Description
This paper explain corporate entrepreneurship in network organizations.
European Management Journal Vol. 16, No. 3, pp. 355–364, 1998
© 1998 Elsevier Science Ltd. All rights reserved
Pergamon
Printed in Great Britain
0263-2373/98 $19.00 + 0.00 PII: S0263-2373(98)00012-7
Corporate
Entrepreneurship in
Network Organizations:
How Subsidiary Initiative
Drives Internal Market
Ef?ciency
JULIAN BIRKINSHAW, Stockholm School of Economics
Julian Birkinshaw believes that corporate
entrepreneurship can play an important strategic
roˆ le in the business development of large, multi-
national ?rms, and he particularly identi?es
internal initiatives by subsidiary companies. The
growth of network organizations is relevant here,
and he puts forward a framework for an ‘internal
market’ to understand how they work. The internal
market model is used to examine the various roles
that corporate entrepreneurs can play in the large
?rm, with the aim of achieving network optimiz-
ation. The author assesses the costs and bene?ts of
internal initiatives, and points out the importance
of arriving at an appropriate level of subsidiary
entrepreneurship, agreed by head of?ce and sub-
sidiary, rather than applying blanket prescriptions.
© 1998 Elsevier Science Ltd. All rights reserved
The concept of corporate entrepreneurship has often
been hard to get across to managers who have spent
their working lives in large organizations. The idea
is innately appealing, but for the most part the
organizational infrastructure that grows up around
large ?rms ends up constraining and frustrating the
efforts of would-be entrepreneurs. Most ideas that
are conceived within large ?rms, it seems, either
wither on the vine or come to fruition once the frus-
trated entrepreneur has set up on his own.
European Management Journal Vol 16 No 3 June 1998 355
My purpose in this article is to argue that we should
be more receptive to the idea of corporate
entrepreneurship in large ?rms. Certainly, there con-
tinues to be a lot of resistance to corporate
entrepreneurship, and a lot of reasons why it should
be managed with care, but there are several reasons
why the mid–late 1990s are seeing a resurgence of
interest in the phenomenon.
1
One is the much-dis-
cussed shift towards a knowledge society, and thus
the need for ?rms to manage their knowledge assets
more effectively. Important trends such as the
empowerment movement, delayering, re-engineering
and network organizations are all shifting power
away from the centre and towards front-line
employees. A second important change is the increas-
ing geographical dispersal of many large organiza-
tions, and a corresponding need to take into account
and act on the ideas of subsidiary managers in
foreign lands. More and more corporations, it seems,
are acknowledging such changes and developing
organizational responses that actively encourage
entrepreneurial behaviour. ABB and Cable & Wire-
less, for example, describe themselves as federative
organizations. GE sees itself as a ‘boundaryless’ com-
pany. Volvo and many others use the term ‘net-
work organization’.
As a ?rst step, then, I simply want to suggest that
the trend towards network organizations represents
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
a major opportunity for the corporate
entrepreneurship movement. First, network organi-
zations provide many of the preconditions that are
necessary for corporate entrepreneurs to thrive: a
license to build relationships laterally, horizontally
and with external parties, as a means of getting
things done; a reasonable level of discretion to pur-
sue an idea before having to justify it; and a greater
openness in head of?ce to new ideas. Second, and
more relevant for this paper, the network organiza-
tion provides a ready metaphor for understanding
the sorts of roles that corporate entrepreneurs can
take. My suggestion is that we can use one of the
key features of the network organization, namely the
‘internal market,’ to describe the way that the various
parts of the large ?rm ?t together, and to look at cor-
porate entrepreneurship as one of the main mech-
anisms through which the internal market functions.
This article is in four parts. First, I provide a little
background on my own research in order to explain
what species of corporate entrepreneurship I am
interested in. Second, I put forward the ‘internal mar-
ket’ framework as a means of understanding how the
network organization works. Third, I use this frame-
work to examine the various roles that corporate
entrepreneurs can play in the large ?rm, and fourth,
I identify and comment on some of the costs of cor-
porate entrepreneurship to large ?rms.
Background: Subsidiary Initiatives in
Multinational Corporations
My own interest in corporate entrepreneurship dates
back to 1993 (see Birkinshaw, 1995), when I began to
study the plight of US-owned subsidiary companies
in Canada in the wake of the Free Trade Agreement.
For these Canadian subsidiaries that had been used
to decades of tariff protection, Free Trade was a
threat, because it gave their parent companies in the
US the complete freedom to source products from
their giant US factories and even to sell them out of
the States. What was the use, observers asked, of a
Canadian head of?ce or a branch plant factory in
such a situation? Well, not much at ?rst glance, so
the Canadian managers I studied began – some as
early as the mid 1980s when Free Trade was just a
distant threat – to look very carefully at their existing
activities and ask themselves: what unique value-
added do we provide for our parent company? 3M
Canada, for example, had developed an impressive
capability in small-lot manufacturing; Honeywell
Canada had the engineering expertise and the pri-
mary manufacturing responsibility for two global
products; and Westinghouse Canada had a stand-
alone business unit that developed and marketed
computer terminals for airline reservations systems
on a worldwide basis.
But identifying such capabilities was, of course, only
European Management Journal Vol 16 No 3 June 1998 356
part of the story. The more dif?cult part was the sell-
ing process – bringing the Canadian subsidiary to the
attention of top management in the US, and then
when the moment was right, convincing them that
they should make a further investment in the subsidi-
ary. I termed this process a subsidiary initiative – it
began with the identi?cation of a new business
opportunity, proceeded with a major selling process
to the head of?ce and to other parts of the corpor-
ation, and ?nished with the commitment of resources
to the business opportunity in question. 3M Canada,
for example, was able to build on its proven strength
in small-lot manufacturing to push through a series
of initiatives in the 1988–90 period for new manufac-
turing plants.
Subsidiary initiatives, I argue, are an important form
of corporate entrepreneurship. Obviously the circum-
stances behind these Canadian subsidiaries’ actions
were very context-speci?c, but the basic idea of ident-
ifying a new opportunity and pushing it through the
levels of management to get funding is exactly the
same as what Kanter and Pinchott were talking about
in the mid-1980s. Moreover, it seems to me that sub-
sidiary initiatives represent a rather rare form of cor-
porate entrepreneurship, because they face two
obstacles: (1) the usual resistance to anything new or
unproven, and (2) additional resistance because the
sponsoring unit is ‘foreign’. Many multinational cor-
porations are still somewhat ethnocentric in their
world-view, which results in an extra layer of scepti-
cism being attached to any proposal that comes from
outside the home country. This multi-faceted resist-
ance to subsidiary initiative I have called the Corpor-
ate Immune System (Birkinshaw and Ridderstråle,
1996), in that its implicit role is to reject all alien bod-
ies that attempt to penetrate the inner core of the
organization. As with all metaphors, this one cannot
be taken too far. Its value is that it paints a clear pic-
ture of the challenge that the subsidiary manager
faces in the pursuit of his/her initiative. It also
reminds us of an important point, which is that many
initiatives should be rejected by corporate gatekeepers
as a means of ensuring that resources are tied to the
most fruitful opportunities. But the metaphor should
not be taken too literally either, because the ‘alien
bodies’ in question seek to improve the corporation,
not to damage it.
The second important insight I gained from this
research was the observation that subsidiary initiat-
ives can be both externally- and internally-oriented.
Figure 1 illustrates this concept, in terms of the two
generic forms of initiative that can be found within
large multinational ?rms (see Birkinshaw and Fry,
1998). Externally-oriented initiatives seek to identify
new customer needs, develop new suppliers or forge
new alliance relationships. The rationale behind them
is a simple one, namely that subsidiary companies
around the world are exposed to an enormous var-
iety of stimuli that could never be sensed by man-
agers sitting in head of?ce. It is therefore the
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
Figure 1 Two Forms of Subsidiary Initiative
responsibility of subsidiaries to seek out and respond
to those opportunities that present themselves in the
local marketplace. The focus, in other words, is on
Market Development. Internally-oriented initiatives, by
contrast, seek to make the existing set of relationships
within the multinational corporation work more
ef?ciently. They are directed towards building new
relationships, challenging existing ones, and ident-
ifying un-met opportunities, all within the con?nes
of the existing network. I call this process Network
Optimization.
The interesting point here is that while externally-
oriented initiatives are well known, the internally-
oriented variety seem thus far to have escaped atten-
tion. What follows, then, is an explorative account of
internal initiatives – the various forms they can take,
their impact on large organizations, and their costs
and bene?ts.
Initiatives and Internal Markets
To make sense of the different types of internal
initiative, it is useful to model the large network
organization on an Internal Market.
2
By internal mar-
ket, all I mean is that market-like forces are recre-
ated – to some degree – inside the ?rm. Consider the
following examples, all of which are quite commonly
seen in large ?rms:
O A production unit can source parts from outside
suppliers if it believes that the internal supplier’s
offering is not competitive.
O Two or more units within the multinational ?rm
make competing products: it is up to each national
sales operation to decide which product it will
push in its own market.
O The sales unit can elect to sell a competitor’s pro-
European Management Journal Vol 16 No 3 June 1998 357
ducts if it does not like the products offered by its
own product divisions.
O R&D units have to ‘sell’ their services to the vari-
ous product divisions before they can undertake
a project.
The logic behind these sorts of arrangements is com-
pelling. If an automobile manufacturer is compelled
to buy engines from an inef?cient internal supplier
rather than the world leader in engines, the cars it
produces will not be competitive. Such mandated
sourcing relationships bear a remarkable resem-
blance to the old central-planning system in commu-
nist Russia, in which every company was told what
to make and who to sell it to. A much better
approach, one can argue, is to give the manufacturer
the option of buying engines from outside. This
ensures that the cars it produces will have the best
engines, and more to the point, it provides the
impetus that the internal engine manufacturer needs
to keep its products up-to-date and competitive.
There are dangers in the internal market approach as
well. An obvious concern is that internal markets cre-
ate competition between units, and thus hinder the
development of cooperation and resource sharing. A
second concern is that internal markets lead man-
agers to focus all their efforts on transfer pricing and
internal negotiations rather than adding value to the
end customer. The challenge, as always, is to design
an organization that gets the maximum bene?ts out
of internal competition whilst avoiding the worst of
the costs. This involves building a system that mot-
ivates people to share ideas and cooperate, but at the
same time to compete with one another. Not an easy
balancing act, of course, but one several ?rms –
including ABB, Hewlett Packard and Sharp – have
come some way towards achieving.
Once one accepts the basic logic of the internal mar-
ket, it becomes clear that the subsidiary’s role
changes dramatically. Using the terminology of econ-
omist Kirzner (1973), the subsidiary manager
becomes an entrepreneur, whose role is essentially
to seek out ‘hitherto unnoticed opportunities’ in the
internal market and act upon them. In so doing, the
subsidiary manager helps to resolve inef?ciencies in
the internal market and thus gradually push it
towards equilibrium. Of course, the subsidiary man-
ager still his or her ‘mandated’ job to do, such as sell-
ing the ?rm’s products in the local market, but
according to Kirzner this work is done alongside the
entrepreneurial work. Every individual, in other
words, has a latent dual role: manager and entrepren-
eur.
The internal market framework, and the concept of
the subsidiary manager as an entrepreneur, gives us
a good theoretical foundation for looking at the types
of internal initiatives I have identi?ed during my
research. While there are many ways of categorizing
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
them, Figure 2 provides a simple breakdown into
four groups, based on the nature of the business
opportunity and the level of head-of?ce support. The
opportunity can be of two types: it can be an existing
internal-market arrangement, or it can be an emerg-
ing business area. In the former case the subsidiary
initiative aims to recon?gure existing activities
within the ?rm; in the latter case it aims to enhance
the allocation of new activities.
3
In terms of head
of?ce support, the initiative is either sanctioned by
head of?ce or it is not. When it is sanctioned, that
means there is an existing set of rules or procedures
the subsidiary can follow. When it is not, that means
head of?ce mangers are either unaware of the initiat-
ive, or they are turning a blind eye towards it. Both
of these dimensions will be elaborated upon in the
paragraphs below.
Recon?guration Initiative
A recon?guration initiative can best be understood
as an effort by the subsidiary to alter the existing con-
?guration of activities within the ?rm to enhance
their ef?ciency. Consider the case of IBM in Scotland.
It began manufacturing PCs in Greenock in 1982.
Towards the end of the 1980s, though, plant manage-
ment realized that there was a limited future in just
being a ?nal assembly operation, so they began look-
ing for ways to extend their ‘charter’ i.e. the set of
activities they were responsible for at a corporate
level. In 1991 they identi?ed a small monitor devel-
opment group near London as a logical complement
to their manufacturing, and succeeded in getting it
relocated to Greenock. Subsequently, they identi?ed
two service functions, the order ful?lment process
(whereby orders are collected, organized and sent to
the factory) and help centres (e.g. for people with
installation problems) as Europe-wide activities that
could potentially be centralized in Greenock. Again,
they were successful.
The key point here is that the initiatives all resulted
in changes in the con?guration of activities for IBM
Figure 2 Types of Internal Initiative
European Management Journal Vol 16 No 3 June 1998 358
Europe as a result of efforts by the Scottish subsidi-
ary. Such changes could be – and often are – initiated
by head of?ce management, but it is fair to assume
that the management on location understand the
implications of such changes better than the head
of?ce managers 3000 miles away. Moreover, the IBM
case was not just about recon?guration of activities.
As Gordon Brown, the Greenock controller com-
mented, “by extending the value chain and linking
various activities together in one location, millions of
dollars were saved, customer satisfaction improved,
and market share increased”. The process, in other
words, created a new vitality in the initiative-taking
subsidiary, which led to a signi?cant performance
improvement.
Another important element of the recon?guration
initiative is that it typically results in a loser as well
as a winner. A Honeywell Canada manager, for
example, was touring the plant of a sister unit in the
US back in 1972. He observed their production pro-
cess for a control switch, and realized that his plant
back in Toronto “could do it signi?cantly better and
much cheaper”. He proposed such a change to head
of?ce management, and it was quickly passed, with
the result that the control switch production was
transferred up to Canada. But of course his col-
leagues in the losing plant were not happy – “I was
not welcome back” he admitted. Perhaps they should
have had the opportunity to improve their perform-
ance, but as this manager explained, “the corporate
philosophy at the time was: you could have done a
better job, so you should have done before somebody
took the work off you. And that’s what kept us com-
petitive.”
The IBM and Honeywell examples are cases where
the subsidiary has facilitated a recon?guration of
physical activities – production, R&D operations, and
help centres. But it is important to bear in mind that
many of the internal ?ows in a multinational corpor-
ation are of intangibles such as market knowledge,
technology and organizing principles. If we extend
the de?nition of an internal market to cover a market
for knowledge, then we also open up several other
forms of initiative. Consider, for example, the case of
3M Sweden which developed a high level of expert-
ise in customer focused marketing and key account
management. Gradually other subsidiaries within 3M
Europe got to know about this, so that when they
had questions about customer-focused marketing,
they called Sweden. 3M Sweden thus emerged as a
‘Centre of Excellence’, and the individual heading up
the Swedish marketing operation found himself trav-
elling extensively and disseminating his knowledge.
In addition a number of other centres of excellence
were also designated within 3M Europe.
The 3M story may appear very different from the
IBM and Honeywell examples, but it actually has a
number of common themes: it was inspired by the
subsidiary, it was sanctioned by head of?ce, and it
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
resulted in the more ef?cient use of existing
resources, i.e. 3M Sweden’s excellence in customer
focused marketing. The key difference, of course, is
that knowledge resources can be transferred without
loss to the transferor, whereas physical resource
transfers tend to be a zero-sum game.
Another example that emerges from looking at the
?rm as an internal knowledge market is Skandia
AFS, a Swedish insurance ?rm that specializes in
unit-linked life assurance products. Skandia AFS has
about 20 subsidiaries around the world which all act
very independently. However it also has four sub-
sidiaries called ISUs (international Sales Units) whose
role is to identify opportunities to sell products from
one country into other countries. The ISU subsidi-
aries, in other words, act as brokers in the internal
market, continuously evaluating the existing product
portfolios of sales subsidiaries, and making them
available to other subsidiaries if they look attractive.
There are, then, many types of recon?guration initiat-
ives. They differ substantially in the type of internal
market – in the IBM and Honeywell cases the market
was for corporate charters, in the Skandia and 3M
cases the market was for knowledge. But they all
serve one critical function, namely enhancing the
ef?ciency of the internal market without head of?ce
management having to become actively involved.
Maverick Initiatives
Maverick initiatives are similar to recon?guration
initiatives to the extent that they result in an enhance-
ment in the ef?ciency of the internal market. How-
ever, they are also undertaken without head of?ce
sanctioning, which makes things quite tricky when
the charter ‘steal’ involves another unit losing out.
Such initiatives appear to be the least common of the
four, and I was only able to identify two examples
during my research, both of which requested anon-
ymity. The ?rst was the Swedish subsidiary of Data-
com, a US-based computer manufacturer. Datacom
4
was struggling in the early 1990s because their pro-
ducts were not perceived as competitive. The Swed-
ish subsidiary, on the verge of bankruptcy in 1991,
was aggressively turned around by the new manag-
ing director in part through one simple change: he
began to sell and service competitors’ products, not those
made by Datacom. As the managing director explain-
ed,
‘the US staff concluded that we were prioritizing the sale
of competitors’ equipment. We were not of course – it was
not a choice of A or B, it was B or nothing! To say the least,
this was not understood. Many heated phonecalls, hostile
video-conferences and top level pressure were applied to
stop us selling competitors computers, and go back to sell-
ing what the factory could make’
Whether Datacom Sweden was right to sell competi-
European Management Journal Vol 16 No 3 June 1998 359
tors’ equipment is impossible to judge. But if we go
back to the internal market, it is logical to give the
sales subsidiary freedom to buy products from mul-
tiple sources if its competitiveness is otherwise
compromised. Given that most money in this indus-
try is made in service contracts and not in product
margin, it is possible that the Swedish initiative rep-
resented an important shift in mindset within Data-
com. Today, six years later, the Swedish managing
director has been forgiven for his rule-breaking, and
his turnaround is widely acknowledged within the
?rm as a success story.
The second example of this type of initiative emerged
from a disagreement between Alpha, a Swedish
multinational and its German subsidiary. The Ger-
man subsidiary was demanding some changes to the
product speci?cations in order to sell it in Germany;
the Swedes responded that they could not make the
requested changes in a cost-ef?cient manner. The
manager of the US subsidiary came to hear about this
dispute, and saw a great opportunity. He arranged
to buy several hundred products from Sweden, got
them shipped over to America, retro?tted them with
the German subsidiary’s requested changes, and
shipped them on to Germany, making a tidy pro?t
along the way. This story is a fascinating example of
subsidiary initiative, undertaken directly against the
wishes of head of?ce. It is also a very good example
of entrepreneurship in a Kirznerian sense, in that the
internal market between Sweden and Germany was
clearly failing. Through his alertness to a market
opportunity (in this case an unsatis?ed German
subsidiary), the US subsidiary manager was able to
step in and make a pro?t, and at the same time
enhance the ef?ciency of the internal market.
Of course many people reading this will see the
above two examples as blatantly subversive. Such
actions, they will conclude, lead to internal anarchy
and cause multinational ?rms to lose their strategic
focus. My argument is more balanced. There is cer-
tainly a ?ne line between initiative and subversion,
as the ?nal section of this paper discusses, and only
time will tell if the Datacom and Alpha initiatives
crossed that line. But in the interests of organizational
experimentation, these sorts of initiative are
important to recognize, because they can potentially
have signi?cant bene?ts, as well as obvious costs.
Bid Initiatives
Bid initiatives are those that are directed towards
emerging business areas for the multinational ?rm.
Consider the case of Monsanto Canada, whose top
management identi?ed an interesting proposal in the
corporation"s long-range plan in 1991. Monsanto was
developing a new formulation for its best-selling
agrochemical product, with the intention of bringing
it to market around 1996. Canadian management saw
a great opportunity to argue for the investment to
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
be made in Canada because of the country’s strong
agricultural industry. They persuaded corporate
management to bring forward the planned invest-
ment, and they lobbied hard to ensure that a Canad-
ian site was actively considered. After a careful
analysis of the three competing sites, the Canadian
one was selected.
It is important to compare the Monsanto Canada case
with the IBM case described earlier. Both ended up
following similar processes, in terms of identifying
possible extensions to the subsidiary’s business, lob-
bying head of?ce managers, and doing a careful cost–
bene?t analysis. But the key difference is that IBM
Scotland were pushing to recon?gure existing activi-
ties, while Monsanto Canada were trying to attract a
new activity. This difference is critical for one main
reason, namely that recon?guration initiatives rely
exclusively on the subsidiary to identify the opport-
unity whereas bid initiatives are usually identi?ed
jointly by the head of?ce and the subsidiary. In Hew-
lett Packard Canada, for example, I saw a couple of
examples of initiatives that were originally identi?ed
by the subsidiary but which were subsequently
fought out on an internal basis between the Canadian
subsidiary and a couple of US-based divisions.
The best way of making sense of bid initiatives, I
believe, is by thinking in terms of both the internal
and external market. While the opportunity itself
(such as Monsanto’s need for a new production
plant) is typically oriented towards the external mar-
ket, the subsidiary has to win the right to act on that
opportunity through the internal market. For that
reason, the process tends to resemble that seen in the
recon?guration initiatives – a lobbying process
directed towards head of?ce decision-makers as well
as other in?uential entities within the ?rm.
Moreover, the internal market perspective also indi-
cates another important dimension to bid initiatives
that was not observed in recon?guration initiatives,
namely the need for both push and pull strategies.
The push strategy was what was observed in Mon-
santo Canada, with the top management becoming
actively involved in the speci?c initiative. The pull
strategy is a little more gentle, in that it involves the
subsidiary ‘broadcasting’ its capabilities to in?uential
managers throughout the ?rm, in the hope that it will
be approached when the right investment opport-
unity arises. Take the example of Motorola’s East Kil-
bride plant in Scotland. This is one of 16 Motorola
plants around the world that fabricates silicon wafers
for semiconductors. All the key performance meas-
ures such as cost, service and quality are explicitly
compared on a monthly basis, and new investments
are (obviously) made in the top performing sites.
Because of this comparability, East Kilbride’s man-
agement do not need to think in terms of push stra-
tegies. Their approach, rather, is to stay ahead of
internal competitors on the key performance metrics,
and ‘pull’ new investments in as they occur.
European Management Journal Vol 16 No 3 June 1998 360
Push and pull are by no means alternative strategies.
While management at Motorola’s East Kilbride plant
can focus on the pull approach, they still have to be
very responsive and keen whenever new investments
are planned. Monsanto Canada worked hard on the
pull strategy in the late eighties by broadcasting their
interest in attracting investment, and then used a
push strategy when speci?c opportunities arose. But
the point is that the relative emphasis is likely to
vary, and this will be a function of the ?rm’s system
for managing new investments as well as the individ-
ual manager’s preferred style. For example, Ericsson,
the Telecommunications equipment ?rm, has man-
aged its international R&D activities through an
internal market process for many years. Ericsson’s
corporate culture is typically Swedish, in that it
places a great deal of emphasis on consensus build-
ing and group decision-making. As a result, R&D
charters are handed out through a rather interactive
bid process in which subsidiaries use mostly pull
strategies coupled with active top-down involve-
ment.
Finally, we can also extend the bid initiative termin-
ology to internal knowledge markets. Here the
emphasis is on emerging knowledge areas that cer-
tain subsidiary units actively pursue in order to
become thought-leaders. As before, the term most
?rms use for such a unit is a Centre of Excellence.
For example, consulting ?rms such as Andersen Con-
sulting and McKinsey encourage the various units
around the world to develop leading-edge knowl-
edge in emerging practise areas or industries. Often
such units are formally designated as centres of excel-
lence, with a view to their knowledge being dissemi-
nated throughout the ?rm. An alternative approach
is for the ?rm to use its information technology to
make this leading-edge knowledge available elec-
tronically. In both cases, of course, it is a question of
the ef?ciency of the internal market system (word of
mouth versus electronic) whether such knowledge
?nds its way to the individuals throughout the ?rm
who need it.
‘Leap of Faith’ Initiatives
The fourth and ?nal type of initiative I have called
‘leap of faith’ initiatives because it typi?es the atti-
tude of the managers responsible for them. Such
initiatives are directed towards emerging areas of
business, but they are undertaken without head
of?ce sanction. However, unlike the Maverick initiat-
ive discussed earlier this type is somewhat less con-
tentious because there is no obvious losing party.
Essentially it represents a bet by the subsidiary man-
ager on the emergence of a certain technology or
business area, in the hope that it will really take off.
A well-known example of this is NCR’s Dundee
operation. This plant was on the verge of closure in
1980, largely because its key product, the automatic
teller machine (ATM) had serious quality problems.
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
Jim Adamson, the newly-appointed general man-
ager, worked on improving manufacturing quality
and restoring the con?dence of major customers,
while at the same time developing a vision for
Dundee as NCR’s strategic centre for the ATM busi-
ness. Product development responsibility of?cially
lay with HQ in Dayton, but Adamson began
directing resources towards upgrading and renewing
the Dundee product line to meet the emerging
demands of the big British banks, Dundee’s key cus-
tomers. Faced with active resistance from the devel-
opment group in Dayton, Adamson pursued a delic-
ate strategy of co-operating with them while
continuing privately to sponsor Dundee’s inde-
pendent research program. A successful product
upgrade in 1982 was followed 18 months later by a
next-generation ATM that set new standards in func-
tionality, reliability and serviceability. Dundee’s glo-
bal market share reached 20 per cent in 1984. The
following year, responsibility for the global ATM
business was of?cially transferred from Dayton to
Dundee.
In this case, the leap of faith was not so much in
terms of the ATM market (which was already grow-
ing rapidly) but in terms of Dundee’s ability to com-
pete with IBM and a host of other established com-
petitors. Certainly, there were many within NCR
who doubted that Dundee would ever return to
pro?tability, let alone become a global leader in
ATMs, yet Adamson persisted in his vision and
eventually won out.
A second example of a leap-of-faith initiative is
Honeywell’s Scottish subsidiary in Newhouse, which
manufactures a range of control valves and related
items for the European market. During the seventies
and early eighties control valves were mechanical
devices, but the general manager of Newhouse
became aware of the impending shift towards elec-
tronic devices and he invested discretionary funds in
the development of an electronic manufacturing
capability. Thus, when in 1991 corporate manage-
ment decided to invest in a global-scale facility for
electronic fan-coils, Newhouse was the obvious
location because it had already developed the neces-
sary capabilities. This case was far less dramatic, but
it emphasizes the importance of a far-sighted subsidi-
ary manager who was willing to bet on the emerg-
ence of a certain technology. It also illustrates the
judicious use of discretionary funds in the subsidiary.
While not of?cially sanctioned by head of?ce, it
seems unlikely that such investments would have
caused trouble (unlike the maverick initiatives
described earlier).
In sum then, leap-of-faith initiatives represent an
important fourth category in which – like bid initiat-
ives – subsidiary managers are actively looking for
opportunities in the external market while at the
same time trying to get them resourced and legi-
timized in the internal market. The challenge in such
European Management Journal Vol 16 No 3 June 1998 361
cases appears to be one of effectively managing both
arenas. In the case of the Honeywell Newhouse man-
ager, for example, one must assume that he was con-
scious of the level of investment in digital systems
among the other Honeywell plants, as well as con-
scious of emerging market demands.
To summarize, this section should be seen primarily
as a descriptive account of the forms that internal
initiative can take. Each initiative type requires very
different tactics on the part of the subsidiary man-
ager, from the rather Machiavellian approach in mav-
erick initiatives, through to the by-the-book logic of
bid initiatives. And moreover, each suggests a very
different set of expectations on the part of head of?ce
management, such that in any given corporation
there may only be one or two of these types of initiat-
ive that could conceivably be pursued. Notwith-
standing the differences, it seems that the similarities
are also quite striking. All four require subsidiary
managers to be entrepreneurial, looking for new
ways of adding value and proactively approaching
head of?ce with a proposal, and all four require a
much greater level of awareness and openness on the
part of head of?ce managers to ‘foreign’ ideas than
is normally the case. Clearly the sort of cultural shift
I am talking about does not come easily, but it is a
gradual process and one which many of the ?rms I
have studied seem to be getting to grips with. But
one of the biggest stumbling blocks is a simple and
fundamental scepticism towards corporate
entrepreneurship. So it is to this issue that I now turn.
The Costs of Corporate
Entrepreneurship
As I observed at the outset, many managers remain
suspicious of corporate entrepreneurship. They are
suspicious in the ?rst instance because it sounds like
an oxymoron: perhaps an attractive idea in principle,
but something which just does not happen very
often. But the second reason for suspicion, and the
one I want to address here, is that corporate
entrepreneurship has signi?cant and often hidden
costs. Entrepreneurship is, most of us would agree,
an important engine of growth and renewal whether
it occurs inside the ?rm or in the economy as a whole.
But as with most things, it is possible to have too
much entrepreneurship. Below I have listed four of
the principal costs of corporate entrepreneurship that
I think managers need to be cognizant of, and a num-
ber of suggestions about how those costs can be man-
aged.
Empire building. We are all familiar with the stories.
The country manager in France, for example, con-
vinces head of?ce that he needs to build a local
manufacturing operation to keep the government
happy. A few years later he comes out with some
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
product changes that are ‘essential to our continued
success in France’ but which cut against the world-
wide positioning of the product. He then argues that
a larger development team is needed to support the
new product line, because the R&D people back at
head of?ce do not understand the changes that have
been made. And so on.
Is the French country manager cleverly adapting the
product to the unique French marketplace, or is he an
empire-builder, interested only in building his own
power base at the expense of the ?rm? Similarly, if
we take the Datacom Sweden manager or the US sub-
sidiary manager of Alpha mentioned earlier, where
do their priorities lie? The answer to these question
depends essentially on the preconceptions of head
of?ce management. One preconception argues that
subsidiary managers are prone to act opportun-
istically and build their own little empires. The other
is to trust subsidiary management, and to believe that
they are acting in the best interests of the corporation.
In my experience, the ‘facts’ of the speci?c case can
be interpreted to reach whichever conclusion ?ts the
manager’s preconceptions. Obviously in a few cases
there is solid evidence that the subsidiary’s actions
were well-intentioned or oriented towards empire-
building, but most cases fall in the grey area where
opinion is what matters.
So what can be done to guard against empire build-
ing? My opinion is that the head of?ce – subsidiary
relationship has to be built on trust, otherwise the
strict control systems that ensue will make opportun-
istic behaviour a self-ful?lling prophecy. However,
that trust-based relationship must also be tempered
with a system for evaluating and monitoring the sub-
sidiary’s initiatives, to ensure that they are consistent
with corporate goals. And it also needs to be
undergirded and reinforced by a strong corporate
culture that allows head of?ce managers to be con?-
dent about the abilities and objectives of its subsidi-
ary managers.
Lack of focus. This problem is related to empire-build-
ing, in that it is also the result of ‘too much’ corporate
entrepreneurship. Imagine that every employee takes
the idea of corporate entrepreneurship to heart, and
follows up on their new idea. Soon the ?rm is getting
into new businesses in every conceivable direction,
with the result that ?nancial resources are stretched
too thin and there is no coherence to the ?rm’s pro-
duct line. Hewlett Packard reputedly found itself in
this position in the late 1980s, because it had actively
encouraged individual initiative but it had not put in
place the necessary controls. Changes were made to
reduce the autonomy of subsidiary managers, but
inevitably this approach also choked off the ?ow of
initiatives, with the result that the company probably
moved too far the other way.
Corporate entrepreneurship can end up being pur-
sued outside the ?rm. Ethicon, Johnson & Johnson’s
European Management Journal Vol 16 No 3 June 1998 362
Scottish subsidiary, is a good example of this: it has
on numerous occasions spun off businesses that did
not ?t its strategy, several of which have gone on to
become major ?rms in their own right.
Costs of administrating the internal market. Unlike the
market economy which appears to work best when
left alone, the internal market system needs a con-
siderable amount of administration to work effec-
tively. One set of administrative costs result from set-
ting up and maintaining the internal market system,
e.g. de?ning the conditions under which sourcing
relationships can be challenged, and setting rules for
transfer prices. There are also transaction costs that
are accrued by the participating units every time an
initiative is undertaken. Such costs are not only
higher than would be the case in a ‘centrally-plan-
ned’ organization, they are also potentially higher
than in an external market. In one case I studied, for
example, head of?ce management had decided
against sending out a Request For Proposal to subsidi-
ary companies because they did not want to spend
the time going through all the proposals and justify-
ing their decision to each one in turn. They chose,
instead, to create their own shortlist of two or three
subsidiaries, and evaluate them in detail.
As with the last point, there is a tricky balance here
between maximizing the level of subsidiary initiative
and keeping the cost of managing it under control.
If head of?ce managers draw up their own shortlist
of subsidiaries, they keep the process under control
but they may send the wrong signals to other entrep-
reneurially-minded subsidiary managers. Again, the
rule of thumb appears to be one of keeping a very
broad perspective at ?rst, but fairly quickly selecting
out those initiatives that are uncompetitive or not
aligned with the ?rm’s needs.
Coping with internal ‘unemployment’. The ?nal cost of
corporate entrepreneurship is that it can result in
unemployment, either literally or ?guratively. If we
think about the market economy for a second, it is
generally accepted that closing an inef?cient coal
mine is good for the economy as a whole but results
in hundreds of layoffs in the old coal town. The same
can be argued for the internal marketplace. If we con-
sider the Honeywell case discussed earlier, the
decision to move a production line from Minneapolis
to Toronto was good for Honeywell as a whole
(because it resulted in more ef?cient production) but
resulted in the loss of some jobs in Minneapolis. This
leads to certain costs for the ?rm. Either the
employees in question are laid off, which results in
a one-time cost to the ?rm and more importantly the
loss of some well-trained experienced people, or they
are retained by the ?rm and assigned new responsi-
bilities, a process which inevitably takes time.
Of course, the costs of closing down factories and
reallocating people occurs in all ?rms, but the point
to emphasize here is that a system which encourages
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
the use of an internal market will experience much
more frequent changes in the allocation of activities
to operating units, and will therefore suffer from a
higher level of ‘internal unemployment’. Just as the
central planning system in Russia was able to boast
full employment until it all fell apart, centrally-
planned ?rms can keep the number of unassigned
workers to a minimum for a long time, and then pay
the price later through massive job losses. The costs
of internal unemployment, in other words, are quite
substantial, but probably worth paying because they
are better than the alternative. The managerial chal-
lenge then becomes one of managing the reassign-
ment of employees to other parts of the ?rm, and of
training them so that their skills match future needs.
Concluding Comments
This paper has focused on the perspective of subsidi-
ary managers, both at a conceptual level, and with
regard to the speci?c initiative strategies they can
pursue. My assumption is that many ‘head of?ce’ ori-
ented managers will have found my approach worry-
ing, because my implicit advice is that subsidiary
managers should take more initiatives, even initiat-
ives that go against the of?cial sanctioning of head
of?ce. In this ?nal paragraph I want to allay such
suspicions slightly, by emphasizing that I would like
the ideas presented here to be seen as descriptive of
an interesting phenomenon rather than a how-to
guide for subsidiary managers. The fact is that we
simply do not know if subsidiary initiatives, of the
type described here are really good for the multi-
national ?rm. Obviously I have chosen to focus on
success stories, but for every case like NCR Dundee
there are a large number of failures in which money
was wasted on a poorly-thought-out investment, or
precious time was spent arguing the pros and cons
of a marginal idea. Rather than offering simple rules-
of-thumb about what level of initiatives should be
encouraged, I would prefer to argue that the appro-
priate level is extremely dependent on the speci?cs
of the ?rm and industry. In some of the companies
I have studied, head of?ce managers have attempted
to suppress the entrepreneurial endeavours of their
subsidiary managers, for fear of being swamped with
proposals. And in others, I have heard head of?ce
managers complain that their subsidiary managers
do not take enough initiative. Getting the level of cor-
porate entrepreneurship right is clearly a process of
mutual adjustment, and one that is unlikely to be
achieved without some pain.
The purpose of this article, then, was to provide a
European Management Journal Vol 16 No 3 June 1998 363
way of thinking about the different forms that
internal initiatives can take, and to explain their
likely impact. My hope is that it will have stimulated
both subsidiary and head of?ce managers to re-
evaluate their approach to corporate
entrepreneurship, and to look with fresh eyes on its
possibility for enhancing the ef?ciency of their
internal market systems.
Notes
1. There was a wave of enthusiasm for corporate
entrepreneurship in the mid-1980s, fuelled by Rosabeth
Moss Kanter’s book The Change Masters (1985: Simon and
Schuster, New York) and Gifford Pinchott’s Intrapreneuring
(1985: Harper and Row, New York). As I read the litera-
ture, we can see something of a second-wave of enthusiasm
at the moment. Notable recent contributions include, Gho-
shal and Bartlett (1997); Birkinshaw (1997); Block and
McMillan (1996), and Ginsberg and Hay (1994).
2. On the issue of internal markets, see also the work of
Halal (1994).
3. One important point of clari?cation is in order here,
namely that initiatives directed towards new market
opportunities are really ‘internal – external hybrids’ in
terms of Figure 1. More speci?cally, while the ‘locus of
opportunity’ is external to the ?rm, it is typically the case
that head of?ce management identi?es the opportunity, so
that the ‘locus of pursuit’ of the initiative is internal.
Viewed in this way, such initiatives are still part of the
internal initiative phenomenon.
4. Datacom and Alpha are disguised names.
References
Birkinshaw, J. (1995) Entrepreneurship in multinational cor-
porations: the initiative process in foreign subsidiaries.
Unpublished doctoral dissertation, University of West-
ern Ontario.
Birkinshaw, J. (1997) Corporate entrepreneurship in multi-
national corporations. Strategic Management Journal 18,
207–229.
Birkinshaw, J. and Fry, N. (1998) Fighting the corporate
immune system: initiative strategies for subsidiary man-
agers. Sloan Management Review (forthcoming).
Birkinshaw, J. and Ridderstråle, J. (1996) Fighting the corpor-
ate immune system: a process study of peripheral initiat-
ives in multinational corporations. In Proceedings of the
European International Business Academy Annual Confer-
ence, Stockholm.
Block, Z. and McMillan, I. (1996) Corporate Venturing.
Ghoshal, S. and Bartlett, C. (1997) The Individualized Corpor-
ation.
Ginsberg, A. and Hay, M. (1994) Confronting the challenges
of corporate entrepreneurship. European Management
Journal 12(4), 382–389.
Halal, W. (1994) From hierarchy to enterprise: internal mar-
kets are the new foundation of management. Academy of
Management Executive 8, 69–83.
Kirzner, I. (1973) Competition and Entrepreneurship. University
of Chicago Press, Chicago.
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
JULIAN
BIRKINSHAW, Insti-
tute of International
Business, Stockholm
School of Economics, PO
Box 6501, S113 83,
Stockholm, Sweden.
Julian Birkinshaw is
Assistant Professor of
Business at the Institute
of International Business
in Stockholm. His research is focused on the internal
workings of large multinational corporations,
including issues of entrepreneurship in foreign sub-
sidiaries, knowledge transfers between units, and co-
ordination of international marketing operations. He
received his PhD from the Richard Ivey School of
Business, University of Western Ontario.
European Management Journal Vol 16 No 3 June 1998 364
doc_222565135.pdf
This paper explain corporate entrepreneurship in network organizations.
European Management Journal Vol. 16, No. 3, pp. 355–364, 1998
© 1998 Elsevier Science Ltd. All rights reserved
Pergamon
Printed in Great Britain
0263-2373/98 $19.00 + 0.00 PII: S0263-2373(98)00012-7
Corporate
Entrepreneurship in
Network Organizations:
How Subsidiary Initiative
Drives Internal Market
Ef?ciency
JULIAN BIRKINSHAW, Stockholm School of Economics
Julian Birkinshaw believes that corporate
entrepreneurship can play an important strategic
roˆ le in the business development of large, multi-
national ?rms, and he particularly identi?es
internal initiatives by subsidiary companies. The
growth of network organizations is relevant here,
and he puts forward a framework for an ‘internal
market’ to understand how they work. The internal
market model is used to examine the various roles
that corporate entrepreneurs can play in the large
?rm, with the aim of achieving network optimiz-
ation. The author assesses the costs and bene?ts of
internal initiatives, and points out the importance
of arriving at an appropriate level of subsidiary
entrepreneurship, agreed by head of?ce and sub-
sidiary, rather than applying blanket prescriptions.
© 1998 Elsevier Science Ltd. All rights reserved
The concept of corporate entrepreneurship has often
been hard to get across to managers who have spent
their working lives in large organizations. The idea
is innately appealing, but for the most part the
organizational infrastructure that grows up around
large ?rms ends up constraining and frustrating the
efforts of would-be entrepreneurs. Most ideas that
are conceived within large ?rms, it seems, either
wither on the vine or come to fruition once the frus-
trated entrepreneur has set up on his own.
European Management Journal Vol 16 No 3 June 1998 355
My purpose in this article is to argue that we should
be more receptive to the idea of corporate
entrepreneurship in large ?rms. Certainly, there con-
tinues to be a lot of resistance to corporate
entrepreneurship, and a lot of reasons why it should
be managed with care, but there are several reasons
why the mid–late 1990s are seeing a resurgence of
interest in the phenomenon.
1
One is the much-dis-
cussed shift towards a knowledge society, and thus
the need for ?rms to manage their knowledge assets
more effectively. Important trends such as the
empowerment movement, delayering, re-engineering
and network organizations are all shifting power
away from the centre and towards front-line
employees. A second important change is the increas-
ing geographical dispersal of many large organiza-
tions, and a corresponding need to take into account
and act on the ideas of subsidiary managers in
foreign lands. More and more corporations, it seems,
are acknowledging such changes and developing
organizational responses that actively encourage
entrepreneurial behaviour. ABB and Cable & Wire-
less, for example, describe themselves as federative
organizations. GE sees itself as a ‘boundaryless’ com-
pany. Volvo and many others use the term ‘net-
work organization’.
As a ?rst step, then, I simply want to suggest that
the trend towards network organizations represents
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
a major opportunity for the corporate
entrepreneurship movement. First, network organi-
zations provide many of the preconditions that are
necessary for corporate entrepreneurs to thrive: a
license to build relationships laterally, horizontally
and with external parties, as a means of getting
things done; a reasonable level of discretion to pur-
sue an idea before having to justify it; and a greater
openness in head of?ce to new ideas. Second, and
more relevant for this paper, the network organiza-
tion provides a ready metaphor for understanding
the sorts of roles that corporate entrepreneurs can
take. My suggestion is that we can use one of the
key features of the network organization, namely the
‘internal market,’ to describe the way that the various
parts of the large ?rm ?t together, and to look at cor-
porate entrepreneurship as one of the main mech-
anisms through which the internal market functions.
This article is in four parts. First, I provide a little
background on my own research in order to explain
what species of corporate entrepreneurship I am
interested in. Second, I put forward the ‘internal mar-
ket’ framework as a means of understanding how the
network organization works. Third, I use this frame-
work to examine the various roles that corporate
entrepreneurs can play in the large ?rm, and fourth,
I identify and comment on some of the costs of cor-
porate entrepreneurship to large ?rms.
Background: Subsidiary Initiatives in
Multinational Corporations
My own interest in corporate entrepreneurship dates
back to 1993 (see Birkinshaw, 1995), when I began to
study the plight of US-owned subsidiary companies
in Canada in the wake of the Free Trade Agreement.
For these Canadian subsidiaries that had been used
to decades of tariff protection, Free Trade was a
threat, because it gave their parent companies in the
US the complete freedom to source products from
their giant US factories and even to sell them out of
the States. What was the use, observers asked, of a
Canadian head of?ce or a branch plant factory in
such a situation? Well, not much at ?rst glance, so
the Canadian managers I studied began – some as
early as the mid 1980s when Free Trade was just a
distant threat – to look very carefully at their existing
activities and ask themselves: what unique value-
added do we provide for our parent company? 3M
Canada, for example, had developed an impressive
capability in small-lot manufacturing; Honeywell
Canada had the engineering expertise and the pri-
mary manufacturing responsibility for two global
products; and Westinghouse Canada had a stand-
alone business unit that developed and marketed
computer terminals for airline reservations systems
on a worldwide basis.
But identifying such capabilities was, of course, only
European Management Journal Vol 16 No 3 June 1998 356
part of the story. The more dif?cult part was the sell-
ing process – bringing the Canadian subsidiary to the
attention of top management in the US, and then
when the moment was right, convincing them that
they should make a further investment in the subsidi-
ary. I termed this process a subsidiary initiative – it
began with the identi?cation of a new business
opportunity, proceeded with a major selling process
to the head of?ce and to other parts of the corpor-
ation, and ?nished with the commitment of resources
to the business opportunity in question. 3M Canada,
for example, was able to build on its proven strength
in small-lot manufacturing to push through a series
of initiatives in the 1988–90 period for new manufac-
turing plants.
Subsidiary initiatives, I argue, are an important form
of corporate entrepreneurship. Obviously the circum-
stances behind these Canadian subsidiaries’ actions
were very context-speci?c, but the basic idea of ident-
ifying a new opportunity and pushing it through the
levels of management to get funding is exactly the
same as what Kanter and Pinchott were talking about
in the mid-1980s. Moreover, it seems to me that sub-
sidiary initiatives represent a rather rare form of cor-
porate entrepreneurship, because they face two
obstacles: (1) the usual resistance to anything new or
unproven, and (2) additional resistance because the
sponsoring unit is ‘foreign’. Many multinational cor-
porations are still somewhat ethnocentric in their
world-view, which results in an extra layer of scepti-
cism being attached to any proposal that comes from
outside the home country. This multi-faceted resist-
ance to subsidiary initiative I have called the Corpor-
ate Immune System (Birkinshaw and Ridderstråle,
1996), in that its implicit role is to reject all alien bod-
ies that attempt to penetrate the inner core of the
organization. As with all metaphors, this one cannot
be taken too far. Its value is that it paints a clear pic-
ture of the challenge that the subsidiary manager
faces in the pursuit of his/her initiative. It also
reminds us of an important point, which is that many
initiatives should be rejected by corporate gatekeepers
as a means of ensuring that resources are tied to the
most fruitful opportunities. But the metaphor should
not be taken too literally either, because the ‘alien
bodies’ in question seek to improve the corporation,
not to damage it.
The second important insight I gained from this
research was the observation that subsidiary initiat-
ives can be both externally- and internally-oriented.
Figure 1 illustrates this concept, in terms of the two
generic forms of initiative that can be found within
large multinational ?rms (see Birkinshaw and Fry,
1998). Externally-oriented initiatives seek to identify
new customer needs, develop new suppliers or forge
new alliance relationships. The rationale behind them
is a simple one, namely that subsidiary companies
around the world are exposed to an enormous var-
iety of stimuli that could never be sensed by man-
agers sitting in head of?ce. It is therefore the
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
Figure 1 Two Forms of Subsidiary Initiative
responsibility of subsidiaries to seek out and respond
to those opportunities that present themselves in the
local marketplace. The focus, in other words, is on
Market Development. Internally-oriented initiatives, by
contrast, seek to make the existing set of relationships
within the multinational corporation work more
ef?ciently. They are directed towards building new
relationships, challenging existing ones, and ident-
ifying un-met opportunities, all within the con?nes
of the existing network. I call this process Network
Optimization.
The interesting point here is that while externally-
oriented initiatives are well known, the internally-
oriented variety seem thus far to have escaped atten-
tion. What follows, then, is an explorative account of
internal initiatives – the various forms they can take,
their impact on large organizations, and their costs
and bene?ts.
Initiatives and Internal Markets
To make sense of the different types of internal
initiative, it is useful to model the large network
organization on an Internal Market.
2
By internal mar-
ket, all I mean is that market-like forces are recre-
ated – to some degree – inside the ?rm. Consider the
following examples, all of which are quite commonly
seen in large ?rms:
O A production unit can source parts from outside
suppliers if it believes that the internal supplier’s
offering is not competitive.
O Two or more units within the multinational ?rm
make competing products: it is up to each national
sales operation to decide which product it will
push in its own market.
O The sales unit can elect to sell a competitor’s pro-
European Management Journal Vol 16 No 3 June 1998 357
ducts if it does not like the products offered by its
own product divisions.
O R&D units have to ‘sell’ their services to the vari-
ous product divisions before they can undertake
a project.
The logic behind these sorts of arrangements is com-
pelling. If an automobile manufacturer is compelled
to buy engines from an inef?cient internal supplier
rather than the world leader in engines, the cars it
produces will not be competitive. Such mandated
sourcing relationships bear a remarkable resem-
blance to the old central-planning system in commu-
nist Russia, in which every company was told what
to make and who to sell it to. A much better
approach, one can argue, is to give the manufacturer
the option of buying engines from outside. This
ensures that the cars it produces will have the best
engines, and more to the point, it provides the
impetus that the internal engine manufacturer needs
to keep its products up-to-date and competitive.
There are dangers in the internal market approach as
well. An obvious concern is that internal markets cre-
ate competition between units, and thus hinder the
development of cooperation and resource sharing. A
second concern is that internal markets lead man-
agers to focus all their efforts on transfer pricing and
internal negotiations rather than adding value to the
end customer. The challenge, as always, is to design
an organization that gets the maximum bene?ts out
of internal competition whilst avoiding the worst of
the costs. This involves building a system that mot-
ivates people to share ideas and cooperate, but at the
same time to compete with one another. Not an easy
balancing act, of course, but one several ?rms –
including ABB, Hewlett Packard and Sharp – have
come some way towards achieving.
Once one accepts the basic logic of the internal mar-
ket, it becomes clear that the subsidiary’s role
changes dramatically. Using the terminology of econ-
omist Kirzner (1973), the subsidiary manager
becomes an entrepreneur, whose role is essentially
to seek out ‘hitherto unnoticed opportunities’ in the
internal market and act upon them. In so doing, the
subsidiary manager helps to resolve inef?ciencies in
the internal market and thus gradually push it
towards equilibrium. Of course, the subsidiary man-
ager still his or her ‘mandated’ job to do, such as sell-
ing the ?rm’s products in the local market, but
according to Kirzner this work is done alongside the
entrepreneurial work. Every individual, in other
words, has a latent dual role: manager and entrepren-
eur.
The internal market framework, and the concept of
the subsidiary manager as an entrepreneur, gives us
a good theoretical foundation for looking at the types
of internal initiatives I have identi?ed during my
research. While there are many ways of categorizing
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
them, Figure 2 provides a simple breakdown into
four groups, based on the nature of the business
opportunity and the level of head-of?ce support. The
opportunity can be of two types: it can be an existing
internal-market arrangement, or it can be an emerg-
ing business area. In the former case the subsidiary
initiative aims to recon?gure existing activities
within the ?rm; in the latter case it aims to enhance
the allocation of new activities.
3
In terms of head
of?ce support, the initiative is either sanctioned by
head of?ce or it is not. When it is sanctioned, that
means there is an existing set of rules or procedures
the subsidiary can follow. When it is not, that means
head of?ce mangers are either unaware of the initiat-
ive, or they are turning a blind eye towards it. Both
of these dimensions will be elaborated upon in the
paragraphs below.
Recon?guration Initiative
A recon?guration initiative can best be understood
as an effort by the subsidiary to alter the existing con-
?guration of activities within the ?rm to enhance
their ef?ciency. Consider the case of IBM in Scotland.
It began manufacturing PCs in Greenock in 1982.
Towards the end of the 1980s, though, plant manage-
ment realized that there was a limited future in just
being a ?nal assembly operation, so they began look-
ing for ways to extend their ‘charter’ i.e. the set of
activities they were responsible for at a corporate
level. In 1991 they identi?ed a small monitor devel-
opment group near London as a logical complement
to their manufacturing, and succeeded in getting it
relocated to Greenock. Subsequently, they identi?ed
two service functions, the order ful?lment process
(whereby orders are collected, organized and sent to
the factory) and help centres (e.g. for people with
installation problems) as Europe-wide activities that
could potentially be centralized in Greenock. Again,
they were successful.
The key point here is that the initiatives all resulted
in changes in the con?guration of activities for IBM
Figure 2 Types of Internal Initiative
European Management Journal Vol 16 No 3 June 1998 358
Europe as a result of efforts by the Scottish subsidi-
ary. Such changes could be – and often are – initiated
by head of?ce management, but it is fair to assume
that the management on location understand the
implications of such changes better than the head
of?ce managers 3000 miles away. Moreover, the IBM
case was not just about recon?guration of activities.
As Gordon Brown, the Greenock controller com-
mented, “by extending the value chain and linking
various activities together in one location, millions of
dollars were saved, customer satisfaction improved,
and market share increased”. The process, in other
words, created a new vitality in the initiative-taking
subsidiary, which led to a signi?cant performance
improvement.
Another important element of the recon?guration
initiative is that it typically results in a loser as well
as a winner. A Honeywell Canada manager, for
example, was touring the plant of a sister unit in the
US back in 1972. He observed their production pro-
cess for a control switch, and realized that his plant
back in Toronto “could do it signi?cantly better and
much cheaper”. He proposed such a change to head
of?ce management, and it was quickly passed, with
the result that the control switch production was
transferred up to Canada. But of course his col-
leagues in the losing plant were not happy – “I was
not welcome back” he admitted. Perhaps they should
have had the opportunity to improve their perform-
ance, but as this manager explained, “the corporate
philosophy at the time was: you could have done a
better job, so you should have done before somebody
took the work off you. And that’s what kept us com-
petitive.”
The IBM and Honeywell examples are cases where
the subsidiary has facilitated a recon?guration of
physical activities – production, R&D operations, and
help centres. But it is important to bear in mind that
many of the internal ?ows in a multinational corpor-
ation are of intangibles such as market knowledge,
technology and organizing principles. If we extend
the de?nition of an internal market to cover a market
for knowledge, then we also open up several other
forms of initiative. Consider, for example, the case of
3M Sweden which developed a high level of expert-
ise in customer focused marketing and key account
management. Gradually other subsidiaries within 3M
Europe got to know about this, so that when they
had questions about customer-focused marketing,
they called Sweden. 3M Sweden thus emerged as a
‘Centre of Excellence’, and the individual heading up
the Swedish marketing operation found himself trav-
elling extensively and disseminating his knowledge.
In addition a number of other centres of excellence
were also designated within 3M Europe.
The 3M story may appear very different from the
IBM and Honeywell examples, but it actually has a
number of common themes: it was inspired by the
subsidiary, it was sanctioned by head of?ce, and it
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
resulted in the more ef?cient use of existing
resources, i.e. 3M Sweden’s excellence in customer
focused marketing. The key difference, of course, is
that knowledge resources can be transferred without
loss to the transferor, whereas physical resource
transfers tend to be a zero-sum game.
Another example that emerges from looking at the
?rm as an internal knowledge market is Skandia
AFS, a Swedish insurance ?rm that specializes in
unit-linked life assurance products. Skandia AFS has
about 20 subsidiaries around the world which all act
very independently. However it also has four sub-
sidiaries called ISUs (international Sales Units) whose
role is to identify opportunities to sell products from
one country into other countries. The ISU subsidi-
aries, in other words, act as brokers in the internal
market, continuously evaluating the existing product
portfolios of sales subsidiaries, and making them
available to other subsidiaries if they look attractive.
There are, then, many types of recon?guration initiat-
ives. They differ substantially in the type of internal
market – in the IBM and Honeywell cases the market
was for corporate charters, in the Skandia and 3M
cases the market was for knowledge. But they all
serve one critical function, namely enhancing the
ef?ciency of the internal market without head of?ce
management having to become actively involved.
Maverick Initiatives
Maverick initiatives are similar to recon?guration
initiatives to the extent that they result in an enhance-
ment in the ef?ciency of the internal market. How-
ever, they are also undertaken without head of?ce
sanctioning, which makes things quite tricky when
the charter ‘steal’ involves another unit losing out.
Such initiatives appear to be the least common of the
four, and I was only able to identify two examples
during my research, both of which requested anon-
ymity. The ?rst was the Swedish subsidiary of Data-
com, a US-based computer manufacturer. Datacom
4
was struggling in the early 1990s because their pro-
ducts were not perceived as competitive. The Swed-
ish subsidiary, on the verge of bankruptcy in 1991,
was aggressively turned around by the new manag-
ing director in part through one simple change: he
began to sell and service competitors’ products, not those
made by Datacom. As the managing director explain-
ed,
‘the US staff concluded that we were prioritizing the sale
of competitors’ equipment. We were not of course – it was
not a choice of A or B, it was B or nothing! To say the least,
this was not understood. Many heated phonecalls, hostile
video-conferences and top level pressure were applied to
stop us selling competitors computers, and go back to sell-
ing what the factory could make’
Whether Datacom Sweden was right to sell competi-
European Management Journal Vol 16 No 3 June 1998 359
tors’ equipment is impossible to judge. But if we go
back to the internal market, it is logical to give the
sales subsidiary freedom to buy products from mul-
tiple sources if its competitiveness is otherwise
compromised. Given that most money in this indus-
try is made in service contracts and not in product
margin, it is possible that the Swedish initiative rep-
resented an important shift in mindset within Data-
com. Today, six years later, the Swedish managing
director has been forgiven for his rule-breaking, and
his turnaround is widely acknowledged within the
?rm as a success story.
The second example of this type of initiative emerged
from a disagreement between Alpha, a Swedish
multinational and its German subsidiary. The Ger-
man subsidiary was demanding some changes to the
product speci?cations in order to sell it in Germany;
the Swedes responded that they could not make the
requested changes in a cost-ef?cient manner. The
manager of the US subsidiary came to hear about this
dispute, and saw a great opportunity. He arranged
to buy several hundred products from Sweden, got
them shipped over to America, retro?tted them with
the German subsidiary’s requested changes, and
shipped them on to Germany, making a tidy pro?t
along the way. This story is a fascinating example of
subsidiary initiative, undertaken directly against the
wishes of head of?ce. It is also a very good example
of entrepreneurship in a Kirznerian sense, in that the
internal market between Sweden and Germany was
clearly failing. Through his alertness to a market
opportunity (in this case an unsatis?ed German
subsidiary), the US subsidiary manager was able to
step in and make a pro?t, and at the same time
enhance the ef?ciency of the internal market.
Of course many people reading this will see the
above two examples as blatantly subversive. Such
actions, they will conclude, lead to internal anarchy
and cause multinational ?rms to lose their strategic
focus. My argument is more balanced. There is cer-
tainly a ?ne line between initiative and subversion,
as the ?nal section of this paper discusses, and only
time will tell if the Datacom and Alpha initiatives
crossed that line. But in the interests of organizational
experimentation, these sorts of initiative are
important to recognize, because they can potentially
have signi?cant bene?ts, as well as obvious costs.
Bid Initiatives
Bid initiatives are those that are directed towards
emerging business areas for the multinational ?rm.
Consider the case of Monsanto Canada, whose top
management identi?ed an interesting proposal in the
corporation"s long-range plan in 1991. Monsanto was
developing a new formulation for its best-selling
agrochemical product, with the intention of bringing
it to market around 1996. Canadian management saw
a great opportunity to argue for the investment to
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
be made in Canada because of the country’s strong
agricultural industry. They persuaded corporate
management to bring forward the planned invest-
ment, and they lobbied hard to ensure that a Canad-
ian site was actively considered. After a careful
analysis of the three competing sites, the Canadian
one was selected.
It is important to compare the Monsanto Canada case
with the IBM case described earlier. Both ended up
following similar processes, in terms of identifying
possible extensions to the subsidiary’s business, lob-
bying head of?ce managers, and doing a careful cost–
bene?t analysis. But the key difference is that IBM
Scotland were pushing to recon?gure existing activi-
ties, while Monsanto Canada were trying to attract a
new activity. This difference is critical for one main
reason, namely that recon?guration initiatives rely
exclusively on the subsidiary to identify the opport-
unity whereas bid initiatives are usually identi?ed
jointly by the head of?ce and the subsidiary. In Hew-
lett Packard Canada, for example, I saw a couple of
examples of initiatives that were originally identi?ed
by the subsidiary but which were subsequently
fought out on an internal basis between the Canadian
subsidiary and a couple of US-based divisions.
The best way of making sense of bid initiatives, I
believe, is by thinking in terms of both the internal
and external market. While the opportunity itself
(such as Monsanto’s need for a new production
plant) is typically oriented towards the external mar-
ket, the subsidiary has to win the right to act on that
opportunity through the internal market. For that
reason, the process tends to resemble that seen in the
recon?guration initiatives – a lobbying process
directed towards head of?ce decision-makers as well
as other in?uential entities within the ?rm.
Moreover, the internal market perspective also indi-
cates another important dimension to bid initiatives
that was not observed in recon?guration initiatives,
namely the need for both push and pull strategies.
The push strategy was what was observed in Mon-
santo Canada, with the top management becoming
actively involved in the speci?c initiative. The pull
strategy is a little more gentle, in that it involves the
subsidiary ‘broadcasting’ its capabilities to in?uential
managers throughout the ?rm, in the hope that it will
be approached when the right investment opport-
unity arises. Take the example of Motorola’s East Kil-
bride plant in Scotland. This is one of 16 Motorola
plants around the world that fabricates silicon wafers
for semiconductors. All the key performance meas-
ures such as cost, service and quality are explicitly
compared on a monthly basis, and new investments
are (obviously) made in the top performing sites.
Because of this comparability, East Kilbride’s man-
agement do not need to think in terms of push stra-
tegies. Their approach, rather, is to stay ahead of
internal competitors on the key performance metrics,
and ‘pull’ new investments in as they occur.
European Management Journal Vol 16 No 3 June 1998 360
Push and pull are by no means alternative strategies.
While management at Motorola’s East Kilbride plant
can focus on the pull approach, they still have to be
very responsive and keen whenever new investments
are planned. Monsanto Canada worked hard on the
pull strategy in the late eighties by broadcasting their
interest in attracting investment, and then used a
push strategy when speci?c opportunities arose. But
the point is that the relative emphasis is likely to
vary, and this will be a function of the ?rm’s system
for managing new investments as well as the individ-
ual manager’s preferred style. For example, Ericsson,
the Telecommunications equipment ?rm, has man-
aged its international R&D activities through an
internal market process for many years. Ericsson’s
corporate culture is typically Swedish, in that it
places a great deal of emphasis on consensus build-
ing and group decision-making. As a result, R&D
charters are handed out through a rather interactive
bid process in which subsidiaries use mostly pull
strategies coupled with active top-down involve-
ment.
Finally, we can also extend the bid initiative termin-
ology to internal knowledge markets. Here the
emphasis is on emerging knowledge areas that cer-
tain subsidiary units actively pursue in order to
become thought-leaders. As before, the term most
?rms use for such a unit is a Centre of Excellence.
For example, consulting ?rms such as Andersen Con-
sulting and McKinsey encourage the various units
around the world to develop leading-edge knowl-
edge in emerging practise areas or industries. Often
such units are formally designated as centres of excel-
lence, with a view to their knowledge being dissemi-
nated throughout the ?rm. An alternative approach
is for the ?rm to use its information technology to
make this leading-edge knowledge available elec-
tronically. In both cases, of course, it is a question of
the ef?ciency of the internal market system (word of
mouth versus electronic) whether such knowledge
?nds its way to the individuals throughout the ?rm
who need it.
‘Leap of Faith’ Initiatives
The fourth and ?nal type of initiative I have called
‘leap of faith’ initiatives because it typi?es the atti-
tude of the managers responsible for them. Such
initiatives are directed towards emerging areas of
business, but they are undertaken without head
of?ce sanction. However, unlike the Maverick initiat-
ive discussed earlier this type is somewhat less con-
tentious because there is no obvious losing party.
Essentially it represents a bet by the subsidiary man-
ager on the emergence of a certain technology or
business area, in the hope that it will really take off.
A well-known example of this is NCR’s Dundee
operation. This plant was on the verge of closure in
1980, largely because its key product, the automatic
teller machine (ATM) had serious quality problems.
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
Jim Adamson, the newly-appointed general man-
ager, worked on improving manufacturing quality
and restoring the con?dence of major customers,
while at the same time developing a vision for
Dundee as NCR’s strategic centre for the ATM busi-
ness. Product development responsibility of?cially
lay with HQ in Dayton, but Adamson began
directing resources towards upgrading and renewing
the Dundee product line to meet the emerging
demands of the big British banks, Dundee’s key cus-
tomers. Faced with active resistance from the devel-
opment group in Dayton, Adamson pursued a delic-
ate strategy of co-operating with them while
continuing privately to sponsor Dundee’s inde-
pendent research program. A successful product
upgrade in 1982 was followed 18 months later by a
next-generation ATM that set new standards in func-
tionality, reliability and serviceability. Dundee’s glo-
bal market share reached 20 per cent in 1984. The
following year, responsibility for the global ATM
business was of?cially transferred from Dayton to
Dundee.
In this case, the leap of faith was not so much in
terms of the ATM market (which was already grow-
ing rapidly) but in terms of Dundee’s ability to com-
pete with IBM and a host of other established com-
petitors. Certainly, there were many within NCR
who doubted that Dundee would ever return to
pro?tability, let alone become a global leader in
ATMs, yet Adamson persisted in his vision and
eventually won out.
A second example of a leap-of-faith initiative is
Honeywell’s Scottish subsidiary in Newhouse, which
manufactures a range of control valves and related
items for the European market. During the seventies
and early eighties control valves were mechanical
devices, but the general manager of Newhouse
became aware of the impending shift towards elec-
tronic devices and he invested discretionary funds in
the development of an electronic manufacturing
capability. Thus, when in 1991 corporate manage-
ment decided to invest in a global-scale facility for
electronic fan-coils, Newhouse was the obvious
location because it had already developed the neces-
sary capabilities. This case was far less dramatic, but
it emphasizes the importance of a far-sighted subsidi-
ary manager who was willing to bet on the emerg-
ence of a certain technology. It also illustrates the
judicious use of discretionary funds in the subsidiary.
While not of?cially sanctioned by head of?ce, it
seems unlikely that such investments would have
caused trouble (unlike the maverick initiatives
described earlier).
In sum then, leap-of-faith initiatives represent an
important fourth category in which – like bid initiat-
ives – subsidiary managers are actively looking for
opportunities in the external market while at the
same time trying to get them resourced and legi-
timized in the internal market. The challenge in such
European Management Journal Vol 16 No 3 June 1998 361
cases appears to be one of effectively managing both
arenas. In the case of the Honeywell Newhouse man-
ager, for example, one must assume that he was con-
scious of the level of investment in digital systems
among the other Honeywell plants, as well as con-
scious of emerging market demands.
To summarize, this section should be seen primarily
as a descriptive account of the forms that internal
initiative can take. Each initiative type requires very
different tactics on the part of the subsidiary man-
ager, from the rather Machiavellian approach in mav-
erick initiatives, through to the by-the-book logic of
bid initiatives. And moreover, each suggests a very
different set of expectations on the part of head of?ce
management, such that in any given corporation
there may only be one or two of these types of initiat-
ive that could conceivably be pursued. Notwith-
standing the differences, it seems that the similarities
are also quite striking. All four require subsidiary
managers to be entrepreneurial, looking for new
ways of adding value and proactively approaching
head of?ce with a proposal, and all four require a
much greater level of awareness and openness on the
part of head of?ce managers to ‘foreign’ ideas than
is normally the case. Clearly the sort of cultural shift
I am talking about does not come easily, but it is a
gradual process and one which many of the ?rms I
have studied seem to be getting to grips with. But
one of the biggest stumbling blocks is a simple and
fundamental scepticism towards corporate
entrepreneurship. So it is to this issue that I now turn.
The Costs of Corporate
Entrepreneurship
As I observed at the outset, many managers remain
suspicious of corporate entrepreneurship. They are
suspicious in the ?rst instance because it sounds like
an oxymoron: perhaps an attractive idea in principle,
but something which just does not happen very
often. But the second reason for suspicion, and the
one I want to address here, is that corporate
entrepreneurship has signi?cant and often hidden
costs. Entrepreneurship is, most of us would agree,
an important engine of growth and renewal whether
it occurs inside the ?rm or in the economy as a whole.
But as with most things, it is possible to have too
much entrepreneurship. Below I have listed four of
the principal costs of corporate entrepreneurship that
I think managers need to be cognizant of, and a num-
ber of suggestions about how those costs can be man-
aged.
Empire building. We are all familiar with the stories.
The country manager in France, for example, con-
vinces head of?ce that he needs to build a local
manufacturing operation to keep the government
happy. A few years later he comes out with some
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
product changes that are ‘essential to our continued
success in France’ but which cut against the world-
wide positioning of the product. He then argues that
a larger development team is needed to support the
new product line, because the R&D people back at
head of?ce do not understand the changes that have
been made. And so on.
Is the French country manager cleverly adapting the
product to the unique French marketplace, or is he an
empire-builder, interested only in building his own
power base at the expense of the ?rm? Similarly, if
we take the Datacom Sweden manager or the US sub-
sidiary manager of Alpha mentioned earlier, where
do their priorities lie? The answer to these question
depends essentially on the preconceptions of head
of?ce management. One preconception argues that
subsidiary managers are prone to act opportun-
istically and build their own little empires. The other
is to trust subsidiary management, and to believe that
they are acting in the best interests of the corporation.
In my experience, the ‘facts’ of the speci?c case can
be interpreted to reach whichever conclusion ?ts the
manager’s preconceptions. Obviously in a few cases
there is solid evidence that the subsidiary’s actions
were well-intentioned or oriented towards empire-
building, but most cases fall in the grey area where
opinion is what matters.
So what can be done to guard against empire build-
ing? My opinion is that the head of?ce – subsidiary
relationship has to be built on trust, otherwise the
strict control systems that ensue will make opportun-
istic behaviour a self-ful?lling prophecy. However,
that trust-based relationship must also be tempered
with a system for evaluating and monitoring the sub-
sidiary’s initiatives, to ensure that they are consistent
with corporate goals. And it also needs to be
undergirded and reinforced by a strong corporate
culture that allows head of?ce managers to be con?-
dent about the abilities and objectives of its subsidi-
ary managers.
Lack of focus. This problem is related to empire-build-
ing, in that it is also the result of ‘too much’ corporate
entrepreneurship. Imagine that every employee takes
the idea of corporate entrepreneurship to heart, and
follows up on their new idea. Soon the ?rm is getting
into new businesses in every conceivable direction,
with the result that ?nancial resources are stretched
too thin and there is no coherence to the ?rm’s pro-
duct line. Hewlett Packard reputedly found itself in
this position in the late 1980s, because it had actively
encouraged individual initiative but it had not put in
place the necessary controls. Changes were made to
reduce the autonomy of subsidiary managers, but
inevitably this approach also choked off the ?ow of
initiatives, with the result that the company probably
moved too far the other way.
Corporate entrepreneurship can end up being pur-
sued outside the ?rm. Ethicon, Johnson & Johnson’s
European Management Journal Vol 16 No 3 June 1998 362
Scottish subsidiary, is a good example of this: it has
on numerous occasions spun off businesses that did
not ?t its strategy, several of which have gone on to
become major ?rms in their own right.
Costs of administrating the internal market. Unlike the
market economy which appears to work best when
left alone, the internal market system needs a con-
siderable amount of administration to work effec-
tively. One set of administrative costs result from set-
ting up and maintaining the internal market system,
e.g. de?ning the conditions under which sourcing
relationships can be challenged, and setting rules for
transfer prices. There are also transaction costs that
are accrued by the participating units every time an
initiative is undertaken. Such costs are not only
higher than would be the case in a ‘centrally-plan-
ned’ organization, they are also potentially higher
than in an external market. In one case I studied, for
example, head of?ce management had decided
against sending out a Request For Proposal to subsidi-
ary companies because they did not want to spend
the time going through all the proposals and justify-
ing their decision to each one in turn. They chose,
instead, to create their own shortlist of two or three
subsidiaries, and evaluate them in detail.
As with the last point, there is a tricky balance here
between maximizing the level of subsidiary initiative
and keeping the cost of managing it under control.
If head of?ce managers draw up their own shortlist
of subsidiaries, they keep the process under control
but they may send the wrong signals to other entrep-
reneurially-minded subsidiary managers. Again, the
rule of thumb appears to be one of keeping a very
broad perspective at ?rst, but fairly quickly selecting
out those initiatives that are uncompetitive or not
aligned with the ?rm’s needs.
Coping with internal ‘unemployment’. The ?nal cost of
corporate entrepreneurship is that it can result in
unemployment, either literally or ?guratively. If we
think about the market economy for a second, it is
generally accepted that closing an inef?cient coal
mine is good for the economy as a whole but results
in hundreds of layoffs in the old coal town. The same
can be argued for the internal marketplace. If we con-
sider the Honeywell case discussed earlier, the
decision to move a production line from Minneapolis
to Toronto was good for Honeywell as a whole
(because it resulted in more ef?cient production) but
resulted in the loss of some jobs in Minneapolis. This
leads to certain costs for the ?rm. Either the
employees in question are laid off, which results in
a one-time cost to the ?rm and more importantly the
loss of some well-trained experienced people, or they
are retained by the ?rm and assigned new responsi-
bilities, a process which inevitably takes time.
Of course, the costs of closing down factories and
reallocating people occurs in all ?rms, but the point
to emphasize here is that a system which encourages
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
the use of an internal market will experience much
more frequent changes in the allocation of activities
to operating units, and will therefore suffer from a
higher level of ‘internal unemployment’. Just as the
central planning system in Russia was able to boast
full employment until it all fell apart, centrally-
planned ?rms can keep the number of unassigned
workers to a minimum for a long time, and then pay
the price later through massive job losses. The costs
of internal unemployment, in other words, are quite
substantial, but probably worth paying because they
are better than the alternative. The managerial chal-
lenge then becomes one of managing the reassign-
ment of employees to other parts of the ?rm, and of
training them so that their skills match future needs.
Concluding Comments
This paper has focused on the perspective of subsidi-
ary managers, both at a conceptual level, and with
regard to the speci?c initiative strategies they can
pursue. My assumption is that many ‘head of?ce’ ori-
ented managers will have found my approach worry-
ing, because my implicit advice is that subsidiary
managers should take more initiatives, even initiat-
ives that go against the of?cial sanctioning of head
of?ce. In this ?nal paragraph I want to allay such
suspicions slightly, by emphasizing that I would like
the ideas presented here to be seen as descriptive of
an interesting phenomenon rather than a how-to
guide for subsidiary managers. The fact is that we
simply do not know if subsidiary initiatives, of the
type described here are really good for the multi-
national ?rm. Obviously I have chosen to focus on
success stories, but for every case like NCR Dundee
there are a large number of failures in which money
was wasted on a poorly-thought-out investment, or
precious time was spent arguing the pros and cons
of a marginal idea. Rather than offering simple rules-
of-thumb about what level of initiatives should be
encouraged, I would prefer to argue that the appro-
priate level is extremely dependent on the speci?cs
of the ?rm and industry. In some of the companies
I have studied, head of?ce managers have attempted
to suppress the entrepreneurial endeavours of their
subsidiary managers, for fear of being swamped with
proposals. And in others, I have heard head of?ce
managers complain that their subsidiary managers
do not take enough initiative. Getting the level of cor-
porate entrepreneurship right is clearly a process of
mutual adjustment, and one that is unlikely to be
achieved without some pain.
The purpose of this article, then, was to provide a
European Management Journal Vol 16 No 3 June 1998 363
way of thinking about the different forms that
internal initiatives can take, and to explain their
likely impact. My hope is that it will have stimulated
both subsidiary and head of?ce managers to re-
evaluate their approach to corporate
entrepreneurship, and to look with fresh eyes on its
possibility for enhancing the ef?ciency of their
internal market systems.
Notes
1. There was a wave of enthusiasm for corporate
entrepreneurship in the mid-1980s, fuelled by Rosabeth
Moss Kanter’s book The Change Masters (1985: Simon and
Schuster, New York) and Gifford Pinchott’s Intrapreneuring
(1985: Harper and Row, New York). As I read the litera-
ture, we can see something of a second-wave of enthusiasm
at the moment. Notable recent contributions include, Gho-
shal and Bartlett (1997); Birkinshaw (1997); Block and
McMillan (1996), and Ginsberg and Hay (1994).
2. On the issue of internal markets, see also the work of
Halal (1994).
3. One important point of clari?cation is in order here,
namely that initiatives directed towards new market
opportunities are really ‘internal – external hybrids’ in
terms of Figure 1. More speci?cally, while the ‘locus of
opportunity’ is external to the ?rm, it is typically the case
that head of?ce management identi?es the opportunity, so
that the ‘locus of pursuit’ of the initiative is internal.
Viewed in this way, such initiatives are still part of the
internal initiative phenomenon.
4. Datacom and Alpha are disguised names.
References
Birkinshaw, J. (1995) Entrepreneurship in multinational cor-
porations: the initiative process in foreign subsidiaries.
Unpublished doctoral dissertation, University of West-
ern Ontario.
Birkinshaw, J. (1997) Corporate entrepreneurship in multi-
national corporations. Strategic Management Journal 18,
207–229.
Birkinshaw, J. and Fry, N. (1998) Fighting the corporate
immune system: initiative strategies for subsidiary man-
agers. Sloan Management Review (forthcoming).
Birkinshaw, J. and Ridderstråle, J. (1996) Fighting the corpor-
ate immune system: a process study of peripheral initiat-
ives in multinational corporations. In Proceedings of the
European International Business Academy Annual Confer-
ence, Stockholm.
Block, Z. and McMillan, I. (1996) Corporate Venturing.
Ghoshal, S. and Bartlett, C. (1997) The Individualized Corpor-
ation.
Ginsberg, A. and Hay, M. (1994) Confronting the challenges
of corporate entrepreneurship. European Management
Journal 12(4), 382–389.
Halal, W. (1994) From hierarchy to enterprise: internal mar-
kets are the new foundation of management. Academy of
Management Executive 8, 69–83.
Kirzner, I. (1973) Competition and Entrepreneurship. University
of Chicago Press, Chicago.
CORPORATE ENTREPRENEURSHIP IN NETWORK ORGANIZATIONS
JULIAN
BIRKINSHAW, Insti-
tute of International
Business, Stockholm
School of Economics, PO
Box 6501, S113 83,
Stockholm, Sweden.
Julian Birkinshaw is
Assistant Professor of
Business at the Institute
of International Business
in Stockholm. His research is focused on the internal
workings of large multinational corporations,
including issues of entrepreneurship in foreign sub-
sidiaries, knowledge transfers between units, and co-
ordination of international marketing operations. He
received his PhD from the Richard Ivey School of
Business, University of Western Ontario.
European Management Journal Vol 16 No 3 June 1998 364
doc_222565135.pdf