifications.
-Economy Program:
Top-selling models are
produced by common
forecasts by Compaq and
distributors.
-Vendor Express: Large
business customers have
dedicated web-pages
from where they could
configure and order PCs.
Table 1: Comparison among the three channel assembly initiatives of IBM, Compaq and HP
.
complement to working with dealers. This new
strategy was first introduced in the USA and later
implemented in Europe, initially in relation to small
and medium-sized companies and later to private
consumers. By doing so, IBM wanted to satisfy the
demands of customers who wanted to buy directly
and did not need to integrate the hardware with
additional services. This decision was received in
very different ways by their distribution partners. A
manager at one of the dealers compared the new
approach to going behind your partner’s back or
being unfaithful in a marriage. He argued that IBM
had been very unclear in its strategy, arguing one
day that the channel partner is the company’s most
important resource and selling behind the reseller’s
back the next. (Bohlin, 2000a).
Another dealer argued, on the contrary, that:
“IBM is doing the right thing, starting up direct
sales through the Internet […] It is only resellers
who do not believe in their own capability to add
value who need to be worried” (Bohlin, 2000b). So
this balancing act was not unproblematic for IBM,
and conflicts with dealers were definitely a
problem. Therefore, in 2001 IBM gave up the
consumer market and focused only on business
customers, with direct relationships to large
accounts and relationships through dealers to
smaller businesses. Later, in 2004, IBM sold its PC
business to Chinese PC supplier Lenovo, who
continued to sell PCs under the IBM brand.
In 1998, Compaq announced that it would begin
selling directly to end-customers (Kaplan, 2002).
This decision was based on Compaq’s acknowledgment that “there were customers who preferred
to buy directly, and they approximated that the
direct business models had in excess of 35 percent
overall sales in the USA market in 1998. The
decision on how to buy a computer would rest with
the individual end-customers” (Kaplan, 2002:200).
The basic idea was that Compaq was not the one to
decide how the end-customers should buy a certain
product but this decision should rest with the endcustomer and be based on this end-customer’s
specific needs. In line with this, Compaq would
offer customers the greatest possible variety of
ways to buy their products: through different
dealers, or directly from Compaq. Furthermore,
each end-customer was able to communicate with
Compaq in a number of ways: by telephone, e-mail,
Internet, fax, personal sales force, or through
dealers. This new approach was called the
Customer Choice Model. One new aspect of this
model was that it centered on a reference price, the
same for all customers. For distribution partners,
this meant that in order to make a profit they had to
provide value adding services for customers, for
which they could charge, e.g. delivery, support, and
consulting and integration services (Kaplan, 2002).
Looking back at Compaq and its development
from a distribution network perspective, we can see
The IMP Journal
Volume 4. Issue 3
181
Dell, HP
Distributors
Resellers
Business customers
Retailers
Private consumers
Figure 4: The distribution arrangements of HP and Dell in 2008
that Compaq continuously connected to new types
of actors, all playing important roles in the
distribution network. First, traditional small
resellers were used to reach out to small businesses,
and then when the number of dealers increased
distributors were added and used to handle the
relationships with resellers. When the number of
consumers increased, retailers were also added, and
solution partners and system specialists played
important roles when PCs began increasingly being
interconnected into larger networks. Later, when
Compaq introduced their new business models
(Optimized Distribution Model and Customer
Choice Model) the relationships among the actors
involved changed as described above. The introduction of these models particularly changed the
relationship between Compaq and the endcustomer. In 2002, HP acquired Compaq and
integrated Compaq into HP’s distribution strategy,
which was very similar to the one used by Compaq.
The ‘new’ HP continued using the approach as
described above, with both direct sales through the
Internet and indirect sales through dealers.
Similarly to Compaq, Dell also began to listen
more to end-customers in terms of how they wanted
to buy Dell’s products. Some of Dell’s large
business customers wanted to buy Dell products
from the same supplier, often a value adding reseller, from whom they already bought other IT
related equipment, in order to consolidate
purchases. In 2005, Dell re-organized its distribution arrangement in order to comply with these
demands, and began to approach resellers for sales
to business customers. In 2007, Dell announced the
program Dell Partner Direct, aimed at value-added
resellers. Resellers that joined the program were
promised increased discounts on products and the
possibility of using the Dell name and logo in their
marketing efforts.
This became the starting point for a shift in
Dell’s strategy. Later in 2007 Dell also came to an
agreement with Wal-Mart, and later other retailers,
offering them a limited range of Dell’s products for
sale to SMEs and households. These were for the
most common PC configurations that could be sold
in high volumes. Although Dell began to approach
resellers and retailers, the direct relationship and
sales approach remained most important for Dell.
Retail sales were expanded in 2008, incorporating
more retailers and retail chains into the growing
Dell dealership network. Figure 4 shows that the
distribution arrangements of HP and Dell at this
time were very similar, illustrating a mix of the two
distribution arrangements shown in Figure 2. This
created greater variety from the customer’s
perspective with regard to how PCs could be
bought and from whom.
3.7. Summing up
In summary, this story began with the launch of
the first microcomputers in the mid-1970s. They
were sold as kits by telephone sales or at trade fairs.
When the next generation of ‘complete’ microcomputers was launched in the late 1970s and early
1980s the suppliers of these computers used their
established dealership networks for office equipment and radio outlets as well as department stores
and even toy stores. Specialized computer stores
and computer retail chains like ComputerLand were
also established around this time. When the number
of private consumers increased, superstores were
used, and they sold PCs along with other consumer
electronics. Similarly, when PCs were becoming
The IMP Journal
Volume 4. Issue 3
1975
Microcomputers
Computer stores
1981
1984
IBM’s PC
Dell founded
Standardization of
Component interfaces Direct sales
1975
1977
Apple founded
1982
The first clone:
Compaq
1992
Distributors
introduced
182
1997
Channel
assembly programs
1994
The Internet
1991
Dell approached
dealers
1996
Dell started
Internet sales
1998
Compaq
Direct sales
2005
2002
Dell approached
HP acquired
dealers
Compaq
2004
Lenovo acquired
IBM’s PC division
2008
2008
2000
HP and Dell
IBM
Direct sales Mix of end-cusotmer
interfaces
Figure 5: PC distribution timeline 1975-2008
increasingly connected into larger networks, system
resellers and system integrators emerged that could
provide customers with total IT solutions adapted to
specific purposes and use contexts. Dell entered the
network with a new idea, to sell PCs directly to
customers on a build-to-order basis. Instead of
relying only on title holding actors such as distributors, resellers and retailers, they worked closely
with third party logistics firms for delivery and for
combinations of parts of orders at logistics sites.
Furthermore, when the Internet entered the arena
along with other advances in information technology in the mid-1990s, new possible ways of
relating to each other occurred in the distribution
network. Internet sales were first offered by suppliers such as Dell, using a direct approach.
However, for firms working with a structure
relying on indirect sales this also led to changes in
the relationship among these actors. For example,
advances in information technology were important
enablers of the channel assembly programs.
Furthermore, many of the relationships among distributors and resellers also changed, creating new
divisions of work, all with the aim of becoming
more cost efficient and at the same time providing
high levels of services. This was followed by a
period when PC suppliers began to take more
consideration of the customers’ demands. This
meant that firms that had traditionally used resellers
set up direct sales initiatives, and firms that had
traditionally only relied on direct sales began to use
resellers in parallel with their original strategies.
This was a start towards an increase in multichannels approaches, which provided several paths
to the customers, with regard to physical distribution, communication and sales.
Thus, in the beginning of the microcomputer and
personal computer era, each PC supplier worked
with more or less a single approach for reaching out
to customers. Some, like Compaq, worked strictly
with indirect sales, using distributors and resellers,
while others, especially Dell, worked with a direct
sales model. The firms later started to experiment
with different alternative arrangements. It is interesting to note that many firms have at some time
abandoned their original distribution strategies to
test alternatives and then again turned back to their
initial, possibly adapted, strategies, and then again
experimented with alternatives. Kaplan (2002)
found, for example, that between 1983 and 2000
both Compaq and Dell implemented five different
‘business models’ for their distribution. Furthermore, not only the PC suppliers experimented with
their arrangements. The other actors involved in the
distribution networks, such as various distributors,
resellers, retailers and logistics providers, also
experimented from their respective perspective, in
turn restricting and enabling the efforts of the PC
suppliers. Some key events in the evolution of the
PC distribution network are illustrated in the
timeline in Figure 5.
4. Distribution dynamics as network evolution
The analysis below is divided into four sections.
The dynamics in the activity pattern, resource
constellation, and web of actors make up the three
first sections. This is followed by a section on the
interplay among these three network layers.
4.1. Dynamics in the pattern of activities
Looking back at our empirical example it is clear
that the pattern of activities has changed in several
ways over these three decades. To understand these
dynamics and how distribution arrangements are
organized with regard to activities we refer to two
principles: the principle of speculation and the
principle of postponement (Bucklin, 1965). According to the principle of speculation, differentiation
“in form, and the movement of goods to forward
inventories, should be made at the earliest possible
The IMP Journal
Volume 4. Issue 3
time in the marketing flow” (ibid:68). The principle
of speculation relies on forecasts of demand.
Speculation generally supports the capturing of
economies of scale in manufacturing, but may
result in substantial inventories being held at
several places. The principle of postponement was
originally mostly discussed in terms of time and
place postponement - taking finished products as
the point of departure (van Hoek, 2001). Form
postponement was later discussed, for example, by
Zinn and Bowersox (1988), who identified a
number of different types of form postponement:
labeling, packaging, assembly, and manufacturing.
In the early stage of the microcomputer era, the
suppliers required a certain number of orders from
dealers before starting assembly of the parts of the
computer kit. In other words, a postponement
strategy was applied. The end-customers were
involved in the production activities because they
put the kits together, including soldering some
parts. Later, when volumes increased and assembly
was more industrialized, PC suppliers began to
produce on speculation. The PC manufacturers used
forecasts of demand to estimate the quantity to
produce. Similarly, dealers made forecasts based on
sales figures. This created inventories of finished
products held by both producers and dealers. These
inventories acted as buffers that decoupled the
sequentially related production, distribution and use
activities.
The way activities are organized in the
distribution network thus affects the type and extent
of dependencies among activities. Parallel
dependence arises because activities use the same
resources, i.e. they are similar with reference to
resource utilization. This is described by
Richardson (1972: 888) as similarity. He argues
that activities are similar when they “require the
same capability for their undertaking”.
The principle of speculation means focusing
mainly on activity coordination of each actor to improve efficiency within firm boundaries (Hulthén,
2002). This means that economies of scale can be
taken advantage of by creating similarities among
activities. Furthermore, this means rather loosely
coupled activity structures among the firms
involved and the physical flows can be coordinated
within the boundaries of one firm, without much
consideration for the operations of other firms.
Information exchange in this case is mostly a matter
of order placement and information about prices.
However, as was predicted by Bucklin (1965:
31), the development of new technologies, for
example new means and resources for transportation and information handling “has the effect of
reducing the relative advantage of speculation over
postponement. Hence intermediate inventories will
tend to disappear and be replaced by distribution
channels which have a direct flow.” First, Dell
introduced a totally new way of thinking about
183
production and distribution activities for PCs, based
on the principle of postponement. This was partly a
reaction to the established structure of the network,
primarily based on arrangements relying on the
principle of speculation. However, when removing
inventories of finished goods, serving as buffers,
the sequentially dependent activities needed to be
integrated and coordinated across actors in the
network. Sequential dependence is related to the
fact that activities are serially related. This means
that activities have to be performed in a certain
order. This kind of dependence exists as a result of
activities being what Richardson (1972: 889) refers
to as complementary. Activities are complementary
when they represent “different phases of a process
of production and require in some way to be
coordinated”. Activities are closely complementary
if there is a need to “match not the aggregate output
of a general-purpose input with the aggregate
output for which it is needed but of particular
activities.” (ibid: 891). The implication is that these
activities are directed towards a specific counterpart. Closely complementary activities can be
undertaken within one firm or by several firms. In
the latter organizations have to “agree to match
their related plans in advance” (ibid: 890).
The Internet provided opportunities for efficient
information exchange and electronic commerce.
The fact that information could be exchanged
efficiently and effectively among many actors had
major implications for how distribution could be
organized. With regard to the possibilities for direct
sales based on postponement, the Internet provided
a low cost tool for connecting buyers and sellers
and also for coordination of physical flows. Dell
was able to connect to suppliers, contract manufacturers and logistics providers in a way that
resulted in transparent information exchange, which
was a prerequisite for this arrangement to work
smoothly. In these new arrangements actors specializing in, for example, information or materials
handling emerged and/or took on new and important tasks. For example, third party logistics
providers began to act as coordinators of the
physical flows, including products and components
arriving from different facilities to be combined
into products for delivery to end-customers. The
increasing specialization resulted in a large number
of specialized firms such as: component suppliers,
contract manufacturer, logistics providers, information handlers, etc. In order to coordinate them,
extensive information sharing activities across firm
boundaries is required. This became even more
important in cases of build-to-order arrangements,
implying that each product made and distributed
was directed to a specific end-customer, which
required closely complementary activities.
The analysis shows that distribution arrangements based on speculation are complemented,
rather than replaced, by those based on post-
The IMP Journal
Volume 4. Issue 3
ponement or of arrangements mixing the two
principles in various ways. Thus, a certain distribution set-up can affect adaptations of other
arrangements in the network which, in turn, have to
become related to the existing ones. The reason for
this is that the various distribution arrangements
overlap in the overall distribution network, because
actors are involved in several distribution
structures. This is an aspect of the role of variety in
the evolution of distribution networks. For example,
a logistics service provider or a reseller might be
involved in different types of distribution set-ups,
of diverse natures, thereby acting as a connector of
the various parts of the network. By acting as
connector in this sense, each such actor needs to
take into account ‘all’ their connections. The adjustments made with regard to these considerations
drive evolution in certain directions.
The case also shows how still other arrangements occurred when parts of the network tried to
adjust to the distribution structures based on
postponement (as applied, for example, by Dell).
Based on their existing distribution set-ups relying
on speculation, the actors tried to adapt these
arrangements so that they could also be able to
utilize some of the benefits of postponement. The
development of channel assembly programs
exemplifies this. The case also shows the problems
involved in changing an established pattern of
activities, since the actors also need to consider
other established relationships in the distribution
network that are affected by the new arrangements.
For example, the channel assembly programs meant
that dealers needed to adjust their way of working
in some new directions, which were not always in
line with all the dealers’ own strategies.
Thus, the organization of activities in these
arrangements differs, and this has consequences for
the overall activity pattern. At a certain point in
time, some parts of the activity pattern are
characterized by quite loosely coupled activity
structures while other parts of the network are
characterized by strong activity interdependencies.
However, these patterns are constantly changing in
different directions, in line with our empirical
example. Furthermore, the different activity
structures embedded in the larger activity pattern
are interconnected by the use of common resources
(e.g. for transportation or for information
exchange). This creates interdependencies among
resources as well as among actors. This
interconnectedness is decisive of how evolutionary
changes in one part of the network impact on other
parts, contributing to network dynamics.
These developments in distribution do not only
apply to PC distribution. According to Gadde and
Ford (2008), distribution structures are increasingly
characterized by postponement. New cost efficient
arrangements have been made possible by
information technology, which has enhanced the
184
speed and accuracy of the information flow, in turn
improving the coordination of the physical flow of
materials. (Bello et al., 2003). Developments in
physical distribution have reduced lead times and
distribution costs. This has enabled firms to work
with new types of activity configurations, for
example build-to-order strategies, just-in-time
deliveries, and systems for efficient-consumerresponse (ECR). All these strategies rely on activity
synchronization within and between manufacturing,
distribution and use processes, leading to increased
activity interdependencies across firm boundaries.
Specialized firms are most often involved in these
arrangements, since one single firm cannot be on
the cutting edge of all the technologies on which it
relies. Activity interdependencies require that firms
interact intensively in order to exchange
information to coordinate activities. Furthermore,
the resources on which a firm relies are often
located outside the own firm boundary. The next
section will analyze distribution dynamics from a
resource perspective in greater depth.
4.2. Dynamics in the constellation of resources
Our empirical example illustrates the interplay
among a number of resources. Here we rely on
Håkansson and Waluszewski’s (2002) categorization of resources, distinguishing between two
physical resources: - products and facilities - and
two organizational resources - business units and
business relationships. The development of a
resource depends on how the resources interact.
Our case shows how resource interaction impacts
on distribution network evolution.
Products are exchanged among actors and can be
‘moved around’ and therefore relate to different
resource structures. The product in a distribution
network can be seen from different actors’ perspectives. Alderson (1957: 215) states that “goods that a
producer has for sale are the expression of his skills
and resources”. So, ideal “specifications from his
viewpoint would be those which made most
effective use of his plant capacity and of the
available labor and raw materials.” However, the
“specifications of the ideal product from the
consumer viewpoint are determined by use requirements.” Consequently, the “product appears in a
very different setting at these two levels and may be
said to belong to the technology of production at
one stage and the technology of use at the other.”
(Alderson, 1957: 216).
The case clearly shows how the evolution of the
distribution network is related to interaction among
products, e.g. components and PCs, as well as
interaction among products and other resource
categories. We have identified two aspects related
to resources as interesting to discuss further. The
first concerns whether the resource is developed
internally or using external resources. The second is
The IMP Journal
Volume 4. Issue 3
related to whether a resource is standardized or
specific. For example, the early focus on internally
designed specialized components restricted the
development of a standard for microcomputers.
Apple tried to use standard hardware when
developing the Apple II, but there was no available
operating system for sale, and so they had to use an
internally designed system. Later, this restricted
Apple’s possibility of becoming a part of the IBM
PC community. When IBM began to source
components from external partners to build PCs,
they opened up for standardized interfaces among
components. Most importantly, this enabled other
actors to develop accessories and add-ons to the
PC. It also provided small non-brand producers
with an opportunity to assemble PCs on a small
scale. Hence, the development towards an open
architecture played an important role in the
dynamics of distribution. From a resource
perspective, IBM’s PC could easily interact with
accessories and other PCs, thanks to its clearly
defined and standardized resource interfaces.
Furthermore, the standardized interfaces enabled
modularization, in turn creating opportunities for
mass customization. Thus, standardization at the
component level was a prerequisite for economically efficient customization at the product level.
This standardization also paved the way for simplified service and repair activities, and enabled actors
to specialize in such operations. Standardization
and modularization were also major prerequisites
for certain changes in the distribution network, for
example realization of channel assembly programs.
Another important aspect related to products is the
high pace of technological development of
components, which leads to short life cycles of
products and creates a pressure to reduce inventtories in the distribution network. This characteristic of products has forced suppliers to rely
increasingly on the principle of postponement to
reduce inventories. Consequently, shortening the
time between component manufacturing and the
finished product in the hands of the end-user is
crucial to efficient distribution of PCs.
Thus, the emergence of specialized component
suppliers, designing standardized components, and
the pace of the technological development driven
by these suppliers have played a significant role for
the development of the PC itself, how it is produced
and, not least, how it is distributed. Furthermore,
the modularization and standardization of the PC
meant that it became more of a commodity than had
been the case with the earlier more specialized
computers and components.
Facilities are more stable by nature than products
(Håkansson et al., 2009). Examples of facilities are
warehouses, buildings and plants, equipment for
production, logistics, transportation, materials
handling, and distribution, the Internet, information
systems, etc. For firms in a distribution network it
185
is crucial to exploit facilities (both internally and
externally) in an efficient way. Furthermore, how
facilities are made use of in the distribution network
is a crucial aspect of network evolution. Relating to
the discussion above, the change from internally
designed components to standardized components
developed by ‘independent’ suppliers influenced
how facilities were used. Rather than relying on
their own facilities, firms began to rely on those of
others. The first step concerned the development
and production of components. The next step was
to also outsource assembly activities, which led
many PC suppliers to begin to share facilities,
controlled by specialists in assembly activities.
Later, more and more suppliers also outsourced
logistics operations to third parties. All in all, this
increasing reliance on the facilities of others was
highly related to the standardization and modularization of the products. Furthermore, these shared
facilities connect the distribution arrangements of
different PC suppliers to each other, which, in turn
means that similarities among activities can be
created, and these improve the efficiency of each
facility (cf. Richardson, 1972).
The Internet, a specific facility that played a
major role in the evolution of the distribution
network, was first exploited by Dell. Kaplan (2002)
argues that this was attributable to Dell’s original
business model of selling directly to users. Dell
was already positioned in a situation amenable to
applying the Internet as a facility for electronic
commerce. One important factor of the utilization
of this new tool was that Dell had already
established an organizational resource in the form
of business relationships directly with end-users. In
contrast, firms working according to the principle
of speculation, such as Compaq, IBM and HP, had
their established business relationships with
distribution partners, which made it more difficult
for them to develop electronic commerce with endcustomers. These companies instead used the
Internet as an information channel to promote
suitable distribution partners to customers. Later,
they also used the Internet to set up the channel
assembly programs which, as we have shown,
changed the nature of their relationship to
distribution partners as well as end-customers.
This example points out the importance of the
business relationship as a resource for the
development of distribution networks and their
interactions with other resource categories. Kaplan
(2002) shows how the acquired capabilities, needed
to exploit the Internet, had to be integrated with
other capabilities of the firm. Kaplan found that
even when capabilities to some extent are
developed ‘in-house’ they are to a large extent
developed in interaction with external resource
providers. This is in line with Håkansson and
Waluszewski (2002: 36), who argued that by
“distinguishing business units as a resource it is
The IMP Journal
Volume 4. Issue 3
possible to investigate the skills involved in dealing
with interaction.” The case illustrates that
knowledge concerning how to use internal and
external resources and how to combine them has
played a significant role in the development of the
distribution network.
4.3. Dynamics in the web of actors
Many important characteristics of the actor layer
have already been discussed, as they were of
importance for evolution of activity patterns and
resource constellations. Our case shows that
changes in the activity pattern and the resource
constellation create opportunities in the network for
the entry of new actors, and also create both
opportunities and restrictions for the established
actors. Furthermore, these changes also enable new
relationships and new contents of relationships
between both old and new actors, including actors
who enter from the wider network, as argued above.
The case has also shown how new and established
actors, in their turns, affect evolution related to
activities and resources.
According to Håkansson et al. (2009: 141)
“actors combine and configure their resources and
activities based on those of other actors.” Each firm
tries to find and develop a position in the
distribution network by connecting to other actors.
The position of a firm defines its roles in relation to
the other firms in the network (Mattsson, 1985),
and thus reflects how the actor connects its internal
activities and resources to other actors and their
activities and resources. The position concept
implies that an actor is defined as much by the
resources of other actors to whom it has
relationships as by the resources that are within the
legal boundary of the actor (Henders, 1992). Thus a
firm’s position in a network is defined by how it is
related in the network context. Håkansson et al.
(2009:139) suggest that the “roles that are attributed
to an actor depend on with whom it interacts.” So
“with whom an actor interacts and how it interacts
shapes what it can and cannot do, how it is
constituted and ultimately, who or what it is” (ibid:
140). When firms interact they also discover new
opportunities. This means that firms and their
positions are constantly changing. We illustrate this
below with examples from our case.
First, entry of new actors means that new
connections among activities and resources can be
established and new distribution arrangements
arise. Retail chains carrying a wide assortment of
electronic equipment for households or in some
cases for small businesses, were established and
expanded during the 1990s. Such stores carried
many brands and inventories of pre-assembled,
ready-made products that could be bought directly
off the shelf in the store. These stores contributed to
the ‘commoditization’ of PCs and to price
186
competition. With regard to suppliers of PCs, the
standardized components and interfaces opened up
opportunities for many firms to enter the network.
This was possible since it did not require much
internal technological development skills, but rather
the ability to connect to and cooperate with external
partners. There were also opportunities for actors to
handle new customer needs. For example, new
actors referred to as systems integrators emerged to
handle new customer needs when IT equipment
was increasingly interlinked within and between
organizations. These firms were not always new
actors. Sometimes it was established resellers who
changed focus. Other examples are firms that
focused on the opportunities of the Internet for
electronic commerce (see Morris and Morris,
2002), as well as information handling firms,
specialized in creating information support systems
for new distribution arrangements.
Second, sometimes established network actors
take on new roles. For example, resellers adapt to
new customer needs and start Internet sales or turn
into system integrators. Our case also illustrates
how actors integrate these ‘new products’ into their
assortment and thereby change their position.
Furthermore, when IBM began PC distribution they
were able to take advantage of the distribution
network they had developed for their microcomputers. For IBM, however, this was not uncomplicated owing to their history of having an
internal business focus. In the same way Dell tried
to change its position in the 1990s, which proved to
be hard to accomplish owing to its earlier position
in the network. The establishment of the required
relationships to resellers to achieve the new position was a problem since these actors predicted a
short term focus from Dell’s side. Another example
is related to the channel assembly programs, where
some logistics providers took over assembly
operations, a totally new type of activity for them.
Logistics actors also performed an important role in
traditional distribution arrangements, providing an
infrastructure for transportation that was also
undergoing change, further contributing to network
dynamics. Changing roles in the distribution
network thus create opportunities for new
connections in the activity pattern and resource
constellation.
Third, changing business relationships are also
an important factor for distribution dynamics. For
example, the channel assembly programs changed
relationships within the network. They became
more partner-like arrangements, focusing more on
cooperation and information sharing than before.
Furthermore, by working more closely together
some resellers and distributors aimed at improving
efficiency in the physical flow. These arrangements
meant changes in the division of work among the
actors and thereby eliminating duplication of work
such as storing, packing and materials handling.
The IMP Journal
Actors
Activities
Resources
Volume 4. Issue 3
Type 1
-Logistics service providers
-Information specialists
-System intergrators
-Build-to-order
-Direct delivery
-Internet
-IT-equipment
Changes
Type 2
-Changing roles for, e.g, brand
owners, distributors, resellers,
logistic firms
-Channel assembly programs
-Standardized components and
interfaces
-Production equipment
187
Type 3
-New types of relationships
to end customers
-Changed relationships
between distributors and
resellers
-Increasing interdependence
-External resource access
Table 2: Types of changes in the three network layers
In distribution arrangements based on speculation, the relationships among the actors involved
do not have to be close. Each actor can focus on the
operations within their own boundaries because of
the decoupling of activities made possible by the
inventories. However, when firms began relying
more on arrangements based on the principle of
postponement and build-to-order, the relationships
among firms had to be of a more cooperative
nature. For example, Dell uses “technology and
information to blur the traditional boundaries in the
value chain among suppliers, manufactures, and
end-users” Magretta (1998: 74). To work in this
way, cooperative rather than arms’ length relationships are required. This also relates to the discussion above of increasing specialization in the
distribution network. Specialization and decoupling
of activity patterns also require coordination and reintegration of these activity patterns.
Another aspect relating to the web of actors is
how ‘ideas’ about alternative distribution arrangements (e.g. direct sales, use of the Internet, chain
store development), which we have shown to exist
at a specific point in time, are disseminated and
implemented throughout the network. Experiments
involving some actors might, if they are successful,
stimulate efforts by other, competing or cooperating
actors to try to imitate the new ideas. There may
also be influences from application of a specific
arrangement in distribution networks for other
products that overlap with the PC network, for
instance TPL solutions or arrangements relying on
build-to-order. However, as we have shown, it often
proved hard to implement these ideas since
established distribution arrangements involve connections in the resource, activity and actor layers
that need to be re-considered and changed. Such
connections may be related to the past and present
as well as to considerations of a preferred future
network structure. So the complexity of networks
makes it impossible for firms to have more than a
very limited view of the web of actors in which
they are embedded. This means that their know-
ledge and understanding of the development of the
web of actors can only be partial. However, by
sharing these partial pictures, firms’ views and
ideas on ‘how to conduct business’ often stretch
across the web of actors (Håkansson et al., 2009).
Obviously, firms constantly try to change positions
in the distribution network, both by making radical
changes and by fine tuning activities and resources
and the way they interrelate. In line with this, Narus
and Anderson (1996: 112) found that “companies
are experimenting with their distribution channels
to make them more flexible and responsive.” As a
result of these efforts, Curry and Kenney (1999: 9)
state that new “business models are constantly
being introduced.”
4.4. Interplay among the three layers
As outlined in the analytical model in figure 1,
we identify three different changes that can occur in
a distribution network. The first, type 1, relates to
the entry of new actors, resources, and activities in
the network as well as to the exit of established
ones. Type 2 deals with how established actors,
resources and activities are adapted in the network
over time. Finally, type 3 concerns changes in the
connections within one of the layers, for instance
among two activities, or among the layers, for
instance among a resource and an activity. Based on
the case analysis we outline some of the most
important changes identified (see table 2).
As our case illustrates, new actors are constantly
entering the network and established actors change
roles in order to adapt to actions of others as well as
to improve their own positions in the network. The
changes in the web of actors also result in new
relationships being formed and established
relationships changing. Furthermore, changes in the
actor dimension create opportunities as well as
obstacles for changes in the activity and resource
dimensions and, at the same time, they are
prerequisites for some changes to take place. For
example, activity structures relying on build-to-
The IMP Journal
Volume 4. Issue 3
order arrangements and channel assembly programs
require new types of relationships among actors in
order to handle the increasing activity interdependence resulting from these activity configurations. These types of activity configurations were
made possible by the standardization of components and component interfaces and the modularization of the PC as such. This enabled firms to
rely increasingly on external resources rather than
keeping operations controlled internally. Furthermore, new actors were enabled to enter the network
or change current operations by specializing in
some resource or resource interface. This could
involve, for example, component production,
production of accessories and add-ons, software
development, assembly operations, integration of
PCs into systems, or logistics. It fostered specialization and the capturing of similarities among
activities, which was crucial to gaining economies
of scale in operations relating to production and
distribution. These changes led to overall changes
in the division of work in the network, and
specialization required activity coordination across
firm boundaries. These distribution set-ups required
close cooperation among firms in order to exchange
information and synchronize activities characterized by close complementarity. In such structures
resources related to information handling and
sharing became particularly important since
information sharing concerning order flows and
physical flows needs to be coordinated among
many specialized firms. This required more cooperative relationships.
New activity configurations require new or
modified resources. New resources provide opportunities for new activities and for new ways of
coordinating activities. For example, the Internet as
a resource came to play an important role for how
activities in the network could be performed and
linked together. Internet sales changed the activity
structures concerning how orders were placed and
transferred. Furthermore, these sales played
different roles for various firms depending on their
established positions in the distribution network. As
indicated before, Compaq and Dell made use of this
resource in different ways. Changes of facilities are
also closely related to the organization of activities.
Facilities are often adapted in order to increase the
economization of resources, relating to the
capturing of similarities among activities in
changing activity structures. For example, when
firms tried to change from speculative to more
postponement based strategies by introducing
channel assembly programs, facilities at the firms
involved needed to be adapted to this new way of
working.
From our analysis we identify one characteristic
of each of the network layers to discuss further in
the concluding section. We find that variety in the
web of actor is an important aspect of distribution
188
network dynamics. This concerns both variety in
terms of different types of actors, how distribution
can be organized and in the nature of relationships
among actors. The activity pattern is, in turn,
characterized by various interdependencies. The
case shows that changes in interdependency are a
major aspect of dynamics in distribution networks.
Finally, control, for example internal versus
external access to resources, is an important aspect
of the resource collection and the dynamics in
distribution networks. These three characteristics
are further discussed in the concluding section. This
section is held more general than the previous ones.
Consequently, we now leave the dynamics in the
PC distribution network to discuss distribution
dynamics from the perspective of some general
topics.
5. Concluding discussion
5.1. Variety in the web of actors
Creation of variety is closely related to
specialization in industry. Firms specializing in
certain distribution functions have been able to
strengthen their positions. ‘Traditional’ distributors
perform a broad range of activities since they
“contact customers and makes the product available
by providing necessary supporting services, such as
delivery, credit, technical advice, repair service,
assembly, and promotion” (Herbig and O’Hara,
1994: 199). It is difficult for firms to maintain and
update their capabilities within all these diverse
areas, since technical development constantly
creates new opportunities for firms to be on the
cutting edge. It comes as no surprise therefore that
specialized firms are increasingly important in the
evolving distribution arrangements, for example
logistics service providers (Carbone and Stone,
2005; Panayides, 2005). In a similar way firms
specializing in information exchange have improved their positions in terms of, for example,
portal services (Clarke and Flaherty, 2003) and
serving as electronic hubs (Shevchenko and
Shevchenko, 2005). Further examples of specialization include call-centre operations (Duder and
Rosenwein, 2001), warehousing (Faber et al., 2002)
and personal selling (Marchetti and Brewer, 1996).
Specialization increases efficiency primarily
because it enables economies of scale in each
function. Moreover specialization breeds specialization, since the existence of one type of specialist
makes it possible for others to specialize in complementary functions. The trend towards specialization
provides opportunities to combine distribution
resources and thereby to create variety in distribution networks. At the same time, however, it also
requires increasing coordination, since specialization at one point in an industrial system will
The IMP Journal
Volume 4. Issue 3
require reintegration at some other point. The
specialized tasks have to be reassimilated “into a
cognitive form wherein they have economic
meaning” (Piore, 1992:442). In order to handle this
coordination effectively, cooperative relationships
among actors in the distribution network have
emerged. According to Hoyt and Huq (2000),
channel relationships have evolved from transactional processes based on arm’s-length conditions
towards collaborative arrangements. A similar
opinion is expressed, for example, by Buzzel and
Ortmeyer (1995:85), who state that “formerly
adversarial relationships between retailers and their
suppliers are giving way to cooperative partnerships”.
Variety appears on the supply side, with many
specialized suppliers, as well as on the user side,
with customers having diverse needs and requirements. This, in turn, implies a variety of distribution arrangements. Variety, however, is not a
clear-cut concept, but can be applied to options for
intermediation of a specific product category
offered by an individual brand manufacturer or to
options available in the product category when all
brand manufacturers are considered. We label the
first type of variety micro-variety, and the second
macro-variety. Variety might be seen as an outcome
of the evolutionary process or as an intrinsic aspect
of the process. Variety is therefore strongly linked
to experimentation and to innovations in all three
network layers. Variety in these two respects can
vary over time.
From a micro-perspective, variety is strongly
related to that which, in the channel literature, has
been discussed as multi-channel marketing (see e.g
Alptekinoglu and Tang, 2005; Wirtz, 2007). Multichannel marketing refers to the situation where a
firm provides multiple interfaces towards customers, e.g. direct sales, resellers, retailers, system
integrators, telephone, and internet sales, meaning
that each customer can choose how to interact with
the supplier. Such a strategy also enables the firm to
meet various customer demands with regard to
distribution options. Owing to diverse end-user
requirements, different approaches are needed to
satisfy various end-users, for example, in terms of
communication, delivery, and sales.
As pointed out above, increasing specialization,
requiring increased coordination across firm
boundaries, calls for cooperative relationships.
However, the emergence of multi-channel arrangements point in other directions. For example,
Hibbard et al. (2001: 45) conclude that the growth
of multi-channel systems has “significantly
increased the potential for discord between
suppliers and channel intermediaries,” illustrated in
among other ways by brand owners’ decisions to
initiate direct sales, an approach disliked by their
partners in distribution. This type of situation has
often been referred to as channel conflicts. One way
189
of dealing with that has been to encourage the
distribution partners to change roles, for example
specializing in logistics or other services, thereby
adapting their positions in the network, enabling
them to remain important partners in distribution.
Variety may be seen as an outcome of the
evolutionary process, which makes it strongly
linked to experiments and to innovation in all three
network layers. Variety is, however, not only an
outcome. It also affects evolution because it creates
different paths for future development. Some
arrangements might of course fail (from one or
many actors’ perspectives) or need to be adapted.
Variety is related to experimentation, and the
efforts to imitate successful experiments are
important to evolution. Experiments with new alternative distribution arrangements are communicated
and sometimes circulated in the network, increasing
variety and influencing evolution. However, one
problem with increased variety in the available
options for customers is that it is costly. This means
that variety in its most extreme form will seldom be
possible to realize since customers will not be
willing to pay the cost. So for firms involved in the
distribution network it becomes decisive to find a
suitable level of variety in order to be both cost
efficient and at the same time to satisfy variety on
the user side of the network. The way this is
accomplished is tightly linked to how the
distribution network is organized with regard to
activities and resources, as discussed below.
5.2. Interdependencies in the activity pattern
As pointed out above, firms try to differentiate
distribution solutions in accordance with various
customer requirements. Some of these arrangements are customized and involve arrangements
based on just-in-time (JIT) deliveries (Kannan and
Tan, 2005), and efficient consumer response (ECR)
(Kurnia and Johnston, 2005). The ultimate
approach in this respect is build-to-order (BTO)
production, which might be a prerequisite for
handling the problems with expanding assortments
(Gunase-karan and Ngai, 2005). The effect of these
three processes (JIT, ECR, and BTO) is increasing
interdependence among activities. Previous arrangements relied on inventories as buffers, which
imposed only loose couplings between activities.
When these inventories are reduced, the resulting
interdependencies have to be handled through
integration of processes and thus increased coordination of activities. In many cases these
integrative efforts span the boundaries of firms.
In this paper, we show how interdependence
changes over time and in different parts of
networks. For example, depending on whether distribution arrangements rely on the logic of
speculation or postponement, the activity interdependencies within and across firm boundaries
The IMP Journal
Volume 4. Issue 3
will differ as will the required coordination. When
distribution networks are organized according to the
principle of speculation, the activity interdependencies are mainly a firm-internal matter. The
inventories act as buffers in the activity structure
and thereby decouple the different firms’ activity
structures from each other. Sequential interdependencies are to a large extent decreased. In this
respect the actors also have a low degreee of
interdependence. However, within each firm it
becomes important to handle the parallel interdependencies stemming from the need to capture
similarities among activities within the firm in
order to obtain economies of scale. In distribution
networks relying on postponement there are no
inventories acting as buffers among the sequentially
interdependent activities. Therefore, coordination
among firms is necessary in order to handle this
kind of sequential interdependencies resulting from
activities being closely complementary with regard
to a specific actor.
In retrospect, we have witnessed changes with
regard to interdependencies in distribution networks. Before the ‘Modern Business Enterprise’
was established around 1900 (see Chandler, 1977),
most distribution structures relied on postponement.
This is so since speculative arrangements were not
easily realized with the kind of technology available
at that time. The Modern Business Enterprise
enabled mass production and mass distribution,
which, in turn, required standardization of products
and processes. This led to distribution structures
based on speculation. Then new information
technology and innovations in production and
logistics equipment enabled efficient distribution
arrangements based on postponement. However,
although postponement strategies seem very
‘trendy’ in today’s business environment, speculation will always be important since it facilitates
economies of scale in production and distribution.
Many firms also try to mix these two in, for
example, mass-customization arrangements (see
e.g. Salvador et al., 2009).
So none of these arrangements are ‘the one’, but
they coexist, even though one type of arrangement
might be dominant at a certain period of time. The
important point we are making here is that they coexist because they can provide a variety of solutions
to end-customers. When distribution networks are
rearranged it has implications for the activity interdependencies and how this is managed by the firms
involved. This also means that one type of
interdependence might be exchanged for another
type of interdependence, i.e. interdependence
should, and could, not be avoided in the network.
Consequently, independence in one part of the
network results in interdependence in another part
of the network.
190
5.3. Control in the resource constellation
Control is, among other things, about influencing
others, whether discussed from a channel or a
network perspective. Weitz and Jap (1995)
distinguish between three control mechanisms:
authoritative, contractual, and normative. Authoritative control is achieved through ownership and
vertical integration or through the use of power.
Contractual control relies on incentives and
compensation regulated by contracts – explicit or
implicit. Normative control, finally, is based on
shared visions, common norms and cultures, and
trust among the parties in a relationship.
Traditionally firms were recommended to avoid
dependence on individual counterparts, which
meant that so called arm’s-length relationships
became the norm in distribution (Anderson et al.,
1997). In the 1980’s there was a shift towards what
is often referred to as Vertical Marketing Systems
(VMS) (Bowersox and Morash, 1987). A VMS is a
distribution channel where the members, for
example a producer, distributor and reseller, work
jointly in relation to their customers. Bowersox and
Morash (1987) distinguish between three types of
VMS: (1) corporate, where control is obtained by
ownership; (2) contractual where contracts creates
control; (3) administrative, where one of the
channel members is the dominant member,
legitimizing that members take control. A vertical
marketing system hence always relies on some kind
of channel captain – whom the others are expected
to obey.
Based on our discussion above it seems reasonable to believe that this type of arrangement is less
applicable in today’s business environments, made
up of specialized firms in need of joint coordination
of activities and mutual resource sharing. These
specialized actors probably do not see themselves
as ruled by a channel captain. On the contrary, each
actor has its own agenda and tries to influence the
other actors in the network in certain directions.
What we have analyzed in this article is a development process in a network context, a network
composed of a large number of specialized and
interconnected actors rather than a number of
isolated channel arrangements ruled by channel
captains.
Channel relationships have evolved from
transactional processes to collaborative relationships, heavily dependent on information sharing
(e.g. Hoyt and Huq, 2000). In the words of Weitz
and Jap (1995: 305), there has been a shift from
authoritative to contractual or normative control as
well as from discrete to relational exchange,
implying “exchanges between parties who have an
exchange history and plans for future interaction”.
Frazier and Antia (1995) give examples of how re-
The IMP Journal
Volume 4. Issue 3
tailers, distributors, and manufacturers are developing closer relationships to improve performance.
What to control and how is thus a crucial issue
for any firm in a distribution network. When they
give up authoritative control by ownership and rely
instead on relationships with external partners and
their resources, firms lose control from one point of
view. However, the internal control restricts the
firm from making use of the most appropriate
resources in the network. The issue concerning
what to have in-house and what to outsource is one
that firms need to reconsider constantly in order to
be able to economize in the resource constellation.
Finally, as pointed out at the beginning of this
paper, distribution literature has often treated
distribution dynamics as either initiated by a
producer in the form of a channel captain or by
external forces. This paper shows how changes are
initiated from all kinds of actors in the network and
in all three network dimensions. The paper
illustrates how activity interdependencies, resource
control, and variety in the web of actors interplay
with distribution dynamics.
References
Alderson, W. (1957). Marketing Behaviour and
Executive Action. Homewood, Ill.: Richard D.
Irwin.
Behaviour. A Functionalist Theory of Marketing. Homewood, Ill.: Richard D. Irwin.
Alptekinoglu, A. and Tang, C. (2005). A model for
analyzing multi-channel distribution systems.
European Journal of Operational Research, 163
(3): 802-824.
Review, 38 (4): 59-69.
Andersson, P. (1992). Analysing distribution
channel dynamics: Loose and tight coupling in
distribution networks. European Journal of
Marketing, 26 (2): 47-68.
Bello, D., Chelariu, C. and Zhang, L. (2003). The
antecedents and performance consequences of
relationalism in export distribution channels.
Journal of Business Research, 56 (1): 1-16.
Bohlin, S. (a) (2000). IBM går bakom ryggen på
åf-ledet. Computer Sweden, 63: 2000-06-19.
Bohlin, S. (b) (2000). Återförsäljare kan inte
stoppa utvecklingen. Computer Sweden, 67:
2000-07-11.
Bowersox, D. and Morash, E. (1987). Marketing
concept integration and the division of labor.
Proceedings of the 3rd Conference on
Historical Research in Marketing, 215-224.
East Lansing: Michigan State University.
191
Bucklin, L.P. (1965). Postponement, speculation,
and the structure of distribution channels.
Journal of Marketing Research, 2 (1): 26-31.
Buzzel, R. and Ortmeyer, G. (1995). Channel
partnerships streamline distribution. Sloan
Management Review, 36 (3): 85-96.
logistics service providers: Rationale and
outcomes. Transportation Research Part E, 41
(6): 495-510.
Ceruzzi, P. E. (2003). A History of Modern
Computing. Cambridge: MIT Press.
Chandler, A.D. (1977). The Visible Hand. The
Clarke III, I. and Flaherty, T. (2003). Web-based
B2B portals. Industrial Marketing Management,
32 (1): 15-23.
Cox, A. (1999). A research agenda for supply
chain and business management thinking.
Supply Chain Management: An International
Journal, 4 (4): 209-211.
Curry, J. and Kenney, M. (1999). Beating the
clock: Corporate responses to rapid change in
the PC industry. California Management
Review, 42 (1): 8-36.
Duder, J. and Rosenwein, M. (2001). Towards
"zero abandonments" in call center performance. European Journal of Operational
Research, 135 (1): 50-56.
Dwyer, W. (1982). The IBM Standards as
Marketing Strategy. Computers & Standards 1:
137-144.
Faber, N., de Koster, R. and van de Velde, S.
(2002). Linking warehouse complexity to
warehouse planning and control structure.
International Journal of Physical Distribution
& Logistics Management, 32 (5): 381-395.
Frazier, G. and Antia, K. (1995). Exchange
relationships and interfirm power in channels of
distribution. Journal of the Academy of
Marketing Science, 23 (4): 321-326.
Gabrielsson, M. (1999). Sales Channels Strategies
for International Expansion - The case of large
companies in the European PC industry.
Doctoral thesis. Helsinki: Helsinki School of
Economics and Business Administration.
Gabrielsson, M., Kirpalani, M. and Luostarinen, R.
(2002). Multiple channel strategies in the
European personal computer industry. Journal
of International Marketing, 10 (3): 73-95.
Gadde, L-E. (1993). Evolution Processes in
Distribution Networks. In. Sharma, D. (ed.)
Advances in International Marketing, (5): 4366. Greenwich: JAI Press.
IMP Journal, 2 (3): 36-52.
The IMP Journal
Volume 4. Issue 3
Gadde, L-E. and Håkansson, H. (1992). Analysing
change and stability in distribution channels - a
network approach. In: Axelsson, B. and Easton,
G. (eds.) Industrial Networks - A New View of
Reality, 166-183. London: Routledge.
Gunasekaran, A. and Ngai, E. (2005). Build-toorder supply chain management: a literature
review and framework for development.
Journal of Operations Management, 23 (5):
423-451.
Håkansson, H. (1992). Evolution processes in
industrial networks. In Axelsson, B. and Easton,
G. (eds) Industrial Networks - A new view of
reality, 129-143. London: Routledge.
Routledge.
Håkansson, H. and Waluszewski, A. (2002).
Managing Technological Development: IKEA,
the environment and technology. London:
Routledge.
Networks. Chichester: John Wiley & Sons Ltd.
Hayes, M.H., Jenster, P.V. and Aaby, N-E. (1996).
Business Marketing: A Global Perspective.
Boston: Richard D. Irwin, R.R. Donnelley &
Sons Company.
Henders, B. (1992). Positions in Industrial
Networks - Marketing Newsprint in the UK.
Doctoral thesis. Uppsala: Uppsala University.
Herbig, P. and O’Hara, S. (1994). Industrial
distributors in the twenty-first century.
Industrial Marketing Management, 23 (3): 199203.
Hibbard, J., Kumar, N. and Stern, L. (2001).
Examining the impact of destructive acts in
marketing channel relationships. Journal of
Marketing Research, 38 (1): 45-61.
Hoyt, J. and Huq, F. (2000). From arm’s-length to
collaborative relationships in the supply chain.
International Journal of Physical Distribution
& Logistics Management, 30 (9): 750-764.
Hulthén, K. (2002). Variety in Distribution
Networks: A Transvection Analysis. Doctoral
thesis. Gothenburg: Chalmers University of
Technology.
Kannan.V. and Tan, K. (2005). Just-in-time, total
quality management, and supply chain
management; understanding their linkages and
impact on business performance. Omega, 33
(2): 153-162.
Kaplan, M. (2002). Acquisitions of Electronic
Commerce Capability: The Cases of Compaq
and Dell in Sweden. Doctoral thesis. Stockholm:
Stockholm School of Economics,.
Kurnia, S. and Johnston, R. (2001). Adoption of
efficient consumer response: the issue of
mutuality. Supply Chain Management: An
International Journal, 6 (5): 230-241.
192
Langlois, R.N. (1990). Creating external
capabilities:
Innovation
and
vertical
disintegration in the microcomputer industry.
Business and Economic History, Second Series,
19: 93-102.
Langlois, R.N. (1992). External economies and
economic progress: The case of the microcomputer industry. Business History Review, 66
(1): 1-50.
Magretta, J. (1998). The power of virtual
integration: An interview with Dell Computer’s
Michael Dell. Harvard Business Review, 76 (2):
73-84.
Marchetti, M. and Brewer, G. (1996). Nurturing
relationships with independent reps. Sales and
Marketing Management, 148 (9): 43.
Mattsson, L-G. (1985). An application of a
network approach to marketing: Defending and
changing positions. In: Dholakia, N. and Arndt,
J. (eds.) Changing the Course of Marketing:
Alternative Paradigms for Widening Marketing
Theory, 263-288. Greenwich: JAI Press.
McCammon, B. (1964). Alternative explanations
of institutional change and channel evolution. In
Greyser, S. (ed.) Toward Scientific Marketing,
477-490. Chicago: American Marketing
Association.
Morris, L.J. and Morris, J.S. (2002). The changing
role of middlemen in the distribution of personal
computers. Journal of Retailing and Consumer
Services, 9 (2): 97-105.
Narus, J. and Anderson, J. (1996). Rethinking
distribution - adaptive channels. Harvard
Panayides, P. (2005). The impact of organizational
learning on relationship orientation, logistics
service
effectiveness
and
performance.
Piore, M. (1992). Fragments of a cognitive theory
of technological change and organizational
structure. In Nohria, N. and Eccles, R. (eds.)
Networks and Organizations. Structure, Form
and Action. Boston: Harvard Business School
Press.
Richardson, G.B. (1972). The organization of
industry. The Economic Journal, 82 (327): 883896.
Salvador, F., Martin de Holan, P. and Miller, F.
(2009). Cracking the code of mass
customization. MIT Sloan Management Review,
50 (3): 70-78.
Samli, C. and Wills, J. (1986). Strategies for
marketing computers and related products.
Shevchenko, A. and Shevchenko, O. (2005). B2B
e-hubs in emerging landscape of knowledge
based economy. Electronic Commerce Research
and Applications, 4 (2): 113-123.
The IMP Journal
Volume 4. Issue 3
Stern, L. W. and El-Ansary, A. (1992). Marketing
Channels, 4th edition. Englewood Cliffs:
Prentice Hall.
Thompson, A. and Strickland, A.J. (1999).
Strategic Management – Concepts and Cases,
11th edition. Boston: Irwin McGraw-Hill.
van Hoek, R. (2001). The rediscovery of
postponement - a literature review and
directions for research. Journal of Operations
Management, 19 (2): 161-184.
Weitz, B. and Jap, S. (1995). Relationship
marketing and distribution channels. Journal of
the Academy of Marketing Science, 23 (4): 305320.
193
Wirtz, B. (ed.) (2007). Handbuch Multi-ChannelMarketing. Wiesbaden.: Gabler.
Zinn, W. and Bowersox, D. (1988). Planning
physical distribution with the principle of
postponement. Journal of Business Logistics, 9
(2): 117-136.
Additional sources:
www.pc-history.org
www.commadore.ca
Network World, June 21, 2010.
Creative Computing, Vol. 10, No11, November
1984, Page 135.
The IMP Journal
Volume 4. Issue 3
194
Opportunities and constraints for intermediaries
in distribution: The challenge of variety
Leif-Magnus Jensen
Department of Marketing and Logistics, Jönköping International Business School, Jönköping, Sweden.
E-mail: leif-magnus.jensen@jibs.hj.se
Abstract
The variety in contemporary distribution systems is large and reflects complex customer demands and new
technological and organizational possibilities. Changes in the distribution system lead to new challenges and
constraints for intermediaries trying to establish consistent roles for themselves. The challenge for theory is how
to describe these changed and complex roles.
This article is based on a case study of one intermediary (a third-party logistics provider) in the car
distribution industry. The article presents a framework of six roles, four of which find parallels in functionalist
discussions of roles, and two of which appear more closely tied to new developments in distribution. These are
specifically tied to the way intermediaries have increased opportunities to act as resource providers and
organizers in contemporary distribution.
The opportunities and constraints for intermediaries in the type of distribution context studied are analyzed
using the IMP literature. This proves fruitful in particular through discussing the impact of actor bonds and
resource ties. The resources the intermediary controls may be less important in defining its role than the fit with
resources possessed by others and the resulting opportunities.
Keywords: Variety in distribution, roles of intermediaries, car industry, case study, opportunities and constraints
1. Introduction
Contemporary distribution systems incorporate a
number of new features, combined with features
that have been part of distribution systems for a
long time. Taken together, these features represent
variety because there are many alternative ways of
structuring the system. The label “modern business
systems” is intentionally wide, with some of the
features pertaining to customers, some to the
organisation of production and distribution and
others to the nature of the firms involved. This
paper divides this label into three: the degree of
customisation to the end user, features relevant to
the system as a whole and, finally, the issue of
actors and specialists.
Modern business systems frequently have a high
degree of individualisation to end user requirements, which reflects a trend towards increasing
customisation (Hulthén and Gadde, 2007). This
customisation can occur both in terms of the
product choice or the provision of a high degree of
product variety, through distribution channels that
provide goods through alternate routes. The
background for the high degree of customisation
comprises a number of factors. Firstly, customers
are more demanding and expect to have a wide
choice of features for products such as cars,
personal computers or mobile phones. Secondly,
firms may see the ability to offer such variety as a
source of competitiveness and may therefore
increase the variety of their offerings to tap new
consumer segments. Finally, many industries now
have the technological ability to offer such variety.
The tendency to postpone activities until actual
customer demand is known is typical of these
approaches (Hulthén and Gadde, 2007).
A second major feature of modern business
systems is the proliferation of channels used to
reach customers. An instructive example is the IT
industry, which has seen several types of
distribution systems during its relatively short
existence. Because IT competence in the customer
population varies widely, a manufacturer should
serve a range of people, from those who assemble
components from a variety of manufacturers to
those who require complete installation of the
system at home, as well as extensive follow-up
The IMP Journal
Volume 4. Issue 3
(Morris and Morris, 2002). This makes the IT
sector a prime example of serving different
customer segments in different ways, which is often
described as hybrid distribution (Hulthén, 2002).
The final feature of modern business systems
that is raised here is the considerable use of and
reliance on specialists to carry out a range of tasks.
An important result of the variety in the distribution
system as a whole is considerable growth in the
number of specialists and, consequently, in the
degree of complexity (Bot and Neumann, 2003,
Cespedes and Corey, 1990, Coelho and
Easingwood, 2004). The activities and resources
necessary to carry out distribution are increasingly
found outside the firm, which necessitates interorganisational agreements.
This means that counterparts are becoming
increasingly important for firms but also that it may
be necessary to work with different firms in
different ways depending on the end customer and
the specific product. A manufacturer may work
with a range of firms in order to distribute its
products through retail stores as compared to an
online store, which clearly presents a substantial
challenge for manufacturers (Gadde and Hulthen,
2009). However, considering a distribution
specialist instead, such as a logistics service
provider that a manufacturer uses for only a few
specific tasks, reveals an even more challenging
situation. This specialist must create enough
business for itself by carrying out a limited set of
tasks for a number of manufacturers and other
firms, all of whom may have different
requirements, resources and approaches to their
service providers. This is the challenge that this
paper addresses.
The changes in distribution are relevant to all
actors in the distribution system but particularly
those that act as intermediaries between the
manufacturer and customer. This paper uses the
concept of an intermediary as a more general term
for the distribution specialists or logistics service
providers in the empirical case. In general,
increased variety in the distribution system should
lead to increased variety in how these intermediaries operate. In other words, the roles taken
by intermediaries might be expected to change and
become more varied in a more varied distribution
system. However, there is no consensus regarding
the effect of these changes (Agrawalh et al., 2001,
Bakos, 1998, Barnes and Hinton, 2007, Berglund et
al., 1999, Giaglis et al., 2002) and whether they
present new opportunities or a threat to intermediaries overall. Because of the increased variety
and unresolved effects on intermediaries, a study of
their role in present systems is particularly salient.
This paper focuses on the variety of roles for
intermediaries and how this fits with the variety in
the distribution system as a whole.
195
2. Purpose and outline of the paper
The purpose of this paper is to explore and
analyse how variety in distribution systems is
reflected in the variety of roles of intermediaries in
distribution. The question of what opportunities or
constraints appear for intermediaries that are specifically tied to distribution systems with considerable variety will be explored by considering
previous research on roles and analysed using the
IMP literature. Section 3 of this paper presents the
analytical framework, focusing on activities,
resources and actors as well as previous literature
on distribution and roles. Section 4 describes the
method that was employed for the empirical study,
while Section 5 presents the empirical case, focused
on one particular third-party logistics provider
(3PL) in the car distribution industry. Section 6
presents the analysis, tying together the framework
and empirical case and proposing a classification
scheme for the roles of intermediaries and then
connecting these to the issue of variety. Section 7
discusses the roles and variety in more detail.
Section 8 concludes and proposes further research.
3. Analytical Framework
The analytical framework in this paper builds on
two main streams of research. One is the IMP
school, which covers many aspects of industrial
systems, particularly the way in which both
physical and other structures are connected. The
wide focus on actors, activities and resources makes
it suitable for studying the complex systems
described in the introduction. Further, when
considering the dynamics of industrial systems, the
IMP school has various tools for explanation, since
changes can take place in any of the three ARA
dimensions. In the context of IMP, distribution is a
special or more limited case of an industrial system.
The second stream of research deals with the
roles of intermediaries in distribution or historically
in distribution channels. This literature focuses on
the activities that are necessary to successfully
distribute products from manufacturers to
customers and how this distribution system can be
organised in terms of the actors involved. A central
aspect of this process is whether to use intermediaries and what role such intermediaries are to
have. A fundamental question may be if it is even
necessary to have intermediaries at all.
3.1 Industrial networks (IMP)
A challenge for researchers in a setting
characterised by great variety is the considerable
degree of complexity. The number of different
actors involved increases the need for coordination.
The IMP Journal
Volume 4. Issue 3
Therefore, studying contemporary distribution
structures requires the right kind of lens. Industrial
networks, as a theoretical stream, has become even
more appropriate for this type of system, contingent
on a number of the changes that have lead to great
variety in modern distribution systems. A major
trend is the increased specialisation of actors
(Anderson et al., 1997, Mudambi and Aggarwal,
2003). This, in turn, necessitates increased
coordination, which must be inter-organisational in
order to succeed. The overall complexity of
distribution
structures
has
also
become
considerable. The previously mentioned specialisation leading to more actors with more coordination
between them, along with tighter integration
between firms in general due to changes such as
greater emphasis on postponement and increased
use of multiple channels of distribution, has led to a
significant aggregate increase in complexity.
Finally, firms affect each other indirectly and in
untidy ways, which creates a need to consider more
of the distribution system in order to capture its
dynamics.
In short, the picture of a tidy channel with
clearly defined actors at each step of the way is not
descriptive of contemporary distribution systems. A
network view can help draw an alternative version
that is based on the IMP layers – activities,
resources and actors. The complexities that have
been briefly outlined above will be reflected in each
of these layers.
3.1.1 Activities
Activities refer to the processes of combining,
developing, exchanging or creating resources
through other resources. Activities are more or less
tightly coupled to each other and are mutable in the
sense that single activities can be substituted and
activity constellations can be changed without
necessarily changing the outputs significantly.
Changes in activities are to be expected, which has
certainly been seen in distribution systems.
Previously, allocation tended to be by geographical
area or customer groups, with actors such as
industrial distributors carrying out a wide range of
activities (Gadde, 2000). Allocation is now done
more by specialised activities. Specialists may be
more limited in the range of activities carried out
but they typically carry them out more efficiently
by, for example, achieving economies of scale for
limited activities. This can be tightly linked to
investments in developing specialised services and
equipment.
Even if activities can be carried out more
efficiently through specialists, this does not
automatically improve the distribution system. It is
necessary to combine the operations of several such
specialists in order to carry out the full set of
activities. Indeed, some researchers have noted that
196
it is now necessary to focus on essential processes
rather than individual activities (Gadde, 2004). A
firm that might have previously needed to focus on
carrying out many different activities with some
degree of efficiency might now find it more
advantageous to develop skills in coordinating the
activities that specialists are better at carrying out.
Not only have activities become more specialised, they have also changed since more
distribution systems are based upon the principle of
postponement and try to be more customer-driven,
both in terms of customising products and avoiding
excessive inventories (Boone et al., 2007, GarciaDastugue and Lambert, 2007). This leads to a great
increase in the interdependence between firms –
activities must be tightly coupled in order to
respond to the customer in a timely manner and
there are limited speculative inventories to absorb
delays and other problems. In IMP terms, then,
many distribution systems have changed their
activity configurations to a considerable degree.
Activities are carried out more efficiently and there
is a potential gain for the entire distribution system,
although this is dependent upon a high degree of
coordination.
3.1.2 Resources
Focusing on activities, while also being mindful
of actors and resources, is consistent with several
works within the industrial network approach.
Indeed, Gadde (2004) pointed out that although the
coordination of activities has become one of the
main challenges in modern distribution systems,
this also affects resource dependence between
firms. This leads to an increased interdependency
between firms, which further reinforces the value of
a network approach and also leads to further work
specifically on resources in industrial networks
(Jahre et al., 2006).
Resources are considered to be “heterogeneous”
and have “attributes in an unlimited number of
dimensions” (Håkansson and Johanson, 1992: 32).
This fits with a changeable and adaptive distribution system: “when heterogeneous resources are
combined their joint performance increases through
experiential learning and adaptation” (Ibid). This
refers both to combination of resources within the
firm and between firms.
If activities are coordinated across firm
boundaries, it follows that resources may also be
employed across firm boundaries. For example,
“The main argument in our discussion of the
resource substance in business relationships has
been that relationships are not only a way to acquire
resources but also a way to develop resources”
(Håkansson and Snehota, 1995: 182). Sharing and
combining the resources of several firms can create
new opportunities, especially when each of these
firms employs resources adapted to carrying out
The IMP Journal
Volume 4. Issue 3
certain specialised activities. In a setting in which
resources are becoming more specialised and
controlled by different firms, the opportunities for
development through inter-organisational combination of resources becomes proportionately
greater. These specialists operate the resources at a
minimum efficient scale (Chandler, 1990), although
the success of the distribution system overall is
given by the combinations of their resources with
other firms. For example, there is no point in
developing a transport system that supports a justin-time system if there is no such system to serve.
3.1.3 Actors
Activity specialisation among actors tends to
lead to a proliferation of actors, given that more
specialists are needed to cover a complete set of
activities to carry out the distribution of a product.
The previously mentioned coordination between
actors is increasingly essential in order for the
entire distribution system to function effectively
and to reap the benefits of specialisation. Some
authors now draw a picture of “networks of valueadding partnerships like confederations of specialists” (Andersson et al. 1997). There is a
potential danger, however, that poor coordination
between the actors could nullify the gains made in
individual activities.
In addition to new activity configurations, there
are new ways of working together in the
distribution system, in which the balance of
cooperation and competition has shifted from an
environment that is more concerned with
competition to one where cooperation is also
important. The implication of increased specialisation of actors and a setting in which cooperation is
seen as more natural and legitimate is that the
traditional roles of a manufacturer, wholesaler and
retailer are no longer sufficient to capture the reality
of contemporary distribution systems. In other
words, the business relationships of the firms
involved have changed in nature. One effect of
increased cooperation is that business relationships
become more extensive and stronger for some
actors; this, in itself, should change how actors
relate to each other.
It seems reasonable to suggest that the changes
in activities, resources and the working
environment between actors would lead to
considerable changes in the respective roles that
actors can play in the distribution system. The
conception of an actor is similar to certain role
conceptions, as will be seen shortly. For example,
“Our argument that companies can be considered
actors is based on the notion of identity they
acquire in interaction with others” (Håkansson and
Snehota, 1995: 194).
197
3.2 Distribution, intermediaries and roles
One broad concept of an intermediary is that it
covers all the firms between the manufacturer and
the end-customer. Historically, this referred to a
traditional channel structure with a vertically
integrated manufacturer, a wholesaler, retailer and
an end-customer (Alderson, 1954, Alderson, 1965,
Shaw, 1912). Recently, however, it has become
more likely that the intermediary level consists of
more specialised firms providing a range of
services. “Emerging distribution constellations thus
seem to be more adequately characterised as
networks rather than channels” (Gadde, 2004: 159).
These firms may or may not own the goods they are
handling and their area of responsibility can be
limited or broad, depending on the structure of the
distribution system.
The literature has described these roles in several
different ways (Alderson and Martin, 1965, Merton,
1957, Morris and Morris, 2002, Alderson, 1965,
Johanson and Mattsson, 1992). One basic approach
is that roles say something about the expectations of
behaviour associated with a particular type of firm;
for example, that a wholesaler should focus on
assortment. When a firm violates its role (for
example, if a wholesaler starts to sell directly to end
customers), this tends to create conflict. While this
can be a useful approach , the expectation are less
clear in a complex setting where the roles are fluid
and changing. The present article argues that a role
is based on a set of activities that fulfill a certain
need or provide a specific service to a customer
(Jensen, 2009). Alderson (1965) expanded on these
activities or services in four distinct areas
(functions), which are reviewed briefly here in
order to provide an overview of the types of areas
in which roles can apply. The functions can be
summarized as the reduction of business ties,
achieving scale advantages, task and skill specialisation, and risk sharing.
Conceptually, reducing the number of business
ties when a set of customers wants contact with a
set of suppliers and uses an intermediary is a
mathematical function of the number of actors
involved. For example, through use of an
intermediary, the number of ties is reduced from 16
to eight for a setting with four suppliers and
customers. The effect is much stronger for a higher
numbers of actors. In an empirical setting, the effect
will not be as strong because it is highly unlikely
that all actors maintain ties to everyone else.
Assuming that it is costly to maintain business
links, using an intermediary can create savings. At
the same time, this may give the intermediary some
power because it now holds the interface between
the customer and supplier (Mudambi and Aggarwal, 2003).
The IMP Journal
Volume 4. Issue 3
The introduction of an intermediary into this
system will reduce costs because the number of
business relationships is reduced, assuming some
additional conditions. It must be possible to conduct
the same kind of business through an intermediary
that the customer would have been able to conduct
in direct dealings with the supplier. This raises a
number of questions, both with regard to the
intermediary as an agent of the customer or supplier
and in terms of whether the intermediary will
conduct business on behalf of the customer with the
same force as the customer would itself. There is,
of course, a danger that the intermediary will make
arrangements that are more beneficial to itself than
to its customers. Another potential danger depends
on whether it is actually possible for the
intermediary to understand the needs of the
customer and to bring these to the supplier in a
meaningful way. The intermediary may only have
this function where arms’ length dealings are
sufficient to carry out business. It is certainly
possible for an intermediary to have this function
for certain types of business, with firms doing
business directly with each other where required;
that is, a differentiated system (Wilson and Daniel,
2006).
The cost increase involved in maintaining a large
number of business relationships might not be
linear; instead, the cost may be decreasing at the
margin. A second important role for a distribution
intermediary is to achieve scale advantages for its
customers and suppliers. In a distribution setting,
two ways of achieving scale advantages seem
particularly important.
The first pertains to achieving scale in certain
basic operations, such as goods handling, order
processing and transport planning. For this, it is
necessary to pool orders from a number of
customers in order to achieve a certain critical
mass. This may simply lead to transporting full
containers, which leads to lower per-unit transport
costs, or it may be done by obtaining enough
volume to change the distribution system to a more
efficient one, such as one that employs crossdocking principles (Mudambi and Aggarwal, 2003).
Several authors have pointed to an important
feature of many technologies, namely that certain
operations or processes are, for practical purposes,
indivisible below a certain scale (Richardson,
1972). A machine for making a particular product
may have a certain capacity, below which a loss of
advantages of scale may occur, either because the
machine itself is not operating efficiently or simply
because the capital invested in the machine is not
being fully utilised if it is not working at maximum
efficiency. Conceptually, achieving scale can be
seen both as substituting capital for labour by
investing in specialised machinery and as achieving
minimum efficient scale for the machines used
(Chandler, 1990). This usually means operating the
198
machines at near full utilisation. It is easy to see
that equivalent features exist in distribution, such as
the use of reloading terminals, use of transport
facilities such as ships or trucks and in light
manufacturing carried out in the distribution
system. If this minimum capacity is much larger
than that required by any particular firm, it may be
advantageous to have the operation handled by a
specialist that also takes on the problem of finding
other customers in order to ensure full use.
Alternatively, the manufacturer must find a way to
balance its distribution capacity to manufacturing,
which may be challenging (Chandler, 1990). It is
important to note, however, that this type of feature
is not always present as some activities do not
provide the mentioned benefits of scale, at least not
to an appreciable degree. The minimum scale of
operations argument should not be ignored in
distribution because it may be as important as in
manufacturing.
The third basic function of an intermediary is to
provide task and skill specialisation. This point may
be similar to the economies of scale and reduce the
business ties argument, in that the intermediary has
a specific role with regard to certain tasks.
However, while the previous two functions did not
assume any superior competence on the part of the
intermediary, task and skill specialisation is based
on competence. In order to have a useful role, the
intermediary must either carry out operations with
greater skill and efficiency (if this skill is due to
experience curve effects, it is admittedly very close
to the advantages of scale argument) or represent a
qualitatively better way of doing business. In a
distribution setting, the intermediary would
specialise in certain distribution tasks, such as
coordinating orders, transport, financing or
warehousing operations. A second dimension could
be the customers served or the specific products or
services provided (Porter, 1996). Alderson pointed
to sorting as a fundamentally important skill for
intermediaries: “The justification for the middleman rests on specialised skill in a variety of activities and particularly in various aspects of sorting”
(Alderson, 1954: 14) The impact of IT systems,
however, may mean that some of the sorting
operations are now more easily performed; that is,
they represent a less significant part of what the
intermediaries provide. This point has been
extensively explored in the core competency
literature (Conner and Prahalad, 1996, Prahalad and
Hamel, 1990).
Building on the functionalist literature, the final
area for an intermediary is risk sharing and risk
distribution. Alderson provided three general
strategies for handling risk: shifting risk, pooling or
hedging of risk and elimination of risk through
control of the operating situation (Alderson, 1954).
Shifting of risk means moving risk from one actor
in the distribution system to another. There is a
The IMP Journal
Volume 4. Issue 3
great difference between, for example, receiving a
fee for the handling of a car and having the same
car in inventory with tied-up capital if it does not
sell. The shifting of risk from other actors to the
intermediary is only advantageous for the system if
the intermediary is better able to tolerate the risk.
The pooling and hedging of risk refers to basic risk
handling mechanisms. A typical example is a
central storage facility that holds stocks of goods
that local retailers can call upon when local demand
fluctuations means the goods are sold out. If local
demand variation is at least partially independent,
the aggregate variation is small at the central
storage facility and the total amount of inventory
needed in the system in order to achieve the same
level of availability is less than for each local
retailer.
The third point, the elimination of risk by
controlling the operating situation, is particularly
interesting here. In effect, it means that an
intermediary can reduce the risk of error in the
system by taking over important operations and
making these more reliable through the application
of standards, knowledge and competence. If the
intermediary is able to carry out operations better
through specialisation, this is a fundamental justification for employing the intermediary.
This review of the roles of intermediaries shows
that there are several different concepts of a role
and that there are some fundamental areas in which
an intermediary can contribute to a distribution
system. There is, however, a challenge in selecting
from these different concepts and seeing the
relevance to contemporary distribution systems that
have a great deal of variety and are characterised by
networks. The conceptions of roles developed in the
literature, by Alderson for example, is based on a
particular distribution reality. If we assume that this
reality has changed, it is likely that the theoretical
concepts must also be adapted to reflect a new
reality. In other words, if the reality is becoming
more network-oriented, the theoretical framework
appropriate to it may also be more networkoriented.
3.3 Third-party logistics providers
A central theme in this article is the role of
intermediaries in distribution. However, intermediaries can be considered a very broad topic. For
the purposes of an empirical study, it is better to
focus on one particular type of intermediary or, at
the very least, an easily identifiable type of firm.
This allows for the required depth of study rather
than spreading the empirical inquiry across too
many different types of firms. The present study
focused on logistics service providers, specifically a
third-party logistics provider. The approach was to
study one particular type of intermediary well
rather than cover all possible types of intermediary.
199
Third-party logistics providers are increasingly
important in distribution and have been extensively
studied in their own right (Ashenbaum et al., 2005,
Carbone and Stone, 2005, Hertz and Alfredsson,
2003, Lieb, 1992, Lieb and Bentz, 2004, Lieb and
Randall, 1996). The industry has grown because of
the desire among shippers to outsource logistics
tasks. The 3PL industry grew by 3–5 % annually
from 1996 to 2004 (Ashenbaum et al., 2005) and 80
percent of Fortune 500 companies now rely on TPL
services (Lieb and Bentz, 2004). Indeed, 3PLs are
often described under the label “logistics
outsourcing”. In terms of the previously discussed
specialisation, this is a particularly apt example of
the changes that have taken place, both from the
point of view of significance and activity
constellations.
The development of the 3PL industry further
makes it a particularly relevant empirical area for
study with regard to variety. Initial 3PL services
were provided by firms providing warehousing and
transport services, adding to their basic service
provision. This was followed by a second wave of
firms in the express parcel delivery business and,
finally, firms that focused on financial and IT
services and even management consulting (Gadde
and Hulthen, 2009, Selviaridis and Spring, 2007).
The considerable number of mergers and acquisitions within the sector has also changed and
complicated the picture of what a 3PL provides
(Carbone and Stone, 2005), with a number of large
firms complemented by many smaller specialists.
This not only means that the firms themselves are
different; the business ties that they bring with them
also vary considerably. For example, some 3PL
firms add more and more services to a basic offer
because that is what their customers demand. In
such cases, the existing business ties are important
for explaining what the firm does.
It would be a mistake, however, to only consider
the 3PL industry from the provider side because the
changes that have allowed for the growth of the
industry come largely from the buyer side. The
major trend to outsource logistics services alone has
made the 3PL industry viable but the way buyers
treat logistics is equally important in defining the
nature of the industry. Buyers assign greater
importance to the logistics function, especially due
to changes in the way goods are produced and
distribution through just-in-time systems and
customisation approaches. High quality logistics are
paramount in such approaches because disruptions
are felt more immediately (van Hoek, 2001). The
trend towards seeing logistics as more strategic and
part of the core offer of the firm has also made
buyers more discerning and more demanding in
terms of what they actually buy from 3PLs. Taken
together, the increased importance and more
strategic nature of logistics means that buying
arrangements become more complex and long term.
The IMP Journal
Volume 4. Issue 3
“Current TPL-arrangements are characterized by
considerable variety, and these range today from
the undertaking of single activities to complete
solutions where many competences and resources
are needed” (Gadde and Hulthen, 2009: 637).
In terms of the literature, a basic definition of a
3PL is “the use of external companies to perform
logistics functions that have traditionally been
performed within an organization” (Lieb, 1992: 29).
The definition was later expanded to “a relationship
between a shipper and third part, which […] has
more customized offerings, encompasses a broader
number of service functions and is charac-terized
by a longer term, more mutually beneficial
relationship” (Murphy and Poist, 1998: 26, in
Marasco, 2008: 128) .
This is important for the purposes of the present
study since it recognises the longer term nature of
the ties between the firms. By moving activities
outside the firm, buyers are also necessarily moving
resources outside the firm and changing the actor
structure. All these dimensions should affect the
framework employed here, which consists of
activities, resources and actors as well as their roles.
3PLs reflect specialisation but they also reflect
changes since many firms providing more basic
logistics services are moving into the 3PL market
(Ashenbaum et al., 2005, Hertz and Alfredsson,
2003). Since these firms come from different
backgrounds, they also have different resources and
existing ties to other actors. For a study, 3PLs
represent a particular type of firm that reflects some
of the changes seen in distribution. 3PLs also
represent increased complexity in contemporary
distribution settings and, it can be argued, one
instance of an intermediary. Other authors have
specifically pointed out that the IMP school is
appropriate for the analysis of 3PL firms, including
dynamics and how services are designed (Marasco,
2008, Selviaridis and Spring, 2007).
4. Method
The empirical presentation herein is based on a
combination of material from the Distribution
Networks in Transition project report (Persson et
al., 2008) and a dissertation (Jensen, 2009) tied to
this project. The project studied changes in three
different industries. Car distribution was selected
for the dissertation due to the appropriateness of the
setting. This also meant that a 3PL became an important part of the data collection since this was one
of the main contacts in the project and the differing
ways of working with its customers represented
variety. Data collection for the project and PhD
were connected but carried out in several stages.
Due to limitations of space, only the most relevant
parts are presented in the present article. The
investigation was carried out in a context of
200
discovery since one of the main issues in the project
case was to obtain rich data that dealt with
contemporary roles for intermediaries.
The data collection can usefully be divided into
two parts: an initial round connected to distribution
in the car industry in general, and a later second
round that dealt more specifically with a logistics
service provider and surrounding network. The
first, broad data collection provided a familiarisation with the setting and made the second data
collection both possible and more meaningful.
After initial meetings to explain the scope of the
project and obtain agreement from the firms, the
study proceeded with interviews of key personnel
in the participant firms. Interviews were conducted
according to an interview guide using the main
categories of actors, resources and activities (the
main components of the ARA-model (Håkansson
and Snehota, 1995)). However, this method was
mainly used to structure the interviews when
necessary and to operate as a checklist of which
topics to cover, rather than as a specific set of
questions for the firms to answer. The firms were
asked to list their most important counterparts,
which helped identify the network surrounding the
firm. Any such conception is naturally limited by
what has been called the “network horizon”
(Holmen & Pedersen, 2003); that is, the firm is only
aware of and only deals with firms that are
reasonably close to it in the network or distribution
system.
It is also relevant that a number of secondary
sources were used that described trends in car
distribution in Europe in order to obtain a better
understanding of the overall industry. This made it
possible to compare trends in Norway to more
international movements. This is particularly useful
since the car manufacturers themselves are not
Norwegian.
The second round of data collection differed
from the first in several respects. The topic to be
investigated was now narrower, which meant that it
was easier to identify important interview targets,
with particular focus on the different car importers.
These were selected in order to provide variation in
terms of the size of the importer, whether it had a
regional concept, and the scope of services
purchased. A new “topic structure” that was similar
to the interview guide used in the first round was
formulated. However, in the interest of obtaining
‘active data’ (Dubois and Gadde, 2002), this was
formulated in a general sense to ensure that
important topics were covered. Active data
represents data that does not fit into predetermined
schemes or categories created by the researcher.
This could include relevant data that the researcher
had not even considered. The interview guide was
never shown to interview subjects nor was it used
directly during the interviews. The extent to which
creating such a document can create bias is a
The IMP Journal
Volume 4. Issue 3
relevant issue but it was equally important to be
clear about the main issues that were to be
investigated. Summaries of the case documentation
were presented back to the participating firms in
order to ensure consistency. Other non-interview
contacts involved sitting in on monthly coordination meetings between the 3 PL and importers,
accompanying an importer on a two-day trip to visit
the 3 PLs facility and other site visits. A total of 27
interviews, visits and meetings were carried out,
ranging from roughly 45 minutes to two days in
duration. These are presented in the appendix.
The quality criteria for the study were taken
from Lincoln and Guba (1985) and were
appropriate for an interpretative study. According
to Lincoln and Guba (1985), there are four
components of trustworthiness: credibility, transferability, dependability and confirmability. The
primary technique that the present study used for
data collection is triangulation of investigators by
having three interviewers for most of the first stage
of data collection. The extended site visit can be
referred to as a very limited case of persistent
observation but it would be an exaggeration to say
that this was a significant technique. Several
techniques were used for data analysis. Peer debriefing was used extensively, both internally in the
project process by creating and critiquing case
descriptions, and through the PhD guidance process
itself. Member checking was carried out at several
stages, by presenting the authors’ understanding of
the car distribution system back to member firms
and the focal logistics service provider. An
additional check was carried out by transcribing the
majority of the interviews that involved only one
investigator. This functioned both as a check vis-àvis interviewees, since they could comment on the
interview, and as a more permanent record, which
made it theoretically possible to trace more of the
data collection process.
One key advantage of this study is that it deals
with an important intermediary firm (a logistics
service provider) that is well placed to effect
economies across different distribution systems.
The focal LSP (or 3PL) is responsible for roughly
80 percent of new car transportation in Norway,
which makes it central to the system. This, coupled
with the fairly extensive background research on
car distribution tied to the research project, provides
an excellent overview of some of the most
important elements of the distribution system, not
only in Norway but in its larger setting. The
interviews were conducted mainly with personnel
who were well positioned in terms of the
distribution system and problems related to
distribution. More information could possibly be
gained from a greater understanding of the
marketing function in terms of a background for
some of the decisions made, since the logistics
department in several cases considered the
201
incoming flow of cars to mainly be exogenous
shocks that they have to deal with. However, this
must be balanced against the additional use of time
and resources required to carry out such interviews.
5. Case: Car distribution in Norway
The empirical material presented here is based
on an extensive study of car distribution in Europe
and Norway, with particular focus on one logistics
service provider. In order to focus on elements of
variety, only a short introduction to the overall
setting is provided, followed by a presentation of
the main activities normally carried out in this
distribution system. These activities are quite
general, however, and the final section presents
how the logistics service provider (labelled here as
3PL Comp) provides different services and
activities to different actors in the distribution
system.
5.1 The car distribution system
The topic of car distribution covers the distribution of finished cars from factories to the endcustomer. Car distribution in Europe and Norway
is, by and large, carried out through a system of
franchised dealers. In the basic franchise system,
contracts are granted to franchised dealers, who
receive the exclusive right to sell a particular brand
of car in a particular area. Dealers buy cars from the
manufacturer and risk not being able to sell them.
They are also limited in that they cannot operate
outside their own area and must follow manufacturers’ standards as well as providing after-sales
service and workshop services. On the other hand,
the structure is set up so that dealers only have to
focus on competing with other brands, and the basic
margins for the dealers on sold cars have
traditionally been considered quite high.
The high degree of manufacturer power and
control over the distribution system is perhaps a
dominant feature of the system. Manufacturers
decide on most of the relevant standards in both
distribution and retail, often through a large
logistics department or organisation, and although
most manufacturers do not own the majority of
their retail outlets, they have, to a large extent,
structured the entire distribution system. In 2000
there were 55,000 franchised outlets in Europe,
representing 45,000 owners. The dealers are
generally small and fragmented, although some
chain-like structures are emerging in some
European countries. Compared to the manufacturer,
however, these are still not significant in terms of
size.
Furthermore, large manufacturers normally have
a logistics department or organisation that is responsible for the overall flow of goods and
The IMP Journal
Volume 4. Issue 3
202
Figure 1: General organisation of the franchise system
logistics standards. National importers are typically
responsible for ensuring that standards are followed
in specific markets. Service providers are used
extensively but usually just for limited tasks. Figure
1 illustrates the main actors involved, as well as a
general summary of their responsibilities.
5.2 Activities in the distribution system
Although the activity structure in car distribution
can vary between countries and manufacturers, it is
possible to show a general set of activities that are
normally required in order to distribute a car.
Changes to the distribution system can also change
the activity structure and the division of tasks
between the actors. In this section, the activity
structure in Figure 2 can be seen as the basis for the
system. It is worth emphasising that these activities
mainly pertain to the physical aspects of the
distribution system and that the physical flow has
been used as a way to structure the data throughout.
Finished cars at the manufacturing plant are
loaded on railway or road transport (loading and
transport). They are then normally transported to a
port (in the case of Norway) and loaded onto shortrange feeder ships or on large deep-sea transport
vessels if produced overseas (unloading and
handling).
A shipping company is then responsible for the
cars until they are unloaded at the receiving port in
Norway (transport/shipping unloading, handling).
Until they are unloaded at the port, the cars are
generally managed by the logistics division of the
manufacturer. In most cases, customs clearance is
handled by the importer for a particular brand, but
3PL Comp can also assist with this.
It is only once the cars are unloaded that 3PL
Comp takes over responsibility and physical
possession of the cars. Upon arrival at the port, cars
are unloaded and placed in 3PL Comp’s “48-hour
zone”. The unloading from ships is carried out by a
firm that has an exclusive contract with the port
It is only once the cars are unloaded that 3PL
Comp takes over responsibility and physical
possession of the cars. Upon arrival at the port, cars
are unloaded and placed in 3PL Comp’s “48-hour
zone”. The unloading from ships is carried out by a
firm that has an exclusive contract with the port
authorities. Once cars are in the “48-hour zone”,
they are sorted and moved to transport, storage or
services within 48 hours. The destination depends
on whether the cars are intended for a specific
customer and whether they need PDI (Pre-Delivery
Inspection, preparing the car for the customer) or
other services. For some manufacturers, PDI has
already been carried out at this stage, whereas
Figure 2: Main activities in car distribution, based on Norwegian findings
The IMP Journal
Volume 4. Issue 3
others use 3PL Comp for a full set of services. 3PL
Comp provides storage services at its site in the
Norwegian port. The required destination and
services for each car are based on the service list
that the customer provides to 3PL Comp.
For storage and maintenance, 3PL Comp
operates car-parking facilities with a total capacity
of 11,000 cars, 4500 of which are under roof.
Billing for cars in storage is carried out once a
month for each customer. It is usually not necessary
to categorise the cars because each car is allocated
the same space. The price for storage is agreed as
part of the customer contract.
PDI is a set of activities that involves making the
car ready for the customer. This consists of
removing transport protection on the car (such as
plastic strips to protect corners and removal of the
wax with which some cars are covered when they
leave the factory), washing, polishing and internal
cleaning. It also includes installation of optional
extras such as stereo systems or equipment such as
tow bars in addition to various interior details.
These operations are relatively simple and often
included in the PDI operations but they are not
technically part of the removal of transport
protection.
PDI must be carried out close to the final
customer so as to avoid the car being unprotected
for a significant period of time. There is, however, a
trend towards carrying out PDI at a regional level.
Therefore, the last transportation leg either needs to
be closed (closed railway wagons or trucks) or a
final cleaning by the dealers is required.
Another category of services offered at 3PL
Comp is car modifications and rebuilding. Modifications to cars are firstly carried out in order to
comply with the prevailing tax regime. Changing
the number of seats or the size of the luggage
compartment can allow the car to fall into a more
advantageous tax bracket, making it more attractive
for the customer. A second group of modifications
is cars for trades people or car fleets for firms,
which have mounted workshop tools, extra
windows and other special equipment. A final
category is special equipment for certain regions,
such as extra heaters for cold climates. Some of
these operations are very simple and are carried out
as part of the PDI process, since they do not require
a specialised workshop. For this type of process, the
installation is not included in the regular production
process because it is only relevant for a small
percentage of the factory output (that is,
Scandinavian volume is relatively low by European
standards).
Furthermore, since the operations are quite
similar and can be done for a number of different
manufacturers using the same equipment, a service
provider such as 3PL Comp is likely to achieve
similar economies of scale in these operations as
the manufacturer.
203
It is the planning office’s responsibility to “build
loads”; that is, plan the composition of transport so
that cars arrive according to the lead times and with
the best possible utilisation of trailers. There is
limited spare capacity in the system to deal with
peaks, although there is some flexibility related to
overtime and second drivers of cars. On a day-today basis, however, it is important to ensure that the
trailers going out are as full as possible, with cars
matching the specific trailer features (there are
several different models) in order to accommodate
different car features.
5.3 An intermediary and the actor structure
Section 5.2 provided an overview of the main
activities carried out in the distribution system and
by 3PL Comp in particular. However, the situation
as described is quite general. In fact, 3PL Comp
carries out different activities for different
customers or other actors in the distribution system.
In order to describe these differences, Figure 3
briefly outlines the main actors involved in the
distribution system from the point of view of 3PL
Comp. A subset of these is then selected for further
presentation in Section 5.4.
The focal firm in this part of the study, 3PL
Comp, is a third-party logistics provider responsible
for transporting a large proportion of the new cars
in Norway, as well as a number of associated
services such as handling, storage, PDI (predelivery inspection) and some limited rebuilding of
cars. The firm provides different services to
different manufacturers (represented through the
importers in Figure 3) and directly to dealers
depending on manufacturer requirements.
In this case, the importers are the direct
representative of the manufacturer and each has a
manufacturer system to consider. However, the
importers generally draw the boundaries around the
Norwegian part of the system. The links between
the different firms are those that are directly
uncovered during the study. Clearly, there may be
more informal links between other actors and this
only shows part of the overall car distribution
network.
The two most important groups are the main
subcontractors and customers. The main customers
are the car importers and car dealers and it is the
pooled demand from these parties that forms the
principal basis for the operation of the 3PL Comp
system. In other words, the aggregate demand for
transport and associated services enables 3PL
Comp to operate a large distribution system and
smooth out some of the inevitable fluctuations in
demand. 3PL Comp’s planning office is responsible
for dealing with order planning on a daily basis and,
in this sense, it is a core activity for 3PL Comp. The
main subcontractors are transport operators, which
can be small trucking firms or railway providers.
The IMP Journal
Volume 4. Issue 3
204
Figure 3: 3PL Company and its main counterparts
Some railway services are contracted through
3PL Comp agents. Specialised repair shops may
provide some services directly to the customers or
to 3PL Comp if required.
The two remaining actors in the simplified
picture are the regional competitors and the Port
Authorities, an important counterpart in terms of
3PL Comp’s location at the Port and the expansion
of the facilities there. The flexibility of the Port
Authorities in allowing 3PL Comp to expand and
modify their operations is, of course, important
since it would otherwise present a significant
obstacle for the firm. 3PL Comp sometimes uses
the agents of its main competitor, and vice versa, in
order to ensure delivery reliability.
The most significant aspect of this picture is that
3PL Comp organises a large part of the distribution
system vis-à-vis the main customers, both in terms
of daily operations and system maintenance. Their
customers range considerably in size but none of
them are large enough to efficiently organise their
own distribution system. If 3PL Comp was
removed from the picture (and not replaced by an
equivalent firm), the necessary amount of interorganisational contracting and communications
would increase greatly, since every dealer or
importer would then have to deal with many
different subcontractors, agents and even repair
shops. It is not possible to predict exactly what such
a system would look like but it is clearly adventageous for 3PL Comp’s customers to only have to
deal with 3PL Comp rather than all the subcontractors. The arrangement is also advantageous
for 3PL Comp in that it cements the position of the
firm in the distribution system and makes it a
natural provider of additional services, since so
many of the basic services are already in place. This
does not mean that 3PL Comp is the only
alternative, rather that its position makes it well
placed to obtain new business, especially where the
importer does not have a large organisation to carry
out activities such as PDI, modification and storage.
It became clear from the interviews with the car
manufacturers that they preferred a single logistics
interface for distribution and other services,
especially one that operates regionally and
particularly where the manufacturer has a regional
concept. This clearly counteracts the considerable
splitting of activities seen above but the strength of
the trends is not obvious here.
5.4 Counterparts and variety
Variety in the distribution system is reflected in
how 3PL Comp interacts with other actors in the
system, which is typically reflected in the activities
it carries out for customers. Such variety can be
studied by considering 3PL Comp in its dealings
with different types of actors but also in how it is
used differently by similar actors. The core of the
empirical material here pertains to three different
importers and the different ways in which they
employ 3PL Comp. This is supplemented by
considering how the group of small transport
providers interact with 3PL Comp. The previous
figures of standard activities in car distribution
(Figure 2) and the main actors involved (Figure 3)
provide an important backdrop to this section.
The IMP Journal
Volume 4. Issue 3
205
Figure 4: Importer 1 and 3PL comp
Importer 1 represents one of the smaller brands in
terms of the Norwegian market but has experienced
steady growth over the last few years. The
manufacturer has established a regional system in
the Scandinavian countries, with a hub in Sweden,
but the Norwegian volume bypasses this hub and
goes directly to Norway. Due to the regional
system, however, there is only a sales office and
dealer network in Norway. Some support for this
network is provided by the regional hub, with the
remainder provided by 3PL Comp. This leads to a
range of different services. 3PL Comp provides
transport services from the port to the dealers,
creating benefits in terms of frequency and quality
for Importer 1. Quality can be related to how well
the drivers handle the cars and adhere to standards
in order to prevent damage to the cars. Ensuring
that the drivers handle the cars well is one of the
main tasks for 3PL Comp since experience shows
that inexperienced or careless drivers can vastly
increase the damage rate for vehicles, thereby
affecting profitability.
3PL Comp also provides PDI services for all of
Importer 1 brand cars. All car manufacturers have
quite exact standards for PDI but, in this case,
Importer 1 has allowed some deviation from its
standards in order to fit with 3PL Comp’s PDI
system.
Importer 1 cars are also rebuilt by 3PL Comp
where appropriate, leading to a centralisation of this
process. The advantage of this is an accumulation
of volume and access to an existing workshop,
which also handles traffic from other manufacturers. Rebuilding of cars is always specific and
requires specific parts, but the general competence
in rebuilding is still applicable.
In summary, Importer 1 can be said to “piggyback” on the existing systems of 3PL Comp to a
large degree, at the cost of not having all aspects of
the system perfectly adapted to its own
requirements. However, for an importer with
relatively low volume, the benefits in terms of
access to a specialist with economies of scale are
considerable. Figure 4 shows Importer 1’s division
of activities.
Importer 2 represents one of the top three brands
in Norway in terms of sales and is very well
established. For historical reasons, this importer
owns and operates its own facility at Oslo Port,
which receives, stores and prepares cars for the
customer. The tasks for 3PL Comp are largely
limited to transport from Importer 2’s port facility
to various dealers and customers.
Three important points can be made regarding
this relatively limited selection of services. Firstly,
Importer 2 is large and the actual volume of
business is much greater than for Importer 1, even
though the range of services purchased is much
narrower. Although significant, the storage capacity
at the port facility is periodically under pressure,
which implies that it is highly relevant how well the
flow of cars from the facility to the customer is
managed. Secondly, it is important for the importer
to ensure the standards from the manufacturer are
followed. A significant part of these standards
relate to how cars are handled by the truck drivers
and, as previously mentioned, an important task for
3PL Comp is to manage and educate these drivers
to handle the cars according to the standards.
Despite its large volume, Importer 2 does not
believe it is large enough to organise such a system
itself.
The IMP Journal
Volume 4. Issue 3
206
Figure 5: Importer 2 and 3PL Comp
Thirdly, 3PL Comp is used as a benchmark for
many processes because it is seen as a professional
and well-run firm, which enables Importer 2 to
evaluate its own performance. Importer 2 has a
much smaller PDI and rebuilding operation than
3PL Comp but it nevertheless serves as an
important indicator of how well processes are run.
The opportunity for this is strengthened because
3PL Comp conducts tours of its facilities. At
several points, evaluations have been carried out to
see whether some activities should be transferred to
3PL Comp; this is clearly an incentive for the
service provider.
In summary, although the activities carried out
for Importer 2 are much narrower, the importance
of 3PL Comp can be seen here as well. Figure 5
illustrates the division of activities.
Importer 3 is another large importer but has
more brands than the other two, meaning that it
already has a challenge dealing with different
standards in its internal systems. Like Importer 1, it
buys many different services, and like Importer 2, it
represents large volume of business. However, it
uses several providers for many of the services and
often allows dealers to choose which provider to
use for different types of services.
Furthermore, it uses several KPIs, such as leadtime, as a major way of allocating contracts
between alternative transport providers. This
creates a somewhat complex situation for 3PL
Comp since it is evaluated according to other
criteria in this system and must deal with many
actual customers (mostly the dealers) on a day-today basis. In this sense, the agreements with
Importer 3 must be seen as framework agreements
for specific services such as transport, PDI
operations and rebuilding, although the actual
extent of these activities is determined by changing
needs of dealers. For example, a dealer can choose
to employ a regional transporter to deliver the cars,
or a local provider (for example a repair shop) to
carry out PDI. In some cases the dealers may have
workshops and can carry out some activities
themselves. It is also significant that most of the
rebuilding in this system takes place at the more
centralised European level by dedicated service
providers at the large hubs. This changes the order
of activities. Figure 6 illustrates this division of
activities.
A final issue not only tied to the importers but
also the subcontractors is that most of the actual
trucks used in car distribution in Norway are driven
by small one- or two-person trucking firms. As seen
from the discussion of the importers, 3PL Comp is
important from the importers’ point of view for
managing these firms. However, looking from the
other side, 3PL Comp is even more important for
the trucking firms themselves. It accumulates
business in a way that the firms themselves could
not possibly do since they represent a small part of
the overall transport capacity. 3PL Comp carries
out transport planning for almost the entire
Norwegian network, which is something the
trucking firms could not carry out themselves.
Indeed, the ability to carry out administrative
planning is an important task for 3PL Comp and,
while it sells specific transports from point to point,
one of its main contributions is to organise and
balance the transport operations for the entire
Norwegian market. In order for the distribution
system to work, it is essential that one or more
actors take on this task.
A final and perhaps unexpected feature is that
3PL Comp carries out financial buffering, paying
the trucking firms on completion for a transport
route, but invoices dealers and importers periodi-
The IMP Journal
Volume 4. Issue 3
207
Figure 6: Importer 3 and 3PL Comp (Gradient circles indicate a split
where dealers can choose whether to use 3PL Comp or local suppliers).
cally for this with relatively long deadlines. This is
helpful for the dealers but critical for the trucking
firms, which typically have very limited financial
resources to deal with late payments.
6. Analysis
When analysing this case, it is useful to follow
the main categories of activities, resources and
actors from the industrial network school. This
allows for a familiar structure and also allows this
study to introduce the concept of roles for intermediaries where appropriate rather than treating it
separately. Activities are considered first, with
particular emphasis on the overall structure of
activities in the distribution system. The discussion
then moves to the issue of resources, followed by a
more detailed analysis of actors and respective
roles. Finally, the paper discusses the issue of how
the different roles can be combined. The sections
are separated for ease of analysis only since they
are strongly related.
6.1 Activities
A core aspect of the setting and firm described
herein is that there is a large range of potential
activities that have to be carried out. Although this
is by no means unique to the setting, the clear
division into some main services, such as car
transport, handling, warehousing, PDI services,
rebuilding and administrative services, means that a
“menu-style” approach is feasible. Importers and
dealers do make different choices in terms of the
services they require and this creates variety in the
system. These choices are important for understanding how service providers such as 3PL Comp
can find business.
A service provider in this distribution system has
to decide what services to provide and what other
firms to provide them to. The question of the
services provided is partially a matter of what
activities naturally fit together. That is, the
activities the firm is most suited to carry out, given
its resources. Richardson provided a useful
definition of this type of activity: “Activities which
require the same capability for their undertaking I
shall call similar activities” (Richardson, 1972:
888). Carrying out such “similar activities” for
many different actors should enable a firm to
achieve advantages of scale, which in itself is an
important issue for intermediaries. There is logic in
finding new activities that make use of present
resources and activities that are similar to the
activities that the firm is already carrying out. For a
firm such as 3PL Comp, this might mean additional
transport services or additional modifications to
cars. Many activities are also possible because 3PL
Comp has possession of the cars, and adding more
activities fits in with the overall pattern of
activities. That is, from one point of view it makes
sense for the firm that transports and stores cars to
also carry some rebuilding or preparation, since it
already has the car ready at the facility.
In a network setting, the reverse question must
also be asked; namely, how the individual actor can
select activities that fit together and allow it to
achieve advantages of scale. Specifically, the way
in which activities are structured overall in the
network limit or create opportunities for a third
party or service provider. PDI services illustrate this
The IMP Journal
Volume 4. Issue 3
in the present case. Some manufacturers carry out
PDI services at a regional level prior to 3PL Comp
handling the cars. Others allow individual dealers to
determine whether they want to use 3PL Comp,
while yet others assign all PDI operations for their
Norwegian volume to 3PL Comp. If PDI operations
are not carried out by 3PL Comp, they can be
carried out by local workshops close to the dealers.
In all three of these instances, the choice of whether
to use 3PL Comp may be determined by the
existing activity structure, which the manufacturer
is used to and is reflected in the network, rather
than the actual capacity of 3PL Comp. These
activity structures are not static, as demonstrated by
the moves that several manufacturers have made
towards more regional distribution systems. For a
specific third party, however, they create unique
opportunities and restrictions. The same applies for
rebuilding cars, where the fact that manufacturers
do not include all modifications as part of the basic
building process means that there are residual
activities that must be carried out in the distribution
system. This is an obvious opportunity for a third
party.
6.2 Resources
The present resources and current actors’ bonds
to the intermediaries are highly important in
defining their opportunities. At one level, this is
wholly unsurprising – a car dealer needs a retail
facility and access to certain support services in
order to function adequately. More in line with IMP
thinking, certain resources for intermediaries, such
as placement of facilities and the integration of IT
systems over time, mean that resources are difficult
to copy and their value depends on others. The
location of a distribution hub is, to a large degree,
dependent on whether customers use it, and in car
distribution, where regional approaches are
important, the value of such a hub is very uncertain.
If several manufacturers move their hubs to another
country, such as Sweden, the value of a Norwegian
hub is greatly reduced, even if the resources at the
hub are unchanged.
Considering the distribution system as a whole,
there are very real constraints for the firms. The
placement of large fixed points such as ports,
warehouses and PDI facilities affect all the
participants in distribution. An intermediary that
has made a large investment in a PDI facility, for
example, must obtain consistent use of this in order
to be successful. However, this depends on whether
manufacturers and dealers want to carry out PDI
operations themselves. This may come down to a
question of costs, where the intermediary is useful
through providing more efficient operations; that is,
as a specialist for certain tasks. It can also be the
case that the manufacturer has a general policy
208
regarding PDI, perhaps one that allows local dealers
to make decisions themselves. In this case, the
possible cost advantage of the intermediary can be
matched up against non-economic considerations
such as the dealers preferring to get cars earlier and
carrying out PDI operations themselves, since this
gives them more influence over priority, thereby
improving their customer service.
The intermediary’s resources are important and
may explain why counterparts use the intermediary
instead of other companies. The ties between the
intermediary and other firms are even more
significant, however. There are good reasons why it
is even more important to consider the ties between
the intermediary and other actors than the actual
stock of resources that the intermediary possesses,
especially when considering its opportunities.
Even though an intermediary can offer certain
scale advantages, specialised knowledge and
competitive prices, it cannot control the services
required or the resources that others possess. Using
the example of 3PL Comp, this company has a
good position in the Norwegian distribution system
and can provide several high-quality services, some
of which are enhanced because it can draw on
advantages of scale. However, there are numerous
examples of the opportunities available to the
company, which are determined more by what a
specific counterpart possesses in terms of its own
resources and strategies. One of the importers has a
general policy of using several providers and a
specific KPI in order to select which one to use for
a particular transport route. Another importer
already has a significant investment in physical
infrastructure and only uses 3PL Comp for the few
services it cannot provide itself, such as transport.
Even Importer 1, which has an exclusive agreement
to use 3PL Comp for a number of services in
Norway, has a regional hub in Sweden that carries
out many tasks. The decision on whether to include
the Norwegian volume at this regional hub is made
according to criteria that are unrelated to 3PL Comp
but which completely determine the services it is
asked to undertake.
Dealers fit with the traditional definition of an
intermediary since they are part of the distribution
system between the manufacturer and end
customer. The possibility of dealers combining into
dealer groups and dealer chains, thereby accumulating business, opens up the possibility of
taking over some of the tasks that other actors, such
as LSPs, have carried out. This includes economies
of scale in PDI operations and some local or lastmile transport. A further and intriguing possibility
is that dealers will work across different
manufacturer systems and sell a number of different
cars. While this is already the case for some brands,
having one dealer group carry out the same distribution for larger regions and more brands opens up
The IMP Journal
Volume 4. Issue 3
209
Roles
(1) Hub
(4) Risk carrier
(2) Broker
(5) Resource provider
(3) Specialist competence
(6) Organiser
Table 1: Roles of intermediaries (Jensen, 2009)
new possibilities according to the logic here.
Specialised repair shops are a more limited case,
with repair shop chains now able to repair cars from
a number of different brands and areas. There is a
challenge in terms of the different technical
standards and specialised equipment involved, but
if this can be overcome, the repair shops will be in a
good position to accumulate business. It is
conceivable that they will also move into the
rebuilding and PDI operations but, again, this will
depend on what other actors choose to do. For both
of these other types of potential intermediaries,
their opportunities may be defined by the present
resources of other actors and the ties that they do
not possess compared to the established intermediaries.
6.3 Actors
An important part of this article’s approach is to
describe the roles that intermediaries can take in a
distribution system. In terms of the empirical
material, this specifically refers to the roles that
3PL Comp plays in the distribution system with
regard to certain other actors. In this sense, the
roles, as presented here, are all tied in to
observations in the case that were judged to be
relevant to the conception of roles. Where the
observations were relevant to the theoretical
framework, this was used as a point of departure. In
this sense, the first four roles are based on the
functions discussed by Alderson but they also
incorporate insights from the case and the way in
which roles are defined in this paper. The final two
roles were not found in the original literature and
were based more heavily on the case, although they
did refer to extant literature where relevant. Initial
observations here were tied to how other firms
depended on decisions that 3PL Comp made with
regard to investments in resources and the use of
resources, as well as the degree of latitude 3PL
Comp had in organising certain parts of the
distribution system. However, the connection to the
literature was far less clear for these observations
and so the final two roles are based on the case to
an even greater degree. Based on the empirical
material, it is possible to suggest a number of roles
that 3PL Comp plays in the distribution system with
regard to certain other players. The roles are listed
in Table 1 and described below.
The hub role is based on the firm being a central
node in the distribution system in terms of
information exchange, physical transport routes
and, possibly, decisions. The type of distribution
setting studied is likely to be closely tied to the
physical structure of the distribution system,
although this is not always the case, especially if
the main service is information exchange. The
firm’s role as a hub allows it to create economies in
terms of reducing the number of business ties
(Alderson, 1957, Alderson and Martin, 1965) but
also that it is able to absorb fluctuations in transport
volumes from different manufacturers when there
are many customers with varying demands in the
short term. It does not apply to systematic
variations, such as those due to low overall demand.
This situation is advantageous for individual
customers because they do not need to consider the
transport capacity for their own peaks in demand to
the same extent.
The absorbing variation argument and the role of
organising the hub is partially related to resources
in that 3PL Comp must have the systems and
competence required to coordinate daily and longterm flows. However, it is based primarily on how
the activities and actors are structured, with a
number of importers having varying volumes and a
very large number of small one- or two-person
firms carrying out the main direct transport
operations. Other transport firms that only carry out
local transport for some importers do not have the
same kind of role with regard to running the system
or acting as a hub.
As defined here, the broker role is closely tied to
the achieving scale argument described by
Alderson; that is, a broker creates different types of
advantages of scale by acting as a go-between for
different suppliers and customers. This applies both
where the intermediary is quite clearly placed
between a customer and supplier group, and where
The IMP Journal
Volume 4. Issue 3
the accumulation of scale allows the intermediary
to offer services in a different way because it
achieves scale. The main argument here is that a
broker creates scale in operations by accumulating
business. The issue is not whether activities are
actually carried out by a supplier or the
intermediary itself but that the accumulation itself
allows for better operations, for example by
investing in new machinery. Furthermore, by
representing many clients, the broker can demand
concessions from suppliers and pass some of these
on to its customers.
It is clear from the case that the hub and broker
role are tied closely together, so one cannot be
properly discussed without reference to the other.
The broker role is most obvious in terms of some of
the scale advantages achieved as an effect of
accumulating customer volume. The hub role is, to
a large extent, informational and physical, with 3PL
Comp taking on the considerable task of communicating with the small transport providers, the
large number of dealers and a moderate number of
importers. As shown in the theoretical framework,
this has the effect of greatly reducing the amount of
communication needed. The degree to which the
two roles are connected is more obvious from the
empirical case than the theoretical conceptualisation of these roles. 3PL Comp’s hub role is
strengthened because it can use the size it has
achieved by additional business to become a broker.
At the same time, the ability to absorb customer
variation under the hub role is greatly facilitated
because the firm is large and controls a large proportion of the new car transport business in
Norway. This is not to say that these two roles are
always closely connected, but this is the case here.
The specialist competence role is proposed as an
intermediary that focuses on carrying out certain
specialised tasks with particular efficiency. These
competences should be related to distribution in the
present setting and can be exemplified by 3PL
Comp’s skill in coordinating and planning the dayto-day movement of cars and fulfilling different
PDI standards without excessive loss of efficiency
in processes.
Ideally, it should be difficult for other firms to
match this competence. However, since roles are
not meant to be absolute, other firms may have
similar competencies or the competencies may not
be unique in nature. It should, however, be the case
that it will be costly for others to develop the same
competencies. It is also likely that other firms will
expect this and will defer to the specialist in terms
of certain decisions. The discussion in the case
regarding how others use 3PL Comp as a benchmark is relevant in this context.
This role is related to the specialisation of an
intermediary discussed in the theoretical
framework. If an intermediary or a third party in a
210
distribution system is to carry out certain tasks,
these tasks would be expected to be carried out
more cheaply and/or more effectively than the
actors themselves. At the same time, such a firm
becomes a source of information for other actors
who either use the firm as a benchmark to improve
their own operations or test the firm against their
own operations to see whether it is operating more
efficiently or if is there potential to re-assign the
tasks.
The risk carrier role means the intermediary
takes on some of the risk inherent in the distribution
system as part of its business. This may also imply
that the intermediary is particularly well suited to
carrying this risk. According to Alderson (1965),
the three main aspects of risk sharing are shifting,
pooling or hedging, and elimination through control
of the operating situation. Shifting risk can be a part
of the role in that the intermediary takes on
additional risk but here it is equally interesting to
consider why the intermediary can carry this risk.
Part of the reason is that the intermediary may have
superior processes or operational knowledge, which
it ties to the specialist competence role. If the
intermediary can carry out risky operations more
safely – that is, with lower damage rates – this
could be a sufficient reason to employ it. The
second argument, pooling or hedging of risk, is
inherently tied to the hub role in which variation is
evened out and tied to the size of the intermediary
compared to its customers.
The resource provider role, as presented here,
means that the intermediary provides specialised
resources in which the customer cannot or does not
want to invest. There are two aspects to this role.
One is that the intermediary makes substantial
investments in specialised infrastructure or equipment and the efficient use and high degree of
utilisation of this equipment becomes an important
issue for the intermediary. If the intermediary
successfully obtains high and efficient use, it can
provide services at a relatively low cost. The
literature has referred to this as an asset-based
approach (Persson and Virum, 2001). The second,
related aspect is an outsourcing argument. Even if
the resources are not greatly specialised, they may
be especially useful to a customer because they
provide resources that the customer does not have.
The matching of resources between the intermediary and customer may have more explanatory
power in this situation rather than the exact level of
operational efficiency the intermediary is achieving.
This can also explain why the customer chooses
one intermediary over another – namely, better
matching of resources.
In the present setting, it is possible for an
importer to integrate the concept of a resource
provider with its own resources and use certain 3PL
Comp services, thereby removing the need to
The IMP Journal
Volume 4. Issue 3
Role
Hub
Main features
Reducing business ties
Placed at central physical location
Broker
Scale advantages
Power through aggregating demand
Specialist/competence
Specialising in certain tasks necessary for
distribution/taking on outsourced activities
Risk carrier
Taking on risk in the distribution system,
improving operations to reduce risk
Resource provider
Matching resources
Making investments
Organiser
Partial 3PL/4PL argument – organising a part of
the distribution system, but no real control of flow
and does not take title
Organising involves performance of other actors
211
Observations in case
Connecting numerous customers
and suppliers
Physical location at Drammen
port
PDI machine for scale
Scale of Drammen and Malmö
operations
Day-to-day running of
system/sorting competence in
using transport resources
Modifying competence
Insurance system and clear
responsibility takes care of most
risk
Importance of operational risk
seen in BOS case
Carrying financial risk for small
providers
Different matches in the different
cases
Large investments in
warehousing, PDI, railway and
trucking
Making the system work across
different standards
Ensuring day-to-day operation
and maintaining population of
suppliers
Some influence in determining
structure of the system
Handling shocks to the system
and ensuring capacity
Table 2: Summary of the roles and their main features
develop such resources themselves. In most cases,
3PL Comp must make adaptations and compromises with regards to the standards of the different
manufacturers. However, there is some evidence of
adaptation going the other way with regards to
resources. Importer 1 has made some adaptations to
3PL Comp’s operating conditions in Norway based
on existing facilities, whereas the equivalent
operation in Sweden is held to a slightly different
standard that matches the standards of Importer 1
more exactly. In this sense, we can say that
although some of the tasks taken by 3PL Comp are
made for strategic reasons as well as for price, some
tasks are also performed with the available
resources of both 3PL Comp and the importer
strongly in mind, and these show definite
adaptation to local resources.
The final role proposed here is the organiser role,
but that is not to say that this role should “fill in the
blanks”. The role is tied to some of the arguments
made regarding 3PL and 4PL (Fourth-party
logistics providers) firms in the wider 3PL literature. The organiser role can be defined as being
responsible for making a substantial part of the
distribution system work, beyond just service
provision. This can include organising and
monitoring one or more suppliers, akin to the
tiering concept used by automobile manufacturers,
for example (Womack et al., 2007). This role is
unlikely to be reflected in the actual contracts made
with suppliers and is more likely to be shown by the
expectations of the customers when problems arise.
It is also made explicit in the present case, since
3PL Comp is expected to “deal with” the plethora
The IMP Journal
Volume 4. Issue 3
of small firms carrying out the actual transportation,
both in the short and long term. In many ways, this
role is supported by several of the other roles, given
that the capacity to organise parts of the system can
be founded on knowledge acquired through the
other roles. In the present framework, it could also
be said that the organiser role is built on having a
sufficient number of business ties to influence the
network. Care should be taken, however, to specify
that this organising applies to limited tasks and
within areas where other actors, such as the
importers, have initially allowed the intermediary
room to develop. For an intermediary without
ownership of the goods, however, it is a substantial
responsibility.
There is a significant cost to occupying this role
since the intermediary may be held responsible for
external shocks that it cannot actually influence.
This is especially the case because customers may
have lost contact with some of the goings on in the
distribution system through giving responsibility to
the intermediary. There are, however, considerable
positive effects from this situation. Most notably, it
allows 3PL Comp to make decisions regarding
operations under its designated role. This type of
autonomy makes it possible to compromise among
the different requirements from various manufacturers and customer groups.
Within the limits imposed by standards of the
different manufacturers and other organisations,
such as the EU, it is possible for 3PL Comp to
organise distri-bution in the best way it can, making
use of its considerable scale in operations. It also
anchors the firm strongly in terms of competitive
position; if it is one of the main actors making the
distribution system work, then it is likely to retain
its competitive edge for some time. This
competitive strength is presumably enhanced if the
firm has strong ties to suppliers and the skill to
make the suppliers perform.
The roles as described here can be summarised
to provide a better overview of what the authors
consider to be the main features of each role and
what observations in the case prompted the
formulation of the role. This should not be seen as a
complete description of each role, especially since
it does not summarise all the theoretical arguments,
but Table 2 provides a useful summary nevertheless.
7. Discussion
The analysis in Section 6 covered a number of
areas regarding roles and variety. However, some
over-arching issues remain. This section discusses
the issue of combining the roles developed in
Section 6. It then discusses the limits of variety and,
finally, ways of dealing with the challenges posed
by variety in a distribution system, with particular
212
focus on how intermediaries in general can deal
with these issues.
7.1 Combining roles
It is important to note that in this categorisation,
it is possible, and indeed expected, that a firm can
take on several roles, especially when these roles
are considered in relation to a number of different
customers or other firms. In this sense, the presence
of a firm in several roles in the same distribution
system is consistent with the present conception
and can be related to the idea of a role-set (Merton,
1957). That is, each role that is taken on can be
justified in terms of customer needs or some
fundamental economic purpose served, although it
is expected that each firm serves a number of such
purposes. Indeed, this may be the only way for the
firm to survive.
The framework used here to describe the roles of
intermediaries is largely used to describe what an
intermediary does for specific other firms and was
generated on the basis of 3PL Comp presented here.
For a specific counterpart, the intermediary may
have one role or a combination of roles, which can
vary from firm to firm. Not surprisingly, this study
found that the intermediaries play different roles
vis-à-vis different general types of firms, such as
manufacturers and importers. More significantly,
they play different roles between different firms,
which might be considered similar. For example,
the roles that 3PL Comp plays vary between
different manufacturers, even though these
manufacturers might be expected to have similar
requirements since they are essentially distributing
new cars into the same market. Connecting this
framework of roles with the IMP literature and the
case presented herein proves useful in a number of
dimensions.
An important point with regard to the roles is
that not only can one intermediary take on several
roles, but the roles will tend to reinforce each other.
The degree to which some actors use the
intermediary can further reinforce the usefulness of
the intermediary to others. At a very basic level, the
more actors make use of the intermediary, the
easier it is for it to fulfill some of the roles
described in the theoretical section. Many users
tend to make scale effects more pronounced and
this is tied to the broker role. This role can
sometimes be fulfilled by rearranging activities and
resources, such as investing in specific types of
equipment and using this at its minimum efficient
scale (Chandler, 1990). Likewise, the hub role, in
which an intermediary is responsible for information exchange and physical transport routes, is
more attractive when there are many customers.
Some of these advantages could also be achieved
by having a single large customer, but in car
distribution no single manufacturer can completely
The IMP Journal
Volume 4. Issue 3
213
Figure 7: Combining roles
dominate the distribution system so that having
many customers is significantly related to scale. In
this sense, it is reasonable to conclude that the
number and quality of ties that an intermediary has
to other firms is important in terms of the roles it
can play vis-à-vis specific counterparts.
Figure 7 shows some of the main connections
between the roles observed in the empirical case.
The figure should not be interpreted as making
general statements about how the roles are
connected. The resource provider and specialist
competence roles seem to be quite basic in this
case. Some specialist skills in core activities, such
as the knowledge of transport resources, scheduling
and planning transports, form the basis of 3PL
Comp’s activities. Similarly, the resource provider
role is key to the way in which the firm operates,
making investments in transport resources with
limited alternative uses. This serves as a good
foundation for 3PL Comp and these roles can be
observed for all of the importers.
The next two roles, broker and hub, are seen as
closely tied to each other. The general advantage of
doing the same things for many customers is
strongly present here – the more importers and
dealers who buy services from 3PL Comp, the
easier it is to create scale advantages. The fact that
creating scale generally depends on investments
(Chandler, 1977) means that there is a close tie to
the resource provider role.
The risk carrier role builds on specialist
competence by improving operational effectiveness
in the system and through the hub role, since this
allows the company to absorb some of the local
variation.
The “top” role in the figure, that is, the
organiser, builds on the hub role in particular, since
this gives access to information and results in
interaction with many other actors. It can be argued
that most of the other roles feed into this role. The
ability to organise part of the system requires
specialist competence. It can be based upon
providing special resources (although it does not
need to be) and it is facilitated by a certain scale.
7.2 The challenges of variety
Most of this article’s focus on variety has been
positive, in the sense that it can create opportunities
for intermediaries that are able to overcome its
challenges. However, these challenges remain
significant. When dealing with variety, the issues of
costs and limits to how much variety a distribution
system can tolerate are important. Perhaps even
more important are the tradeoffs created by variety
between different aspects of performance.
The costs of variety are the extra costs incurred
in providing greater diversity to the customer, or
meeting the customers’ preferences for product
options and variety in distribution channels. For
example, using multiple channels to reach the
customer creates maintenance costs for these
different channels. Operating both direct and
indirect channels, such as an agent network,
franchises, online purchasing or a presence at
online auctions, represents costs that may be
considerable. It is costly to have a high number of
product models and options and, although lean
manufacturing techniques can overcome some of
these issues, the costs can still affect distribution
because they involve many different products. A
further argument is that maintaining several
channels can lead to channel conflict (Webb and
Lambe, 2007). This can have several effects,
stemming from fundamental issues such as which
channel “owns” the customer and how the benefits
should be appropriated to the different channels.
In terms of a more complex system involving
many firms, there is the underlying challenge that,
the more complex the system and the higher the
The IMP Journal
Volume 4. Issue 3
214
Figure 8: Importers and activities in the case
demands on it, the more difficult it is to make the
system work. A distribution system that fits the
customer well and works for one manufacturer may
not be appropriate for another. The reason for this is
simply that there is not the capacity to make it work
properly, either because of internal limitations or
because the specialists necessary to make the
system work are not available. The demands and
standards of the different actors in a distribution
system may not be compatible. Operating standards
may be too different, demands too difficult to
combine or information available too late for an
intermediary to fulfill the roles discussed above.
The result can be that the opportunities for intermediaries are limited and the efficiency of distribution may be significantly reduced.
Figure 8 shows the activities that are carried out
for the different counterparts, which represents
opportunities for achieving at least some
advantages of scale. All counterparts require
transport services, whereas rebuilding is in the main
limited to one importer. In this sense, there is an
opportunity to create scale in transport services,
although this is limited by the different
requirements that each manufacturer demands in
terms of how its cars are to be handled. The same
applies for PDI where procedures are highly
specified. The investments 3PL Comp has made
both in transport, warehousing and PDI
infrastructure matches the picture of these activities
as central.
The basic tradeoff observed here is that
specialists can carry out activities for many
customers, leading to scale efficiencies, but this
advantage is reduced by adaptations to customers
with too many differing demands. The customers
themselves may be willing to adapt their demands
in order to achieve efficiencies but two more
obstacles immediately appear in the pursuit of this
goal. One is that considerable exchange of information and negotiation is necessary in order to
define exactly how demands can be adjusted to
create efficiencies overall. The second is that if the
roles of intermediaries are, to a great extent, defined
by the resources and activities carried out by others,
then adaptations for efficiency must happen in the
ongoing business relationships between the firms.
7.3 Handling the challenges
For an intermediary in general, the question
becomes how to handle the challenges posed by
variety, being aware that they cannot always be
fully overcome. The arguments made here can be
summarised and related to the motivation for the
study as follows.
The increasing variety in customer demand
requires adaptation through the use of multiple
channels, modularisation of products and greater
emphasis on the principle of postponement. In order
to make the resulting complicated distribution
system work, it is necessary to specialise, which
leads to firms such as intermediaries taking on
important roles in the distribution system. This
increases the need for coordination and places
greater demand on all the participants in the
distribution system, since there are reduced buffers
in the system.
Sub-optimal use of the system represented by
channel conflicts may lead to very poor performance. If the system is operated successfully,
however, it is possible to keep costs down while
simultaneously serving existing and new customer
groups better. At the same time, variety in products
and distribution channels has no inherent value;
rather, it must be matched with customer requirements and the capabilities of focal firms and the
distribution system.
Table 3 summarises some of the main features
seen in distribution systems for handling variety,
their main benefits and also some of the challenges
and costs that arise as a result.
The IMP Journal
Volume 4. Issue 3
Feature to handle variety
Multiple channels and
products
Use of specialists to handle
increased complexity
Modularisation and
customisation approaches
Increased postponement
Benefits
Reaches new customers
and serves present
customers better
Benefits of specialisation
in scale, scope and core
skills
Serving customer better
and more cheaply
Less obsolescence, better
fit with actual demand
215
Challenges/Costs
Increased complexity and increased
costs, including manufacturing,
channel conflict
Increased need for coordination,
higher demands on system
Challenge of complexity in the
system, higher performance
standards
Higher performance standards
needed, potential out of stock
Table 3: Main tensions in coping with variety
Table 3 shows some of the main areas of conflict
in this system but some of the main ways of
overcoming these areas can also be presented. The
case offered one such approach: negotiating with
one of the importers to relax the standards for PDI
services in order to achieve better overall
efficiency. This can be tied to the interaction
between the intermediary and its customers; that is,
without an ongoing business relationship with some
degree of communication, it may not be possible to
identify opportunities tied to slight adaptations by
both parties. Again, this demonstrates the
importance of ongoing business ties to the intermediary. In car distribution, manufacturers have
tended to maintain a large degree of control and
only outsource highly defined services. A better
approach might be to allow the intermediary to
offer larger and specialized bundles of services.
This can allow the intermediary to offer a
customized offer based on the needs of the
customer, while still achieving scale in specific
activities (Gadde and Hulthen, 2009). The customization creates variety and challenges for the firm,
but as shown in the case intermediaries can
overcome many of these challenges, especially if
done in concert with their customers. A mutual
adaptation can bring substantial benefits for both
parties in this case.
A second way to increase the amount of business
is to provide more of the existing services to new or
existing actors. Although the option is not explored
in the present case, 3PL Comp also provides its set
of services to a number of other importers and is
continually seeking to win new contracts. This is an
obvious way of achieving success but, following
the arguments above, may take time and depend on
developing business relationships over time in order
to understand how to provide services to the
customer. A variation of this strategy is to follow
larger customers to other countries, especially
where there is a regional concept. In this sense, 3PL
Comp is also active in terms of buying facilities in
other Scandinavian locations.
The third approach to increasing the amount of
business gained may be referred to as changing the
activity or resource constellations. 3PL Comp
purchased shares in a railway company and a set of
railway carriages that can be converted to carry
generic cargo for return trips. This has potentially
widened the opportunity to provide services, based
on the features of the present system (that is, there
is almost no return traffic of new cars and railway
carriages are normally specialised). Effectively, this
means connecting entirely standardised transport
services for generic goods to the existing system. In
this sense, the approach is not necessarily overly
demanding but can still prove very useful for the
intermediary.
8. Conclusions
This study has considered the case of a logistics
service provider (3PL) as a type of intermediary
and its roles in a particular distribution system. A
characteristic of the setting is that the intermediary
must play different roles with regard to different
types of counterparts in the distribution system but
also vis-à-vis counterparts that seem very similar.
The need to play such different roles is created by
the variety in the system, which is significant
despite the distribution of cars appearing to be
relatively uniform in the case. The research
provides a categorisation of the roles that are taken
by the 3PL. It is useful to explore the roles that
3PLs or service providers play as intermediaries
because the traditional literature has not specifically
considered service providers as relevant channel
members (e.g. Bucklin, 1966). In terms of the four
functions described by Alderson, this creates a
more updated framework that more closely reflects
the situation of this type of firm. Note, however,
that the four functions are still applicable but, as
might be expected, a framework of roles based on
what firms do for others leads to more roles in a
distribution setting with considerable variety.
The IMP Journal
Volume 4. Issue 3
For all the three dimensions of analysis used
herein – activities, actors and resources – the study
has shown that the opportunity to take on different
roles is made possible or constrained by the
structure; in other words, what others do. This is a
central point to the research in that features of the
network and the actions of others can make certain
roles untenable and others possible. This is not to
argue that it fully determines the roles taken since
problems can be solved in several ways and it is by
no means a given that an intermediary has the
resources to make use of an opportunity. It does,
however, mean that the role of an intermediary is
determined less by the resources it controls or the
activities it is capable of undertaking and more by
the resources that others have and the activities that
others are capable or incapable of undertaking.
The present framework enables one firm to take
on several roles. It has also discussed how the roles
in the present case fit together, with the organiser
role as a kind of “top level” construct, which can
only be held by firms that are already very
knowledgeable about the distribution system and
have considerable capabilities, as evidenced by
holding a number of other roles. The relations
between the roles presented here should be seen as
strongly tied to the present case, although some of
the relations may be quite typical. The broker and
hub roles, for example, seem to be closely related,
although they fulfil different functions for the
customers. On a more general level, the roles
largely seem to reinforce each other in that having
one or several roles makes it more possible to fulfil
further roles. However, this should be seen in
relation to the role framework being developed on
the basis of one large actor. Analysing the role
combinations of firms (3PLs and others) in other
settings and possible role conflicts would be an
interesting avenue for further research.
Variety may consist of different manufacturers
organising their distribution systems in different
ways to fit with strategy or historical developments.
For an individual firm in these systems, variety can
be both a challenge and a path to success. In
general, there are many permutations for
distributing cars, far beyond the limited selection
shown in the case. For an intermediary, this can be
an advantage. If cars are distributed in a number of
slightly different ways and manufacturers use
slightly different systems, this creates an opportunity for those intermediaries that can provide
similar services as a component to the different
systems. For example, working across different
handling and PDI standards may require experience
with many different systems and existing ties to
different manufacturers or importers, which, in
itself, can create an opportunity for an intermediary.
If the systems were identical, it would be far easier
for others to provide the same services or for the
manufacturers themselves to pool some of their
216
activities. Variety then increases the complexity of
the distribution system and may mean that
manufacturers require many different specialists to
carry out distribution.
The discussion shows that there are distinct
limits to how well variety can be handled, even
when using specialists. For example, having
multiple channels and a number of different
standards within a distribution system creates costs
and operational challenges that are difficult to
overcome. This paper suggests that firms have and
could employ three different strategies for handling
these challenges. One is to work through the
existing business ties in order to define the scope of
outsourcing and identify where small compromises
from several actors can lead to significant benefits.
A second is to follow present customers geographically and identify new ones that require similar
services. Although this is close to what many firms
are already doing, for intermediaries the opportunities to do so may be greater than for many other
types of firms. Thirdly, there may be particular
features of a specific distributions system that
enable creative adaptations (such as multifunctional railway carriages) to allow access to
other types of business.
Finally, there are indications in the empirical
material that the current conception of roles can be
expanded to other types of intermediaries beyond
the 3PL studied here, such as dealers, importers or
even sea carriers. However, this must be more
thoroughly explored in order to compare it with the
present classification or develop it further. This task
is left for further research.
References
Agrawalh, M., Kumaresh, T. V. and Mercer, G. A.
(2001). The false promise of mass customization. The McKinsey Quarterly, (3): 62-71.
Alderson, W. (1954). Factors governing the
development of marketing channels. In:
Clewett, R. (ed.) Marketing Channels for
Manufactured Products. Homewood, Illinois:
Richard D. Irwin.
Alderson, W. (1957). Marketing Behavior and
Executive Action: A Functionalist Approach to
Marketing Theory. Homewood, Illinois: Richard
D. Irwin.
Behavior: A Functionalist Theory of Marketing.
Homewood, Illinois: Richard D.Irwin
Alderson, W. and Martin, M. W. (1965). Toward a
formal theory of transactions and transvections.
Anderson, E., Day, G. S. and Rangan, V. K. (1997).
Review, 38 (4): 59-69.
The IMP Journal
Volume 4. Issue 3
Ashenbaum, B., Maltz, A. and Rabinovich, E.
(2005). Studies of trends in third-party logistics
usage: What can we conclude? Transportation
Journal, 44 (3): 39-50.
Bakos, Y. (1998). The emerging role of electronic
marketplaces on the Internet. Communications
of the ACM, 41 (8): 35-42.
Barnes, D. and Hinton, M. (2007). Developing a
framework to analyse the roles and relationships
of online intermediaries. International Journal
of Information Management, 27 (2): 63-74.
Berglund, M., Laarhoven, P. V., Sharman, G. and
Wandel, S. (1999). Third-party logistics: Is
there a future? International Journal of
Boone, C. A., Craighead, C. W. and Hanna, J. B.
(2007). Postponement: an evolving supply chain
concept. International Journal of Physical
Distribution & Logistics Management, 37 (8):
594-611.
Bot, B. L. and Neumann, C.-S. (2003). Growing
pains for logistics outsourcers. McKinsey
Quarterly, (2): 68-77.
Bucklin, L. P. (1966). A Theory of Distribution
Channel Structure. Berkeley: IBER Special
Publications, University of California.
Carbone, V. and Stone, M. A. (2005). Growth and
logistics service providers: Rationale and
outcomes. Transportation Research Part E, 41
(6): 495-510.
Cespedes, F. V. and Corey, E. R. (1990). Managing
multiple channels. Business Horizons, 33 (4):
67-77.
Chandler, A. (1977). The Visible Hand: The
Chandler, A. D. (1990). Scale and Scope: The
Dynamics of Industrial Capitalism. Cambridge:
The Belknap Press of Harvard University Press.
Coelho, F. J. and Easingwood, C. (2004). Multiple
channel systems in services: Pros, cons and
issues. The Service Industries Journal, 24 (5): 129.
Conner, K. R. and Prahalad, C. K. (1996). A
resource-based theory of the firm: Knowledge
versus opportunism. Organization Science, 7
(5): 477-501.
Dubois, A. and Gadde, L-E. (2002). Systematic
combining: An abductive approach to case
research. Journal of Business Research, 55 (7):
553-560.
Gadde, L-E. (2000). From marketing channels to
differentiated networks - distribution dynamics
in a historical perspective. In: Dahiya, S. (ed.)
The Current State of Business Disciplines.
Rohtak: Spellbound Publications.
resource combining in distribution networks Implications for relationship involvement and
217
the relationship atmosphere. Journal of
Marketing Management, 20 (1/2): 157-184.
Gadde, L-E. and Hulthén, K. (2009). Improving
logistics outsourcing through increasing buyerprovider interaction. Industrial Marketing
Management, 38 (6): 633-640.
Garcia-Dastugue, S. J. and Lambert, D. M. (2007).
Interorganizational time-based postponement in
the supply chain. Journal of Business Logistics,
28 (1): 57-81.
Giaglis, G. M., Klein, S. and O'Keefe, R. M.
(2002). The role of intermediaries in electronic
marketplaces: Developing a contingency model.
Info Systems Journal, 12 (3): 231-246.
Hertz, S. and Alfredsson, M. (2003). Strategic
development of third party logistics providers.
Hulthén, K. (2002). Variety in Distribution
Networks: A Transvection Analysis. Doctoral
thesis. Gothenburg: Chalmers University of
Technology.
Hulthén, K. and Gadde, L-E. (2007). Understanding
the 'new' distribution reality through 'old'
concepts: a renaissance for transvection and
sorting. Marketing Theory, 7 (2): 185-208.
Routledge.
Jahre, M., Gadde, L-E., Håkansson, H., Harrison,
D. and Persson, G. (2006). Resourcing in
Business Logistics: The art of systematic
combining. Copenhagen: Liber & Copenhagen
Business School Press.
Jensen, L.-M. (2009). The Role of Intermediaries in
Changing Distribution Contexts: A Study of Car
Distribution. Doctoral thesis. Oslo: BI
Norwegian School of Management.
Johanson, J. and Mattsson, L-G. (1992). Network
positions and strategic action - an analytical
framework. In: Axelsson, B. and Easton, G.
(eds.) Industrial Networks: A New View of
Lieb, R. C. (1992). The use of third-party logistics
services by large American manufacturers.
Lieb, R. C. and Bentz, B. A. (2004). The use of
third-party logistics services by large American
manufacturers: The 2003 Survey. Transportation Journal, 43 (3): 24-33.
Lieb, R. C. and Randall, H. L. (1996). A
comparison of the use of third-party logistics
service by large American manufacturers, 1991,
1994 and 1995. Journal of Business Logistics,
17 (1): 305-320.
Lincoln, Y. S. and Guba, E. G. (1985). Naturalistic
Inquiry, Newbury Park: Sage Publications.
Marasco, A. (2008). Third-party logistics: A
literature review. International Journal of
The IMP Journal
Volume 4. Issue 3
Merton, R. K. (1957). The role-set: Problems in
sociological theory. The British Journal of
Sociology, 8 (2): 106-120.
Morris, L. J. and Morris, J. S. (2002). The changing
role of middlemen in the distribution of personal
computers. Journal of Retailing and Consumer
Services, 9 (2): 97-105.
Mudambi, S. and Aggarwal, R. (2003). Industrial
distributors: Can they survive in the new
economy? Industrial Marketing Management,
32 (4): 317-325.
Persson, G., Jensen, L.-M., Engebrethsen, E. S. and
Flygansvær, B. (2008). Distribution Networks
in Transition. BI Norwegian School of
Management Research Reports. Oslo: BI
Norwegian School of Management.
Persson, G. and Virum, H. (2001). Growth
strategies for logistics service providers: A case
study. International Journal of Logistics
Management, 12 (1): 53-64.
Porter, M. E. (1996). What is strategy? Harvard
Prahalad, C. K. and Hamel, G. (1990). The core
competence of the corporation. Harvard
218
Richardson, G. B. (1972). The organisation of
industry. The Economic Journal, 82 (327): 883896.
Selviaridis, K. and Spring, M. (2007). Third party
logistics: a literature review and research
agenda. The International Journal of Logistics
Management, 18 (1): 125-150.
Shaw, A. W. (1912). Some problems in market
distribution. Quarterly Journal of Economics,
26 (4): 703-765.
van Hoek, R. (2001). The rediscovery of postponement - a literature review and directions for
research. Journal of Operations Management,
19 (2): 161-184.
Webb, K. L. and Lambe, C. J. (2007). Internal
multi-channel conflict: An exploratory investigation and conceptual framework. Industrial
Wilson, H. and Daniel, E. (2006). The multichannel challenge: A dynamic capability
approach. Industrial Marketing Management,
36 (1): 10-20.
Womack, J. P., Jones, D. T. and Roos, D. (2007).
The Machine that Changed the World: How
Lean Production Revolutionized the Global Car
Wars. London: Simon & Schuster.
The IMP Journal
Volume 4. Issue 3
219
Appendix: Overview of interviews and data sources
Interview or other
Presentation and feedback 3PL
Comp & Importer 2
Monthly meeting 3PL Comp &
Importer 2
Importer 2
Importer 2
Interview
Telephone interview
Interview
Position of main subject
3PL Comp and Importer 2
management
management
management
management
Sales, Importer 1
Head of Sales, Importer 1
Logistics manager, Importer 3
DNet Meeting
Importer 2
DNet Meeting
Importer 2 Retail
DNet Meeting
Head of Car Logistics, Importer 2
Interview
Planning meeting
On-site observation and talks at
Importer 2 during “boat-day”
Passive observer, Importer 2
Group meeting
Tour of 3PL Comp Facility
Sweden
Interview
Importer 2 dealer
Head of Car Logistics, Importer 2
Interview
Interview
Short talk
Tour of Port Management
Authority facilities
Meeting/presentation
Interview
Interview
Interview
Interview
Interview
Feedback meeting
Duration
2 hours
(approx.)
1 hour
1 hour
1 hour
90 minutes
30 minutes
100
minutes
45 min
(approx.)
45 min
(approx.)
45 min
(approx.)
90 minutes
2 hours
5 hours
1 ½ days
Head of Marketing
2 hours
Head of Administration Importer 2
Logistics
Head of Importer 4
General Managers, Cars Port
Management Authority
Deputy General Manager, Cars Port
Management Authority
Head of Marketing
2 hours
Managing Director, 3PL Comp
Head of Marketing, 3PL Comp
Head of Domestic Transport, 3PL
Comp
Head of Services, 3PL Comp
Sales, 3PL Comp
Head of production foreign
transport/Head of logistics, 3PL
Comp
Head of Marketing, 3PL Comp
2 hours
90 minutes
1 hour
90 minutes
1 hour
20 min
1 hour
1 hour
70 minutes
90 minutes
90 minutes
The IMP Journal
Volume 4. Issue 3
220
Organizing a network within the network:
The case of MC Elettrici
Chiara Cantù
Francesca Montagnini
Roberta Sebastiani
SeGeSTa (Department of Management), Università Cattolica del Sacro Cuore, Milan. E-mail: chiara.cantu@unicatt.it
SeGeSTa (Department of Management), Università Cattolica del Sacro Cuore. E-mail: francesca.montagnini@unicatt.it
SeGeSTa (Department of Management), Università Cattolica del Sacro Cuore. E-mail: roberta.sebastiani@unicatt.it
Abstract
In recent years, new driving forces are significantly affecting intermediation and emphasizing the centrality of
its role. From a traditional role of physical interconnection, intermediation has taken on a critical new function –
first of all by supporting interaction and facilitating customer participation within a network, and then by
creating the conditions for enhancing relationships between different organizations.
We have empirically investigated this evolution by looking at the case of an Italian electrical equipment
consortium that revised its intermediation role to better support actors involved in its distribution network and to
allow the survival of its partners, which are primarily local SMEs. The main aim of this paper is to analyze the
impact of the intermediation evolutionary process on the distribution network by focusing on the organizing of
actors, resources, and activities in a “network within the network” perspective.
Keywords: intermediation, distribution network, electrical equipment industry
1. Introduction
Current changes in distribution practices have
affected the conditions for intermediaries and
intermediation. The increasing attention to
outsourcing implies that activities undertaken by
different firms needs to be coordinated and
synchronized. This change also makes resources
increasingly spread among firms, which calls for a
focus on processes for connecting and combining
resources. These conditions give new meaning to
the word ‘intermediation’, and change the
conditions for both the intermediation and the
evolution of its functions.
From a traditional perspective, intermediation
has always played a key role in the process of
distribution, facilitating the matching of supply and
demand, supporting their interconnection and
managing the convergence between the provider’s
needs and the buyer’s needs.
The main activities of intermediation, in this
sense, are the collection, structuring and dissemination of information, stock management and the
routinizing of transactions in the exchange process
(Bucklin, 1966 and 1972). According to this view,
the main role of intermediaries in distribution is to
provide economies of distribution, by increasing
the efficiency of the process, creating the right
product at the right place and at the right time
(Alderson, 1965). The main aim of an intermediary
can be thus identified in the search for efficiency,
based on standardization of activities that characterize mass production and mass distribution
(Woollard, 1954; Piore and Sabel, 1984). Mass
distribution is based on standardized products
characterized by a predefined form, acceptable
quality, low price and minimal value-added
services. The main task of intermediation in this
system used to be to “channel” products from the
factory to the customer and, still, one of its primary
functions is to reduce the cognitive and physical
distance between supply and demand. Intermediaries adjust the discrepancy of assortments,
routinize the transaction-minimizing cost of distribution and facilitate the search processes of both
The IMP Journal
Volume 4. Issue 3
producers and customers by structuring the information that is essential to both parties (Pitt et al.,
1999).
Historically, intermediaries have been involved
in transport, warehousing, inventory, packaging and
information (Rushton et al., 2006: 6), but during the
previous decades, three main driving forces
influenced them: customer evolution, technological
changes and globalization.
First, the transition process was generated by
customer evolution (Meyer-Waarden, 2008), which
significantly influenced the change in intermediation. As users became more acquainted with
the products/services offered, they tended to require
customized solutions (Davis, 1987; Pine, 1993;
Lampel and Mintzberg, 1996; Pine and Davis,
2001, Da Silveira et al., 2001). Firms are drawn to
these exchanges “not just for the reduction in
transaction costs, but also because they create
added value by providing extended services such as
logistics scheduling, credit and payment options,
aggregated data to analyze own buying habits, and
systems integration” (Lichtenthal and Eliaz, 2003:
5). Increasingly, firms emphasize the consumption
experience and “turning the distribution activity
into a more efficient customer orientated process as
well,” outlining mass customized distribution
(Mason and Lalwani, 2008: 71). Strategies and
logistics organization have changed, focusing on
customer service and cost efficiency (see e.g.
Groothedde et al., 2005; Mason and Lalwani, 2008)
as well as on economies of scope and customized
product configuration (Pine, 1993).
The second force is that technology development
(Farshchi et al., 2009) made customization possible.
New technologies have allowed intermediation to
facilitate customer participation in the distribution
process, making the whole process of distribution
not only more efficient but also more effective.
Technology changes have affected information
technology and integrated logistics systems that
have allowed companies to monitor the availability
of products and services (Pfohl and Buse, 2000)
and to set up rapid delivery services from distant
locations to customer sites (Webb, 2002).
Furthermore, the Internet and other new
technologies have been considered key tools for
intermediation revolution. The pervasiveness of
new technologies has led to consequences for the
intermediation to overcome the barriers of space
and time. However, the Internet has not eliminated
intermediaries, but rather it has led to a phase of
reassessment and realignment between producers,
intermediaries and technology (Dewan et al., 2003).
Moreover, this realignment has been requiring
not only the evolution of traditional intermediation,
but it has also determined the entry of special kinds
of intermediaries with specific expertise concerning
products and management of information, such as
infomediaries and metamediaries (Sarkar et al.,
221
1996; Hagel and Rayport, 1997; Evans and
Wurster, 2000; Viswanathan et al., 2007).
Furthermore, the third of the driving forces,
globalization, defined as the production and distribution of products and services of a homogenous
type and quality on a worldwide basis, has really
impacted intermediation (see e.g. Scholte, 2000;
Waters, 2001; Drori et al., 2006): intermediation is
increasingly influenced by the development of
internationalization strategies with regards to the
management of global supply chains (Fraser and
Opphenheim, 1997; Flint, 2004). Consequently, this
development, together with the continuous search
for company differentiation within a global context,
required a change in intermediation.
A closer interconnection between supply and
demand was needed to support a more individualized offering system, or customized products
(Cox and Ruffin, 1998; Duray et al., 2000; Da
Silveira et al., 2001; Hvam et al., 2008). As a
further stage of development, the potentialities of
new technologies and the increasing complexity in
distribution have led to the creation of collaborative
relationships in distribution (Buzzel and Ortmeyer,
1995; Rosenbloom, 1995; Golicic et al., 2003). As
emphasized by different researchers in economics
and business, intermediation includes the linking of
different elements by considering the system of
partnerships and alliances that a firm creates to
source, augment, and deliver its offering system
(Anderson et al., 1997; Heide, 2003; Wathne and
Heide, 2004). In particular, the shift from standardization to customization requires improvements
in coordination (Gadde, 2004) as well as greater
differentiation in distribution networks (Gadde and
Ford, 2008).
As an output of our research, we pose that from a
traditional interconnecting role, intermediation is
evolving towards supporting interaction and
facilitating customer participation within a network
and – most thought-provoking – towards creating
the conditions to enhance the relationships among
different organizations. The main aim of this paper
is to explore how these changing conditions in
distribution impact on intermediation and the
business opportunities for intermediaries.
Our analysis is based on an empirical investigation of an Italian electrical equipment consortium
that revised its intermediation role to better support
actors involved in its distribution network and to
allow the survival of its partners, which are
primarily local wholesalers and retailers.
This work is articulated into two parts. The first
part, based on a review of the literature, outlines the
evolution of intermediation and, in particular,
focuses on the issues connected with organizing
actors, resources and activities. The second part is
devoted to the analysis of the case study regarding
the MC Elettrici consortium. Thereafter, research
findings and main conclusions are presented.
The IMP Journal
Volume 4. Issue 3
2. Intermediation from an industrial network
perspective
The new conditions for intermediation, outlined
above, require and support the evolution of
traditional intermediaries and the development of a
new style of intermediation (Narus and Andersson,
1996). In particular, intermediaries are rethinking
their role in accordance with the driving forces
previously described, among which the Internet
plays a key role in the distribution system (Pitt et
al., 1999; Lancioni et al., 2000). However, new
types of actors also enter, for example, firms
specialized in handling the flow of materials, such
as logistics service providers, (Carbone and Stone,
2005; Christopher, 2005) or the flow of information, such as hubs (Clarke and Flaherty, 2003),
portals and brokers of various kinds (Sarkar et al.,
1996; Sashi and O’Leary, 2002; Nault and Dexter,
2006). Furthermore, the sudden development of
new technologies has influenced the evolution of
intermediation, not necessarily only in terms of
disintermediation but also by supporting direct
relationships between suppliers and customers. This
ongoing evolution of intermediation has been
confirmed not only in theory but also in practice
(Gadde and Ford, 2008).
In addition, increasing outsourcing and
specialization also means that the structure becomes
increasingly fragmented. Greater specialization
requires greater coordination.
This means that there is room for someone to
connect (organize) activities, resources and actors
within these increasingly specialized structures.
Moreover, the evolution of intermediation generates
an evolution of the traditional model towards a
distribution network-like model. Therefore, we
notice that, in practice, distributors, previously
considered passive ‘collectors of orders,’ are in fact
becoming ‘partners in purchasing and selling’
(Lambert and Cooper, 2000; Whipple et al., 2007;
Fiocca et al., 2009).
Several scholars have claimed that distribution
practice has become network-like (i.e., Narus and
Anderson, 1996; Anderson et al., 1997; Gadde and
Ford, 2008) when considering the interrelationships
among the actors involved and their structure.
In the network approach, based on legally
autonomous organizations that have high levels of
interdependence and cooperative collaboration
(Håkansson and Johanson, 1992; Håkansson and
Snehota, 1995), each actor is considered as an
intermediary between two or more different nodes
of the network (Gadde and Snehota, 2000, 2001).
According to IMP scholars, a network framing
seems a more appropriate conceptualization than a
channel perspective to approach the evolving
supply side structures (Ford et al., 1998). In fact, a
firm develops relationships with actors to share
resources and carry out activities, overcoming
222
traditional company boundaries and the traditional
chain perspective. As emphasized by Gadde and
Ford (2008: 43), the network literature is more
appropriate as “developed during a much less stable
period when many new types of companies entered
the network providing a wider range of offerings,
less strongly based on physical products […]The
network literature relates much more closely to an
era of postponement and flexibility in production
and distribution”.
Furthermore, the industrial network approach is
more appropriate for a better understanding of
intermediation considering the centrality of
interaction and relationships that characterize the
nature of intermediation and its evolution.
Similarities between the concepts and models in
early distribution research and central features of
the industrial network approach can be summarized
into a holistic view of the world and the nature of
business exchange and the perspective of resource
heterogeneity and the subsequent call for
coordinated action in terms of adaptations (Gadde
and Ford, 2008). Despite these similarities, the
network approach emphasizes cooperative action,
complementarity of activities across firm borders
(Gadde, 2004) and inter-firm coordination (Easton,
1992).
Moreover, in the network approach, specifically,
in the distribution network, interacting is
considered the most fundamental activity of a
company (Håkansson et al., 2009). This interaction
is rooted in the exchange of products and services
and is concerned with how two companies choose
to organize the flow of goods and information
between them (Gadde et al., 2003).
The main function of intermediation can thus be
identified in the supporting interaction and the
organizing process of products and information
flow. In addition to this, IMP scholars have
emphasized the impact of interrelationships in
activity links, resource ties, and actor bonds. The
three layers provide a holistic view of the business
landscape where business relationships play a
crucial role (Gadde and Hulthén, 2009). Consequently the organizing of actors, resources and
activities becomes a critical aspect of intermediation. Actors organize activities by dividing
and coordinating them; they also organize resources
by selecting and combining them to develop
activities. When organizing the actors, it is
important to identify the actors belonging both to
the network horizon and the network context by
their different degrees of relevance to develop and
enhance the relationships (Johanson and Mattsson,
1992; Anderson et al., 1994; Håkansson and Ford,
2002; Holmen and Pedersen, 2003; Nyström et al.,
2008). Network context outlines all firms and
relationships considered most relevant by the focal
firm (Håkansson and Snehota, 1989; Anderson et
al., 1994), while the network horizon is defined in
The IMP Journal
Volume 4. Issue 3
terms of network boundaries that characterize the
focal firm (Anderson et al., 1994; Holmen and
Pedersen, 2003). As the network context and
network horizon denote the actor’s view of the
network, their arbitrary boundaries can change over
time (Huemer et al., 2004).
As underlined in previous literature (Ford et al.,
2002; Gadde, 2004), the Actors Resources
Activities (ARA) model illustrates, in distribution
networks as well, the evolution in the relationships
among actors, underlining the interdependencies
among the changes in actor bonds, activity links
and resource ties (Håkansson and Snehota, 1995).
Therefore, the ARA model supports the investigation of intermediation by outlining implications
related to the organizing of activities, resources,
and actors.
Moreover, organizing is based on the model of
business interaction that involves the three layers of
the ARA model (activities, resources, and actors)
by considering also the time and space dimensions.
Each layer of interaction is developed within the
wider activity patterns, resource constellations and
actor webs (Håkansson et al., 2009: 42). This
perspective may have an important impact on the
evolution of intermediation through business
interaction.
The change affecting interaction, in fact, cannot
be predetermined, and it requires considering the
influence generated by the parallel interaction of
the actors involved. At that very point in time, a
specific episode takes place, and it is considered
part of the change process based on learning,
adaptation, commitment and distance-reduction
(Ford et al., 2003). In a development process,
several episodes are linked together, producing a
cumulative effect over time “affecting subsequent
interaction in multiple direction” (Håkansson et al.,
2009: 36). Different episodes can be characterized
by great variation in the intensity of interaction
(Håkansson et at., 2009). Moreover, interaction
affects and is affected by the way that actors,
resources and activities are positioned in relation to
each other (Johanson and Mattsson, 1992). Every
actor, in fact, has a specific position within the
network. What occurs during interaction between
two entities in a specific episode is thus dependent
on the time dimension, i.e., any previous interaction
between the two. It is also dependent on the space
dimension in which this interaction takes place.
This means that the specific interaction between the
two is affected by their respective interaction with
others.
2.1 Organizing of activities
In the network approach, activities “brings life to
a network: goods are produced, delivered and
displayed; services are provided, accounts are
calculated and bills are paid” (Håkansson et al.,
223
2009: 93). Key activities can be identified in
procurement-purchasing, warehousing, inbound and
outbound logistics, exchange of information,
knowledge-sharing and trust-building.
By interconnecting its own activities to the
activities of its counterparts, a company exploits the
interdependencies that exist among the activities of
the different actors (Gadde et al., 2003). All
activities are connected to other internal and
external activities; the pattern of activities is
extended within and beyond the company and its
organizational boundaries, and they “stretch across
the landscape of the network” (Håkansson et al.,
2009: 99). The division of activities together with
their coordination between different organizational
units (Mintzberg 1979) characterizes the organization of activities (Corswant et al., 2001). As
outlined by Gadde and Hulthén (2009: 635)
“synchronization of activities within and between
manufacturing and logistic processes has become a
priority, resulting in considerable interdependencies
among activities”. These conditions call for just-intime deliveries (e.g. Vokurka and Lummas, 2000;
Kaneko and Nojiri, 2008), extensive use of
customization (Suomala et al., 2004; Su et al.,
2005) and increasing reliance on build-to-order
production arrangements (Gunasekaran and Ngai,
2005; Fredriksson and Gadde, 2005).
In a network approach, the organizing of
activities implies continuous configuration and
reconfiguration of activity patterns based on
activity interdependency in space and increasing
specialization over time. In a space dimension, the
firm’s activities depend on activities occurring in
the other firms of the network (Gadde et al., 2003).
In a time dimension, activity patterns are constantly
evolving through the adjustment of several activities in interaction Furthermore, specialization in
the activity pattern favors some types of change
while constraining other modifications (Håkansson
et al., 2009: 99).
Intermediation implies a shift from transferring
efficiency to coordinating effectiveness (Gadde and
Snehota, 2001; Dubois et al., 2004). To work
efficiently, the single activity must be adjusted,
through interaction, to other activities. The diversity
in the configuration of the activities enables
particular requirements to be met for specific
actors, in terms of a customized solution. “The call
for diversity in an activity pattern is always
accompanied by requirements for increasing the
similarity in the way activities are undertaken”
(Håkansson et al., 2009: 115). Similarity is reached
through standardization that supports economies of
scale.
Linking activities between two firms gives both
companies the opportunity to rationalize operations
that are important and that extend beyond
ownership boundaries (Gadde et al., 2003). Consequently, systematic building of interdependencies
The IMP Journal
Volume 4. Issue 3
becomes crucial (Dubois, 1998; Håkansson and
Ford, 2002). Generally, interdependencies occur
between production, distribution and service
activities. In turn, these activities form part of other
activities at a more aggregate level. A traditional
activity pattern configuration involves furnishing,
warehousing, inbound and outbound logistics,
assemblage and production.
The extensive use of outsourcing and the
disaggregation of activity patterns outline the fact
that the coordination of activities increasingly goes
beyond company boundaries. The coordination of
interdependent activities across the boundaries of
several connected companies cannot occur through
market mechanisms but rather requires extensive
interaction among firms (Gadde and Håkansson,
1994; Dubois, 1998). In this framework, the
coordination of activities can be described as the
configuration of activities in order to reach specific
objectives.
The intermediary operates to better fit the
activities of two actors in relationships, and it
provides specific services. For instance, a
traditional retailer/wholesaler provides services
related to a stock adjustment from provider to user
and services to reduce the geographic distance
between the provider and user. Increasingly,
innovative retailers and wholesalers are providing
other services, such as virtual community, ecommerce and customized solutions. From a
traditional perspective, the intermediary supports
the activity of several actors that make up supply
chains and distribution channels.
In a network approach, the intermediary supports
the management of interdependencies among
several activities: the performance of one activity is
dependent on another because both are dependent
on a third. Coordination of the activities in the
network shapes the firm’s productivity: it is
important for a company to relate its activities to
those of other firms to enhance its performance, and
these adaptations make the company dependent on
its counterparts (Gadde et al., 2003). Therefore, the
intermediary with its activity supports the
coordination of the pattern of activities that
involves several actors, providing tools to reach
efficiency and effectiveness.
All activities affect and are affected by the other
layers in interaction, namely, actors and resources
(Håkansson et al., 2009).
2.2 Organizing of resources
Resources are identified in terms of, for
example, raw materials, physical facilities,
components, operating systems and finance as well
as human knowledge and ability. A single
economic resource alone is passive and without
value. Value is determined by the interconnections
and combinations of heterogeneous resources
224
(Penrose, 1959) related to different actors.
Moreover, the interaction model suggests that
resources are characterized by heterogeneity (the
space dimension), and their development is based
on a specific path over time (the time dimension)
(Håkansson et al., 2009).
In the space dimension, a change in how one
resource is combined and related to some other
resources affects its value and the value of the
resource constellation in which it is embedded
(Håkansson et al., 2009: 87).
The resources that characterize an actor and the
resources that characterize its counterparts outline
the position of the first actor. Every position in a
network is based on some resources, but it is also
determined by the positions of the counterparts and
their resources (Johanson and Mattson, 1992).
At the same time, the position depends on the
efficiency of the resource deployment of the firm
and also on its effectiveness in relating to other
actors. Moreover, the network position is a relative
concept, as it means different things to different
parties related to the focal organization. At a more
in-depth level, the performance of an organization
in a relationship is perceived and evaluated by
another party on the basis of previous experience
and present expectations. It is thus enacted rather
than given by the amount and type of resources
directly controlled (Håkansson and Snehota, 2006).
Over time, resources move in relation to each
other or become more or less close. Interaction can
thus provide opportunities for learning where and
how resources can be adapted to each other. Every
resource is loaded with features from its previous
interactions with other resources: a resource is
embedded in an earlier path (Håkansson et al.,
2009: 86). A resource item is always a part of
several resource collections, following different
paths. Furthermore, a new situation can create a
new crossroads, where different kinds of path
dependencies meet, creating new possibilities for
utilizing resources. Any attempt to utilize a certain
feature of a resource in a new way will affect other
related resources (Håkansson and Waluszewski,
2002).
Interaction in the relationships shapes the flow of
resources (e.g., Baraldi and Strömsten 2006, 2009).
The 4R Model (Håkansson and Waluszewski,
2002) divides resources into products, facilities,
organizational units and relationships. Physical
resources are articulated in product (an artifact
exchanged between and within firms, components,
finished and semi-finished goods) and production
facilities (equipment, machinery, IT systems and
tools utilized to produce or transform products),
while social resources involve organizational units
(competences, capabilities, and skills) and business
relationships. In interaction, resources are
combined, developed, exchanged, or created
through the use of other resources, generating a
The IMP Journal
Volume 4. Issue 3
complex web of interacting resource interfaces
(Harrison and Håkansson, 2006). The value of the
resource depends on how it is combined with other
resources and resources of other organizations
(Penrose, 1959). Through the development of
relationships, resources are adapted to each other,
and features of one resource become embedded in
other resources (Håkansson and Waluszewski,
2002). Every relationship can thus be considered a
reflector of connected relationships and their
resources (Gadde et al., 2003).
Consequently, the organizing of resources is
based on resource selection and resource combination. Key issues in combining resources are the
combination of physical and social resources as
well as the combination of internal and external
resources (Baraldi and Strömsten, 2006).
The way in which resources are used and
developed depends on the interactions among the
organizations throughout the network (Ford et al.,
2003). Change or stability in one or more relationships can lead to different and/or more extensive
outcomes in the network as a whole (Ford et al.,
2002). In a network perspective, “intermediation is
not only about connecting two individual firms to
each other but also about connecting all of the
relationships of these two firms and with those of
the intermediary”. In this way, a network view
emphasizes that what happens between actors is a
process of interaction rather than simple exchange
(Gadde and Ford, 2008: 48).
A company’s success in activating interactions
that support the organization of resources will
define its strategic positioning and its role within
the network.
2.3 Organizing of actors
Actors play a key role in intermediation because
of the connections existing between them and their
counterparts in the configuration of activities and
the combining of resources. In the network, each
firm is characterized by a position that depends on
the image perceived by other actors and on the
interconnection of their activities (Håkansson and
Ford, 2002; Nyström et al., 2008). This position can
change with the development of new relationships
with new actors. On the one hand, a firm can
confirm its position and consolidate it by increasing
its effectiveness. On the other hand, an actor can
change its position by combining the existing
relationships in a new way or by activating new
relationships (Ford et al., 2003: 181).
An actor is always defined by his interaction
with other actors through a linking process that is
expressed as jointness (the space dimension). In
addition, the actor is involved in a process of coevolution (the time dimension) through which each
225
actor “seeks to relate its own problems, resources
and activities to those of others” (Håkansson et al.,
2009: 44).
In the space dimension, jointness not only
involves closeness between actors across
geographical space but also includes links between
the respective resources, such as knowledge,
technologies, capital investments and cultures of
the two actors. To complement each other, firms
relate to each other selectively. In a time dimension,
co-evolution is manifest in the small world of direct
relationships that makes actors into what they are
(Håkansson et al., 2009: 144): in a co-evolution
process, firms work closely with multiple alliance
partners (Wilson and Hynes, 2009). In interaction
the two parties develop reciprocal knowledge and
capabilities jointly and in mutual dependence
(Håkansson and Snehota, 2006). Relationship
development is based on joint action (Håkansson
and Ford, 2002).
From this perspective the organizing of actors
may be defined as the organization of relationships
between specific actors belonging to the network
context and the network horizon (Holmen and
Pedersen, 2003; Nyström et al., 2008). Bonds
among actors determine the organization’s identity
and influence the ability to relate to other
organizations. Actor bonds influence the ways
firms engage in activity coordination and resource
exchange (Håkansson and Snehota, 1989). The
position of a company is contingent on how the
company relates to the firms with which it actually
is involved in business exchanges (Gadde et al.,
2003).
In particular, we are interested in addressing two
main issues. First, we investigate how relationships
are developed by a focal firm with actors belonging
to a network context and network horizon,
considering that the network context is made up of
the actors that are considered most relevant for the
focal firm (Håkansson and Snehota, 1989; Huemer
et al., 2004). Second, we outline how actors can
move from network horizon to network context,
considering that there is no natural network
boundary (Gadde and Håkansson, 2001; Huemer et
al., 2004). Moving a firm from position to position
involves the evolution of network horizon,
considered to be the extent of an actor’s view of the
network (Holmen and Pedersen, 2003).
A firm may also interact with organizations that
are directly or indirectly linked to it within the
network horizon. The relationships in a network can
be considered a mechanism for knowledge
coordination and information-sharing. Therefore,
the main function of intermediation can be
identified as supporting knowledge coordination to
manage relationships and widening network context
and network horizon.
The IMP Journal
Volume 4. Issue 3
3. Research context and methodology
Our research is exploratory in nature: our
objective is not to demonstrate a particular theoretical framework but rather to improve the
understanding of a specific phenomenon (Dubois
and Gadde, 2002; Dubois and Arujo, 2004).
The empirical investigation focuses on the
evolution of an Italian electrical equipment consortium, MC Elettrici. This organization was
chosen because, over the years, it has impressively
rethought the traditional intermediation model to
better interpret the dynamics of a rapidly changing
environment. It is a particular consortium that was
established as an interface between manufacturers
and retailers. It not only provides centralized
purchasing on behalf of its almost 100 members
(representing approximately 170 points of sale),
mainly SMEs, but also offers a wide range of
services, from HR training to consulting (e.g., store
restyling, marketing activities, debt collection, and
law consulting).
The members of the consortium are retailers and
wholesalers that operate in the electrical equipment
industry and share certain phases of their activity
(procurement, warehousing, logistics, training and
marketing) through the consortium to be efficient
and effective. The consortium allows members to
focus their attention on their core activity and to
delegate to the consortium selected activities. The
consortium offers specific support and expertise to
small and medium sized retailers. Generally, the
retailer is rooted in a limited territory and is often a
specialist but has no national or international
visibility. For a retailer, it becomes very difficult to
keep abreast of the industry, changes in trade
policy, promotion and legislation. MC supports
innovation in a static industry.
In our opinion, this represents an emblematic
experience of how companies are able to transform
their business model by adopting a distribution
network perspective. We also think this case can
help us pinpoint the main implications of this
perspective in terms of the organizing of activities,
resources, and actors. In the case analysis, our
attention focuses on the evolutionary process
adopted, which contrasts with the traditional
approach to intermediation, creating benefits and
increasing the value for all of the consortium
members through the active construction of a sort
of “network within the network.”
We adopted a case study approach as a suitable
method for studying business-to-business relationships and networks. The case study approach
investigates the dynamics of the phenomenon and
provides a multidimensional view of the situation
(Easton, 1995; Eisenhardt, 1989; Halinen and
Törnroos, 2005). Case studies enable us to study
contemporary phenomena in their real-life settings
characterized by blurred borders between context
226
and phenomenon (Eisenhardt, 1989; Yin, 1994,
2003).
The case study method allowed us to collect data
from the key consortium informants and the key
members of its network context. Adopting a case
study approach, we investigated how networks
work in different settings and contexts (Ford et al.,
2002; Dubois and Gadde, 2002; Halinen and
Törnroos, 2005), considering operational links
(Dubois and Gibbert, 2010). As especially suitable
for case studies in business network research
(Dubois and Gadde, 2002), we adopted an
abduction process that enables data-driven theory
generation (Järvensivu and Törnroos, 2010):
choices related to the theoretical framework
influenced the empirical investigation. The study
became more theoretically and empirically focused.
Furthermore, a constant interaction between theory
and empirical observation characterizes the process
of systematic combining involved in case studies
(Dubois and Gadde, 2002) based on abductive
process (Piekkari et al., 2010).
Over a period of four years, we interviewed the
three co-founders (President, CEO, and Logistic
Manager) to identify the evolution of the MC
Elettrici business model. In particular, we analyzed
general company data, offering systems (width and
depth of product assortments and services provided)
and the nature of the relationships with its members
(levels of affiliation). Data were collected through
semi-structured interviews lasting from 60 to 120
minutes. Interviews were recorded and taped. The
internal validity of the method was ensured by the
fact that interviews were read and analyzed by all
three researchers independently. These primary data
were combined with secondary data gathered from
the firm’s website, reports, trade press and other
internal documents. A holistic description of the
network generated by multiple sources of evidence
(Järvensivu and Törnroos, 2010) has been required
to reach the aim of the research that is to analyze in
greater depth the impact generated by the evolution
of MC Elettrici on the relationships of its members
with suppliers, customers and other consortium
members. In this way we also gathered descriptive
data from the consortium members to consider their
perspective as well and define the different levels of
evolution of its members, according to the
development of a MC formula. For interviews, we
selected 40 members that have developed
significant relationships with the MC consortium,
considering the frequency of relationships, the
frequency and importance of their requests to MC
and their participation in MC initiatives and
meetings. We chose these organizations on the
basis of information provided by the MC
consortium. We have argued that these 40 members
belong to the network context because they are
considered key actors for MC. Interviews were
conducted with the entrepreneurs in charge of the
The IMP Journal
Volume 4. Issue 3
companies belonging to the consortium member.
Interviews included general company data
(employees, turnover, and date of foundation),
targets and offering systems, relationships with MC
Elettrici (the main objectives that determined the
consortium’s participation and evolution and the
use of services and their evolution, particularly
considering innovative services), and relationships
with other members (the main aim of these
relationships and the role of MC Elettrici in
developing
relationships).
Interviews
were
conducted by the three researchers and then
transcribed. Verbatim interviews were analyzed by
each of the three researchers in tabular displays
separately. After having coded several interviews,
the authors discussed their initial interpretations of
data and then reflected on previous empirical ideas
from existing research.
First, we will give a brief general description of
MC Elettrici. Then we will analyze the three stages
of its business model development. To pinpoint the
implications of MC Elettrici’s evolution on the
relationships and network of each consortium
member, we will describe the paradigmatic
experience of the “Alpha” member and analyze it as
the focal firm. To investigate the influence of
consortium membership on the evolution of the
members, we conducted interviews with Alpha’s
founder and top manager (Table 1).
Main actors
MC Elettrici
MC members
Alpha
Key
informants
Founder
Top manager
Wholesalers
Retailers
Installers
Founder
Top manager
Number of
interviews
5
5
15
20
5
3
2
Table 1: In-depth interviews
Alpha was founded in the 1970s with the aim of
commercializing products and equipment in the
electromechanical and electronics industry. In the
beginning, products sold were: electric motors, fans
industrial control systems, and lighting. Following
the development of the electronics industry, the
product range was expanded with the introduction
of industry automation.
Currently, the firm is based on a structure
composed of about 20 people organized in warehouse management, sales and purchases; a qualified
staff to assist customers in technical and
commercial applications is also available. In the
middle of 2000, Alpha entered MC Elettrici,
receiving fundamental support with regular
procurement and effective distribution of the best
brands in the electrical business. Alpha is one of the
40 key wholesalers/retailers belonging to the con-
227
sortium that experienced all of the stages of MC
Elettrici’s evolution and belongs to the network
context. Alpha is a key actor for the Consortium,
considering its high level of participation in the MC
project as well as the frequency of interaction.
In each stage of evolution of the MC Elettrici
model of intermediation, we will explain the
fundamental changes that occurred in terms of
features and coordination of the activities, as well
as the resources that were shared and combined
within the network.
Preliminary and final research results were
presented in a manager workshop and to a scientific
audience through seminars and conferences; this
helped us to improve the transparency and
acceptance of our evidence-based arguments
(Järvensivu and Törnroos, 2010). Throughout the
research process we tried to uphold transparency,
openness, and consistency in our research work
Through the interviews we collected information
to compare the perceived dimensions of MC top
managers and the perceived dimensions of
managers of MC members.
The interviews conducted with the 40 members
allowed us to isolate the main dynamics in the
network evolution and thus to identify the
emblematic case as representative of these
dynamics affecting the consortium members
belonging to the MC network context.
4. MC Elettrici Case
MC Elettrici began its activities in 1999 with 67
members, based on an idea of its founder, who
found inspiration from its previous business
experience in the electrical equipment supply chain.
Its origins date back to 1993 on a regional basis,
with a number of local consortia that were the
original members of MC Elettrici. In 1999, it
became a joint stock company and a fast start-up
was required. “We started in 1999. I chose the
software that we would use, recruited personnel and
built up the logistic network. Finally, we set up our
operation in a centralized warehouse near Milan,”
said the C.E.O. The original aim was to create scale
economies for associated companies (i.e., rates and
delivery conditions) by intermediating higher levels
of aggregate demand to manufacturers and
originating long-lasting relationships with the main
producers. Members are both wholesalers and
retailers; the wholesalers ensure that products are
distributed to single retailers scattered throughout
Italy, while the retailers are responsible for
distributing products to electricians and installers
through their stores. Since it began, the consortium
has been made up of small- and medium-sized
companies that had previously held a relatively
weak position at the bottom of the electrical
equipment supply chain: in fact, they were not com-
The IMP Journal
Volume 4. Issue 3
228
Figure 1: The main actors related to MC
petitive when up against larger companies due to
their low purchasing power and the consequent
imbalance in the prices offered to customers.
Moreover, they were struggling with scarce results
in the face of increasing pressure from manufacturers, who were facing reductions in their
margins.
MC Elettrici entered quite an articulated industry
in terms of actors and primary activities, as
summarized below:
- electrical equipment manufacturers: production
activity;
- warehouse keeper: storage, management, order
preparation, and delivery of goods activities;
- transport and logistics firms: transport and logistic
activity;
- wholesaler/distributor: “handling system” activity;
- retailers: sales to installers and electricians.
Although the primary service offered to
members is the chance to “buy” from a centralized
warehouse with competitive conditions and
immediate availability, the MC Elettrici formula is
characterized by an integrated system of “soft”
services that helps the customization of standard
products in personalized solutions and supports the
development of collaboration among different
actors belonging to the electrical equipment
industry. By building a complex system of
relationships, the business model adopted by MC
Elettrici not only provides an opportunity to
rationalize the entire distribution network, thus
benefiting all actors involved, but also guarantees
the survival and independence of small retailers that
are otherwise destined to disappear.
The network of competences and activities
mobilized by MC facilitates the improvement of the
value created, enriching the supply of each actor to
shape its own value proposition, adding and
revising different services and activities available
before interacting with its customers (installers and
electricians). We can summarize the main valueadded activities that MC Elettrici developed over
time to support its members in the following way:
demand management; order fulfillment; procurement; product commercialization and service development; training (technical and managerial); and
sales and returns control. From the beginning,
traditional activities that characterized MC were
identified in warehouse management and logistics.
Over time, these traditional activities were
influenced by a greater orientation to customer
needs and development of technology that allowed
for transformation of warehouse management and
logistic activities. Furthermore, other activities
related to knowledge-sharing and community
development were involved in the MC offering
system.
Summing up, the MC Elettrici business model is
to be considered from a distribution network
perspective (Figure 1), in which different kinds of
actors are related with MC Elettrici and to each
other. Initially, the main actors related to MC were
wholesalers, retailers and electrical equipment
manufactures. Over time, the activity of MC began
The IMP Journal
Volume 4. Issue 3
to involve relationships also with installers and
specialized actors, such as warehouse keepers,
transport and logistics firms and other specialized
actors. After this brief description of MC Elettrici,
we shall now outline the evolution of MC to
interpret intermediation, as far as actors, resources
and activities are concerned.
It is possible to identify three phases of the
evolution of MC: phase 1 (1999-2003), phase 2
(2004-2005), and phase 3 (2006 to present).
4.1 MC phase 1: Facilitating physical connections
among the actors
In a context characterized by globalization and a
high level of competitiveness, the original aim of
MC Elettrici was to create scale economies for its
members by developing the traditional intermediation function aimed at interconnecting the
different actors involved in the electrical equipment
distribution.
In this phase, the main activities of MC were
articulated in central procurement, central warehouse and incoming logistics. In this sense, the
main resources offered to the consortium members
related to procurement management, warehouse and
logistics management, and documentation flow
management. From this perspective, the main aim
of MC Elettrici was to develop an advanced form of
the traditional functions of intermediation: MC
Elettrici provided services to support with a high
degree of efficiency the stock management in the
central warehouse and logistic processes.
The key role of the central warehouse in this
phase is also testified by the brand “MC” adopted
by consortium, which in Italian means central
warehouse. This was done to be able to exploit
economies of scale. To optimize the fixed costs
associated with freight handling, a key role was
also recognized for logistics. Through 20,000
square meters of space and 8,000 cubic meters of
product, MC provided a rigorous and efficient
logistics system for its members. Since that time,
MC has been able to provide 2-3 deliveries to its
members each week throughout Italy. Furthermore,
members have been supported in accessing more
information about market conditions and better
communicating with different actors of the supply
chain.
In this phase, all of the efforts were devoted to
developing the consortium through the affiliation of
new members and the establishment of long-lasting
contracts with suppliers and business partners. At
that time, the consortium counted 67 members and
approximately 30 different suppliers. The 67
members and 30 suppliers derived from previous
local consortia with which the MC founder had
previously developed relationships. These relationships were characterized by strong ties, as the
actors involved were well-known on the basis of
229
previous activities and projects. The suppliers
involved both international firms and medium-sized
Italian firms.
4.2 MC phase 2: Developing new knowledge
services and a common marketplace
In the second phase of development, MC
Elettrici realized that the traditional approach to
intermediation was insufficient to support the
consortium members and the sustainability of the
formula. The new logic underscoring the project
was the willingness to provide new tools to develop
the expertise of wholesalers and retailers. Each
member of the consortium was characterized by
specificities related to geographical localization.
Indeed, in the Italian electrical equipment industry,
the key success factor was to make retailers
“specialists,” able to customize their offers through
the multiple services co-created with wholesalers
while also remaining strictly connected to the
territory. In this sense, in this second phase of
evolution, the consortium started supporting its
members as they developed the customization of
the offering system for their customers. Moreover,
the MC Elettrici mission shifted to training a
professional force determined to “get in touch with
the customers’ needs and to solve their requests” by
adding more personal and informed services than in
large impersonal stores, whose staffs often lacked
in-depth knowledge of products and their uses. This
was primarily accomplished through intensive
training services for associated companies and their
personnel. The professional growth also went with
investments in store management training and in
more sophisticated systems to control sales (e.g.,
ERP software and accounting).
MC Elettrici provided a new function in
intermediation that mobilized new resources
through innovative activities: it started providing
new customized solutions for its members,
collected and shared new kinds of information with
them, and, thanks also to the launch of a virtual
marketplace, MC Elettrici intended to build a
formal network of competences. The affiliated
retailers and wholesalers could fill a virtual
shopping trolley with products they need by
accessing MC Elettrici’s website, which was
directly connected with a central warehouse; in this
way they could obtain favorable conditions and
prices as well as numerous services that could
improve their work and effectiveness in meeting
customers’ needs.
The virtual marketplace was created to involve
all electrical operators, from manufacturers to
retailers and installers. Thus, the marketplace was
not only devoted to trading but also offered plenty
of related information/training advice. The intention
of MC was to become a key referent for its
members for a variety of information services and
The IMP Journal
Phases
Members
Suppliers
Volume 4. Issue 3
1
67
30
2
90
40
230
3
100
70
Specification
In Context
Within the
Horizon
70
wholesalers
40
wholesalers
30
wholesalers
30
Retailers
20
retailers
10
retailers
35
industrial
suppliers
20
industrial
suppliers
15
industrial
suppliers
25
civil
suppliers
15
civil
suppliers
10
civil
suppliers
10
automation
suppliers
5
automation
suppliers
5
automation
suppliers
Table 2: The development of the MC network
organizational processes. Through the website,
members may look up statistics and access the
database. The offering of MC also involved IT and
marketing, indoor and outdoor activities, and teambuilding. MC also developed a training program,
the “toolbox.” The project started considering the
technical and cultural skills necessary for every
professional to do their work, creating a training
program tailored to each of them. In this sense, MC
Elettrici was gradually becoming a “knowledge
connector,” able to improve the final value that the
downstream supply chain could perceive and pass
on to its own customers. In this sense, MC
Elettrici’s logic was shifting from traditional
consortium services, focused on central warehousing and logistics, to innovative services, such
as the various types of management training,
category management and retail marketing tools,
and a complete store redesign, offered to members.
MC members increased to approximately 90 (net
amount between exits and new entrances), and 40
suppliers were considered long-term partners (see
Table 2). The missing alignment between members’
aims and MC aims generated the exit of some
members. Notwithstanding, other wholesalers and
retailers heard about the activity and services
provided by MC and, in accordance with its
philosophy, became members of the consortium.
4.3 MC phase 3: Creating an effective MC
community
According to the CEO, “The only way that
Europe has to resist Chinese and Indian competition
is to embrace innovative ideas and to interpret the
market in a way that is difficult to copy.” The result
of a continuous process of formula rejuvenation is
represented by the launch (2007) of a common
brand to use in all stores (the Identity Project). In
fact, in the most recent phase of MC Elettrici’s
evolution, the consortium has focused its efforts on
activities that allow members and customers to
build up a community and develop relationships
among “peers.” The objective is to reinforce brand
awareness in the final market as well as to support
the wholesalers and retailers belonging to the MC
Elettrici consortium in perceiving themselves as
partners in a group characterized by high quality
and customer orientation.
Through the Identity Project, members are
allowed to redesign the store according to common
criteria relating to signs, benches, shelves, and to
better manage their remote warehouses and
improve their staff training (both on a technical and
managerial basis). The introduction of the private
label “Gli Elettrici” for niche products (i.e., antiintrusion, timer, and fixing systems) is another step
The IMP Journal
Volume 4. Issue 3
231
Figure 2: MC Elettrici: its network context and network horizon
of this evolution. This brand is highly qualified but
non-competing with supplier partners. The main
objectives of this brand are to strengthen the
identity project and to complete the product range
of its members without overlapping with the
products provided by the main supply partners.
In reinforcing its consortium, MC Elettrici seems
to emphasize the individual attributes of its
members while simultaneously developing their
common vision, based on a willingness to enhance
the future of the retailers and to provide services
with high added value. Moreover, the Identity
Project aims at facilitating relationships in a peerto-peer logic to support the development of shared
innovative projects. A renewed partnership between
manufacturers and consortium members occurs,
with lots of new co-branding, in-store activities and
new value-added services available on the MC
online marketplace (i.e., technical information, a
TV lab to explain how to install and use products,
and corners for suggestions and improvements).
Today, more than 100 members—70 wholesalers
and 30 retailers—form the MC Consortium.
Approximately 70 suppliers and partners are linked
to the MC distribution network. Each actor is
characterized by specific resources and competences and is identified as a node in the network,
generated by MC.
Two key aspects are related to this situation. The
first is the evolution over time of members and
suppliers. The second is the characteristics of the
MC network at the end of the phase 3. The
evolution over time of the number of members and
suppliers is shown in Table 2.
In this table, we also show for the phase 3 how
wholesalers, retailers and suppliers belong to either
the network context or the network horizon.
Finally, the table also makes a distinction between
different types of suppliers. Suppliers belonging to
the network context include firms such as Gewiss,
ABB, Italiana Corrugati, Prisma, Teaflex (industrial
suppliers), B Ticino, Philips, and Beghelli (civil)
The IMP Journal
Volume 4. Issue 3
and Bpt, Elvex, and Logisty (automation suppliers). MC seems to develop lots of relationships
outside its network context.
Considering the evolution over time, we observe
the huge increase of members in the phase 2, which
seems to have slowed in the phase 3, and the huge
increase of suppliers in the phase 3.
Focusing on the MC network at the end of the
phase 3 (Figure 2), we observe all of the actors of
the electrical equipment distribution network that
are strictly connected by long-term relationships to
the consortium. To clarify the figure, we have
indicated the main nodes for each typology of
actors in both the network context and network
horizon.
If we consider MC from a network perspective,
we may understand in more depth the manifold
relationships MC has been able to create through its
model of intermediation. Relationships now occur
not only among the consortium members but also
between them and the other actors belonging to the
electrical equipment distribution industry, allowing
them to share resources and develop traditional and
innovative distribution activities. This interpretation
helps us to better focus on the main implications of
MC Elettrici’s evolution, in terms of new resources
and accessible actors, in the subsequent description
of the evolution of one of its members.
We observe the increasing number of members
(wholesalers and retailers) and of suppliers. In
particular, we outline the belonging of main actors
to the network context and network horizon of MC.
We also observe the differences existing between
relationships developed within the context or within
the horizon. In the context, we can find strong ties
that link MC to its members, while on the network
horizon, MC is linked through weak ties to other
actors. Strong ties require more investment for MC
and its counterparts than weak ties. In addition to
this, strong ties are referred to long-term relationships, which also implies a high frequency of
contact. These relationships are based on warehousing activity, training courses and community
development actions. Additionally, we highlight the
increasing number of secondary relationships
between the members, considering wholesalers,
retailers and installers. This is due to the capacity of
MC to provide services that have supported the
development of the MC community, based on the
sharing of aims, visions and the search for value
added solutions.
5. The evolution of MC Elettrici’s members
After considering the peculiarities of MC
Elettrici, we explored how the evolution of the MC
Elettrici partnering formula affected networking at
the consortium member level. We considered the
232
“Alpha member”, an emblematic case, within the
MC Elettrici network context. Alpha is a
wholesaler, with shops open to both small retailers
and installers. It sells electromechanical and
electronic products and equipment. Initially, the
company sold motors, industrial fans, control
systems, and lighting equipments. Over the years,
the company has expanded its range of products to
include automation (e.g., photocells, sensors, PLCs,
and inverters). Alpha is a typical consortium
member, with an average turnover of €3 million.
Like other actors belonging to the MC Elettrici
network context, Alpha experienced four main
phases of evolution during the development of the
MC Elettrici formula, as described in the following
paragraphs.
5.1 Alpha - phase 0: Before joining MC Elettrici
Before joining MC Elettrici, Alpha had
developed relationships directly, mainly with its
suppliers and retailers. Just like other traditional
wholesalers operating in the Italian electrical
equipment supply field, Alpha’s industrial
counterparts were often medium-sized, and Alpha’s
contractual power was usually low. Its customers
belonged to a small geographical area.
In this phase, Alpha managed privileged direct
relationships with nine suppliers – of which four
belonged to the network context and five to the
network horizon. The corresponding figures for
wholesalers were five in the context and seven
within the horizon, and for retailers, eight in the
context and eight within the horizon. All in all, this
means that Alpha did business with 12 wholesalers
and 16 retailers.
The small size together with the localization in a
specific area characterizing Alpha and its features
influenced its ability to exert contractual power
with suppliers and represented a constraint when
granting a flexible offer to customers and when
facing the dynamics affecting distribution.
5.2 Alpha - phase 1: Joining MC Elettrici
In this new phase, Alpha decided to access MC
Elettrici, being interested in the services provided
by the consortium (central warehouse, facilities in
prices and volumes, logistics). The Alpha
warehouse was backed up by the MC warehouse
(30,000 sq.m. area), which ensured a more regular
supply and more competitive distribution of the
best brands of electrical equipment. In the MC
Elettrici warehouse, new technologies enabled the
checking of product availability in real time. Access
to a unique warehouse and centralized service
allowed the member to draw on a complete range of
solutions, which otherwise would not have been
conceivable.
The IMP Journal
Volume 4. Issue 3
Phase 0
D/I
Phase 1
Direct
NC/NH
NC
233
Direct
Phase 2
Indirect
Direct
Indirect
NH
NC
NH
NC
NH
NC
NH
NC
NH
Suppliers
4
5
4
5
9
19
9
15
16
24
Wholesalers
5
7
5
7
6
8
8
12
15
25
Retailers
8
8
8
8
8
9
9
14
9
11
10
20
Installers
NC: network context
NH: network horizon
Indirect relationship = relationship supported by MC
Table 3: Alpha’s direct and indirect relationships during phases 0-2
By joining MC Elettrici, Alpha was able to access
contacts with several new suppliers namely,
manufacturers belonging to the main industrial,
civil, and automation areas, including Siemens,
Lovato, AEG, Socomec, Vemer Siber, Gruppo
Italtrafo, CAM Logic, Comechi, and Camour
(industrial suppliers) and Orieme, Vortice, and
Fracaro (civil suppliers) as well as Moeller, Zeca,
ITC, and CEAM (automation suppliers). Alpha was
able to develop two kinds of relationships with
these suppliers. MC Elettrici mediated the negotiation with some members that belonged to the
qualified group of partners. Thanks to its position,
MC Elettrici was able to achieve better contractual
conditions. Meanwhile, Alpha directly managed
exchanges (for small or residual quantities) with
several suppliers, thanks to its MC Elettrici
membership.
Considering the network context of Alpha, we
notice that in this phase it was enlarged to new
wholesalers and retailers. At the same time, Alpha
had reinforced its relationships with existing
suppliers (already belonging to its network
context), but it had also started developing further
relationships with new suppliers that entered into its
network horizon and/or its network context. The
indirect relationships were developed by Alpha
with actors that were in relation to MC. Thanks to
the initial relationship with MC, in fact, Alpha
could develop new relationships with several
actors. Evidently, the stronger relationships
characterize the network context, while the weaker
relationships characterize the network horizon.
5.3 Alpha - phase 2: Development of knowledge
services in MC Elettrici
During the second phase of evolution, MC
Elettrici’s attention shifted to relationships between
its members and their customers. MC Elettrici
supports its various members in achieving
efficiency and effectiveness objectives in relationships with their customers.
In fact, MC Elettrici developed an integrated
system of “soft” services to help the customization
of standard products into personalized solutions and
supports the development of collaborations among
different actors belonging to the electrical equipment supply chain. MC Elettrici provides not only
warehouse and logistic services but also services
aimed at supporting retail management, as well as
administrative, financial, and information services.
Training services play a key role in the offering
system. As such, MC Elettrici supports the
knowledge management of its members, increasing
the development of each firm from organizational
learning to distribution network learning.
Every actor is involved in a process of
continuous upgrading of knowledge, thanks to the
sharing of processes and products. Now, the main
new resources provided by MC Elettrici are training
competences and support systems, documentation
flow management, and managerial and technical
knowledge. As such, MC Elettrici provides
managerial capabilities to support members’
outbound logistics and warehouse management.
Thanks to the new services, Alpha can support
customers with technical and commercial advice.
Staff training now plays a major role in Alpha’s
strategy. Technical seminars are held by MC
Elettrici in partnership with its main suppliers to
help staff provide the customer with accurate and
up-to-date pre-sales and after-sales services. In
addition, managerial capabilities are continually
developed, thanks to central training activities, instore management support and new systems to
control sales (e.g., ERP software). These services
enable Alpha to become an efficient reliable
The IMP Journal
Volume 4. Issue 3
Phase 0
D/I
Phase 1
Direct
Direct
234
Phase 2
Indirect
Direct
Phase 3
Indirect
Direct
Indirect
NC/NH
NC
NH
NC
NH
NC
NH
NC
NH
NC
NH
NC
NH
Suppliers
4
5
4
5
9
19
9
15
16
24
21
8
30
10
Wholesalers
5
7
5
7
6
8
8
12
15
25
13
9
30
10
Retailers
8
8
8
8
8
9
9
14
9
11
15
10
15
5
10
20
30
10
Installers
NC
NH
Table 4: The evolution of Alpha’s relationships
and expert partner in electrical equipment supply.
Moreover, there is a real reinforcement of the
relationships not only regarding the suppliers, MC
Elettrici and the members of the consortium, but
also of the development of relationships with
installers (Table 3). In the phase 2, Alpha increased
the number of direct relationships with suppliers,
wholesalers and retailers. The total number of
wholesalers in phase 1 was 26, while in phase 2 it
expanded to 60. Additionally, suppliers increased
considerably, particularly in terms of direct relationships. These direct relationships were generated
by the new capabilities of Alpha due to resourcesharing and combining of competences realized in
different relationships.
Thanks to the competences and capabilities
developed, Alpha was able also to activate direct
relationships with 30 new installers.
5.4 Alpha - phase 3: the development of the MC
Elettrici community
In the last phase of MC Elettrici’s evolution, the
consortium set up an actual community and
developed relationships among its “peers.” From
this perspective, the various actors (suppliers, MC
members, and Alpha’s customers) that belonged to
Alpha’s network context reinforce their relationships with the firm. The activities performed by the
consortium (central warehouse, logistics, training
processes, and the community) provide new means
for Alpha to manage its interconnections with other
actors, both with suppliers as well as in peer-to-peer
relationships (among members).
In this phase, Alpha further enlarges its network
context through the connection with the other
members of the consortium. Moreover, MC
Elettrici’s community services facilitate Alpha’s
efforts to exchange both products and information.
In this phase, Alpha further increases the frequency
and quality of both direct and mediated MC
relationships with retailers, wholesalers and
suppliers. Through peer to peer relationships
supported by MC, Alpha can share with other
members information and knowledge related to the
market and technology as well as information
related to activities such as procurement,
warehousing, marketing, communication and
training (Table 4).
Focusing on the shift between the second and
third phases, we observe substantial increasing
changes in the proportion of direct relationships in
relation to all types of business partners.
Furthermore, both for direct and indirect relationships, there is a significant shift towards increasing
proportions of suppliers in the context and thus a
reduction in the horizon.
In addition to this, we observe that, initially,
Alpha increased the number of relationships, in
particular, with suppliers. From the first to second
phase, Alpha increased the relationships with
wholesalers, retailers and installers. We can thus
outline the current network configuration (Figure
3). We have indicated the first and last nodes for
each actor category in the network context, from
phase 0 to phase 3. The network context is made up
of the key actors that entered in different phases
and still are part of it.
The IMP Journal
Volume 4. Issue 3
235
Figure 3: Alpha’s network context
Today, Alpha can offer more than 20,000 items
from major national and international manufacturers that range from industrial to civil and
automation areas. Thanks to its participation in MC
Elettrici, Alpha has indirectly developed relationships with big suppliers that would otherwise have
been far from its supplier portfolio and is able to
relate more easily with wholesalers, retailers and
installers. Furthermore, cooperation with the
consortium has enabled Alpha to enter new markets
and develop its business according to the main aim
of MC. The figure outlines the increasing number
of strong relationships between Alpha and the other
actors involved (suppliers, wholesalers, retailers
and installers). The development of these relationships is supported by the increasing ability of Alpha
to optimize the management of relationships on the
basis of previous experience.
6. Analysis
The case analyzed outlines some of the main
features characterizing the evolution of intermediation and its functions, as described in the
literature and previous empirical research on
network-like contexts. As shown in our analysis,
the need to customize and differentiate the offering
system as well as to introduce new technologies has
generated specific consequences for intermediation.
This seems to be particularly true in contexts where
intermediation plays a key role, such as in the
electrical equipment industry.
Moreover, our longitudinal study outlines the
evolution that has affected the intermediation
function and the three network layers and that has
activated the organizing of activities, resources and
actors.
Starting with the activities, the analysis
highlights, in an evolutionary perspective, the
increasing importance of support activities related
to human resources, technologies and development
of community.
In particular, in the first phase, activities are
identified in procurement management, warehouse
management and incoming logistics management.
Procurement management, in this phase, is related
to identifying the goods to procure, complete
purchase orders, and agree on delivery timeframes
and methods. In addition, warehouse management
The IMP Journal
Volume 4. Issue 3
involves physical quantities as well as the pricing of
the goods as they flow.
In the second phase, these main activities evolve,
becoming more complex and interconnected. In this
phase, procurement management is focused on
better identifying supplier contract milestones.
Thanks to ERP software and other software, warehouse management includes forecasts for safety
stock estimation, programming and monitoring
missions. Cross-docking, trans-shipments, and
specific processes for handling increase the speed
of stock rotation. Furthermore, management of the
virtual marketplace and training of the employees
began to be of key importance.
In the last phase, we observe a transformation of
previous activities and the emerging of a new one:
the development and managing of the community.
In this phase, procurement management is more
focused on reviewing supplier performance with
respect to the contract, identifying and resolving
supplier performance issues, and communicating
the status to management. Community development
is related to social activities, outdoor training and
professional activities that increase the sharing of
aims and the search for common solution.
Taking a more in-depth view of resources, the
findings outline the emerging of social resources
(units and business relationships) rather then
physical resources (products and facilities).
In the first phase, the main resources are the
structure of the central warehouse together with
competences in warehouse and incoming logistics,
competences in procurement management and store
cards. In this phase, physical resources, such as the
central warehouse, play a key role.
In the second phase, the transformation of
previous resources is associated with administration, information services, training competences,
intranet system and the virtual marketplace. As we
observe, physical resources are integrated with
social resources that have begun to take a key role.
The attention is focused on stock control and stock
taking competences related to the central warehouse
as well as the cross-competences between MC and
its members. MC has continuously developed an
accurate audit of existing inventory and stock
valuation, also providing stock discrepancy
information. Other key important resources are
represented by the virtual platform to support
communication and transactions related to the
warehouse and logistics. Furthermore, training
resources involve team-building games and
activities for employee motivation together with
workshops, meetings and videoconferences.
In the last phase, the additional transformation of
previous resources is integrated by the community
project (Identity project). In this phase, social
resources take a more central role than the physical
236
ones. MC supports the sharing of knowledge,
information and values. The attention is focused on
common branding and the capabilities to sustain the
common brand. Furthermore, MC organizes several
workshops and unconventional conferences to
develop a common vision.
With regards to actor layers, we observe the
increasing number over time of actors involved and
their different roles in supply chain. This transformation has led to an increased interconnection
and to co-evolution processes. These have had a
strong impact on both the quality and effectiveness
of the relationships among the actors in the MC
network and Alpha network. Focusing on actors,
the analysis emphasizes the increase in the number
of consortium members (and suppliers) and the
multiplication of the key relationships considering
the outline of the network context.
In the first phase, the main actors amount to 67
members and 30 suppliers. As related to Table 2,
the number of members increased from 67 to 100,
and the number of suppliers from 30 to 70.
By considering together the three layers (Figure
4), we can derive that one of the layers has been in
focus in each of the phases.
In the first phase of evolution, the attention of
MC Elettrici is mainly focused on organizing
activities,
especially
on
managing
their
interdependence (space dimension) and specialization (time dimension). In the first phase of
evolution, referred to as the traditional function of
interconnection, MC Elettrici develops services to
ensure its customers’ efficiency, focalizing its
activities on central warehouse management and
logistic support. New technologies and services
mainly aim at managing physical flows and
answering other basic needs (with tangible
resources) expressed by the counterparts (in MC’s
case, first of all, the consortium members). In this
phase, the intermediation of MC supports the
connection of activities of suppliers, wholesalers,
retailers and installers. In addition, MC promotes a
better combining of resources between different
types of suppliers, wholesalers, retailers and
installers.
This interconnection is very important for the
different actors and also for the end-user of their
products and services. Over time, its role evolves
towards the development of new activities to
support interaction among the network members
and within their networks. Thus, MC may be
considered as an activity developer that operates
upstream (suppliers) and downstream (wholesalers,
retailers and installers).
The second phase of evolution is mainly focused
on organizing resources, that is, considering their
heterogeneity (space dimension) and their
evolutionary path (time dimension).
The IMP Journal
Volume 4. Issue 3
237
Figure 4: Central features of the three network layers in the evolution of MC
The second phase of evolution represents a shift
in the intermediation’s attention to intangible
resources, as the knowledge services developed by
MC Elettrici underlines, though relationships within
the network, are still mainly dyadic. An evolution
in the distribution of electrical equipment requires
MC to provide new services and activities that are
not part of the core business of traditional
wholesalers but that create an increase in MC
Elettrici’s offering value. The impact of new
technologies is considerable: members connect with
the company website and fill a virtual shopping cart
with the products they need. They can choose the
biggest manufacturers’ products and, thanks to their
combined purchasing power, can obtain favorable
rates and delivery conditions that they can then
transfer to their b2b customers. The main function
of intermediation is to support interaction between
the different actors so as to give customization.
Furthermore, MC Elettrici supports its
members in achieving efficiency and effectiveness
in relationships with their customers. Over and
above the traditional warehouse and logistic
services, MC Elettrici develops activities based on
intangible resources, such as training services,
outbound logistic skills and warehouse management skills. It supports knowledge management and
the development of each firm, from organizational
learning towards distribution network learning.
Moreover, a consortium member such as Alpha can
benefit from relationships developed with other MC
members, which become partners with whom to
develop new projects and to better customize
solutions. The key functions recognized in
intermediation can be summed up as the
coordination of activities and the establishment of a
suitable context for the autonomous combining of
resources made available by the actors. In this
sense, MC Elettrici is an emblematic example of a
competence integrator.
In the last phase of evolution, the key intermediation function is now identified in facilitating
relationships among different actors and in
supporting their enhancement within community.
The attention of MC Elettrici is mainly focused on
jointness (space dimension). In addition, actors
evolve in a process of co-evolution (time). The last
phase of evolution embodies, in our opinion, the
actual exploitation of the distribution network. The
multiple interconnections of different activities
push firms to accept a wider perspective of
analysis, not limited to single and isolated relationships but to the distribution network. Intermediation, as represented by MC Elettrici, progressively
The IMP Journal
Volume 4. Issue 3
Before
entering
Suppliers
Wholesalers
Retailers
Installers
total
238
Phase 1
9
12
16
37
26
33
37
96
Phase 2
64
60
43
30
197
Phase 3
69
62
45
40
216
Table 5: The expansion of counterparts in the phases of Alpha’s development
loses its function as provider of resources to
become a relationships facilitator among other
actors: it supports relationships between members
and suppliers; it also supports relationships between
its members and their customers through the tools
and services provided to enhance customer
satisfaction (e.g., training courses and in-store
communication). Furthermore, the differentiation of
MC Elettrici stems from the development of
services and activities that emphasize the centrality
of its partners’ community. The MC Elettrici
Identity Project reinforces the sharing of a common
vision and common values among its members.
Periodical meetings, cross-training on the job,
website communication, travel incentives, and
community activities facilitate and reinforce
relationships among members (wholesalers and
retailers). In this process, MC Elettrici sustains the
organizing of knowledge by supporting every actor
in being involved in the process of continuous
knowledge upgrading, thanks to the sharing of
learning processes.
To better understand distribution dynamics, we
also analyzed the impact that this evolution
generates on an emblematic member belonging to
the network context: Alpha. This helped us to draw
more considerations about the possible reconfigureations of the three layers in a distribution network
perspective (Table 5).
The table shows the tremendous increase of
actors that provide considerable benefits. At the
same time, however, it is not unproblematic. Too
many suppliers cause a lot of problems. They are
costly to handle and demand a lot of resources.
Thus, somewhere there is an obvious limit where
the benefits of expanding numbers will be
outweighed by the disadvantages of many
suppliers, especially for the direct ones. However,
this has to do with how these potential partners are
used. Findings emphasize how the set of suppliers
outline a suppliers portfolio that requires a
relationship management strategy (Zolkiewski and
Turnbull, 2000; Roseira and Brito, 2009).
The potentialities of each partner influences the
choices of Alpha to develop a specific relationship
characterized by time and cost investment,
frequency of interaction, information and
knowledge sharing, resources provided and
combined. By entering the MC network, Alpha had
the chance to establish relationships with new
actors. The analysis shows that the transformation
in MC networking generated a transformation for
Alpha networking as well. Alpha, as a result of
entering MC, was able to identify new suppliers,
new customers and also reinforce relationships with
its existing partners. In fact, Alpha had the chance
to expand into new final markets, for instance,
integrating civil solutions with industrial solutions
as well as automation and lighting, which were not
sustainable before. Both MC Elettrici and Alpha
organized their network by identifying the old and
new actors that operated in the network horizon and
network context. These firms organized their
activities, coordinating internal and external ones,
and combined the resources that characterized the
different actors.
Moreover, MC allows its members to develop
better relationships with their customers, by
enriching the offer. For instance, MC supports its
members in creating training activities for their
customers. In this sense, the evolution of MC
generates new relationships with different actors.
The Alpha member has, as such, been able to
specialize its competences and gain heterogeneous
resources (mainly intangible) within the network.
From a distribution network perspective, the
interdependence of activities requires coordination
that is efficiently and effectively achieved only by
implementing the new function of intermediation,
which is focused on customization, valorization of
intangible assets and uniqueness of the offering
system.
In this sense, the organizing of the network and
its three layers becomes a crucial issue for
intermediation and its actors. It can be identified as
the subsequent phase of strategizing, as influenced
by its choices and properties. A firm’s strategy,
which influences and is influenced by others in the
network (Baraldi et al., 2007), requires a phase of
organizing to facilitate its implementation.
The IMP Journal
Volume 4. Issue 3
7. Conclusions
Over time, intermediation has shifted in its
function from the pure physical interconnection of
supply and demand to the interaction and
involvement of customers in relationships. In
addition, thanks to the development of new
technology and the increasing complexity generated
by globalization, intermediation has focused its
attention to cooperation, by developing relationships with actors involved in distribution networks.
From an evolutionary intermediation perspective, attention has been shifting from efficiency
objectives, based on the reorganization of activities,
to effectiveness objectives, thanks to the organizing
of three layers.
At the beginning of the intermediation evolution
process, the key organizing actor (MC) was
primarily connected to tangible resources and
activities devoted to improving efficiency
objectives. Intermediation was forced to move to
offering systems based not only on efficiency
objectives but also increasingly on the valorization
of intangible assets connected with effectiveness
improvement. According to our study, this process
of evolution directly affects the role of the actors
involved, activities coordinated and resources
mobilized. Moreover, the evidences related to the
organizing of the three layers are strictly connected,
from the intermediation evolutionary perspective, to
time and space dimensions.
As outlined in our case study, the evolution of
intermediation has generated implications for the
organizing of each of the three layers; in this
regard, the intermediary assumes a new role as
activity developer, competence integrator and
relationship facilitator.
In the case of MC Elettrici, each layer plays
different parts in each phase of evolution. To
increase the offering value, actors in the first phase
of intermediation evolution mainly focus their
attention on organizing activities. In the second
phase, the shift to a new intermediation function is
mainly related to the organizing of resources. In the
third phase, intermediation is mainly related to the
organizing of actors. This is a description of what
MC did in relation to those that were members
when everything was initiated. However, the whole
analysis might easily be twisted around, depending
on which perspective is applied. It is possible to
argue that the order was completely the opposite:
MC needed to mobilize companies as members i.e.,
to start with actors; they needed the resources
required for supporting these members; when they
had the members and the resources they could start
making their activities more efficient. This is a
vicious circle. Once they had been able to establish
relationships with consortium members through
239
enhanced efficiency in the first phase, they could do
more to the members: this leads to the second
phase, which in turn leads to the third phase. This
means that what a firm can do in one of the layers is
always dependent on what there are in other two.
Thus, the evolution of the three must match.
We may outline a double perspective in
considering the evolution of intermediation: in the
first one (pro-active approach), the evolutionary
process of MC influences the evolution of several
actors that decide to become its members. From the
second perspective (reactive approach), the
evolution of MC is subordinated to the actions
developed by the members that belong to its
network. In both approaches, MC can develop
actions related to intermediation on the basis of
resources provided and shared, activities realized
and coordinated, and actors’ positions in the
network.
The key elements in organizing activities
concern
the
division,
coordination
and
synchronization of activities. The organizing of
resources is focused on resource selection and
combination. Furthermore, actors are characterized
by increasing specialization and interdependence.
Consequently, the organizing of actors concerns the
supporting of relationships with specific actors by
considering their role and the position in the
network and by widening the network context and
network horizon. Specifically, the boundaries and
configuration of network horizon and network
context vary in time and space. The evolution of the
position of focal firm and other firms is connected
to resources provided and resources accessed,
where relationships are the key drivers of this
process.
In this framework, the key actor (MC) in the
organizing of activities, resources and actors plays a
crucial role in outlining what we define as “a
network within the network.” This leads to a new
function for intermediation that we define as
“organizing a network within a network”: each
member of the network can develop its own ability
for organizing its network through the support
provided by the key actor. We can identify two
particular types of actors: a network organizer and a
member of this network within the network.
The organizing of the three layers arranged by
the main key actor (MC) enables actors belonging
to its distribution network to pursue a higher degree
of flexibility in organizing their own sub-network
of actors, activities and resources. In particular, the
influence of the organizing process seems to be
greater for actors belonging to a key network
context. In intermediation, the organizing of
activities, resources and actors can outline the
organizing of a plurality of networks within the
main network.
The IMP Journal
Volume 4. Issue 3
In conclusion, our work offers some new
investigative routes to the configuration of the three
layers of a network by pinpointing their different
degrees of importance in relation to the different
phases and environmental conditions.
The entrance of MC indicates a new middleman
in this distribution system. This is not what might
be expected in today’s mainstream view of efficient
distribution. Our case shows that the main route to
performance improvements is not always the
elimination of intermediaries. A new intermediary
will always be able to get a foothold as long as it
can contribute to value generation for other firms
References
Alderson, W. and Martin, M.W. (1965). Toward a
formal theory of transactions and transvections.
Review, 38 (4): 59-69.
Anderson, J.C., Håkansson, H. and Johanson, J.
(1994). Dyadic business relationships within a
business network context. Journal of Marketing,
58 (1): 1-15.
Baraldi, E. and Strömsten, T. (2006). Embedding
and utilizing low weight: Value creation and
resource configuration in the networks around
IKEA’s Lack table and Holmen’s newsprint.
IMP Journal, 1 (1): 52- 97.
Baraldi, E. and Strömsten, T. (2009). Controlling
and combining resources in networks - from
Uppsala to Stanford, and back again: The case of
a biotech innovation. Industrial Marketing
Management; 38 (5): 541-552.
Baraldi, E., Brennan, R., Harrison, D., Tunisini, A.
and Zolkiewski, J. (2007). Strategic thinking and
the IMP approach: A comparative analysis.
Bucklin, L.P. (1966). A Theory of Distribution
Channel Structure. Berkeley: Institute of
Business and Economic Research, University of
California.
Bucklin, L.P. (1972). Competiton and Evolution in
the Distributive Trades. Upper Saddle River,
Prantice Hall.
Buzzell, R.D. and Ortmeyer, G. (1995). Channel
partnerships streamline distribution. Sloan
Management Review, 36 (3): 85-95.
logistics providers: Rationale and outcomes.
Transportation Research Part E, 41 (6): 495510.
240
Christopher, M. (2005). Logistics and Supply Chain
Management - Creating Value Adding Networks.
London: Prentice Hall, Financial Times Pearson
Education.
Clarke, I. and Flaherty, T. (2003). Web-based B2B
portals. Industrial Marketing Management, 32
(1): 15-23
Corswant, F., Dubois, A. and Fredriksson, P.
(2001). Organising activities in industrial
networks - The case of Volvo S80. Proceedings
of the 17th IMP Conference. Oslo: BI Norwegian
School of Management.
Cox, W.M. and Ruffin, R.J. (1998). What Should
Economists Measure? The implications of mass
production vs. mass customization. Dallas:
Federal Reserve Bank of Dallas.
Da Silveira, G., Borenstein, D. and Foglietto, F.S.
(2001). Mass customization: Literature review
and research directions. International Journal of
Davis, S. M. (1987). Future Perfect. New York:
Addison-Wesley.
Dewan R., Jing, B. and Seidmann, A. (2003).
Product customization and price competition on
the Internet. Management Science, 49 (8): 10551070.
Drori, G.S., Meyer, J.W. and Hwang, H. (2006).
Globalization and Organization: World Society
and Organizational Change. Oxford: Oxford
University Press.
Dubois, A. (1998). Organising Industrial Activities
Across Firm Boundaries. London: Routledge.
Dubois, A. and Araujo, L. (2004). Research
methods in industrial marketing studies. In.
Håkansson, H., Harrison, D. and Waluszewski,
A. (eds.) Rethinking Marketing: Developing a
New Understanding of Markets, 207-228.
Chichester: John Wiley & Sons.
Dubois, A. and Gadde, L-E. (2002). Systematic
combining. An abductive approach to case
research. Journal of Business Research, 55 (7):
553−560.
Dubois, A. and Gibbert, M. (2010). From
complexity to transparency: Managing the
interplay between theory, method and empirical
phenomena in IMM case studies. Industrial
Marketing Management, 39 (1): 129–136.
Dubois, A., Gadde, L-E., Hulthén, K., Jonsson, P.
and Sundquist, V. (2004). Supply network
flexibility. Proceedings of the 20th IMP
Conference. Copenhagen: CBS Business School.
Duray, R., Ward, P.T., Milligan, G.W. and Berry,
W.L. (2000). Approaches to mass customization:
configurations and empirical validation. Journal
of Operations Management, 18 (6): 605–625.
Easton, G. (1992). Industrial networks: A review.
In Axelsson, B. and Easton, G. (eds.) Industrial
Networks: A New View of Reality, 1-27. London:
Routledge.
The IMP Journal
Volume 4. Issue 3
Easton, G. (1995). Methodology and industrial
networks. In: Möller, K. and Wilson, D. (eds.)
Business Marketing: An Interaction and Network
Perspective, 411-492. Norwell: Kluwer Academic Publishing.
Eisenhardt, K.M. (1989). Building theory from case
study research. Academy of Management
Review, 14 (4): 532-550.
Evans, P. and Wurster, T.S. (2000). Blown to Bits:
how the new economics of information
transforms strategy. Boston: Harvard Business
School Press.
Farshchi, M.A., Janne, O.E.M. and McCann, P.
(2009). Technological Change and Mature
Regions: Firms, Knowledge and Policy.
Cheltenham: Edward Elgar Publishing.
Fiocca, R., Snehota, I. and Tunisini, A. (2009).
Marketing Business to Business. Milano:
McGraw-Hill.
Flint D.J. (2004). Strategic marketing in global
supply chains: four challenges. Industrial
Ford, D., Gadde, L-E., Håkansson H. and Snehota,
I. (2003). Managing Business Relationships,
second edition. Chichester: John Wiley & Sons.
Ford, D., Gadde, L-E., Håkansson, H. and Snehota,
I. (2002). Managing networks. Proceedings of
the IMP-Asia Conference. Perth: Curtin University.
Ford, D., Gadde, L-E., Håkansson, H., Lundgren,
A., Snehota, I., Turnbull, P. and Wilson, D.
(1998). Managing Business Relationships, first
edition. Chichester: John Wiley & Sons.
Fraser, J. and Oppenheim, J. (1997). What's new
about globalization. McKinsey Quarterly, (2):
168-179.
Fredriksson, P. and Gadde, L-E. (2005). Flexibility
and rigidity in customization and build-to-order
production. Industrial Marketing Management,
34 (7): 695-705.
resource combining in distribution networks implications for relationship involvement and the
relationship atmosphere. Journal of Marketing
Management, 20 (1-2): 157-184.
IMP Journal, 2 (3): 36-52.
Gadde, L-E. and Håkansson, H. (1994). The
changing role of purchasing: reconsidering three
strategic issues. European Journal of Purchasing
and Supply Management, 1 (1): 27-35.
Gadde, L-E. and Håkansson, H. (2001). Supply
Network Strategies. Chichester: John Wiley.
Gadde, L-E. and Håkansson, H. (2008). Business
relationships and resource combining. IMP
Journal, 2 (1): 31-45.
Gadde, L-E. and Hulthén, K. (2009). Improving
logistics outsourcing through increasing buyer–
241
provider interaction. Industrial Marketing
Management, 38 (4): 633-640.
Gadde, L-E. and Snehota, I. (2000). Making the
most of supplier relationships. Industrial
Gadde, L-E. and Snehota, I. (2001). Rethinking the
role of middlemen. Proceedings of the 17th IMP
Conference. Oslo: BI Norwegian School of
Management.
Gadde, L-E., Huemer, L. and Håkansson, H.
(2003). Strategizing in industrial networks.
Golicic, S.L., Foggin, J.H. and Mentzer, J.T.
(2003). Relationship magnitude and its role in
interorganizational
relationship
structure.
Groothedde, B., Ruijghrok, C. and Tavasszy, L.
(2005). Towards collaborative, intermodal hub
network: a case study in the fast moving consumer goods market. Transportation Research
Part E, 41 (6): 567-583.
Gunasekaran, A. and Ngai, E. (2005). Build-toorder supply chain management:. Journal of
Operations Management, 23 (5): 423−451.
Hagel, J. and Rayport, J. F. (1997). The new
infomediaries. McKinsey Quarterly, (4): 54-70.
Håkansson H. and Harrison D. (2002). Activation
in resource networks: a comparative study of
ports. Proceedings of the IMP-Asia Conference.
Perth: Curtin University.
Håkansson H. and Snehota I. (1989). No business is
an island. Scandinavian Journal of Management
Studies, 4 (3): 187-200.
Håkansson, H. and Ford, D. (2002). How should
companies interact in business networks. Journal
of Business Research, 55 (2): 113-139.
Håkansson, H. and Johanson, J. (1992). A model of
industrial networks. In Axelsson, B. and Easton,
G. (eds.) Industrial Networks: A New View of
Routledge.
Håkansson, H. and Snehota, I. (2006). No business
is an island: The network concept of business
strategy. Scandinavian Journal of Management,
22 (3): 256-270.
Håkansson, H. and Waluszewski, A. (2002).
Managing Technological Development: IKEA,
the environment and technology. London:
Routledge.
Networks. Chichester: Wiley.
Halinen, A. and. Törnroos, J.-Å (2005). Using case
methods in the study of contemporary business
networks. Journal of Business Research, 58 (9):
1285-1297.
The IMP Journal
Volume 4. Issue 3
Heide, J.B. (2003). Plural governance in industrial
purchasing. Journal of Marketing, 67 (4): 18-29.
Holmen, E. and Pedersen, A.C. (2003). Strategizing
through analysing and influencing the network
horizon. Industrial Marketing Management, 32
(5): 409-418.
Huemer, L., Becerra, M. and Lunnan, R. (2004).
Organizational identity and network identification: relating within and beyond imaginary
boundaries. Scandinavian Journal of Management, 20 (1–2): 53−73.
Hvam. L., Mortensen, N.H. and Riis J. (2008).
Product Customization. Berlin: Springer.
Ishii, K and Ichimura, T. (2008). Information
behaviour in value characteristics deployment
approach to increasing customer satisfaction by
combining product and service. International
Journal of Entrepreneurship and Innovation
Management, 8 (2): 154-167.
Järvensivu, T. and Törnroos, J-Å. (2010). Case
study research with moderate constructionism:
Conceptualization and practical illustration.
Kaneko, J., and Nojiri, W. (2008). The logistics of
just-in-time between parts suppliers and car
assemblers in Japan. Journal of Transport
Geography, 16 (3): 155-173.
Lambert, D.M. and. Cooper, M. (2000). Issues in
supply chain management. Industrial Marketing
Management, 29 (1): 65-83.
Lampel, J. and Mintzberg, H. (1996). Customizing
customization. Sloan Management Review, 38
(1): 21-30.
Lancioni, R.A., Smith, M.F. and Oliva, T.A.
(2000). The role of the Internet in supply chain
management. Industrial Marketing Management,
29 (1): 45-56.
Lichtenthal, D. and Eliaz, S. (2003). Internet
integration in business marketing tactics.
Industrial Marketing Management, 32 (1): 3-13.
Mason, R. and Lalwani, C. (2008). Mass
customised distribution. International Journal of
Mattsson, L.-G. and Johanson, J. (1992). Network
positions and strategic action - an analytical
framework. In Axelsson, B. and Easton, G.
(eds.). Industrial Networks. A new view of
reality, 205-217. London: Routledge.
Meyer-Waarden, L. (2008). The influence of
loyalty programme membership on customer
purchase behaviour. European Journal of
Marketing, 42 (1-2): 87-114.
Mintzberg M. (1979). The Structuring of
Organizations. Englewood Cliffs: Prentice-Hall.
Narus, J.A. and Anderson, J.C. (1996). Rethinking
distribution: Adaptive channels. Harvard
242
Nault, S. and Dexter, A. (2006). Agent
intermediated electronic markets in international
freight transportation. Decision Support Systems,
41 (4): 787-802.
Nyström, A-G., Ramström, J. and Törnroos, J-Å.
(2008). Coping with business network dynamics
- strategizing through role and position. Proceedings of the 24th IMP Conference. Uppsala: University of Uppsala
Penrose, E.T. (1959). The Theory of the Growth of
the Firm. New York: Oxford University Press.
Pfohl, H-C. and Buse, H. (2000). Inter-organizational logistics systems in flexible production
networks. International Journal of Physical
Distribution and Logistics Manage-ment, 30 (5):
388-408.
Piekkari, R., Plakoyiannaki, E. and Welch, C.
(2010). 'Good' case research in industrial
marketing: Insights from research practice.
Pine, B. (1993). Mass Customization - the New
Frontier in Business Competition, first edition.
Boston: Harvard Business School Press.
Pine, B. and Davis, S. (2001). Mass Customization:
the New Frontier in Business Competition,
second edition. Boston: Harvard Business
School Press.
Piore, M. and Sabel, C. (1984). The Second
Industrial Divide: Possibilities for Prosperity.
New York: Basic Books.
Pitt, L., Berthon, P. and Berthon, J.P. (1999).
Changing channels: the impact of the Internet on
distribution strategy. Business Horizons, 42 (2):
19-28.
Pruden, D.R. and Vavra, T.G. (2000). The needed
evolution in marketing research is respecting
relationships with customers. Marketing
Research, 12 (2): 15-19.
Roseira, C. and Brito, C. (2009). Managing
interdependencies in supplier portfolios.
Proceedings of the 25th IMP Conference.
Marseille: Euromed.
Rosenbloom, B. (1995). Channels of distribution.
In: Baker, M. (ed.). Companion Encyclopaedia
of Marketing. London: Routledge.
Rushton, A., Croucher, P., and Baker, P. (2006).
The Handbook of Logistics & Distribution
Management. London: Kogan Page Publishers.
Sarkar, M., Butler, B. and Steinfield, C. (1996).
Intermediaries and cybermediaries: A continuing
role for mediating players in the electronic
marketplace. Journal for Computer Mediated
Communication, 1 (3): 1-11.
Sashi, C.M. and O’Leary, B. (2002). The role of
Internet auctions in the expansion of B2B
markets. Industrial Marketing Management, 31
(2): 103-110.
The IMP Journal
Volume 4. Issue 3
Scholte, J.A. (2000). Globalization: a critical
introduction. New York: ST Martin's Press.
Su, J., Chang, Y. and Ferguson, M. (2005).
Evaluation of postponement structures to
accommodate mass customization. Journal of
Operations Management, 23 (3-4): 315-318.
Suomala, P., Sievanen, M., and Paranko, J. (2004).
Customization from the after sales point of view
- implications of product and item customization
for spare-part business. Technovation, 24 (10):
831–840.
Viswanathan, S., Kuruzovich, J., Gosain, S. and
Agarw, R. (2007). Online infomediaries and
price discrimination: Evidence from the
automotive retailing sector. Journal of
Marketing, 71 (3): 89-107.
Vokurka, R. J. and Lummas, R. R. (2000). The role
of just-in-time in supply chain management. The
International Journal of Logistics Management,
11 (1): 89-98.
Waters, M. (2001). Globalization. New York:
Routledge.
Wathne, K.H. and Heide, J.B. (2004). Relationship
governance in a supply chain network. Journal
of Marketing, 68 (1): 73-89.
243
Webb, K. (2002). Managing channels of
distribution in the age of electronic commerce.
Whipple, J.M. and Russell, D. (2007). Building
supply chain collaboration: A typology of
collaborative relationship strategies. The
International Journal of Logistics Management,
18 (2): 174-196.
Wilson J. and Hynes N. (2009). Co-evolution of
firms and strategic alliances: Theory and
empirical evidence. Technological Forecasting
and Social Change, 76 (5): 620-628.
Woollard, F.G. (1954). Principles of Mass and
Flow Production. London: Iliffe & Sons, Ltd.
Yin, R.K. (1994). Case Study Research, Design,
and Methods, 2nd edition. Newbury Park. CA:
Sage Publications.
Yin, R.K. (2003). Case Study Research: Design
and Methods, 3rd edition. London: Sage
Publications.
Zolkiewski, J. and Turnbull, P. (2000). Relationship
portfolios - past, present and future. Proceedings
of the 16th IMP Conference. Bath: University of
Bath
Author guidelines
Copyright
Authors submitting a manuscript do so on the understanding that if it is accepted for publication,
copyright in the paper exclusive shall be assigned to the IMP Group. However, the IMP Group will
not put any limitations on the personal freedom of the author to use material contained in the article
in other works. A paper is accepted on the understanding that it has not been or will not be published
elsewhere in the same form, in any language, and that it is not under consideration by any other publication at the time of submission. It is the author’s responsibility to obtain clearance to use copyright
material within the article.
Manuscript requirements
• An electronic version of the manuscript (in an editable text file, e.g. Word) including the complete
text as well as any graphics/illustrations and tables.
• A portable document file of the complete manuscript (text, images, image captions, tables, etc) for
reference purposes.
• All required permissions for reproduction of elements originating from other sources, in particular
graphics.
• Submissions should generally be of 4000-6000 words but if a case description is of significant importance the submission can be longer.
• The first page should consist of the title, authors’ names and affiliations. The second page should
comprise the title and an abstract of the paper (100-150 words).
• References are indicated in the text by the Harvard (name and date) system and all references
should be listed at the end of the paper in alphabetical order.
Submitting the paper
Please, submit your paper through our website (http://www.impjournal.org) or send it as an e-mail to
the editor.
Review process
Each paper is reviewed by the Editor and, if it is judged suitable for the IMP Journal, it is then sent to
two referees for double blind peer review. Based on their recommendations, the Editor then decides
whether the paper should be accepted as is, revised or rejected.
About the IMP Journal
The research of the IMP Group is since it started in the 1976 based on the idea that business is not an
isolated activity that occurs within independent organisations, but that it consists of interaction between interdependent companies; whether as customers, suppliers, development partners, facilitators or
competitors.
Over the years a large number of research studies have investigated the content and effect of interaction between individual companies and organisations, as well as how these processes are related to a
larger surrounding network. This research has led to the development of ideas on the characteristics of
interdependencies; on business and organizational relationships and networks, on the processes within
them and on how individual companies can operate in these arenas. To put it short, the research has
lead to ideas on the light and dark sides of networks; for companies and those who are dependent on
them, as well as for the larger society.
The IMP Journal is a forum for high quality of research into business interaction. The ambition is
to publish the best articles of an increasing volume and standard of research, but will also stimulate
further development in the field. We welcome contributions from researchers in a wide area of research; marketing, purchasing, technological development and innovation, policy, strategy, operations,
logistics, information systems, or human resources, etc.; which share the interest in the phenomenon of
interaction and interdependencies between companies and other organisations.
Contemporary ideas on business interaction have mainly appeared through empirical studies within
and between companies. Although we welcome excellent submissions of conceptual work, we would
like the journal to continue this strong empirical tradition.
Contributions may be of any length appropriate to their content, which will give opportunities to include rich case studies.
All submissions will be double blind refereed by members of the editorial review board.
As well as completed manuscripts, we would also like to welcome contributors to contact us at the early
stage of development of manuscripts. We will then be happy to work with those contributors in the
development of their work to achieve the highest possible quality of publication.
The journal is publishing three issues every year and is in an electronic version and is located on its
own website www.impjournal.org. It can also be reached through www.impgroup.org. This means that
it will be freely available to all researchers, without subscription. Every full volume is also printed and
included in the proceedings from the yearly main IMP conference.
ISSN: 0809-7259
Author guidelines
Copyright
Authors submitting a manuscript do so on the understanding that if it is accepted for publication,
copyright in the paper exclusive shall be assigned to the IMP Group. However, the IMP Group will
not put any limitations on the personal freedom of the author to use material contained in the article
in other works. A paper is accepted on the understanding that it has not been or will not be published
elsewhere in the same form, in any language, and that it is not under consideration by any other publication at the time of submission. It is the author’s responsibility to obtain clearance to use copyright
material within the article.
Manuscript requirements
• An electronic version of the manuscript (in an editable text file, e.g. Word) including the complete
text as well as any graphics/illustrations and tables.
• A portable document file of the complete manuscript (text, images, image captions, tables, etc) for
reference purposes.
• All required permissions for reproduction of elements originating from other sources, in particular
graphics.
• Submissions should generally be of 4000-6000 words but if a case description is of significant importance the submission can be longer.
• The first page should consist of the title, authors’ names and affiliations. The second page should
comprise the title and an abstract of the paper (100-150 words).
• References are indicated in the text by the Harvard (name and date) system and all references
should be listed at the end of the paper in alphabetical order.
Submitting the paper
Please, submit your paper through our website (http://www.impjournal.org) or send it as an e-mail to
the editor.
Review process
Each paper is reviewed by the Editor and, if it is judged suitable for the IMP Journal, it is then sent to
two referees for double blind peer review. Based on their recommendations, the Editor then decides
whether the paper should be accepted as is, revised or rejected.
About the IMP Journal
The research of the IMP Group is since it started in the 1976 based on the idea that business is not an
isolated activity that occurs within independent organisations, but that it consists of interaction between interdependent companies; whether as customers, suppliers, development partners, facilitators or
competitors.
Over the years a large number of research studies have investigated the content and effect of interaction between individual companies and organisations, as well as how these processes are related to a
larger surrounding network. This research has led to the development of ideas on the characteristics of
interdependencies; on business and organizational relationships and networks, on the processes within
them and on how individual companies can operate in these arenas. To put it short, the research has
lead to ideas on the light and dark sides of networks; for companies and those who are dependent on
them, as well as for the larger society.
The IMP Journal is a forum for high quality of research into business interaction. The ambition is
to publish the best articles of an increasing volume and standard of research, but will also stimulate
further development in the field. We welcome contributions from researchers in a wide area of research; marketing, purchasing, technological development and innovation, policy, strategy, operations,
logistics, information systems, or human resources, etc.; which share the interest in the phenomenon of
interaction and interdependencies between companies and other organisations.
Contemporary ideas on business interaction have mainly appeared through empirical studies within
and between companies. Although we welcome excellent submissions of conceptual work, we would
like the journal to continue this strong empirical tradition.
Contributions may be of any length appropriate to their content, which will give opportunities to include rich case studies.
All submissions will be double blind refereed by members of the editorial review board.
As well as completed manuscripts, we would also like to welcome contributors to contact us at the early
stage of development of manuscripts. We will then be happy to work with those contributors in the
development of their work to achieve the highest possible quality of publication.
The journal is publishing three issues every year and is in an electronic version and is located on its
own website www.impjournal.org. It can also be reached through www.impgroup.org. This means that
it will be freely available to all researchers, without subscription. Every full volume is also printed and
included in the proceedings from the yearly main IMP conference.
ISSN: 0809-7259

Volume 4 - The IMP Journal

Transcription

Similar documents

Jacqui Kaese poster - Screen Actors Studio

Appreciation for the Biz

Teacher Pages

Mary Kate Zihar: Design and Illustration

Laxmi Khattri lkhattri@union-city.k12.nj.us

www.yvonnedabbracciomanagement.com

Press Release WWII Foundation

Jack Ampuja Supply Chain Optimizers Niagara University June 8

ATT Contract

Divimage Newsletter 2013 WEB