Description
Accounting and the classical coordination forms of hierarchy and market are closely related concepts. New forms of
coordination, including alliances and close cooperation between companies have opened up new challenges to how
accounting should be designed and practised. In the first part of this paper the theoretical basis for the connection
between coordination forms and the need for information is explored. Two key problems seem to be that the coordination
forms are not just alternatives, but can be used in combination, and that relationships between companies can
be viewed in different ways. These issues are studied through a case study of an extensive relationship between Ericsson
and Telia Mobile. The case demonstrates that all of the three basic coordination forms are used in creating a complex
network relating a set of sub-units from the two companies to each other and to other related parties. This means that
accounting methods developed earlier can be used in several ways to enhance the development of the relationships and
that the latter should not be perceived from one level or considered to be a uniform form. Companies can combine the
different forms on one level to achieve a coordination form that is unique at a higher level.
Accounting and network coordination
Ha? kan Ha? kansson
a
, Johnny Lind
b,
*
a
Norwegian School of Management, Department of Marketing, Box 580, N-1302 Sandvika, Norway
b
Stockholm School of Economics, Department of Accounting and Managerial ?nance,
Box 6501, 113 83 Stockholm, Sweden
Abstract
Accounting and the classical coordination forms of hierarchy and market are closely related concepts. New forms of
coordination, including alliances and close cooperation between companies have opened up new challenges to how
accounting should be designed and practised. In the ?rst part of this paper the theoretical basis for the connection
between coordination forms and the need for information is explored. Two key problems seem to be that the coordi-
nation forms are not just alternatives, but can be used in combination, and that relationships between companies can
be viewed in di?erent ways. These issues are studied through a case study of an extensive relationship between Ericsson
and Telia Mobile. The case demonstrates that all of the three basic coordination forms are used in creating a complex
network relating a set of sub-units from the two companies to each other and to other related parties. This means that
accounting methods developed earlier can be used in several ways to enhance the development of the relationships and
that the latter should not be perceived from one level or considered to be a uniform form. Companies can combine the
di?erent forms on one level to achieve a coordination form that is unique at a higher level.
#2003 Elsevier Ltd. All rights reserved.
Introduction
‘Do you want me to explain the organization of
Ericsson? That is impossible. Everything changes
all the time. Our operations are too complex. We
have important customers with di?erent and in
some cases con?icting claims on a cellular system.
We have to manage several technologies at the
same time. We have units everywhere in the world.
It is not an easy task to control this muddle. Some
months ago we put a lot of e?ort into creating
accounting information for a new organizational
unit. However, when we had ?nished, the top
management closed the unit. It is like going on
water logged marshland, the more e?ort you put
in, the more you sink. It is important that you can
?nd the right balance between all the di?erent
demands all the time.’ (Senior accountant)
The complexity confronting a large interna-
tional company has become more complicated
over time. Joint ventures, alliances, strategic part-
nerships, outsourcing, and decentralized organi-
zations change the ?rm’s boundaries continuously.
Management methods such as just in time (JIT),
time based management (TBM), lean production
and business process reengineering (BPR) aim at
eliminating all bu?ers within companies and
0361-3682/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0361-3682(02)00058-2
Accounting, Organizations and Society 29 (2004) 51–72
www.elsevier.com/locate/aos
* Corresponding author. Tel.: +46-8-736-93-17; fax: +46-
8-32-01-09.
E-mail addresses: [email protected] (H. Ha? kansson),
[email protected] (J. Lind).
thereby increase the companies’ dependence on
customers and suppliers. The products and services
consist of technologies that are in signi?cantly
di?erent stages of development. The customers
impose di?erent and sometimes con?icting
demands. The company has in this situation to
function within a network of relationships with
quite di?erent characteristics.
The telecom industry is a typical example of an
industry having the above features. It has under-
gone rapid change since the early 1980s. The
computer industry and telecom industry have
converged into a single industry. The customer
bases have changed from nationally monopolistic
operators to a more heterogeneous bunch of
operators ranging from small local players to large
multinational companies. Ericsson is one of the
largest telecom companies in the world having
extensive collaborative relationships with compa-
nies in di?erent areas. Ericsson has extensive rela-
tionships with some major customers as well as
suppliers but also regarding mobile phones with
Sony, operation support systems with Hewlett
Packard, and WCDMA technologies with Telia.
Ericsson uses relationship coordination to a
considerable extent. Such coordination is char-
acterized by cooperation and quasi integration.
Resources and activities are connected across
organizational boundaries to facilitate ?ows of
products, to increase accuracy and speed, and to
facilitate technical development. This cooperative
form is claimed to increase e?ciency but another
consequence is that the boundary of the single
?rm becomes blurred.
Accounting, information, e?ciency and coordi-
nation forms have long been closely related
(Chandler & Daems, 1979; Tomkins, 2001). Some
writers even claim that the development of
accounting methods has been a prerequisite for
hierarchical coordination (Chandler & Daems,
1979; Johnson, 1981, 1983). According to these
writers, accounting and hierarchical coordination
have been developed in symbiosis. The argument
is that hierarchical coordination has become a
substitute for market coordination because it is
more e?cient in some situations. The hierarchies
have grown larger and it has been possible to
control them through the development of new
sophisticated accounting methods, which give the
managers access to the information required to
control the internal activities. Accounting and the
classical market-hierarchy dichotomy are well
adjusted to each other.
Relationship coordination may cause a problem
from an accounting point of view as contemporary
accounting depends on de?ned, limited entities.
This new form of coordination blurs the clearly
de?ned boundaries which accounting presupposes
and requires. Thus, one can anticipate a problem
when combining the new form of coordination
with the way accounting is applied in companies.
As a consequence, companies such as Ericsson
should have a problem in the design and use of
accounting as it has to combine the use of rela-
tionship coordination with the contemporary
accounting methods. This article is devoted to an
analysis of this problem and how Ericsson has
resolve it.
The remainder of the paper is organized as fol-
lows. In section 2 the theoretical basis for the
connection between coordination forms and the
need for information is explored. Section 3 reviews
previous studies on accounting and the basic
coordination forms and identi?es key problems
within the cooperative coordination. Section 4
presents the research methods used. The ?fth sec-
tion presents the results of the case study. The
sixth section analyzes the case data in more detail
and relates the ?ndings to existing literature.
Conclusions and suggestions for future research
are presented in a ?nal section.
E?ciency, information needs and di?erent forms
of coordination
To achieve e?ciency, di?erent production
activities are coordinated within the industrial
structure (Dubois, 1998; Richardson, 1972). Three
theoretical forms—hierarchy, market and co-oper-
ation (relationship)—were identi?ed by Richard-
son (1972) and, since then, have been discussed and
compared in a large number of studies (e.g. Rind-
?eish & Heide, 1997; Williamson, 1975; 1985).
Hierarchy is a directed coordination form.
According to Richardson, it is useful when two
52 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
activities are closely complementary and similar,
i.e. when there are scale e?ects. In this case, being
closely complementary is de?ned as representing a
situation where di?erent phases of a production
process require some coordination. Similar activ-
ities are de?ned as those activities that require the
same capability for them to be undertaken (ibid).
In a hierarchy, two activities within a company are
directly coordinated. In principal, they become
one activity, and as far as the organization is con-
cerned, they are meshed together. Consequently,
they can be perfectly adapted to each other, but
they can only be jointly related to other activities.
The adaptation is based on information about the
two activities and their closely complementary
relationship. This includes information about the
resources needed to perform the activities. If some
of these resources are heterogeneous (Alchian &
Demsetz, 1972; Ha? kansson, 1993) it is possible to
develop and thereby learn about their interplay.
The existence of learning implies that information
is generated over time. To take advantage of this,
the company needs a continuous supply of very
detailed information concerning technical and
economic aspects of the activities performed and
the use of resources. We can also formulate this in
another way: the better, meaning the more com-
prehensive, the information regarding the activ-
ities and the use of resources, the better the
company can coordinate the two activities
advantageously.
Market is a coordination form where the inter-
face between the two coordinated activities is
standardized to ensure that several customers’
demands are matched with the supply of several
producers of similar products. Thus, there are
large numbers of standardized products creating
an aggregated output controlled by a demand–
supply balance. According to Richardson this is a
suitable coordination form when activities are
complementary and dissimilar. In such a situation,
the market will let each activity be undertaken in
accordance with its optimal scale. The activities
will be performed independently of each other in
any speci?c sense, but they will be highly depen-
dent on their common standardized product.
If we assume that there are some imperfections
in the market in regard to knowledge then a
company will need information regarding the pri-
ces of the standardized product from di?erent
suppliers. Furthermore, it needs information
about the ‘‘consuming’’ activities of other buyers.
The latter information is required as the company
must develop its own ‘‘consumption’’ in accor-
dance with the other buyers in order to demand
the same solution—to retain the standardization
of its products. However, the information needed
is not of the same detailed character at all as in the
hierarchical situation. But the loss of precision
required for the data is achieved at the price of a
very in?exible solution because of the need to
maintain the standardization. However, in
another respect, the market is opened up for new
solutions. If the activity performed by the supplier
or the customer can be combined with another
type of activity, there is no problem switching
counterpart. As long as there is no need for adap-
tation, the activities can be freely combined with
each other. One conclusion is that the information
problem does not a?ect the outcome in the same
way as in the hierarchical situation. Through the
standardization of the products (the solutions) the
information can also be standardized and thereby
become much easier to handle.
The business relationship or cooperation is the
third form of coordination. In this mode, the
coordination of the two activities cannot be left
entirely to the internal coordination of the ?rm or
the market forces. The reason for this is that the
activities are dissimilar and therefore, despite
being closely complementary, they need to be per-
formed by di?erent companies. Thus, the di?erent
companies’ plans must be matched, both qualita-
tively and quantitatively. The market is not sui-
table for this matching (Richardson, 1972). This
situation is much more complicated than the mar-
ket and hierarchy form of coordination, but the
basic feature of this coordination is interaction.
The two units try to ?nd a solution by matching
their plans. In this case, the coordination is not
centrally orchestrated, as in the hierarchical sit-
uation, or carried out by the total structure, as in
the market situation. Instead, we have two units
interactively seeking a suitable solution. In a rela-
tionship, the two companies involved match their
activities and resources to each other. The degree
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 53
to which this is possible is directly dependent on
the information available. As in the hierarchical
situation, each of the two parties must have an
extensive knowledge of their own activities and
their own use of resources, but in this case they
also need some information regarding their coun-
terpart. There must be a certain overlap in the two
parties’ knowledge to make a relationship based
on a common understanding possible. Thus, the
quality of the relationship is directly dependent on
the underlying information.
These three modes of coordination can be taken
to be alternatives, as Richardson and later
research (Bradach & Eccles, 1989; Haugland &
Reve, 1994; Williamson, 1975, 1985) regarding
relationships have done, but they are also clearly
interrelated and interdependent.
In summary, the need for information in the
hierarchical and the relationship modes is crucial
for the quality of the coordination of the activities.
In both of these cases, the degree to which e?-
ciency is created is highly dependent on the infor-
mation available to the decision-makers. This is
not the case for the market situation. But this is
not just a matter of coincidence—it is the very
reason for making the classical distinction between
market and hierarchy. The boundary of the
company separates two di?erent types of inter-
dependencies between activities. Internally, there
are activities that are closely complementary
and/or similar, while externally the activities are
complementary and dissimilar. Hierarchical and
market coordination are not just two alternatives,
they are perfectly complementary and well-mat-
ched from an information point of view. The
boundary of the company divides the world in
accordance with the two coordination modes, which
are also matched by the need for information. On
one side of the boundary is the hierarchy, with a
need for very detailed information. On the other
side is the market with a lesser need for in-depth
information.
When relationships need to be taken into con-
sideration, this clear image is shattered. In the
relationship case, there is neither a clear bound-
ary, nor do we have any matching internal coor-
dination form. Thus, relationships must be a
problem from an accounting point of view, given
that accountancy has been developed in accor-
dance with the hierarchy-market dichotomy.
There must be a need to change accounting when
used in a network situation. Let us ?rst see how
this issue has been handled in the recent develop-
ments within the accounting literature.
Accounting and coordination forms
Accounting and its development is interwined
with models of coordination (Ezzamel, 1992).
Ouchi (1977, 1979, 1980) developed a general fra-
mework for control and coordination forms.
Control and coordination forms are classi?ed
along the two dimensions ability to measure out-
put and knowledge of the transformation process.
Ouchi provides a framework comprised of three
coordination forms: hierarchy, market and clan.
Ezzamel (1992) draws on Ouchi’s classic work
(1979, 1980) when he summarizes the role of
accounting within the coordination forms of hier-
archy, market and clan.
Hierarchical coordination relies on rules and
formal power. Accounting is used to monitor and
in?uence people’s behaviour (Birnberg, 1998;
Ouchi, 1980). Accounting methods, such as mea-
suring direct cost, e.g., the direct cost of labour
and materials, standard costs, break-even charts,
allocation methods, ROI measures, responsibility
accounting, reward structures, transfer prices and
budgets have been developed to support decision
making and performance evaluation within orga-
nizations (Chandler & Daems, 1979; Ezzamel,
1992). Accounting used to focus mainly on the
?nancial aspect, but has gradually incorporated
other non-?nancial aspects (Ittner & Larcker,
1998). Accounting measurements are detailed and
enable di?erent objects such as organizational
units, products, projects and activities to be
described and analyzed in regard to past, present
and future states. Consequently, accounting is in a
good position to support managers by providing
information for decision-making and performance
evaluation in situations where the hierarchy-mar-
ket dichotomy exists (Bromwich, 1990).
Market coordination is dependent on market pri-
ces, and prices are the primary source of information
54 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
supporting decision makers in companies (Ouchi,
1980). However, accounting still has a role within
market coordination. According to Ezzamel
(1992), accounting is vital to ensure the normative
requirement of reciprocity in exchange relation-
ships. In his view this means collecting and sum-
marizing market prices, overseeing the
implementation of contracts, estimating market
opportunities and threats, and developing deci-
sion-making models based on market prices. As
such, it is possible to compare costs and bene?ts
for any given exchange.
Clan coordination relies on traditions and the
necessity of having common values and beliefs
(Ouchi, 1980). Ezzamel (1992) argues that the role
of accounting within clan co-ordination should
contribute to the establishment of common values
and beliefs, and sustain the organizational tradi-
tion. However, occasionally it is necessary to
reconstruct the organizational tradition. Many of
the accounting methods developed and used in
hierarchical coordination will also be utilized in
the clan setting. However, Ezzamel (1992) identi-
?ed three major di?erences: First, the essence of
control is more mutual/consensual. Second, much
e?ort is expended on establishing and continuing a
?nancial tradition but it will be subordinate to
common values and traditions. Third, the focus of
?nancial control is more on the socialization process
than on supporting managers with information for
decision making and performance evaluation.
In recent years, there are several examples of
research in accounting where the need to explore
the e?ects of accounting across organizational
boundaries has been stressed (Hopwood, 1996;
Otley et al., 1995; Shields, 1997). More speci?c
attempts to discuss the role of accounting within
cooperative coordination have been made by
Berry (1994), Gietzman (1996), Seal and Vincent-
Jones (1997), Birnberg (1998), Jones (1998), Seal
et al. (1999), Nicolini et al. (2000), Van der Meer-
Kooistra and Vosselman (2000), Mouritsen et al.
(2001) and Tomkins (2001).
Berry (1994) argues that the clan identi?ed by
Ouchi (1979, 1980) can be related to cooperative
coordination. He exempli?es this by an alliance
between an American and a Japanese company in
the automotive component industry. In his view,
the case shows that the alliance used clan-based
control in which the companies had open discus-
sions with each other, communicated their pro-
blems, and shared their knowledge and
information. One observation was that the two
companies in that study viewed the experience of
the collaboration di?erently. The American com-
pany was dissatis?ed because they felt that they
had not learnt anything. In contrast, the Japanese
company had a more positive experience. The
conclusion Berry (1994) came to was that it can be
necessary to create trans-organizational work-
groups for cooperative coordination. These work-
groups then use clan control as the mode of
control. Berry (1994) argued that clan control is a
suitable mode of control for partnerships and alli-
ances. Such a conclusion seems reasonable when
analyzing just one relationship.
Birnberg (1998) develops a framework for con-
trol in inter?rm cooperative relationships.
According to him (ibid, p. 421) a co-operative
relationship can be described in ?ve dimensions:
‘(1) the degree of absolute and relative commit-
ment; (2) the symmetry of the rewards; (3) the
extent of uncertainty present; (4) the degree of
mutual trust between the parties; and (5) the
length of the relationship’. Four types of inter?rm
cooperative relationships were identi?ed, ranging
from relationships with both low interdependency
and low uncertainty to relationships with both
high interdependency and high uncertainty. Birn-
berg’s analysis focused on single relationships.
Two empirical studies made by Seal et al. (1999)
and Van der Meer-Kooistra and Vosselman (2000)
were conducted in an attempt to try to understand
the role of accounting within the supply chain
from the perspective of transaction cost theory.
Van der Meer-Kooistra and Vosselman (2000)
identi?ed three types of control pattern within
inter?rm relationships: a market based pattern, a
bureaucracy based pattern and a trust based pat-
tern. Seal et al. (1999) indicated that the account-
ing can play a central role in the establishment of
collaborative inter-organizational business rela-
tionships. The empirical studies have concentrated
upon newly established strategic alliances. Thus,
the focus of this research is on a dyadic relation,
which is studied in isolation.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 55
Tomkins (2001) adopted a similar point of
departure to other accounting scholars writing
about cooperative coordination. His main focus
was on the information needed within single busi-
ness relationships. However, the consequences
were also discussed at a network level. In that
way, both the direct and indirect e?ects of the
decisions made by the company were captured
because the decisions not only a?ect the partner
directly, but also the partner’s relationships.
In combination with the earlier discussion
points, the above studies highlight two unresolved
matters. One has to do with how relationships are
conceptualized: if a relationship is seen as an iso-
lated cooperative island, then there is a boundary
around it. This seems to be the case for most of
the studies reviewed. However, if the relationship
is considered to be one element in a network, such
a boundary might create problems. In that case,
an e?cient mode of coordination would require
simultaneous connections to several business rela-
tionships (Anderson et al., 1994). This will put
even greater demands on the supply of infor-
mation. In order to enhance adaptations of activ-
ities and resources across the boundaries of
companies, better knowledge is required about
what is going on in each of the counterparts that
will have an impact on the others. It is not enough
to know about the activities or the resources of
just the counterpart, instead, detailed information
about speci?c activities, i.e. the costs, revenues,
quality, cycle time, reliability or delivery, and
information concerning the use of resources will
be needed for some of the connected parties to
enable ways of improving the operations.
The second problem has to do with the inter-
faces between the three pure forms of coordi-
nation. The relationships a company is involved in
consist of both close and ‘‘arm’s length’’ relation-
ships (Tomkins, 2001). This means that a com-
pany always has a mix of relationships, including
some close-knit associations with a high level of
trust and some more market-like transactions.
Earlier we concluded that the hierarchy and the
market are closely complementary. It is also pos-
sible to ?nd connections between relationships and
these two basic coordination modes. In order for
anyone to have a close and strong relationship it is
important ‘to be someone’. Thus, in the relation-
ship case, there integrating mechanisms are needed
in relation to the individual company. The stron-
ger the relationship, the more important this inte-
gration will have to become (Ha? kansson and
Ford, 2002). Consequently, there seems to be a
positive connection between hierarchy and rela-
tionships. In the same way, in the relationship
case, alternatives are required. It must be possible
for either of the parties to break a relationship, so
the existence of alternative parties is necessary.
Thus, there must be some kind of market ingre-
dients in the relationship case too. If there are no
alternatives whatsoever, the situation will be simi-
lar to vertical integration, i.e. to a hierarchy.
Considered in this way, there are reasons to
assume that the relationship form and its e?ects
might be dependent on the existence of hierarchies
as well as markets. The three forms are not just
alternatives, but are complementary to each other.
In order to study how the use of accounting
methods is further in?uenced by these two issues,
an empirical study was conducted.
Research method
As we have seen, the role of accounting within
cooperative coordination is more complicated
than in hierarchical and market coordination.
Until now the accounting literature has paid little
attention to cooperative coordination, especially
when the company is considered to be intertwined
in a complex network of relationships. An
exploratory case study seems to be a reasonable
research design because of the lack of previous
studies within the area (Yin, 1989). Thus, we have
chosen to study a well developed business rela-
tionship in the telecom industry between Ericsson
and Telia. This is a long lasting and well known
business relationship within the Swedish business
community. The empirical study was partly inten-
ded o identify the relationship and partly to iden-
tify the role of accounting within the relationship.
The point of departure of the relationship part
of the study was the development of standard
software releases within the GSM-area. Thus, the
relationship is primarily captured through the
56 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
standard software release projects and the services
developed in conjunction with the releases. Hence,
four di?erent release projects (R6, R7, R8, R9),
and one end service (Dirigent) have been studied
to identify the interfaces between Ericsson and
Telia. However, to obtain a more comprehensive
understanding of the relationship, some discus-
sions, interviews, and ?eld visits were conducted
with employees from other units, such as those
involved in the basic AXE-research, the plant
producing radio base stations, and key accounts
within Telia. In this way, the study identi?es the
interfaces between units during the entire life cycle
of a service, from the early stages within Ericsson
to the late stages within Telia.
At the time the study was conducted, the GSM
cellular system was the largest business unit within
Ericsson and one of Telia’s prioritized areas.
According to both the purchasing manager in
Telia Mobile and one of the vice presidents in the
GSM business unit, the software development
represented the most important aspect of the
relationship.
The accounting part of the study used the
investigation of the relationship to identify critical
organizational units within the relationship. Four
units were identi?ed, the KAM unit and BSS-unit
within Ericsson and the purchasing and net
development unit in Telia Mobile. The because of
the units chosen, the study focused on radio tech-
nology in both Ericsson and Telia. Accountants,
managers and project members were interviewed
in the four units. The interviews were aimed to
examine the design and use of accounting and the
information ?ow. As such, the intention was to
identify the role of accounting within a unit as well
as outside the unit, and to identify the information
?ow between units within a company as well as
between units in di?erent companies. However, to
expand the understanding of the role of account-
ing, some discussions and interviews were con-
ducted with members from other business units in
Ericsson and in the operation and the market unit
within Telia Mobile.
The empirical study started in December 1997
and the majority of the data collection was done
during 1998 and early 1999. We received the full
support of both companies even to the extent of
having two o?ces assigned to us in Ericsson, one
within the central accounting unit and the other in
the Key Accounts Telia unit. This arrangement
gave us access to Ericsson’s intranet. In addition,
access was given to economic reports, internal let-
ters, detailed statements, organizational charts,
documents relating to strategy and technical and
commercial reports. In Telia, in addition to the
interviews, we were given access to written docu-
ments including organizational charts. As such,
the study tried to use as many sources as possible
to support the empirical data and thereby trian-
gulate the results (Scapens, 1990).
The major sources of information were inter-
views and discussions, which were conducted with
67 employees within the two companies. The
majority of the interviews were with people within
Key Accounts Telia and the BSS unit within
Ericsson, and the purchasing, the development
and the operation units within Telia Mobile.
Those interviewed represented almost all organi-
zational levels from the vice president to develop-
ment engineers and project members. Almost all
functions were covered, from strategic product
management, system development, operations and
marketing, to accounting and top management.
The interviews were semi-structured and lasted
between 1 and 3 h, with an average duration of 1.5
h. A more detailed description of those inter-
viewed is described in the Appendix.
The Ericsson–Telia business relationship
Ericsson is one of the largest companies in
Sweden. It was founded 1876. Telia is the former
public telecommunications operator; it was foun-
ded a few years after Ericsson. Telia had a mono-
poly from 1918 until the early 1980s. Ericsson
grew extensively during the 1990s while Telia
increased its sales volume and reduced its number
of employees at the same time.
The relationship—the global alliance
The relationship between Ericsson and Telia is
important to both companies. Its importance is
manifested by a partnership agreement between
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 57
the companies. The relationship is considered to
be so important that the CEOs in Ericsson and
Telia meet regularly. Ericsson’s sales in Sweden
during 1997 were about 5.5% of its total turnover,
or more than 1 billion US $. Telia is Ericsson’s
largest customer in Sweden. Furthermore Ericsson
is Telia’s largest supplier measured in terms of
purchasing cost. The relationship is global in the
sense that the cooperation is worldwide. For
example, Ericsson delivered new cellular systems
to Telia in Brazil as well as in Denmark. Both
companies have adapted their organization to
each other. Ericsson has special units primarily
engaged in managing the relationship with Telia
and it has a Global Account Manager (GAM)
who is responsible for the relationship as a whole.
Telia has individuals and sub-units dedicated to
work and collaborations with actors within
Ericsson.
The commercial exchange between the compa-
nies consists mainly of hardware and software for
wired and cellular communication systems. Mobile
telephones are not a part of the commercial
exchange, although their existence is important to
both companies as they in?uence the end custo-
mers’ ability to use new services. As a con-
sequence, Telia Mobile has a speci?c unit which is
dedicated to mobile phones and which interacts
with the major mobile phone manufacturers. Over
the years, Telia and Ericsson have cooperated on
the development of new communication systems.
For example, they developed the automated AXE
switch, which is the backbone of Ericsson tele-
communication solutions, together.
There are several di?erent systems involved in
the cellular sector. NMT is the old analogue stan-
dard for the Nordic cellular system ?rst intro-
duced in 1981. Today the NMT-related work in
both Ericsson and Telia is only focused on main-
tenance activities. GSM is the European digital
cellular standard launched in 1991 and it is the
main contemporary cellular system in Sweden.
Telia Mobile and the GSM business unit in Erics-
son are the two counterparts handling the GSM
cellular system. D-AMPS is the American digital
cellular standard, it is mainly handled by Telia
International and the D-AMPS business unit in
Ericsson. WCDMA is the next generation of cellular
systems handled primarily by Telia Research, a
sta? unit. At Ericsson, a newly created business
unit handles the WCDMA system with technology
and research sta? support. The president for the
GSM-standard cellular system and Telia Mobile
meet frequently to discuss problems and potential
opportunities in the relationship.
The interaction between the two companies is
strongly in?uenced by their organizational struc-
tures. The three most important dimensions in
which the organizations have been structured are
technology, relationships and time. Let us start
with the units responsible for the relationships.
Relationship units
The relationship covers both wired and cellular
systems. Formerly the wired system was the cen-
tral part of the relationship, but today the cellular
system is more important. Ericsson has assigned
special organizational units and individuals to be
accountable for the customer relationship. Erics-
son has three di?erent levels for relationship
accountability: Global Account Executive (GAE),
Global Account Manager (GAM) and Key
Account Manager (KAM). One GAE and one
GAM are accountable for the overall corporate
relationship with Telia. They have to develop and
secure the business and coordinate Ericsson’s
activities with Telia. There are several KAMs, one
of which is accountable for the relationship with
Telia Mobile. The KAM for Telia Mobile is an
organizational unit comprised of more than 60
employees.
Ericsson has other GAEs, GAMs and KAMs
accountable for relationships with other important
customers such as Global One, Mannesmann and
Sonera. All of these strive to obtain a closer rela-
tionship between each customer and Ericsson.
Subsequently, each particular KAM’s aim must be
to adapt Ericsson’s technical units to its own cus-
tomer. One obvious problem is that the customers
have di?erent demands. For example, Telia cer-
tainly imposes other demands on a cellular system
than a new operator in China would do. As a
consequence, the technical units within Ericsson
will be confronted with contradictory speci?-
cations. According to the vice president responsible
58 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
for the European market within Ericsson, the di?er-
ence between customer’s needs has increased during
recent years as a consequence of the increased num-
ber of features in the software. A consequence of
these di?erences is that the other KAM units within
Ericsson will view the KAM unit handling Telia
Mobile to an increasing degree as a competitor.
The business units within Telia are organized in
departments on a functional basis, such as pur-
chasing, development, and operations. Telia does
not have a speci?c unit responsible for the rela-
tionship with Ericsson. However, within the func-
tional units in Telia Mobile, speci?c individuals
and sub-units work exclusively with particular sup-
pliers. The purchasing department has dedicated
one speci?c purchaser to Ericsson and another to
Nokia. The sub-units within the development
department at Telia Mobile are divided into two
groups, one of them interacting with Ericsson and
the other with Nokia. Telia Mobile’s cellular sys-
tem in Sweden is comprised of parts from both
Ericsson and Nokia, but with Ericsson being the
dominant supplier. Consequently, Telia Mobile
has had to coordinate the two supplier relation-
ships and cannot entirely adapt its operations to
just one of them. One such example is that Telia
Mobile must coordinate Ericsson’s and Nokia’s
new GSM standard software releases. However
Telia Mobile has adapted its organization to some
critical supplier relationships. Fig. 1) shows the
Fig. 1.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 59
relationship units within Ericsson and Telia, and
other related units and companies.
Technical units
Ericsson and Telia have divided their organi-
zations in accordance with the technical solutions
and products with which they are concerned. On
an aggregate level, Ericsson has three business
units: systems for wired communication, systems
for cellular communication, and mobile phones.
The business unit for cellular communication is
divided into ?ve cellular system standards: NMT,
GSM, D-AMPS, PDC and WCDMA. The busi-
ness unit GSM had 19 000 employees in 1997.
A cellular communication system is divided into
three major systems: switching systems (SS), base
station systems (BSS) and operation support sys-
tems (OSS). The three subsystems consist of hard-
ware, such as the AXE switches and radio base
stations. They also consist of software, which can
be divided according to whether it is standard or
customer or dedicated, that is, speci?c to a parti-
cular customer. The standard software is devel-
oped by Ericsson and released to customers on an
annual basis. The customer dedicated software is
more service-oriented. An example can be the pre-
paid service for mobile phone use that Ericsson
delivered to Telia in 1998.
However, the technical solutions are inter-
twined, and in many cases they need to be coordi-
nated. One example is the connection between
wired communication systems and cellular
communication systems, the AXE switch is an
important component in both wired and cellular
systems. Another example is that the digital cel-
lular system standards—GSM, D-AMPS and
PDC—are built on the same platform, to some
extent requiring coordinated development. The
same applies to the coordination between indivi-
dual products (such as mobile phones) and the
systems (that is, the cellular system for mobile
phones).
Telia was comprised of eight business units and
four marketing units that are subordinate to the
business units. Some of the business units are, for
example, cellular communications, wired network
services and international operations, and the
market units are private customers, the public
sector, small to medium-size companies, and large
companies. Telia has the same need for coordi-
nation between the di?erent telecommunication
systems as described above for Ericsson. Telia has
a wired communication system and four di?erent
cellular communication systems (NMT 450, NMT
900, GSM 900 and GSM 1800) in Sweden.
The GSM unit in Ericsson is divided into mar-
ket units and technical units. The KAM unit
responsible for Telia Mobile is a sub-unit within
one of the market units. The technical units in the
GSM unit are value added services, implemen-
tation services, customer services, switching sys-
tems (SS), base transceiver systems (BTS) and base
station systems (BSS). Value added service is
responsible for the development and maintenance
of more end-customer oriented software, such as
the pre-paid system. The unit for implementation
services is involved when an operator builds a new
physical cellular system. Customer service is
responsible for customer education. SS and BSS
are the core systems, and thus also the backbone
of the cellular system. The SS and BSS systems are
organized within three organizational units SS,
BTS and BSS,
1
respectively. For example, the BSS
unit that we will use as an illustration in this case
is responsible for the development and main-
tenance of hardware and software for the base
station system. These technical units seldom have
any direct contact with the customers, although
they do have their own marketing specialists.
The KAM unit responsible for Telia Mobile is
organized in conjunction with the technical units.
The KAM unit has speci?c units or individuals
that interact on speci?c technical issues with the
technical units within Ericsson as well as with
units within Telia Mobile. The technical units
within Telia Mobile are organized in what is
known as a net development unit which consists of
two units, ‘Radio’ and ‘Switch’ concerned with
radio technology and switches, focused on the
fundamental functionality of a cellular system.
The radio and switch units are responsible for the
1
Base station systems is responsible for the development
and maintenance of base station control systems (BSC) and for
the integration of BTS and BSC to the BSS.
60 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
development, veri?cation and implementation of
hardware and software in their area. The oper-
ation unit is also divided up into a radio unit and a
switch unit. The units within Telia Mobile and
Ericsson responsible for the GSM standard are
designed in a similar way. Both companies have
divided the development and operation of cellular
systems into two parts: radio
2
and switch systems.
Time units
Ericsson delivers a new standard software
release every year for the GSM cellular system.
The time assigned for the development of a new
software release is about 3 years and the imple-
mentation of a release in Telia’s cellular system
takes more than a year. In this way, both compa-
nies are engaged simultaneously in several relea-
ses, each of which is handled by what is referred to
as a separate time unit. One consequence of
handling software development in releases is that
it reduces the interruptions in Telia’s cellular sys-
tems. Delivered releases are updated every week
when Ericsson distributes revisions. A new release
is worked on by di?erent groups, each being
assigned to what is known as a release project.
Every release can be seen as a product in itself, the
di?erent technical parts of which have to be coor-
dinated before they are implemented in a cellular
system. A release project involves the technical
units. At each point in time, all technical units
have at least four ongoing parallel release projects.
The project time and timely delivery are impor-
tant aspects for Telia Mobile. One reason for this
is that both Ericsson and Nokia are suppliers for
Telia Mobile’s cellular system. Consequently, it is
important to Telia Mobile that the two suppliers
develop the same features in their releases and that
they deliver them at almost the same time. If, for
example, Nokia develops a new feature that is not
included in Ericsson’s release, it will not be possi-
ble for Telia Mobile to o?er the new feature to its
customers. Telia Mobile needs to implement a new
feature throughout the cellular system so it has the
same functionality everywhere in Sweden.
If it is to be useful to Telia Mobile, the two
suppliers have to ensure their development is con-
ducted in parallel and at the same pace. Telia’s
strategy is to in?uence both suppliers in the early
stages of a new release project as well as during the
project. Telia Mobile’s development department
has continuous contact with the KAM unit in
Ericsson and with the corresponding unit in
Nokia. Constant updates on the progress of pro-
jects enables Telia to coordinate and in?uence the
work.
The implementation process for a standard
software release in Telia Mobile’s cellular system
lasts more than a year and involves several orga-
nizational units. Both Telia and Ericsson create a
speci?c project for the implementation. These
projects are organized almost identically and can
be divided into three major parts SS, BSS and
OSS. The two projects work closely together. Telia
Mobile runs a parallel implementation project
with the other cellular system supplier, Nokia
(Fig. 2)
Scholars writing about accounting and coop-
erative coordination have focused on isolated
individual relationships and have considered the
relationships to be a distinct coordination form
(Birnberg, 1998; Seal et al., 1999; Van der Meer
Kooistra & Vosselman, 2000). However, the
description of the relationship between Ericsson
and Telia reveals that numerous types of coordi-
nation form involving a number of sub-units
within both companies are used. Some units
within Ericsson have very close relationships with
speci?c units within Telia. At the same time, there
are some units within Ericsson that view other
units within Ericsson—as well as within Telia—
competitively.
Furthermore, some of these units are involved in
relationships with particular units at other produ-
cers, such as Nokia, or with other operators, such
as Mannesman. Sometimes the relationships are
cooperative and sometimes competitive. In this
way, the general picture of the coordination form
can be described not as a relationship, but as a
network of relationships between sub-units. Thus,
it seems essential to include the direct and indirect
e?ects of company decisions on the network level
identi?ed by Tomkins (2001).
2
Ericsson’s units BTS, BSC and BSS all work in the ?eld of
radio technology.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 61
With the intention of analyzing this governance
one step further, we will look more closely at how
these sub-units function, including how they are
related to the control system and how they are
coordinated with other units. Four units with key
roles are the KAM unit in Ericsson, the BSS unit
in Ericsson, the purchasing unit in Telia and the
net development unit within Telia.
Analyzing four important sub-units within the
relationship
The KAM unit
Each KAM unit has overall responsibility for a
speci?c customer. This means that all contact with
Telia Mobile should be handled by the KAM unit
or, at least, the KAM unit should be informed
when contact takes place. The unit is responsible
for marketing, selling and delivering all hardware,
software, support and services to Telia Mobile. In
addition, it supports Telia Mobile’s ongoing cel-
lular system on a continuous basis and it dis-
tributes software revisions on a weekly basis. The
relationship also involves collaboration on the
implementation of new software for the ongoing
cellular system. The KAM unit functions as Telia
Mobile’s ?rst contact in development activities
where the unit acts as an intermediary between
technical units within Ericsson and the develop-
ment’s units in Telia Mobile. This means that the
KAM unit coordinates Ericsson’s activities with
Telia Mobile.
Financially the KAM unit is accountable for the
consolidated market contribution of Telia Mobile
in the Scandinavian countries. This measure cap-
tures the unit’s output and input. The output pri-
ces are established through a negotiation process
with Telia Mobile. For some products and services
the KAM unit can have limits imposed on the
pricing by the technical units. However, no trans-
fer prices are deducted from the revenue. The
consolidated market contribution is a pro?t
measure. The control is oriented towards ?nancial
outcome, customer satisfaction and market shares.
As such, the unit is controlled by output perfor-
mance and not by the operations themselves. The
?nancial accountability focuses on the unit as a
whole and is an important tool in shaping the
KAM unit as a distinct hierarchy. The manager of
the KAM unit has a performance based salary
consisting of three parts: the GSM unit’s return on
investment, the KAM unit’s market contribution,
and ‘‘management by objectives’’ based incentive
scheme.
Fig. 2.
62 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
The KAM unit has a close relationship with
units in Telia Mobile: almost every employee
within the KAM unit interacts with individuals
within Telia Mobile. This means that more than
50 individuals within the KAM unit have daily or
at least weekly contact with their counterparts in
Telia Mobile. In general, an individual within the
KAM unit has more contact with individuals
within Telia Mobile than with other Ericsson
units. Indeed, several individuals emphasized that
they have more in common with their counterpart
than with other Ericsson units. An illustration of
this is the standard software implementation project
which is organized as two similar sub projects within
the two companies. However, the project as a whole
and its members are managed as one project. Telia
Mobile recruited their project manager from Erics-
son to make it easier to integrate the projects.
Another illustration of the close relationship is that
the KAM unit and Telia Mobile have developed an
intranet system together to share information to
which no other units within Ericsson have access.
The close relationship between the KAM unit
and Telia Mobile is also manifested within the
KAM unit’s internal performance evaluation: one
of the targets of the manager of the support unit
within the KAM unit was to encourage the oper-
ation unit within Telia Mobile to use the remote
loading feature. Thus, this Ericsson manager is
evaluated to some extent on what the customer
does, and some of his bonus is dependent on the
outcome. This is not exceptional within the KAM
unit. One of the important targets the manager of
the project management unit has was to implement
a release project on time. However, the implemen-
tation was made with Telia Mobile and it is the cus-
tomer who decides when the release is acceptable.
According to the manager of the KAM unit, the
close relationship with Telia Mobile is a pre-
requisite for obtaining information about Telia
Mobile’s needs. In his view, Telia Mobile is gen-
erally open and clear about its needs. There is no
problem identifying Telia Mobile’s opinion about
what is a reasonable combination of price and
functionality when an o?er is being made. Accord-
ing to him, this transparency is important because
without strong support and clear information from
Telia Mobile it is not possible to persuade the
technical units to develop something that is sui-
table for Telia Mobile. However, the KAM unit
has limited access to formal information systems
within Telia Mobile. Every month Telia Mobile
delivers its forecast for the hardware requirements
for the coming three months. Another formal source
of information comes from the customer satisfac-
tion surveys Ericsson collects from important indi-
viduals in Telia Mobile on a quarterly basis.
All managers interviewed within the KAM unit
have mentioned that the technical features of the
products are central for achieving a good perfor-
mance. However, all these aspects are controlled
by the technical units, not by the KAM unit. The
technical units are responsible for the design and
quality of the product. They also determine the
reference price for the product. Consequently, the
KAM unit is very dependent on the technical units
as well as on the units responsible for the releases.
After a release is implemented in Telia Mobile’s
cellular system the KAM unit takes over respon-
sibility for the release. A bad release will cause
continuous disruptions in the cellular system that
must be handled by the KAM unit.
3
Thus, there
are a number of areas where the KAM unit has no
authority, but which nevertheless have a strong
impact on the relationship with Telia Mobile.
However, the manager of the KAM unit accepts
this dependence and, according to him, it just
means that he needs to be successful in in?uencing
the technical units. Sometimes, the KAM unit and
Telia Mobile together develop a strategy to create
the greatest impact on the technical units. The
result can be that an individual in Telia Mobile is
identi?ed as being the person with the greatest
possibility of in?uencing the technical units within
Ericsson. In this way, units in the two companies
can be cooperating—networking—to create the
greatest in?uence or pressure possible.
BSS unit
Like the KAM units, the technical units
within Ericsson can be characterized as dis-
tinctive hierarchies. The technical units have
3
A release always has some outstanding problems. Erics-
son’s method for resolving this is to deliver weekly revisions.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 63
overall responsibility for a speci?c subsystem in
the GSM cellular system. The BSS unit handles
the base station system; it is responsible for the
development, supply and support of both the hard-
ware and software for it. The unit develops standard
software releases, hardware, and participates in
international standard-setting committees like the
European Telecommunications Standards Institute
(ETSI). Support of the ongoing cellular system is
another responsibility for the BSS unit. However, in
the ?rst instance, within Ericsson it is the KAM
units that handle support, but some problems need
specialists for a particular sub system. The technical
units deliver revisions on a weekly basis.
This means that the BSS unit coordinates Erics-
son’s activities concerning the sub system base
stations. The BSS unit is accountable for the con-
solidated return on sales for their part of the cel-
lular system. The consolidated return on sales
includes all revenue for the base station system
worldwide for both hardware and software. Thus,
the BSS unit is accountable for the revenue gener-
ated by the local companies worldwide. The out-
put prices are decided by the KAM units through
negotiation with the customers. However, the BSS
unit does provide price recommendations for their
products. The KAM units’ sales of the base sta-
tion system are consolidated as sales in the BSS
unit. Transfer prices are not calculated, instead the
sales are counted several times in di?erent organi-
zational units. Furthermore, the consolidated
return on sales also includes all costs connected to
the base station system. Thus, while the BSS unit
is accountable for hardware production costs, the
production plants are organized directly under the
managing director of cellular systems. The plants
produce base station system parts for the GSM,
D-AMPS and PDC cellular systems. Conse-
quently, the BSS unit is accountable for the pro-
duction costs generated by factories outside their
control. The manager of the BSS unit has a per-
formance based salary consisting of three parts
depending on: the GSM unit’s return on invest-
ment, the BSS unit’s return on sales, and a ‘‘man-
agement by objectives’’ based incentive scheme.
The BSS unit is also accountable for the time-
liness and quality of the product supply, main-
tenance and development. The unit focuses on
issues like the time to customer and the time to
market. During the construction of new cellular
sites, it is important that the hardware and soft-
ware products are delivered as expected. The
timeliness and quality are also fundamental to the
development of new software and hardware.
The unit does not have any direct access to sys-
tematic information from Telia Mobile, however,
it does gather some information from Telia and
other customers through personal contacts and
meetings. The BSS unit has meetings with impor-
tant customers in the early stages of a project for a
new release to obtain feedback. The unit could
also sometimes have special projects concerning a
speci?c issue run in conjunction with one or more
of its customers. The BSS unit obtains most cus-
tomer information from the KAM units.
The BSS unit is highly dependent on other
technical units, so there is a need for good coor-
dination between the BSS unit and these units. An
illustration of this is the development of features
in a release project: the BSS unit could focus on a
feature because it is likely to increase the pro?t-
ability of the unit. However, to sell this speci?c fea-
ture it is necessary that the other technical units also
develop the feature. This can be a problem when the
feature in question would be particularly pro?table
for the BSS unit, but not for the other units.
The BSS unit does not have any direct contact
with the customer without being granted permis-
sion by the key account unit. Thus, the BSS unit is
accountable for the return on sales, but the KAM
units e?ectively own the customer.
However, the BSS unit manager accepts this
dependence. In his view, ‘‘it is unavoidable in a
complex organization such as Ericsson.’’ The BSS
unit can control some of the aspects a?ecting both
revenues and costs. It can in?uence the price, time
to market, time to customer, product quality, and
some of the costs. The manager of the BSS unit
indicated that he is able to in?uence the other
units as long as he has good arguments to sub-
stantiate his requests.
The purchasing unit at Telia Mobile
As the purchasing unit is responsible for all pur-
chases made by Telia Mobile, it has an intensive
64 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
interaction with units within Ericsson as well as
Nokia. This means, for example, that several
individuals within the unit have contacts with
individuals within Ericsson on an almost daily
basis. Some individuals in the purchasing unit are
employed full time with the a?airs of either Erics-
son or Nokia. The purchasing unit is responsible
for the commercial part of the relationship with
Ericsson, but not the development and operative
activities.
The purchasing unit is accountable for its own
costs. Thus, the unit is treated as a discretionary
cost center. Control focuses on how the unit can
collaborate with other units within Telia Mobile,
and on the unit’s collaboration with important
suppliers. Furthermore, the control is oriented
towards the ?nancial outcome of Telia mobile as a
whole. As far as the technical units within Telia
Mobile are concerned, they are treated, ?nan-
cially, as discretionary cost centers. Thus, it is only
when considering Telia Mobile as a whole that one
is able to discern a distinct hierarchy.
According to several individuals within the pur-
chasing unit and the technical units, Telia
Mobile has a close relationship with Ericsson.
Some individuals even argued that it would
bene?cial to have a more business-like relation-
ship than they have, that is, that the existing
relationship is too familiar. However, some also
argued that Telia Mobile is just a small customer
so it needs the special relationship to be able to
in?uence the technical units within Ericsson.
According to the purchasing director it would not
be possible for Telia Mobile to end the relation-
ship with Ericsson. Instead, they have to survive
and develop together.
The net development unit within Telia Mobile
The net development unit is responsible for the
development of the underlying functionality of the
cellular system. It comments on release drafts
from Ericsson for several stages of the software
release and hardware development process. It also
handles the veri?cation and implementation of the
software release and hardware. The unit collabo-
rates in international standard setting committees
like the European Telecommunications Standards
Institute (ETSI). It also takes an active part in the
CME user group, which consists of European tel-
ecommunication operators. The net development
unit collaborates closely with the purchasing unit
and supports the unit in its commercial negotia-
tions with the KAM unit in Ericsson.
The net development unit is accountable for its
own costs. However, according to the develop-
ment manager, cost is not a big issue in the per-
formance evaluation as long as it is reasonably
well correlated to the output performance. In his
opinion, the important issues are to make progress
and deliver project results at the expected time.
Another important issue in the performance eval-
uation is the net development unit’s collaboration
with other units within Telia Mobile and with
units outside the company.
The development manager argued that the unit
plays a key role in Telia Mobile because the unit’s
decisions about new base functionality create rev-
enue opportunities, as well as potential cost sav-
ings. For example, the remote loading feature
makes it possible to reduce the consumption of
resources within the operation unit, and the high
speed feature is an essential feature in some of
Telia Mobil’s new end customer services. The
decision making body within the unit is comprised
of members from the net development unit, the
service development unit, the operation unit, the
sales unit, and the IT-unit. Thus, the unit focuses
on Telia Mobile’s overall performance.
Individuals within the net development unit
have an intensive interaction with their counter-
parts within Ericsson and Nokia. The net devel-
opment unit has several interfaces with the KAM
unit, and the technical units within Ericsson.
Thus, there are plenty of individuals within the net
development unit who maintain continuous con-
tact with individuals in the two di?erent units
within Ericsson. For example, the manager of the
R7 (i.e. the release number 7 for GSM) imple-
mentation project was in daily contact with the
project manager within Ericsson’s KAM unit. In
addition, he was in contact with other individuals
within the KAM unit, the BSS unit, and the SS
unit.
Individuals within the net development unit
interact with people within Ericsson in ?ve ways.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 65
Firstly, through the normal development routines
where some individuals within the net develop-
ment unit have technical meetings and frequent
personal contacts with their counterparts within
the KAM unit. Secondly, there is the standard
software implementation project in which the unit
is in contact with all corresponding units within
Ericsson. Thirdly, through special collaboration
projects with some technical unit, focusing on
issues such as GPRS. Fourthly, individuals from
the strategic product management unit within the
technical units have meetings with the net devel-
opment unit in the early stages of a new release.
Fifthly, the development manager participates in
Ericsson’s new technology meetings. These are
small exclusive meetings oriented towards clarify-
ing the speci?cations of the next generation of
cellular communication.
Control and forms of coordination
The more intensive study of the KAM unit, the
BSS unit, the purchasing unit and the net devel-
opment unit showed that the relationship coordi-
nation and its control seems to include some
features of all the classical coordination forms,
including hierarchy, market and cooperation.
Hierarchical coordination is important in several
of the relationships. The KAM unit and the BSS
unit within Ericsson are distinct hierarchies. They
both have a formal power structure and they are
rather autonomous where decision making is con-
cerned. The KAM unit and the BSS unit use
accounting methods such as responsibility
accounting, budgets, reward structures and pro?t
measures, all of which correspond to hierarchical
coordination (Chandler & Dames, 1979; Ezzamel,
1992). However, the KAM unit and the BSS unit
are both very dependent upon other units. The
control system within the two units included the
goals of their important counterparts (Berry,
1994). An example of this is that of the manager of
the support unit within the KAM unit who is
rewarded if Telia Mobile implements the remote
loading feature.
The purchasing unit and the net development
unit within Telia Mobile are treated as discre-
tionary cost centers. The two units function as a
part of the hierarchical coordination within the
distinct hierarchy of Telia Mobile. The purchasing
unit and the net development unit are ?nancially
evaluated on the basis of Telia Mobile’s overall
?nancial performance (Merchant, 1987). This
interest in the company as a whole can be exem-
pli?ed by the decision making forum within the
net development unit, which is comprised of
members from several sub-units within Telia
Mobile. In this way, the overall function of the
relationship between Ericsson and Telia is partly
characterized by the use of hierarchical coordina-
tion mechanisms (Chandler & Daems 1979; John-
son, 1983).
Market coordination is also a part of the total
coordination between Ericsson and Telia. It is
most pronounced within Ericsson between the
KAM unit responsible for Telia Mobile and other
KAM units responsible for other important cus-
tomers. The KAM units represent the needs of
di?erent customers and every KAM unit tries to
in?uence the technical units (BSS, SS and/or BTS)
within Ericsson to get them to develop features
that will meet their customer’s particular needs. As
such, the KAM units compete against each other
to get the attention of the technical units within
Ericsson. That is, the relationship as a whole, is
characterized by the existence of market coordi-
nation mechanisms (Ezzamel, 1992; Ouchi, 1980).
Cooperative coordination occurs within several
of the relationships between individual sub-units.
The KAM unit has close relationships with both
the purchasing unit and the net development unit
within Telia Mobile. They have an intensive col-
laboration within di?erent areas (Nicolini et al.,
2000). They collaborate in their development of
strategies and action plans for how they should
in?uence the technical units within Ericsson. The
KAM unit and the net development unit in Telia
Mobile have common projects (Seal et al., 1999).
Some participants within Telia Mobile even men-
tioned that the relationship was too close and was
too familiar (Tomkins, 2001). Thus the function of
the relationship between Ericsson and Telia as a
whole is characterized by an extensive use of
cooperative coordination mechanisms (Birnberg,
1998, Van der Meer Kooistra, & Vosselman,
2000).
66 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
Conclusions
This study gives a picture—a snapshot—of the
organization of a business relationship at a certain
point in time. The situation does not seem stable
and we have to assume that the relationship is con-
tinually developing. Furthermore, given the existing
characteristics, it seems reasonable to assume that
the relationship is under continuous development.
A ?rst conclusion is that established accounting
methods play a key role in forming the relation-
ship. Such a conclusion supports the analysis
made by Tomkins (2001) showing that existing
accounting techniques are still adequate in busi-
ness alliances and networks. However, one di?er-
ence is that neither collective accounting
information, e.g. inter-organizational budgets and
cost behavior analyses, or transparency through
open book accounting and target costing
approach is used in the relationship. Thus, the
decision makers are not supplied with more for-
malized accounting information about the rela-
tionship or the counterpart. Instead it is through a
systematic combining of accounting in the forma-
tion of organizational units with partly over-
lapping accountability that established accounting
methods support the relationship formation.
Ericsson is using it in relation to technical as well
as customer-de?ned units. One example is the unit
that is responsible for the relationship with Telia
for the GSM sales. This unit has an accounting
system that focuses its activities on one goal—a
goal that clearly includes the counterpart, and is
thus a goal that incorporates the counterpart’s
ambitions without requiring compatibility. In this
way, accounting within this sub-unit is used to
establish a link that bridges over the di?erences
without destroying the need for the two parties to
be unique. In its dealings with other parties, the
KAM unit does not always represent the same
interests: the unit represents Ericsson when it
interacts with Telia, but it clearly also represents
Telia Mobile when it interacts with other Ericsson
units. In this way, accounting has helped to create
new boundaries that do not follow the boundaries
given by ownership. In the same way, the technical
units have partly overlapping domains in their deal-
ings with key account units and where temporal’’
units are concerned. This overlapping accountability
provides the inducement to ?nd new technical solu-
tions as well as the opportunities to develop them.
The overlap forces the units to interact to ?nd com-
promises that do not have to be fully satisfactory for
either one of the parties, because of their temporary
nature. Thus, accounting has helped to establish a
structure that is not a solution to the coordination
problems, but rather, is a base facilitating a process
that can lead to continuous adaptation of the ‘solu-
tions’. This results in the splitting up or distribution
of problems, con?icts and contradictions that,
otherwise, would appear in one single interface
between the producer and the user to several inter-
faces. In this way it gives a depth to the interface that
makes it possible to tolerate a number of seemingly
incompatible di?erences. With all of these contra-
dictory requirements in technical and commercial
terms, there is a tremendous need for continuous
adaptation of a number of dimensions—always
knowing that the solutions are sub-optimal, but also
that they are only temporary.
A second closely related conclusion concerns the
information content within the main relationship
and the associated ones. Some of the information
content regards the exchange taking place, but to
a large extent it is also directed toward the future
in terms of coordinating the development of the
cellular system. This is a highly complex situation
where Ericsson has to handle the customer het-
erogeneity, create economies of scale, and shape a
cellular system that can communicate with the
competitors’ systems. The information ?ow is
clearly not about routine activities: rather, it
concerns unique issues—the creation of features
in new systems. The results from this study show
that neither Ericsson nor Telia has developed
formalized information systems that collect
the required information within each company
or across the company boundaries. The case also
gives reason to question whether such systems can
ever be designed. Instead, knowledge has to be
created jointly by Ericsson and the main custo-
mers in interplay with each other and with some
other key actors (Nonaka & Takeuchi, 1995). It
can be seen as a ‘‘bargain’’ between a number of
units representing technical and commercial fea-
tures, but it should also be seen as a ‘‘problem
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 67
solving process’’ where single units wrestle to com-
bine and confront technical and economical facets.
The two ?rst conclusions lead to a third more
general one: there seems to be good reason for
opening the ‘‘black box’’ of relationship coordi-
nation. The overall coordination in a relationship
like the one in the case study is based on a rather
complicated structure involving hierarchies, rela-
tionships between sub-units and even market
characteristics. The relationship exists in terms of
something ‘‘total’’, which is spoken about and for
which there are managers responsible. But this
totality is not uniform but includes all the three
classical coordination mechanisms. It involves a
mix of relationships whereby the hierarchically
governed sub-units have some very close relation-
ships with some sub-units, but more market like
relationships with others. This is clearly a form of
multi-dimensional coordination where issues rela-
ted to di?erent activities are dealt with in quite
di?erent ways. Thus, it is not a uniform coordi-
nation form. Individual units within each of the
two companies have quite distinct and clearly
de?ned goals. Thus, as a coordination form, hier-
archy is vital within the relationship. These units
have several types of relationships with other units
within their own company, as well as with units
within the counterpart. Some of these relation-
ships are clearly emphasized by the close coopera-
tion—having obvious relationship features—while
others are of a more competitive or hostile type.
Taking the example of a sub-unit within Ericsson,
one ?nds that close and market like relationships
exist with other units both within Ericsson and
within Telia. If we translate this ?nding into the
terms adopted by Richardson, the two ?rms are
performing a number of activities, which, in terms
of similarities and complementarities, are related
to each other in di?erent ways. As there is a var-
iation in these terms there must also be a variation
in the coordination of them. Finally, as these
activities are changing over time, there is also a
need for a more or less continuous change of
coordination forms.
The ?nal conclusion regards the interface
between di?erent coordination forms. To a very
great extent, research concerning governance has
focused on what is special about each mode of
governance, i.e. the di?erences between the three
forms of coordination. Although this is of interest
per se, there might also be very good reason to
study how the forms of coordination can be com-
bined. Such combinatory coordination is probably
closely connected to the existence of ‘‘depth’’ in
the coordination. For example, the two principal
companies in the case have constructed at least
two levels of coordination, which certainly increa-
ses the possibility of ?nding a number of combi-
nations. The existence of two or more ‘‘levels’’ of
coordination justi?es further investigations of
both the interface between hierarchy and relation-
ship and between the market and the relationship/
network. It might be that the largest potential for
development is in these interfaces.
The four conclusions above justify the need for
further research in a very speci?c direction. There
is a need for more intensive case studies of busi-
ness relationships where the role of di?erent orga-
nizational units and the relationships between
them are investigated in detail. The interaction
between companies seems to have a substance and
a variety that need to be conceptualized and
modeled in a much more elaborate way than has
been done to date.
Acknowledgements
This paper has bene?ted from the comments of
Trevor Hopper, participants of the 1999 workshop
on Business Networks and Accounting, Uppsala,
Sweden, two anonymous referees, and seminar
participants at University of Manchester and
Copenhagen Business School. This research has
been supported by grants from Swedish Council
for Research in the Humanities and Social Sci-
ences, Svenska Handelsbanken Foundation for
Social Science Research and Swedish Agency for
Innovation Systems.
68 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
Appendix
Interviews and informal discussions
Ericsson
Key Account Management unit responsible for Telia Mobile
Interviews
1. Mats Gustavsson (project manager, previous responsible for building Telia Mobile’s GSM 1800
cellular system in Stockholm)
2. Maria Agardh (project manager, responsible for the R7 implementation)
3 Anna Langeborg (project controller)
4. Marianne Hansson (business controller)
5. Ann-Christine Karlsson (project manager, responsible for training)
6. Panos Douzoulikopoulos (sales and marketing, responsible for base station systems)
7. Lars Meurling (product management, responsible for the base station systems)
8. Anna-Karin Hansson (product management, responsible for the switching systems)
9. Anders Karlsson (manager, system support)
10. Per Adolfsson (system support, responsible for a subsystem implementation of R7)
11. Sta?an Starin (strategic planning)
12. Mats Granryd (manager, sales and marketing)
13. Orvar Parling (manager, Key Account Telia)
14. Anders Nilsson (manager, support)
Informal discussions
15. Thomas Uhlander (manager, operations)
16. Monika Karlsson (Sales and marketing)
Business unit base station systems (BSS)
17. Aristidis Nikitidis (marketing unit, previous strategic product management unit and responsible
for R7)
18. Cecilia O
¨
berg (strategic product management unit, responsible for R8)
19. Anders Holmring (System design unit, responsible for R8)
20. John Montgommery (manager, strategic product management)
21. Christer Gustafsson (controller)
22. Mattias Karlberg (manager, system design)
23. Martin Bakhuizen (manager, sub unit within system design)
24. Stig-Rune Johansson (manager base station system, vice president GSM-unit)
Other units within Ericsson
25. Sta?an Holme´ r (manager European market, vice president GSM-unit)
26. Krister Jonsson (manager, sub unit within base transceiver stations)
27. Ulf Cige´ n (Senior accountant)
28. Sta?an Hermodsson (Accountant)
29. Peter Lindberg (Senior accountant)
30. Reimo La¨ hthenma¨ ki (Accountant, responsible for production accounting)
31. Anne-Charlotte Jalakas (Accountant)
32. Gary Ethole´ n (Financial analyst, production sta?)
33. Carl-Erik Johansson (AXE software Research)
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 69
Discussions
34. Bjo¨ rn Carlsson (European market, operations support)
35. Sture Jansson (Finance)
36. Mats A
?
rnell (Accountant, Customer service unit)
37. Karin Lundin (Customer service unit)
38. Lars Nylocks (CFO, GSM-unit)
39. Johan Uhlander (Senior accountant)
40. Osakar Ludwig (EDI solutions to suppliers)
41. Erland Lagerlo¨ f (Accountant)
42. Anders Bratt (Operations development, radio base plant)
43. Kaj Nielsen (Project manager, production team project, radio base plant)
44. Ann Hillin (Trade union, radio base plant)
Telia Mobile
Net development unit
Interviews
45. Mari Axell (Radio development unit, responsible for a subsystem implementation of R7)
46. Kim Gro¨ ndahl (Switching development unit, responsible for R7 implementation)
47. Ha? kan Wibergh (Radio development unit, responsible for the technical content of the radio
base system)
48. Lars Andeberg (manager, sub unit within radio development unit)
49. Lars Klasson (manager, development unit, member of Telia Mobile’s executive management
group)
50. Tommy Ljunggren (manager, radio development unit)
Purchasing unit
Interviews
51. Kristina Lundberg (manager, sub unit within purchasing, responsible for purchasing the releases)
52. Monika Ra¨ mme (manager, purchasing, member of Telia Mobile’s executive management group)
Operations unit
Interviews
53. Tord Larsson (net operations, responsible for a sub project within R7 implementation)
54. Lena Sundin (net support, responsible for coordination of release corrections)
55. Martin Blomberg (manager, net operations)
56. Stefan Fa¨ gerhall (manager, operations, member of Telia Mobile’s executive management group)
57. Anders Grahn (manager, sub unit net operations)
58. Lars Olsson (manager, process development)
Service development unit
Interviews
59. Tommy Ytterstro¨ m (manager, service development)
60. Peter Magnusson (manager, development future services)
Market unit
Interviews
61. Johan Freding (product manager, mobile business solutions, responsible for Dirigent)
62. Per Englund (manager, mobile business solutions)
70 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
References
Alchian, A. A., & Demsetz, H. (1972). Production, information
costs, and economic organization. The American Economic
Review, 62, 777–795.
Anderson, J. C., Ha? kansson, H., & Johanson, J. (1994). Dyadic
business relationships within a business network context.
Journal of Marketing, 58, 1–15.
Berry, A. J. (1994). Spanning traditional boundaries: organiza-
tion and control of embedded operations. Leadership &
Organization Development Journal, 15, 4–10.
Birnberg, J. G. (1998). Control in inter?rm co-operation rela-
tionships. Journal of Management Studies, 35, 421–428.
Bradach, J. L., & Eccles, R. G. (1989). Price, authority and
trust: from ideal types to plural forms. Annual Review of
Sociology, 15, 97–118.
Bromwich, M. (1990). The case for strategic management
accounting: the role of accounting information for strategy
in competitive markets. Accounting, Organizations and
Society, 15, 27–46.
Chandler, A. D. Jr., & Daems, H. (1979). Administrative co-
ordination, allocation and monitoring: a comparative analysis
of the emergence of accounting and organization in the USA
and Europe´ . Accounting, Organizations and Society, 4, 3–20.
Dubois, A. (1998). Organising industrial activities. London:
Routledge.
Ezzamel, M. (1992). Corporate governance and ?nancial con-
trol. In M. Ezzamel, & D. Heath?eld (Eds.), Perspectives on
?nancial control (pp. 3–26). London: Chapman & Hall.
Gietzmann, M. B. (1996). Incomplete contracts and the make
or buy decisions: governance design and attainable ?exibility.
Accounting, Organization and Society, 21, 611–626.
Ha? kansson, H. (1993). Networks as a mechanism to develop
resources. In P. Beije, J. Groenewegen, & O. Nuys (Eds.), Net-
working in the Dutch Industry (pp. 207–223). Apeldorn: Garant.
Ha? kansson, H., & Ford, D. (2002). How should companies
interact? Journal of Business Research, 55, 133–139.
Haugland, S., & Reve, T. (1994). Price, authority and trust in
international distribution channel relationships. Scandinavian
Journal of Management, 10, 225–244.
Hopwood, A. G. (1996). Looking across rather than up and
down: on the need to explore the lateral processing of infor-
mation. Accounting, Organization and Society, 21, 589–590.
Ittner, C. D., & Larcker, D. F. (1998). Innovations in perfor-
mance measurement: trends and research implications. Jour-
nal of Management Accounting Research, 10, 205–238.
Johnson, H. T. (1981). Toward a new understanding of nine-
teenth-century cost accounting. The Accounting Review, 56,
510–518.
Johnson, H. T. (1983). The search for gain in markets and ?rms: a
review of the historical emergence of management accounting
systems. Accounting, Organization and Society, 8, 139–146.
Jones, C. S. (1998). Hierarchies, networks and management
accounting in NHS hospitals. Accounting, Auditing &
Accountability Journal, 12, 164–187.
Merchant, K. A. (1987). How and why ?rms disregard the
controllability principle. In W. J. Bruns Jr., & R. S. Kaplan
(Eds.), Accounting and Management. Field Study Perspec-
tives (pp. 316–338). Boston: Harvard Business School
Press.
Mouritsen, J., Hansen, A., & Hansen, C. O. (2001). Inter-
organizational controls and organizational competencies:
episodes around target cost management/functional analysis
and open book accounting. Management Accounting
Research, 12, 221–224.
Nicolini, D., Tomkins, C., Holti, R., Oldman, A., & Smalley,
M. (2000). Can target costing and whole life costing be
applied in the construction industry?: evidence from two case
studies. British Journal of Management, 11, 303–324.
Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating
company. New York: Oxford University Press.
Otley, D., Broadbent, J., & Berry, A. (1995). Research in man-
agement control: an overview of its development. British
Journal of Management, 6, S31–S44.
Ouchi, W. G. (1977). The relationship between organizational
structure and organizational control. Administrative Science
Quarterly, 22, 95–113.
Ouchi, W. G. (1979). A conceptual framework for the design of
organizational control mechanisms. Management Science,
25, 833–848.
Accounting unit
Interviews
63. Anne Janson (business controller)
Other units within Telia
Interviews
64. Johan Stro¨ mba¨ ck (Account Manager, Telia Megacom)
65. Hans Andersson (Account Manager, Telia Megacom)
Discussions
66. Roger Larsson (Telia Research)
67. Thomas Stefansson (Telia Research)
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 71
Ouchi, W. G. (1980). Markets, bureaucracies, and clans.
Administrative Science Quarterly, 25, 129–141.
Richardson, G. B. (1972). The organization of industry. Eco-
nomic Journal, 82, 883–896.
Rind?eish, A., & Heide, J. B. (1997). Transaction cost analysis:
past present and future applications. Journal of Marketing,
61, 30–54.
Scapens, R. W. (1990). Researching management accounting
practice: the role of case study methods. British Accounting
Review, 22, 259–281.
Seal, W., Cullen, J., Dunlop, A., Berry, T., & Ahmed, M.
(1999). Enacting a European supply chain: a case study on
the role of management accounting. Management Accounting
Research, 10, 303–322.
Seal, W., & Vincent-Jones, P. (1997). Accounting and trust in
the enabling of long-term relations. Accounting, Auditing &
Accountability Journal, 10, 406–431.
Shields, M. D. (1997). Research in management accounting by
North Americans in the 1990s. Journal of Management
Accounting Research, 9, 3–61.
Tomkins, C. (2001). Interdependencies, trust and information
in relationships, alliances and networks. Accounting, Organi-
zation and Society, 26, 161–191.
Van der Meer-Kooistra, J., & Vosselman, E. G. J. (2000).
Management control of inter?rm transactional relationships:
the case of industrial renovation and maintenance. Account-
ing, Organization and Society, 25, 51–77.
Williamson, O. E. (1975). Markets and hierarchies: analysis and
antitrust implications. New York: Free Press.
Williamson, O. E. (1985). The economic institutions of capital-
ism. New York: Free press.
Yin, R. K. (1989). Case study research: design and methods.
Newbury Park: Sage Publications.
72 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
doc_679391150.pdf
Accounting and the classical coordination forms of hierarchy and market are closely related concepts. New forms of
coordination, including alliances and close cooperation between companies have opened up new challenges to how
accounting should be designed and practised. In the first part of this paper the theoretical basis for the connection
between coordination forms and the need for information is explored. Two key problems seem to be that the coordination
forms are not just alternatives, but can be used in combination, and that relationships between companies can
be viewed in different ways. These issues are studied through a case study of an extensive relationship between Ericsson
and Telia Mobile. The case demonstrates that all of the three basic coordination forms are used in creating a complex
network relating a set of sub-units from the two companies to each other and to other related parties. This means that
accounting methods developed earlier can be used in several ways to enhance the development of the relationships and
that the latter should not be perceived from one level or considered to be a uniform form. Companies can combine the
different forms on one level to achieve a coordination form that is unique at a higher level.
Accounting and network coordination
Ha? kan Ha? kansson
a
, Johnny Lind
b,
*
a
Norwegian School of Management, Department of Marketing, Box 580, N-1302 Sandvika, Norway
b
Stockholm School of Economics, Department of Accounting and Managerial ?nance,
Box 6501, 113 83 Stockholm, Sweden
Abstract
Accounting and the classical coordination forms of hierarchy and market are closely related concepts. New forms of
coordination, including alliances and close cooperation between companies have opened up new challenges to how
accounting should be designed and practised. In the ?rst part of this paper the theoretical basis for the connection
between coordination forms and the need for information is explored. Two key problems seem to be that the coordi-
nation forms are not just alternatives, but can be used in combination, and that relationships between companies can
be viewed in di?erent ways. These issues are studied through a case study of an extensive relationship between Ericsson
and Telia Mobile. The case demonstrates that all of the three basic coordination forms are used in creating a complex
network relating a set of sub-units from the two companies to each other and to other related parties. This means that
accounting methods developed earlier can be used in several ways to enhance the development of the relationships and
that the latter should not be perceived from one level or considered to be a uniform form. Companies can combine the
di?erent forms on one level to achieve a coordination form that is unique at a higher level.
#2003 Elsevier Ltd. All rights reserved.
Introduction
‘Do you want me to explain the organization of
Ericsson? That is impossible. Everything changes
all the time. Our operations are too complex. We
have important customers with di?erent and in
some cases con?icting claims on a cellular system.
We have to manage several technologies at the
same time. We have units everywhere in the world.
It is not an easy task to control this muddle. Some
months ago we put a lot of e?ort into creating
accounting information for a new organizational
unit. However, when we had ?nished, the top
management closed the unit. It is like going on
water logged marshland, the more e?ort you put
in, the more you sink. It is important that you can
?nd the right balance between all the di?erent
demands all the time.’ (Senior accountant)
The complexity confronting a large interna-
tional company has become more complicated
over time. Joint ventures, alliances, strategic part-
nerships, outsourcing, and decentralized organi-
zations change the ?rm’s boundaries continuously.
Management methods such as just in time (JIT),
time based management (TBM), lean production
and business process reengineering (BPR) aim at
eliminating all bu?ers within companies and
0361-3682/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0361-3682(02)00058-2
Accounting, Organizations and Society 29 (2004) 51–72
www.elsevier.com/locate/aos
* Corresponding author. Tel.: +46-8-736-93-17; fax: +46-
8-32-01-09.
E-mail addresses: [email protected] (H. Ha? kansson),
[email protected] (J. Lind).
thereby increase the companies’ dependence on
customers and suppliers. The products and services
consist of technologies that are in signi?cantly
di?erent stages of development. The customers
impose di?erent and sometimes con?icting
demands. The company has in this situation to
function within a network of relationships with
quite di?erent characteristics.
The telecom industry is a typical example of an
industry having the above features. It has under-
gone rapid change since the early 1980s. The
computer industry and telecom industry have
converged into a single industry. The customer
bases have changed from nationally monopolistic
operators to a more heterogeneous bunch of
operators ranging from small local players to large
multinational companies. Ericsson is one of the
largest telecom companies in the world having
extensive collaborative relationships with compa-
nies in di?erent areas. Ericsson has extensive rela-
tionships with some major customers as well as
suppliers but also regarding mobile phones with
Sony, operation support systems with Hewlett
Packard, and WCDMA technologies with Telia.
Ericsson uses relationship coordination to a
considerable extent. Such coordination is char-
acterized by cooperation and quasi integration.
Resources and activities are connected across
organizational boundaries to facilitate ?ows of
products, to increase accuracy and speed, and to
facilitate technical development. This cooperative
form is claimed to increase e?ciency but another
consequence is that the boundary of the single
?rm becomes blurred.
Accounting, information, e?ciency and coordi-
nation forms have long been closely related
(Chandler & Daems, 1979; Tomkins, 2001). Some
writers even claim that the development of
accounting methods has been a prerequisite for
hierarchical coordination (Chandler & Daems,
1979; Johnson, 1981, 1983). According to these
writers, accounting and hierarchical coordination
have been developed in symbiosis. The argument
is that hierarchical coordination has become a
substitute for market coordination because it is
more e?cient in some situations. The hierarchies
have grown larger and it has been possible to
control them through the development of new
sophisticated accounting methods, which give the
managers access to the information required to
control the internal activities. Accounting and the
classical market-hierarchy dichotomy are well
adjusted to each other.
Relationship coordination may cause a problem
from an accounting point of view as contemporary
accounting depends on de?ned, limited entities.
This new form of coordination blurs the clearly
de?ned boundaries which accounting presupposes
and requires. Thus, one can anticipate a problem
when combining the new form of coordination
with the way accounting is applied in companies.
As a consequence, companies such as Ericsson
should have a problem in the design and use of
accounting as it has to combine the use of rela-
tionship coordination with the contemporary
accounting methods. This article is devoted to an
analysis of this problem and how Ericsson has
resolve it.
The remainder of the paper is organized as fol-
lows. In section 2 the theoretical basis for the
connection between coordination forms and the
need for information is explored. Section 3 reviews
previous studies on accounting and the basic
coordination forms and identi?es key problems
within the cooperative coordination. Section 4
presents the research methods used. The ?fth sec-
tion presents the results of the case study. The
sixth section analyzes the case data in more detail
and relates the ?ndings to existing literature.
Conclusions and suggestions for future research
are presented in a ?nal section.
E?ciency, information needs and di?erent forms
of coordination
To achieve e?ciency, di?erent production
activities are coordinated within the industrial
structure (Dubois, 1998; Richardson, 1972). Three
theoretical forms—hierarchy, market and co-oper-
ation (relationship)—were identi?ed by Richard-
son (1972) and, since then, have been discussed and
compared in a large number of studies (e.g. Rind-
?eish & Heide, 1997; Williamson, 1975; 1985).
Hierarchy is a directed coordination form.
According to Richardson, it is useful when two
52 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
activities are closely complementary and similar,
i.e. when there are scale e?ects. In this case, being
closely complementary is de?ned as representing a
situation where di?erent phases of a production
process require some coordination. Similar activ-
ities are de?ned as those activities that require the
same capability for them to be undertaken (ibid).
In a hierarchy, two activities within a company are
directly coordinated. In principal, they become
one activity, and as far as the organization is con-
cerned, they are meshed together. Consequently,
they can be perfectly adapted to each other, but
they can only be jointly related to other activities.
The adaptation is based on information about the
two activities and their closely complementary
relationship. This includes information about the
resources needed to perform the activities. If some
of these resources are heterogeneous (Alchian &
Demsetz, 1972; Ha? kansson, 1993) it is possible to
develop and thereby learn about their interplay.
The existence of learning implies that information
is generated over time. To take advantage of this,
the company needs a continuous supply of very
detailed information concerning technical and
economic aspects of the activities performed and
the use of resources. We can also formulate this in
another way: the better, meaning the more com-
prehensive, the information regarding the activ-
ities and the use of resources, the better the
company can coordinate the two activities
advantageously.
Market is a coordination form where the inter-
face between the two coordinated activities is
standardized to ensure that several customers’
demands are matched with the supply of several
producers of similar products. Thus, there are
large numbers of standardized products creating
an aggregated output controlled by a demand–
supply balance. According to Richardson this is a
suitable coordination form when activities are
complementary and dissimilar. In such a situation,
the market will let each activity be undertaken in
accordance with its optimal scale. The activities
will be performed independently of each other in
any speci?c sense, but they will be highly depen-
dent on their common standardized product.
If we assume that there are some imperfections
in the market in regard to knowledge then a
company will need information regarding the pri-
ces of the standardized product from di?erent
suppliers. Furthermore, it needs information
about the ‘‘consuming’’ activities of other buyers.
The latter information is required as the company
must develop its own ‘‘consumption’’ in accor-
dance with the other buyers in order to demand
the same solution—to retain the standardization
of its products. However, the information needed
is not of the same detailed character at all as in the
hierarchical situation. But the loss of precision
required for the data is achieved at the price of a
very in?exible solution because of the need to
maintain the standardization. However, in
another respect, the market is opened up for new
solutions. If the activity performed by the supplier
or the customer can be combined with another
type of activity, there is no problem switching
counterpart. As long as there is no need for adap-
tation, the activities can be freely combined with
each other. One conclusion is that the information
problem does not a?ect the outcome in the same
way as in the hierarchical situation. Through the
standardization of the products (the solutions) the
information can also be standardized and thereby
become much easier to handle.
The business relationship or cooperation is the
third form of coordination. In this mode, the
coordination of the two activities cannot be left
entirely to the internal coordination of the ?rm or
the market forces. The reason for this is that the
activities are dissimilar and therefore, despite
being closely complementary, they need to be per-
formed by di?erent companies. Thus, the di?erent
companies’ plans must be matched, both qualita-
tively and quantitatively. The market is not sui-
table for this matching (Richardson, 1972). This
situation is much more complicated than the mar-
ket and hierarchy form of coordination, but the
basic feature of this coordination is interaction.
The two units try to ?nd a solution by matching
their plans. In this case, the coordination is not
centrally orchestrated, as in the hierarchical sit-
uation, or carried out by the total structure, as in
the market situation. Instead, we have two units
interactively seeking a suitable solution. In a rela-
tionship, the two companies involved match their
activities and resources to each other. The degree
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 53
to which this is possible is directly dependent on
the information available. As in the hierarchical
situation, each of the two parties must have an
extensive knowledge of their own activities and
their own use of resources, but in this case they
also need some information regarding their coun-
terpart. There must be a certain overlap in the two
parties’ knowledge to make a relationship based
on a common understanding possible. Thus, the
quality of the relationship is directly dependent on
the underlying information.
These three modes of coordination can be taken
to be alternatives, as Richardson and later
research (Bradach & Eccles, 1989; Haugland &
Reve, 1994; Williamson, 1975, 1985) regarding
relationships have done, but they are also clearly
interrelated and interdependent.
In summary, the need for information in the
hierarchical and the relationship modes is crucial
for the quality of the coordination of the activities.
In both of these cases, the degree to which e?-
ciency is created is highly dependent on the infor-
mation available to the decision-makers. This is
not the case for the market situation. But this is
not just a matter of coincidence—it is the very
reason for making the classical distinction between
market and hierarchy. The boundary of the
company separates two di?erent types of inter-
dependencies between activities. Internally, there
are activities that are closely complementary
and/or similar, while externally the activities are
complementary and dissimilar. Hierarchical and
market coordination are not just two alternatives,
they are perfectly complementary and well-mat-
ched from an information point of view. The
boundary of the company divides the world in
accordance with the two coordination modes, which
are also matched by the need for information. On
one side of the boundary is the hierarchy, with a
need for very detailed information. On the other
side is the market with a lesser need for in-depth
information.
When relationships need to be taken into con-
sideration, this clear image is shattered. In the
relationship case, there is neither a clear bound-
ary, nor do we have any matching internal coor-
dination form. Thus, relationships must be a
problem from an accounting point of view, given
that accountancy has been developed in accor-
dance with the hierarchy-market dichotomy.
There must be a need to change accounting when
used in a network situation. Let us ?rst see how
this issue has been handled in the recent develop-
ments within the accounting literature.
Accounting and coordination forms
Accounting and its development is interwined
with models of coordination (Ezzamel, 1992).
Ouchi (1977, 1979, 1980) developed a general fra-
mework for control and coordination forms.
Control and coordination forms are classi?ed
along the two dimensions ability to measure out-
put and knowledge of the transformation process.
Ouchi provides a framework comprised of three
coordination forms: hierarchy, market and clan.
Ezzamel (1992) draws on Ouchi’s classic work
(1979, 1980) when he summarizes the role of
accounting within the coordination forms of hier-
archy, market and clan.
Hierarchical coordination relies on rules and
formal power. Accounting is used to monitor and
in?uence people’s behaviour (Birnberg, 1998;
Ouchi, 1980). Accounting methods, such as mea-
suring direct cost, e.g., the direct cost of labour
and materials, standard costs, break-even charts,
allocation methods, ROI measures, responsibility
accounting, reward structures, transfer prices and
budgets have been developed to support decision
making and performance evaluation within orga-
nizations (Chandler & Daems, 1979; Ezzamel,
1992). Accounting used to focus mainly on the
?nancial aspect, but has gradually incorporated
other non-?nancial aspects (Ittner & Larcker,
1998). Accounting measurements are detailed and
enable di?erent objects such as organizational
units, products, projects and activities to be
described and analyzed in regard to past, present
and future states. Consequently, accounting is in a
good position to support managers by providing
information for decision-making and performance
evaluation in situations where the hierarchy-mar-
ket dichotomy exists (Bromwich, 1990).
Market coordination is dependent on market pri-
ces, and prices are the primary source of information
54 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
supporting decision makers in companies (Ouchi,
1980). However, accounting still has a role within
market coordination. According to Ezzamel
(1992), accounting is vital to ensure the normative
requirement of reciprocity in exchange relation-
ships. In his view this means collecting and sum-
marizing market prices, overseeing the
implementation of contracts, estimating market
opportunities and threats, and developing deci-
sion-making models based on market prices. As
such, it is possible to compare costs and bene?ts
for any given exchange.
Clan coordination relies on traditions and the
necessity of having common values and beliefs
(Ouchi, 1980). Ezzamel (1992) argues that the role
of accounting within clan co-ordination should
contribute to the establishment of common values
and beliefs, and sustain the organizational tradi-
tion. However, occasionally it is necessary to
reconstruct the organizational tradition. Many of
the accounting methods developed and used in
hierarchical coordination will also be utilized in
the clan setting. However, Ezzamel (1992) identi-
?ed three major di?erences: First, the essence of
control is more mutual/consensual. Second, much
e?ort is expended on establishing and continuing a
?nancial tradition but it will be subordinate to
common values and traditions. Third, the focus of
?nancial control is more on the socialization process
than on supporting managers with information for
decision making and performance evaluation.
In recent years, there are several examples of
research in accounting where the need to explore
the e?ects of accounting across organizational
boundaries has been stressed (Hopwood, 1996;
Otley et al., 1995; Shields, 1997). More speci?c
attempts to discuss the role of accounting within
cooperative coordination have been made by
Berry (1994), Gietzman (1996), Seal and Vincent-
Jones (1997), Birnberg (1998), Jones (1998), Seal
et al. (1999), Nicolini et al. (2000), Van der Meer-
Kooistra and Vosselman (2000), Mouritsen et al.
(2001) and Tomkins (2001).
Berry (1994) argues that the clan identi?ed by
Ouchi (1979, 1980) can be related to cooperative
coordination. He exempli?es this by an alliance
between an American and a Japanese company in
the automotive component industry. In his view,
the case shows that the alliance used clan-based
control in which the companies had open discus-
sions with each other, communicated their pro-
blems, and shared their knowledge and
information. One observation was that the two
companies in that study viewed the experience of
the collaboration di?erently. The American com-
pany was dissatis?ed because they felt that they
had not learnt anything. In contrast, the Japanese
company had a more positive experience. The
conclusion Berry (1994) came to was that it can be
necessary to create trans-organizational work-
groups for cooperative coordination. These work-
groups then use clan control as the mode of
control. Berry (1994) argued that clan control is a
suitable mode of control for partnerships and alli-
ances. Such a conclusion seems reasonable when
analyzing just one relationship.
Birnberg (1998) develops a framework for con-
trol in inter?rm cooperative relationships.
According to him (ibid, p. 421) a co-operative
relationship can be described in ?ve dimensions:
‘(1) the degree of absolute and relative commit-
ment; (2) the symmetry of the rewards; (3) the
extent of uncertainty present; (4) the degree of
mutual trust between the parties; and (5) the
length of the relationship’. Four types of inter?rm
cooperative relationships were identi?ed, ranging
from relationships with both low interdependency
and low uncertainty to relationships with both
high interdependency and high uncertainty. Birn-
berg’s analysis focused on single relationships.
Two empirical studies made by Seal et al. (1999)
and Van der Meer-Kooistra and Vosselman (2000)
were conducted in an attempt to try to understand
the role of accounting within the supply chain
from the perspective of transaction cost theory.
Van der Meer-Kooistra and Vosselman (2000)
identi?ed three types of control pattern within
inter?rm relationships: a market based pattern, a
bureaucracy based pattern and a trust based pat-
tern. Seal et al. (1999) indicated that the account-
ing can play a central role in the establishment of
collaborative inter-organizational business rela-
tionships. The empirical studies have concentrated
upon newly established strategic alliances. Thus,
the focus of this research is on a dyadic relation,
which is studied in isolation.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 55
Tomkins (2001) adopted a similar point of
departure to other accounting scholars writing
about cooperative coordination. His main focus
was on the information needed within single busi-
ness relationships. However, the consequences
were also discussed at a network level. In that
way, both the direct and indirect e?ects of the
decisions made by the company were captured
because the decisions not only a?ect the partner
directly, but also the partner’s relationships.
In combination with the earlier discussion
points, the above studies highlight two unresolved
matters. One has to do with how relationships are
conceptualized: if a relationship is seen as an iso-
lated cooperative island, then there is a boundary
around it. This seems to be the case for most of
the studies reviewed. However, if the relationship
is considered to be one element in a network, such
a boundary might create problems. In that case,
an e?cient mode of coordination would require
simultaneous connections to several business rela-
tionships (Anderson et al., 1994). This will put
even greater demands on the supply of infor-
mation. In order to enhance adaptations of activ-
ities and resources across the boundaries of
companies, better knowledge is required about
what is going on in each of the counterparts that
will have an impact on the others. It is not enough
to know about the activities or the resources of
just the counterpart, instead, detailed information
about speci?c activities, i.e. the costs, revenues,
quality, cycle time, reliability or delivery, and
information concerning the use of resources will
be needed for some of the connected parties to
enable ways of improving the operations.
The second problem has to do with the inter-
faces between the three pure forms of coordi-
nation. The relationships a company is involved in
consist of both close and ‘‘arm’s length’’ relation-
ships (Tomkins, 2001). This means that a com-
pany always has a mix of relationships, including
some close-knit associations with a high level of
trust and some more market-like transactions.
Earlier we concluded that the hierarchy and the
market are closely complementary. It is also pos-
sible to ?nd connections between relationships and
these two basic coordination modes. In order for
anyone to have a close and strong relationship it is
important ‘to be someone’. Thus, in the relation-
ship case, there integrating mechanisms are needed
in relation to the individual company. The stron-
ger the relationship, the more important this inte-
gration will have to become (Ha? kansson and
Ford, 2002). Consequently, there seems to be a
positive connection between hierarchy and rela-
tionships. In the same way, in the relationship
case, alternatives are required. It must be possible
for either of the parties to break a relationship, so
the existence of alternative parties is necessary.
Thus, there must be some kind of market ingre-
dients in the relationship case too. If there are no
alternatives whatsoever, the situation will be simi-
lar to vertical integration, i.e. to a hierarchy.
Considered in this way, there are reasons to
assume that the relationship form and its e?ects
might be dependent on the existence of hierarchies
as well as markets. The three forms are not just
alternatives, but are complementary to each other.
In order to study how the use of accounting
methods is further in?uenced by these two issues,
an empirical study was conducted.
Research method
As we have seen, the role of accounting within
cooperative coordination is more complicated
than in hierarchical and market coordination.
Until now the accounting literature has paid little
attention to cooperative coordination, especially
when the company is considered to be intertwined
in a complex network of relationships. An
exploratory case study seems to be a reasonable
research design because of the lack of previous
studies within the area (Yin, 1989). Thus, we have
chosen to study a well developed business rela-
tionship in the telecom industry between Ericsson
and Telia. This is a long lasting and well known
business relationship within the Swedish business
community. The empirical study was partly inten-
ded o identify the relationship and partly to iden-
tify the role of accounting within the relationship.
The point of departure of the relationship part
of the study was the development of standard
software releases within the GSM-area. Thus, the
relationship is primarily captured through the
56 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
standard software release projects and the services
developed in conjunction with the releases. Hence,
four di?erent release projects (R6, R7, R8, R9),
and one end service (Dirigent) have been studied
to identify the interfaces between Ericsson and
Telia. However, to obtain a more comprehensive
understanding of the relationship, some discus-
sions, interviews, and ?eld visits were conducted
with employees from other units, such as those
involved in the basic AXE-research, the plant
producing radio base stations, and key accounts
within Telia. In this way, the study identi?es the
interfaces between units during the entire life cycle
of a service, from the early stages within Ericsson
to the late stages within Telia.
At the time the study was conducted, the GSM
cellular system was the largest business unit within
Ericsson and one of Telia’s prioritized areas.
According to both the purchasing manager in
Telia Mobile and one of the vice presidents in the
GSM business unit, the software development
represented the most important aspect of the
relationship.
The accounting part of the study used the
investigation of the relationship to identify critical
organizational units within the relationship. Four
units were identi?ed, the KAM unit and BSS-unit
within Ericsson and the purchasing and net
development unit in Telia Mobile. The because of
the units chosen, the study focused on radio tech-
nology in both Ericsson and Telia. Accountants,
managers and project members were interviewed
in the four units. The interviews were aimed to
examine the design and use of accounting and the
information ?ow. As such, the intention was to
identify the role of accounting within a unit as well
as outside the unit, and to identify the information
?ow between units within a company as well as
between units in di?erent companies. However, to
expand the understanding of the role of account-
ing, some discussions and interviews were con-
ducted with members from other business units in
Ericsson and in the operation and the market unit
within Telia Mobile.
The empirical study started in December 1997
and the majority of the data collection was done
during 1998 and early 1999. We received the full
support of both companies even to the extent of
having two o?ces assigned to us in Ericsson, one
within the central accounting unit and the other in
the Key Accounts Telia unit. This arrangement
gave us access to Ericsson’s intranet. In addition,
access was given to economic reports, internal let-
ters, detailed statements, organizational charts,
documents relating to strategy and technical and
commercial reports. In Telia, in addition to the
interviews, we were given access to written docu-
ments including organizational charts. As such,
the study tried to use as many sources as possible
to support the empirical data and thereby trian-
gulate the results (Scapens, 1990).
The major sources of information were inter-
views and discussions, which were conducted with
67 employees within the two companies. The
majority of the interviews were with people within
Key Accounts Telia and the BSS unit within
Ericsson, and the purchasing, the development
and the operation units within Telia Mobile.
Those interviewed represented almost all organi-
zational levels from the vice president to develop-
ment engineers and project members. Almost all
functions were covered, from strategic product
management, system development, operations and
marketing, to accounting and top management.
The interviews were semi-structured and lasted
between 1 and 3 h, with an average duration of 1.5
h. A more detailed description of those inter-
viewed is described in the Appendix.
The Ericsson–Telia business relationship
Ericsson is one of the largest companies in
Sweden. It was founded 1876. Telia is the former
public telecommunications operator; it was foun-
ded a few years after Ericsson. Telia had a mono-
poly from 1918 until the early 1980s. Ericsson
grew extensively during the 1990s while Telia
increased its sales volume and reduced its number
of employees at the same time.
The relationship—the global alliance
The relationship between Ericsson and Telia is
important to both companies. Its importance is
manifested by a partnership agreement between
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 57
the companies. The relationship is considered to
be so important that the CEOs in Ericsson and
Telia meet regularly. Ericsson’s sales in Sweden
during 1997 were about 5.5% of its total turnover,
or more than 1 billion US $. Telia is Ericsson’s
largest customer in Sweden. Furthermore Ericsson
is Telia’s largest supplier measured in terms of
purchasing cost. The relationship is global in the
sense that the cooperation is worldwide. For
example, Ericsson delivered new cellular systems
to Telia in Brazil as well as in Denmark. Both
companies have adapted their organization to
each other. Ericsson has special units primarily
engaged in managing the relationship with Telia
and it has a Global Account Manager (GAM)
who is responsible for the relationship as a whole.
Telia has individuals and sub-units dedicated to
work and collaborations with actors within
Ericsson.
The commercial exchange between the compa-
nies consists mainly of hardware and software for
wired and cellular communication systems. Mobile
telephones are not a part of the commercial
exchange, although their existence is important to
both companies as they in?uence the end custo-
mers’ ability to use new services. As a con-
sequence, Telia Mobile has a speci?c unit which is
dedicated to mobile phones and which interacts
with the major mobile phone manufacturers. Over
the years, Telia and Ericsson have cooperated on
the development of new communication systems.
For example, they developed the automated AXE
switch, which is the backbone of Ericsson tele-
communication solutions, together.
There are several di?erent systems involved in
the cellular sector. NMT is the old analogue stan-
dard for the Nordic cellular system ?rst intro-
duced in 1981. Today the NMT-related work in
both Ericsson and Telia is only focused on main-
tenance activities. GSM is the European digital
cellular standard launched in 1991 and it is the
main contemporary cellular system in Sweden.
Telia Mobile and the GSM business unit in Erics-
son are the two counterparts handling the GSM
cellular system. D-AMPS is the American digital
cellular standard, it is mainly handled by Telia
International and the D-AMPS business unit in
Ericsson. WCDMA is the next generation of cellular
systems handled primarily by Telia Research, a
sta? unit. At Ericsson, a newly created business
unit handles the WCDMA system with technology
and research sta? support. The president for the
GSM-standard cellular system and Telia Mobile
meet frequently to discuss problems and potential
opportunities in the relationship.
The interaction between the two companies is
strongly in?uenced by their organizational struc-
tures. The three most important dimensions in
which the organizations have been structured are
technology, relationships and time. Let us start
with the units responsible for the relationships.
Relationship units
The relationship covers both wired and cellular
systems. Formerly the wired system was the cen-
tral part of the relationship, but today the cellular
system is more important. Ericsson has assigned
special organizational units and individuals to be
accountable for the customer relationship. Erics-
son has three di?erent levels for relationship
accountability: Global Account Executive (GAE),
Global Account Manager (GAM) and Key
Account Manager (KAM). One GAE and one
GAM are accountable for the overall corporate
relationship with Telia. They have to develop and
secure the business and coordinate Ericsson’s
activities with Telia. There are several KAMs, one
of which is accountable for the relationship with
Telia Mobile. The KAM for Telia Mobile is an
organizational unit comprised of more than 60
employees.
Ericsson has other GAEs, GAMs and KAMs
accountable for relationships with other important
customers such as Global One, Mannesmann and
Sonera. All of these strive to obtain a closer rela-
tionship between each customer and Ericsson.
Subsequently, each particular KAM’s aim must be
to adapt Ericsson’s technical units to its own cus-
tomer. One obvious problem is that the customers
have di?erent demands. For example, Telia cer-
tainly imposes other demands on a cellular system
than a new operator in China would do. As a
consequence, the technical units within Ericsson
will be confronted with contradictory speci?-
cations. According to the vice president responsible
58 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
for the European market within Ericsson, the di?er-
ence between customer’s needs has increased during
recent years as a consequence of the increased num-
ber of features in the software. A consequence of
these di?erences is that the other KAM units within
Ericsson will view the KAM unit handling Telia
Mobile to an increasing degree as a competitor.
The business units within Telia are organized in
departments on a functional basis, such as pur-
chasing, development, and operations. Telia does
not have a speci?c unit responsible for the rela-
tionship with Ericsson. However, within the func-
tional units in Telia Mobile, speci?c individuals
and sub-units work exclusively with particular sup-
pliers. The purchasing department has dedicated
one speci?c purchaser to Ericsson and another to
Nokia. The sub-units within the development
department at Telia Mobile are divided into two
groups, one of them interacting with Ericsson and
the other with Nokia. Telia Mobile’s cellular sys-
tem in Sweden is comprised of parts from both
Ericsson and Nokia, but with Ericsson being the
dominant supplier. Consequently, Telia Mobile
has had to coordinate the two supplier relation-
ships and cannot entirely adapt its operations to
just one of them. One such example is that Telia
Mobile must coordinate Ericsson’s and Nokia’s
new GSM standard software releases. However
Telia Mobile has adapted its organization to some
critical supplier relationships. Fig. 1) shows the
Fig. 1.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 59
relationship units within Ericsson and Telia, and
other related units and companies.
Technical units
Ericsson and Telia have divided their organi-
zations in accordance with the technical solutions
and products with which they are concerned. On
an aggregate level, Ericsson has three business
units: systems for wired communication, systems
for cellular communication, and mobile phones.
The business unit for cellular communication is
divided into ?ve cellular system standards: NMT,
GSM, D-AMPS, PDC and WCDMA. The busi-
ness unit GSM had 19 000 employees in 1997.
A cellular communication system is divided into
three major systems: switching systems (SS), base
station systems (BSS) and operation support sys-
tems (OSS). The three subsystems consist of hard-
ware, such as the AXE switches and radio base
stations. They also consist of software, which can
be divided according to whether it is standard or
customer or dedicated, that is, speci?c to a parti-
cular customer. The standard software is devel-
oped by Ericsson and released to customers on an
annual basis. The customer dedicated software is
more service-oriented. An example can be the pre-
paid service for mobile phone use that Ericsson
delivered to Telia in 1998.
However, the technical solutions are inter-
twined, and in many cases they need to be coordi-
nated. One example is the connection between
wired communication systems and cellular
communication systems, the AXE switch is an
important component in both wired and cellular
systems. Another example is that the digital cel-
lular system standards—GSM, D-AMPS and
PDC—are built on the same platform, to some
extent requiring coordinated development. The
same applies to the coordination between indivi-
dual products (such as mobile phones) and the
systems (that is, the cellular system for mobile
phones).
Telia was comprised of eight business units and
four marketing units that are subordinate to the
business units. Some of the business units are, for
example, cellular communications, wired network
services and international operations, and the
market units are private customers, the public
sector, small to medium-size companies, and large
companies. Telia has the same need for coordi-
nation between the di?erent telecommunication
systems as described above for Ericsson. Telia has
a wired communication system and four di?erent
cellular communication systems (NMT 450, NMT
900, GSM 900 and GSM 1800) in Sweden.
The GSM unit in Ericsson is divided into mar-
ket units and technical units. The KAM unit
responsible for Telia Mobile is a sub-unit within
one of the market units. The technical units in the
GSM unit are value added services, implemen-
tation services, customer services, switching sys-
tems (SS), base transceiver systems (BTS) and base
station systems (BSS). Value added service is
responsible for the development and maintenance
of more end-customer oriented software, such as
the pre-paid system. The unit for implementation
services is involved when an operator builds a new
physical cellular system. Customer service is
responsible for customer education. SS and BSS
are the core systems, and thus also the backbone
of the cellular system. The SS and BSS systems are
organized within three organizational units SS,
BTS and BSS,
1
respectively. For example, the BSS
unit that we will use as an illustration in this case
is responsible for the development and main-
tenance of hardware and software for the base
station system. These technical units seldom have
any direct contact with the customers, although
they do have their own marketing specialists.
The KAM unit responsible for Telia Mobile is
organized in conjunction with the technical units.
The KAM unit has speci?c units or individuals
that interact on speci?c technical issues with the
technical units within Ericsson as well as with
units within Telia Mobile. The technical units
within Telia Mobile are organized in what is
known as a net development unit which consists of
two units, ‘Radio’ and ‘Switch’ concerned with
radio technology and switches, focused on the
fundamental functionality of a cellular system.
The radio and switch units are responsible for the
1
Base station systems is responsible for the development
and maintenance of base station control systems (BSC) and for
the integration of BTS and BSC to the BSS.
60 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
development, veri?cation and implementation of
hardware and software in their area. The oper-
ation unit is also divided up into a radio unit and a
switch unit. The units within Telia Mobile and
Ericsson responsible for the GSM standard are
designed in a similar way. Both companies have
divided the development and operation of cellular
systems into two parts: radio
2
and switch systems.
Time units
Ericsson delivers a new standard software
release every year for the GSM cellular system.
The time assigned for the development of a new
software release is about 3 years and the imple-
mentation of a release in Telia’s cellular system
takes more than a year. In this way, both compa-
nies are engaged simultaneously in several relea-
ses, each of which is handled by what is referred to
as a separate time unit. One consequence of
handling software development in releases is that
it reduces the interruptions in Telia’s cellular sys-
tems. Delivered releases are updated every week
when Ericsson distributes revisions. A new release
is worked on by di?erent groups, each being
assigned to what is known as a release project.
Every release can be seen as a product in itself, the
di?erent technical parts of which have to be coor-
dinated before they are implemented in a cellular
system. A release project involves the technical
units. At each point in time, all technical units
have at least four ongoing parallel release projects.
The project time and timely delivery are impor-
tant aspects for Telia Mobile. One reason for this
is that both Ericsson and Nokia are suppliers for
Telia Mobile’s cellular system. Consequently, it is
important to Telia Mobile that the two suppliers
develop the same features in their releases and that
they deliver them at almost the same time. If, for
example, Nokia develops a new feature that is not
included in Ericsson’s release, it will not be possi-
ble for Telia Mobile to o?er the new feature to its
customers. Telia Mobile needs to implement a new
feature throughout the cellular system so it has the
same functionality everywhere in Sweden.
If it is to be useful to Telia Mobile, the two
suppliers have to ensure their development is con-
ducted in parallel and at the same pace. Telia’s
strategy is to in?uence both suppliers in the early
stages of a new release project as well as during the
project. Telia Mobile’s development department
has continuous contact with the KAM unit in
Ericsson and with the corresponding unit in
Nokia. Constant updates on the progress of pro-
jects enables Telia to coordinate and in?uence the
work.
The implementation process for a standard
software release in Telia Mobile’s cellular system
lasts more than a year and involves several orga-
nizational units. Both Telia and Ericsson create a
speci?c project for the implementation. These
projects are organized almost identically and can
be divided into three major parts SS, BSS and
OSS. The two projects work closely together. Telia
Mobile runs a parallel implementation project
with the other cellular system supplier, Nokia
(Fig. 2)
Scholars writing about accounting and coop-
erative coordination have focused on isolated
individual relationships and have considered the
relationships to be a distinct coordination form
(Birnberg, 1998; Seal et al., 1999; Van der Meer
Kooistra & Vosselman, 2000). However, the
description of the relationship between Ericsson
and Telia reveals that numerous types of coordi-
nation form involving a number of sub-units
within both companies are used. Some units
within Ericsson have very close relationships with
speci?c units within Telia. At the same time, there
are some units within Ericsson that view other
units within Ericsson—as well as within Telia—
competitively.
Furthermore, some of these units are involved in
relationships with particular units at other produ-
cers, such as Nokia, or with other operators, such
as Mannesman. Sometimes the relationships are
cooperative and sometimes competitive. In this
way, the general picture of the coordination form
can be described not as a relationship, but as a
network of relationships between sub-units. Thus,
it seems essential to include the direct and indirect
e?ects of company decisions on the network level
identi?ed by Tomkins (2001).
2
Ericsson’s units BTS, BSC and BSS all work in the ?eld of
radio technology.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 61
With the intention of analyzing this governance
one step further, we will look more closely at how
these sub-units function, including how they are
related to the control system and how they are
coordinated with other units. Four units with key
roles are the KAM unit in Ericsson, the BSS unit
in Ericsson, the purchasing unit in Telia and the
net development unit within Telia.
Analyzing four important sub-units within the
relationship
The KAM unit
Each KAM unit has overall responsibility for a
speci?c customer. This means that all contact with
Telia Mobile should be handled by the KAM unit
or, at least, the KAM unit should be informed
when contact takes place. The unit is responsible
for marketing, selling and delivering all hardware,
software, support and services to Telia Mobile. In
addition, it supports Telia Mobile’s ongoing cel-
lular system on a continuous basis and it dis-
tributes software revisions on a weekly basis. The
relationship also involves collaboration on the
implementation of new software for the ongoing
cellular system. The KAM unit functions as Telia
Mobile’s ?rst contact in development activities
where the unit acts as an intermediary between
technical units within Ericsson and the develop-
ment’s units in Telia Mobile. This means that the
KAM unit coordinates Ericsson’s activities with
Telia Mobile.
Financially the KAM unit is accountable for the
consolidated market contribution of Telia Mobile
in the Scandinavian countries. This measure cap-
tures the unit’s output and input. The output pri-
ces are established through a negotiation process
with Telia Mobile. For some products and services
the KAM unit can have limits imposed on the
pricing by the technical units. However, no trans-
fer prices are deducted from the revenue. The
consolidated market contribution is a pro?t
measure. The control is oriented towards ?nancial
outcome, customer satisfaction and market shares.
As such, the unit is controlled by output perfor-
mance and not by the operations themselves. The
?nancial accountability focuses on the unit as a
whole and is an important tool in shaping the
KAM unit as a distinct hierarchy. The manager of
the KAM unit has a performance based salary
consisting of three parts: the GSM unit’s return on
investment, the KAM unit’s market contribution,
and ‘‘management by objectives’’ based incentive
scheme.
Fig. 2.
62 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
The KAM unit has a close relationship with
units in Telia Mobile: almost every employee
within the KAM unit interacts with individuals
within Telia Mobile. This means that more than
50 individuals within the KAM unit have daily or
at least weekly contact with their counterparts in
Telia Mobile. In general, an individual within the
KAM unit has more contact with individuals
within Telia Mobile than with other Ericsson
units. Indeed, several individuals emphasized that
they have more in common with their counterpart
than with other Ericsson units. An illustration of
this is the standard software implementation project
which is organized as two similar sub projects within
the two companies. However, the project as a whole
and its members are managed as one project. Telia
Mobile recruited their project manager from Erics-
son to make it easier to integrate the projects.
Another illustration of the close relationship is that
the KAM unit and Telia Mobile have developed an
intranet system together to share information to
which no other units within Ericsson have access.
The close relationship between the KAM unit
and Telia Mobile is also manifested within the
KAM unit’s internal performance evaluation: one
of the targets of the manager of the support unit
within the KAM unit was to encourage the oper-
ation unit within Telia Mobile to use the remote
loading feature. Thus, this Ericsson manager is
evaluated to some extent on what the customer
does, and some of his bonus is dependent on the
outcome. This is not exceptional within the KAM
unit. One of the important targets the manager of
the project management unit has was to implement
a release project on time. However, the implemen-
tation was made with Telia Mobile and it is the cus-
tomer who decides when the release is acceptable.
According to the manager of the KAM unit, the
close relationship with Telia Mobile is a pre-
requisite for obtaining information about Telia
Mobile’s needs. In his view, Telia Mobile is gen-
erally open and clear about its needs. There is no
problem identifying Telia Mobile’s opinion about
what is a reasonable combination of price and
functionality when an o?er is being made. Accord-
ing to him, this transparency is important because
without strong support and clear information from
Telia Mobile it is not possible to persuade the
technical units to develop something that is sui-
table for Telia Mobile. However, the KAM unit
has limited access to formal information systems
within Telia Mobile. Every month Telia Mobile
delivers its forecast for the hardware requirements
for the coming three months. Another formal source
of information comes from the customer satisfac-
tion surveys Ericsson collects from important indi-
viduals in Telia Mobile on a quarterly basis.
All managers interviewed within the KAM unit
have mentioned that the technical features of the
products are central for achieving a good perfor-
mance. However, all these aspects are controlled
by the technical units, not by the KAM unit. The
technical units are responsible for the design and
quality of the product. They also determine the
reference price for the product. Consequently, the
KAM unit is very dependent on the technical units
as well as on the units responsible for the releases.
After a release is implemented in Telia Mobile’s
cellular system the KAM unit takes over respon-
sibility for the release. A bad release will cause
continuous disruptions in the cellular system that
must be handled by the KAM unit.
3
Thus, there
are a number of areas where the KAM unit has no
authority, but which nevertheless have a strong
impact on the relationship with Telia Mobile.
However, the manager of the KAM unit accepts
this dependence and, according to him, it just
means that he needs to be successful in in?uencing
the technical units. Sometimes, the KAM unit and
Telia Mobile together develop a strategy to create
the greatest impact on the technical units. The
result can be that an individual in Telia Mobile is
identi?ed as being the person with the greatest
possibility of in?uencing the technical units within
Ericsson. In this way, units in the two companies
can be cooperating—networking—to create the
greatest in?uence or pressure possible.
BSS unit
Like the KAM units, the technical units
within Ericsson can be characterized as dis-
tinctive hierarchies. The technical units have
3
A release always has some outstanding problems. Erics-
son’s method for resolving this is to deliver weekly revisions.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 63
overall responsibility for a speci?c subsystem in
the GSM cellular system. The BSS unit handles
the base station system; it is responsible for the
development, supply and support of both the hard-
ware and software for it. The unit develops standard
software releases, hardware, and participates in
international standard-setting committees like the
European Telecommunications Standards Institute
(ETSI). Support of the ongoing cellular system is
another responsibility for the BSS unit. However, in
the ?rst instance, within Ericsson it is the KAM
units that handle support, but some problems need
specialists for a particular sub system. The technical
units deliver revisions on a weekly basis.
This means that the BSS unit coordinates Erics-
son’s activities concerning the sub system base
stations. The BSS unit is accountable for the con-
solidated return on sales for their part of the cel-
lular system. The consolidated return on sales
includes all revenue for the base station system
worldwide for both hardware and software. Thus,
the BSS unit is accountable for the revenue gener-
ated by the local companies worldwide. The out-
put prices are decided by the KAM units through
negotiation with the customers. However, the BSS
unit does provide price recommendations for their
products. The KAM units’ sales of the base sta-
tion system are consolidated as sales in the BSS
unit. Transfer prices are not calculated, instead the
sales are counted several times in di?erent organi-
zational units. Furthermore, the consolidated
return on sales also includes all costs connected to
the base station system. Thus, while the BSS unit
is accountable for hardware production costs, the
production plants are organized directly under the
managing director of cellular systems. The plants
produce base station system parts for the GSM,
D-AMPS and PDC cellular systems. Conse-
quently, the BSS unit is accountable for the pro-
duction costs generated by factories outside their
control. The manager of the BSS unit has a per-
formance based salary consisting of three parts
depending on: the GSM unit’s return on invest-
ment, the BSS unit’s return on sales, and a ‘‘man-
agement by objectives’’ based incentive scheme.
The BSS unit is also accountable for the time-
liness and quality of the product supply, main-
tenance and development. The unit focuses on
issues like the time to customer and the time to
market. During the construction of new cellular
sites, it is important that the hardware and soft-
ware products are delivered as expected. The
timeliness and quality are also fundamental to the
development of new software and hardware.
The unit does not have any direct access to sys-
tematic information from Telia Mobile, however,
it does gather some information from Telia and
other customers through personal contacts and
meetings. The BSS unit has meetings with impor-
tant customers in the early stages of a project for a
new release to obtain feedback. The unit could
also sometimes have special projects concerning a
speci?c issue run in conjunction with one or more
of its customers. The BSS unit obtains most cus-
tomer information from the KAM units.
The BSS unit is highly dependent on other
technical units, so there is a need for good coor-
dination between the BSS unit and these units. An
illustration of this is the development of features
in a release project: the BSS unit could focus on a
feature because it is likely to increase the pro?t-
ability of the unit. However, to sell this speci?c fea-
ture it is necessary that the other technical units also
develop the feature. This can be a problem when the
feature in question would be particularly pro?table
for the BSS unit, but not for the other units.
The BSS unit does not have any direct contact
with the customer without being granted permis-
sion by the key account unit. Thus, the BSS unit is
accountable for the return on sales, but the KAM
units e?ectively own the customer.
However, the BSS unit manager accepts this
dependence. In his view, ‘‘it is unavoidable in a
complex organization such as Ericsson.’’ The BSS
unit can control some of the aspects a?ecting both
revenues and costs. It can in?uence the price, time
to market, time to customer, product quality, and
some of the costs. The manager of the BSS unit
indicated that he is able to in?uence the other
units as long as he has good arguments to sub-
stantiate his requests.
The purchasing unit at Telia Mobile
As the purchasing unit is responsible for all pur-
chases made by Telia Mobile, it has an intensive
64 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
interaction with units within Ericsson as well as
Nokia. This means, for example, that several
individuals within the unit have contacts with
individuals within Ericsson on an almost daily
basis. Some individuals in the purchasing unit are
employed full time with the a?airs of either Erics-
son or Nokia. The purchasing unit is responsible
for the commercial part of the relationship with
Ericsson, but not the development and operative
activities.
The purchasing unit is accountable for its own
costs. Thus, the unit is treated as a discretionary
cost center. Control focuses on how the unit can
collaborate with other units within Telia Mobile,
and on the unit’s collaboration with important
suppliers. Furthermore, the control is oriented
towards the ?nancial outcome of Telia mobile as a
whole. As far as the technical units within Telia
Mobile are concerned, they are treated, ?nan-
cially, as discretionary cost centers. Thus, it is only
when considering Telia Mobile as a whole that one
is able to discern a distinct hierarchy.
According to several individuals within the pur-
chasing unit and the technical units, Telia
Mobile has a close relationship with Ericsson.
Some individuals even argued that it would
bene?cial to have a more business-like relation-
ship than they have, that is, that the existing
relationship is too familiar. However, some also
argued that Telia Mobile is just a small customer
so it needs the special relationship to be able to
in?uence the technical units within Ericsson.
According to the purchasing director it would not
be possible for Telia Mobile to end the relation-
ship with Ericsson. Instead, they have to survive
and develop together.
The net development unit within Telia Mobile
The net development unit is responsible for the
development of the underlying functionality of the
cellular system. It comments on release drafts
from Ericsson for several stages of the software
release and hardware development process. It also
handles the veri?cation and implementation of the
software release and hardware. The unit collabo-
rates in international standard setting committees
like the European Telecommunications Standards
Institute (ETSI). It also takes an active part in the
CME user group, which consists of European tel-
ecommunication operators. The net development
unit collaborates closely with the purchasing unit
and supports the unit in its commercial negotia-
tions with the KAM unit in Ericsson.
The net development unit is accountable for its
own costs. However, according to the develop-
ment manager, cost is not a big issue in the per-
formance evaluation as long as it is reasonably
well correlated to the output performance. In his
opinion, the important issues are to make progress
and deliver project results at the expected time.
Another important issue in the performance eval-
uation is the net development unit’s collaboration
with other units within Telia Mobile and with
units outside the company.
The development manager argued that the unit
plays a key role in Telia Mobile because the unit’s
decisions about new base functionality create rev-
enue opportunities, as well as potential cost sav-
ings. For example, the remote loading feature
makes it possible to reduce the consumption of
resources within the operation unit, and the high
speed feature is an essential feature in some of
Telia Mobil’s new end customer services. The
decision making body within the unit is comprised
of members from the net development unit, the
service development unit, the operation unit, the
sales unit, and the IT-unit. Thus, the unit focuses
on Telia Mobile’s overall performance.
Individuals within the net development unit
have an intensive interaction with their counter-
parts within Ericsson and Nokia. The net devel-
opment unit has several interfaces with the KAM
unit, and the technical units within Ericsson.
Thus, there are plenty of individuals within the net
development unit who maintain continuous con-
tact with individuals in the two di?erent units
within Ericsson. For example, the manager of the
R7 (i.e. the release number 7 for GSM) imple-
mentation project was in daily contact with the
project manager within Ericsson’s KAM unit. In
addition, he was in contact with other individuals
within the KAM unit, the BSS unit, and the SS
unit.
Individuals within the net development unit
interact with people within Ericsson in ?ve ways.
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 65
Firstly, through the normal development routines
where some individuals within the net develop-
ment unit have technical meetings and frequent
personal contacts with their counterparts within
the KAM unit. Secondly, there is the standard
software implementation project in which the unit
is in contact with all corresponding units within
Ericsson. Thirdly, through special collaboration
projects with some technical unit, focusing on
issues such as GPRS. Fourthly, individuals from
the strategic product management unit within the
technical units have meetings with the net devel-
opment unit in the early stages of a new release.
Fifthly, the development manager participates in
Ericsson’s new technology meetings. These are
small exclusive meetings oriented towards clarify-
ing the speci?cations of the next generation of
cellular communication.
Control and forms of coordination
The more intensive study of the KAM unit, the
BSS unit, the purchasing unit and the net devel-
opment unit showed that the relationship coordi-
nation and its control seems to include some
features of all the classical coordination forms,
including hierarchy, market and cooperation.
Hierarchical coordination is important in several
of the relationships. The KAM unit and the BSS
unit within Ericsson are distinct hierarchies. They
both have a formal power structure and they are
rather autonomous where decision making is con-
cerned. The KAM unit and the BSS unit use
accounting methods such as responsibility
accounting, budgets, reward structures and pro?t
measures, all of which correspond to hierarchical
coordination (Chandler & Dames, 1979; Ezzamel,
1992). However, the KAM unit and the BSS unit
are both very dependent upon other units. The
control system within the two units included the
goals of their important counterparts (Berry,
1994). An example of this is that of the manager of
the support unit within the KAM unit who is
rewarded if Telia Mobile implements the remote
loading feature.
The purchasing unit and the net development
unit within Telia Mobile are treated as discre-
tionary cost centers. The two units function as a
part of the hierarchical coordination within the
distinct hierarchy of Telia Mobile. The purchasing
unit and the net development unit are ?nancially
evaluated on the basis of Telia Mobile’s overall
?nancial performance (Merchant, 1987). This
interest in the company as a whole can be exem-
pli?ed by the decision making forum within the
net development unit, which is comprised of
members from several sub-units within Telia
Mobile. In this way, the overall function of the
relationship between Ericsson and Telia is partly
characterized by the use of hierarchical coordina-
tion mechanisms (Chandler & Daems 1979; John-
son, 1983).
Market coordination is also a part of the total
coordination between Ericsson and Telia. It is
most pronounced within Ericsson between the
KAM unit responsible for Telia Mobile and other
KAM units responsible for other important cus-
tomers. The KAM units represent the needs of
di?erent customers and every KAM unit tries to
in?uence the technical units (BSS, SS and/or BTS)
within Ericsson to get them to develop features
that will meet their customer’s particular needs. As
such, the KAM units compete against each other
to get the attention of the technical units within
Ericsson. That is, the relationship as a whole, is
characterized by the existence of market coordi-
nation mechanisms (Ezzamel, 1992; Ouchi, 1980).
Cooperative coordination occurs within several
of the relationships between individual sub-units.
The KAM unit has close relationships with both
the purchasing unit and the net development unit
within Telia Mobile. They have an intensive col-
laboration within di?erent areas (Nicolini et al.,
2000). They collaborate in their development of
strategies and action plans for how they should
in?uence the technical units within Ericsson. The
KAM unit and the net development unit in Telia
Mobile have common projects (Seal et al., 1999).
Some participants within Telia Mobile even men-
tioned that the relationship was too close and was
too familiar (Tomkins, 2001). Thus the function of
the relationship between Ericsson and Telia as a
whole is characterized by an extensive use of
cooperative coordination mechanisms (Birnberg,
1998, Van der Meer Kooistra, & Vosselman,
2000).
66 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
Conclusions
This study gives a picture—a snapshot—of the
organization of a business relationship at a certain
point in time. The situation does not seem stable
and we have to assume that the relationship is con-
tinually developing. Furthermore, given the existing
characteristics, it seems reasonable to assume that
the relationship is under continuous development.
A ?rst conclusion is that established accounting
methods play a key role in forming the relation-
ship. Such a conclusion supports the analysis
made by Tomkins (2001) showing that existing
accounting techniques are still adequate in busi-
ness alliances and networks. However, one di?er-
ence is that neither collective accounting
information, e.g. inter-organizational budgets and
cost behavior analyses, or transparency through
open book accounting and target costing
approach is used in the relationship. Thus, the
decision makers are not supplied with more for-
malized accounting information about the rela-
tionship or the counterpart. Instead it is through a
systematic combining of accounting in the forma-
tion of organizational units with partly over-
lapping accountability that established accounting
methods support the relationship formation.
Ericsson is using it in relation to technical as well
as customer-de?ned units. One example is the unit
that is responsible for the relationship with Telia
for the GSM sales. This unit has an accounting
system that focuses its activities on one goal—a
goal that clearly includes the counterpart, and is
thus a goal that incorporates the counterpart’s
ambitions without requiring compatibility. In this
way, accounting within this sub-unit is used to
establish a link that bridges over the di?erences
without destroying the need for the two parties to
be unique. In its dealings with other parties, the
KAM unit does not always represent the same
interests: the unit represents Ericsson when it
interacts with Telia, but it clearly also represents
Telia Mobile when it interacts with other Ericsson
units. In this way, accounting has helped to create
new boundaries that do not follow the boundaries
given by ownership. In the same way, the technical
units have partly overlapping domains in their deal-
ings with key account units and where temporal’’
units are concerned. This overlapping accountability
provides the inducement to ?nd new technical solu-
tions as well as the opportunities to develop them.
The overlap forces the units to interact to ?nd com-
promises that do not have to be fully satisfactory for
either one of the parties, because of their temporary
nature. Thus, accounting has helped to establish a
structure that is not a solution to the coordination
problems, but rather, is a base facilitating a process
that can lead to continuous adaptation of the ‘solu-
tions’. This results in the splitting up or distribution
of problems, con?icts and contradictions that,
otherwise, would appear in one single interface
between the producer and the user to several inter-
faces. In this way it gives a depth to the interface that
makes it possible to tolerate a number of seemingly
incompatible di?erences. With all of these contra-
dictory requirements in technical and commercial
terms, there is a tremendous need for continuous
adaptation of a number of dimensions—always
knowing that the solutions are sub-optimal, but also
that they are only temporary.
A second closely related conclusion concerns the
information content within the main relationship
and the associated ones. Some of the information
content regards the exchange taking place, but to
a large extent it is also directed toward the future
in terms of coordinating the development of the
cellular system. This is a highly complex situation
where Ericsson has to handle the customer het-
erogeneity, create economies of scale, and shape a
cellular system that can communicate with the
competitors’ systems. The information ?ow is
clearly not about routine activities: rather, it
concerns unique issues—the creation of features
in new systems. The results from this study show
that neither Ericsson nor Telia has developed
formalized information systems that collect
the required information within each company
or across the company boundaries. The case also
gives reason to question whether such systems can
ever be designed. Instead, knowledge has to be
created jointly by Ericsson and the main custo-
mers in interplay with each other and with some
other key actors (Nonaka & Takeuchi, 1995). It
can be seen as a ‘‘bargain’’ between a number of
units representing technical and commercial fea-
tures, but it should also be seen as a ‘‘problem
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 67
solving process’’ where single units wrestle to com-
bine and confront technical and economical facets.
The two ?rst conclusions lead to a third more
general one: there seems to be good reason for
opening the ‘‘black box’’ of relationship coordi-
nation. The overall coordination in a relationship
like the one in the case study is based on a rather
complicated structure involving hierarchies, rela-
tionships between sub-units and even market
characteristics. The relationship exists in terms of
something ‘‘total’’, which is spoken about and for
which there are managers responsible. But this
totality is not uniform but includes all the three
classical coordination mechanisms. It involves a
mix of relationships whereby the hierarchically
governed sub-units have some very close relation-
ships with some sub-units, but more market like
relationships with others. This is clearly a form of
multi-dimensional coordination where issues rela-
ted to di?erent activities are dealt with in quite
di?erent ways. Thus, it is not a uniform coordi-
nation form. Individual units within each of the
two companies have quite distinct and clearly
de?ned goals. Thus, as a coordination form, hier-
archy is vital within the relationship. These units
have several types of relationships with other units
within their own company, as well as with units
within the counterpart. Some of these relation-
ships are clearly emphasized by the close coopera-
tion—having obvious relationship features—while
others are of a more competitive or hostile type.
Taking the example of a sub-unit within Ericsson,
one ?nds that close and market like relationships
exist with other units both within Ericsson and
within Telia. If we translate this ?nding into the
terms adopted by Richardson, the two ?rms are
performing a number of activities, which, in terms
of similarities and complementarities, are related
to each other in di?erent ways. As there is a var-
iation in these terms there must also be a variation
in the coordination of them. Finally, as these
activities are changing over time, there is also a
need for a more or less continuous change of
coordination forms.
The ?nal conclusion regards the interface
between di?erent coordination forms. To a very
great extent, research concerning governance has
focused on what is special about each mode of
governance, i.e. the di?erences between the three
forms of coordination. Although this is of interest
per se, there might also be very good reason to
study how the forms of coordination can be com-
bined. Such combinatory coordination is probably
closely connected to the existence of ‘‘depth’’ in
the coordination. For example, the two principal
companies in the case have constructed at least
two levels of coordination, which certainly increa-
ses the possibility of ?nding a number of combi-
nations. The existence of two or more ‘‘levels’’ of
coordination justi?es further investigations of
both the interface between hierarchy and relation-
ship and between the market and the relationship/
network. It might be that the largest potential for
development is in these interfaces.
The four conclusions above justify the need for
further research in a very speci?c direction. There
is a need for more intensive case studies of busi-
ness relationships where the role of di?erent orga-
nizational units and the relationships between
them are investigated in detail. The interaction
between companies seems to have a substance and
a variety that need to be conceptualized and
modeled in a much more elaborate way than has
been done to date.
Acknowledgements
This paper has bene?ted from the comments of
Trevor Hopper, participants of the 1999 workshop
on Business Networks and Accounting, Uppsala,
Sweden, two anonymous referees, and seminar
participants at University of Manchester and
Copenhagen Business School. This research has
been supported by grants from Swedish Council
for Research in the Humanities and Social Sci-
ences, Svenska Handelsbanken Foundation for
Social Science Research and Swedish Agency for
Innovation Systems.
68 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
Appendix
Interviews and informal discussions
Ericsson
Key Account Management unit responsible for Telia Mobile
Interviews
1. Mats Gustavsson (project manager, previous responsible for building Telia Mobile’s GSM 1800
cellular system in Stockholm)
2. Maria Agardh (project manager, responsible for the R7 implementation)
3 Anna Langeborg (project controller)
4. Marianne Hansson (business controller)
5. Ann-Christine Karlsson (project manager, responsible for training)
6. Panos Douzoulikopoulos (sales and marketing, responsible for base station systems)
7. Lars Meurling (product management, responsible for the base station systems)
8. Anna-Karin Hansson (product management, responsible for the switching systems)
9. Anders Karlsson (manager, system support)
10. Per Adolfsson (system support, responsible for a subsystem implementation of R7)
11. Sta?an Starin (strategic planning)
12. Mats Granryd (manager, sales and marketing)
13. Orvar Parling (manager, Key Account Telia)
14. Anders Nilsson (manager, support)
Informal discussions
15. Thomas Uhlander (manager, operations)
16. Monika Karlsson (Sales and marketing)
Business unit base station systems (BSS)
17. Aristidis Nikitidis (marketing unit, previous strategic product management unit and responsible
for R7)
18. Cecilia O
¨
berg (strategic product management unit, responsible for R8)
19. Anders Holmring (System design unit, responsible for R8)
20. John Montgommery (manager, strategic product management)
21. Christer Gustafsson (controller)
22. Mattias Karlberg (manager, system design)
23. Martin Bakhuizen (manager, sub unit within system design)
24. Stig-Rune Johansson (manager base station system, vice president GSM-unit)
Other units within Ericsson
25. Sta?an Holme´ r (manager European market, vice president GSM-unit)
26. Krister Jonsson (manager, sub unit within base transceiver stations)
27. Ulf Cige´ n (Senior accountant)
28. Sta?an Hermodsson (Accountant)
29. Peter Lindberg (Senior accountant)
30. Reimo La¨ hthenma¨ ki (Accountant, responsible for production accounting)
31. Anne-Charlotte Jalakas (Accountant)
32. Gary Ethole´ n (Financial analyst, production sta?)
33. Carl-Erik Johansson (AXE software Research)
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 69
Discussions
34. Bjo¨ rn Carlsson (European market, operations support)
35. Sture Jansson (Finance)
36. Mats A
?
rnell (Accountant, Customer service unit)
37. Karin Lundin (Customer service unit)
38. Lars Nylocks (CFO, GSM-unit)
39. Johan Uhlander (Senior accountant)
40. Osakar Ludwig (EDI solutions to suppliers)
41. Erland Lagerlo¨ f (Accountant)
42. Anders Bratt (Operations development, radio base plant)
43. Kaj Nielsen (Project manager, production team project, radio base plant)
44. Ann Hillin (Trade union, radio base plant)
Telia Mobile
Net development unit
Interviews
45. Mari Axell (Radio development unit, responsible for a subsystem implementation of R7)
46. Kim Gro¨ ndahl (Switching development unit, responsible for R7 implementation)
47. Ha? kan Wibergh (Radio development unit, responsible for the technical content of the radio
base system)
48. Lars Andeberg (manager, sub unit within radio development unit)
49. Lars Klasson (manager, development unit, member of Telia Mobile’s executive management
group)
50. Tommy Ljunggren (manager, radio development unit)
Purchasing unit
Interviews
51. Kristina Lundberg (manager, sub unit within purchasing, responsible for purchasing the releases)
52. Monika Ra¨ mme (manager, purchasing, member of Telia Mobile’s executive management group)
Operations unit
Interviews
53. Tord Larsson (net operations, responsible for a sub project within R7 implementation)
54. Lena Sundin (net support, responsible for coordination of release corrections)
55. Martin Blomberg (manager, net operations)
56. Stefan Fa¨ gerhall (manager, operations, member of Telia Mobile’s executive management group)
57. Anders Grahn (manager, sub unit net operations)
58. Lars Olsson (manager, process development)
Service development unit
Interviews
59. Tommy Ytterstro¨ m (manager, service development)
60. Peter Magnusson (manager, development future services)
Market unit
Interviews
61. Johan Freding (product manager, mobile business solutions, responsible for Dirigent)
62. Per Englund (manager, mobile business solutions)
70 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
References
Alchian, A. A., & Demsetz, H. (1972). Production, information
costs, and economic organization. The American Economic
Review, 62, 777–795.
Anderson, J. C., Ha? kansson, H., & Johanson, J. (1994). Dyadic
business relationships within a business network context.
Journal of Marketing, 58, 1–15.
Berry, A. J. (1994). Spanning traditional boundaries: organiza-
tion and control of embedded operations. Leadership &
Organization Development Journal, 15, 4–10.
Birnberg, J. G. (1998). Control in inter?rm co-operation rela-
tionships. Journal of Management Studies, 35, 421–428.
Bradach, J. L., & Eccles, R. G. (1989). Price, authority and
trust: from ideal types to plural forms. Annual Review of
Sociology, 15, 97–118.
Bromwich, M. (1990). The case for strategic management
accounting: the role of accounting information for strategy
in competitive markets. Accounting, Organizations and
Society, 15, 27–46.
Chandler, A. D. Jr., & Daems, H. (1979). Administrative co-
ordination, allocation and monitoring: a comparative analysis
of the emergence of accounting and organization in the USA
and Europe´ . Accounting, Organizations and Society, 4, 3–20.
Dubois, A. (1998). Organising industrial activities. London:
Routledge.
Ezzamel, M. (1992). Corporate governance and ?nancial con-
trol. In M. Ezzamel, & D. Heath?eld (Eds.), Perspectives on
?nancial control (pp. 3–26). London: Chapman & Hall.
Gietzmann, M. B. (1996). Incomplete contracts and the make
or buy decisions: governance design and attainable ?exibility.
Accounting, Organization and Society, 21, 611–626.
Ha? kansson, H. (1993). Networks as a mechanism to develop
resources. In P. Beije, J. Groenewegen, & O. Nuys (Eds.), Net-
working in the Dutch Industry (pp. 207–223). Apeldorn: Garant.
Ha? kansson, H., & Ford, D. (2002). How should companies
interact? Journal of Business Research, 55, 133–139.
Haugland, S., & Reve, T. (1994). Price, authority and trust in
international distribution channel relationships. Scandinavian
Journal of Management, 10, 225–244.
Hopwood, A. G. (1996). Looking across rather than up and
down: on the need to explore the lateral processing of infor-
mation. Accounting, Organization and Society, 21, 589–590.
Ittner, C. D., & Larcker, D. F. (1998). Innovations in perfor-
mance measurement: trends and research implications. Jour-
nal of Management Accounting Research, 10, 205–238.
Johnson, H. T. (1981). Toward a new understanding of nine-
teenth-century cost accounting. The Accounting Review, 56,
510–518.
Johnson, H. T. (1983). The search for gain in markets and ?rms: a
review of the historical emergence of management accounting
systems. Accounting, Organization and Society, 8, 139–146.
Jones, C. S. (1998). Hierarchies, networks and management
accounting in NHS hospitals. Accounting, Auditing &
Accountability Journal, 12, 164–187.
Merchant, K. A. (1987). How and why ?rms disregard the
controllability principle. In W. J. Bruns Jr., & R. S. Kaplan
(Eds.), Accounting and Management. Field Study Perspec-
tives (pp. 316–338). Boston: Harvard Business School
Press.
Mouritsen, J., Hansen, A., & Hansen, C. O. (2001). Inter-
organizational controls and organizational competencies:
episodes around target cost management/functional analysis
and open book accounting. Management Accounting
Research, 12, 221–224.
Nicolini, D., Tomkins, C., Holti, R., Oldman, A., & Smalley,
M. (2000). Can target costing and whole life costing be
applied in the construction industry?: evidence from two case
studies. British Journal of Management, 11, 303–324.
Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating
company. New York: Oxford University Press.
Otley, D., Broadbent, J., & Berry, A. (1995). Research in man-
agement control: an overview of its development. British
Journal of Management, 6, S31–S44.
Ouchi, W. G. (1977). The relationship between organizational
structure and organizational control. Administrative Science
Quarterly, 22, 95–113.
Ouchi, W. G. (1979). A conceptual framework for the design of
organizational control mechanisms. Management Science,
25, 833–848.
Accounting unit
Interviews
63. Anne Janson (business controller)
Other units within Telia
Interviews
64. Johan Stro¨ mba¨ ck (Account Manager, Telia Megacom)
65. Hans Andersson (Account Manager, Telia Megacom)
Discussions
66. Roger Larsson (Telia Research)
67. Thomas Stefansson (Telia Research)
H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72 71
Ouchi, W. G. (1980). Markets, bureaucracies, and clans.
Administrative Science Quarterly, 25, 129–141.
Richardson, G. B. (1972). The organization of industry. Eco-
nomic Journal, 82, 883–896.
Rind?eish, A., & Heide, J. B. (1997). Transaction cost analysis:
past present and future applications. Journal of Marketing,
61, 30–54.
Scapens, R. W. (1990). Researching management accounting
practice: the role of case study methods. British Accounting
Review, 22, 259–281.
Seal, W., Cullen, J., Dunlop, A., Berry, T., & Ahmed, M.
(1999). Enacting a European supply chain: a case study on
the role of management accounting. Management Accounting
Research, 10, 303–322.
Seal, W., & Vincent-Jones, P. (1997). Accounting and trust in
the enabling of long-term relations. Accounting, Auditing &
Accountability Journal, 10, 406–431.
Shields, M. D. (1997). Research in management accounting by
North Americans in the 1990s. Journal of Management
Accounting Research, 9, 3–61.
Tomkins, C. (2001). Interdependencies, trust and information
in relationships, alliances and networks. Accounting, Organi-
zation and Society, 26, 161–191.
Van der Meer-Kooistra, J., & Vosselman, E. G. J. (2000).
Management control of inter?rm transactional relationships:
the case of industrial renovation and maintenance. Account-
ing, Organization and Society, 25, 51–77.
Williamson, O. E. (1975). Markets and hierarchies: analysis and
antitrust implications. New York: Free Press.
Williamson, O. E. (1985). The economic institutions of capital-
ism. New York: Free press.
Yin, R. K. (1989). Case study research: design and methods.
Newbury Park: Sage Publications.
72 H. Ha?kansson, J. Lind / Accounting, Organizations and Society 29 (2004) 51–72
doc_679391150.pdf