Wffi'ffi

processes
Knowledge in
global
multinationals

lntroduction

This chapter examines the dynamics of knowledge processeswithin large, global

multinational corporations.The focus here is not on how companiesgrow to become
global, but on the knowledge dynamics within already large, internationalized otganiza-
tions. This representsan interestingand important context for the examination of such
issuesfor a number of reasons.
Firstly, the economic importance of such organizations grew signiflcantly in the last
quarter of the twentieth century. Driven by a combination of interrelated processessuch as
market deregulation,rapid advancesin information and communication technologies,and
growth through meIgeIS and acquisitions, not only has there been a process of
globalization, whereby more and more companiesare becoming globally active, but there
has alsobeen a growth in the number of large organizations,and in the sizeof alreadylarge
organizations(Carchedi1991,ch. 7; Korten 1995;WIR 1999).Von Krogh et al. (1996)char-
acterize this trajectory as involving a two-stage evolution from international to multi-
national flrms, and from multinational firms into global firms. Korten (1995,I21), using an
element of hyperbole, suggeststhat this changehas been so signiflcant that it represented,
,the most rapid and sweepinginstitutional transformation in human history'.
Exemplars of the type of company considered here include: Ernst & Young, the
professionalservicecompany which employs over 105,000people,who work from over
670 locations in over 130 countries; Boeing the aerospacecorporation which employs
over 160,000workersin thirty-eight statesof the USAaswell a seventycountriesglobally;
and IBM the computer company which employs over 300,000 staff in more than
I 00 differentcountriesglobally.T

whichhassitesthroughoutthe
organization
A largemultidivisional is global
worldandwhosebusiness
in character.

7 This information was taken from the following websites on 16 June 2003 (rvwr,v.EY.com,
wn'wBoeing.com,rvwwlBM.com).
 I Z A T I O N ACLO N T E X T S
ORGAN

Secondly,global companies have typically been in the vanguard of attempts to develop

knowledge management solutions/systems, have been some of the most enthusiastic
advocatesof the benefitsof knowledgemanagement,and have generallybeen the earliest
at realizing the potential of knowledge management (KPMG 2000; McAdam and Reid
2001). Thus, for example on Boeing'scorporatewebsite (r,l'vl'w.Boeing.com)8
the third of its
three corporate objectivesis to, 'sharebest practice and technology acrossbusinesses'.
Further, as illustrated by the example considered in Chapter 12, Shell International
Exploration and Production (SIEP)have also become aware of the benefits of internal
knowledge-sharing.
Finally, as will be become apparent as the chapter progresses,becauseglobal multi-
national organizationshave highly dispersedand fragmentedknowledgebases,employ
large numbers of employees,and involve the communication and interaction of people
with diverse sociocultural beliefs, the dynamics of knowledge processesin such organiza-
tions are quite particular.
In examining knowledge processesin this context the chapter links together issues
already examined, with some new themes. Specifically,issuesreturned to include the
distributed nature of organizational knowledge, the d;mamics of knowledge-sharing
acrossboundaries,and how social factors affect knowledge processes. Thesetopics are
then linked with two themes not examined thus far organizational size and cross-
cultural knowledgeprocesses.
The chapter is structuredinto three sections,with the issuesexamined in eachbeing
illustrated and supported by different examples.The first section examines the relation-
ship between the structuring of multinational corporations and the knowledge dynamics
within them. Following this, the secondsectionconsidershow organizationalsizeaffects
the dynamics of organizationalknowledge processes,and generallyconcludesthat the
greaterthe sizeof the organizations,the more complicatedits knowledgedynamics are
Iikely to be. The third section then concludesthe chapter by examining cross-cultural
knowledge processes,and considershow the socioculturalvalues that people possess
affectorganizationalknowledgeprocesses.

The structuring of multinationalsand knowledge processes

As outlined in Chapter 3, the practice-basedperspectiveon knowledge assumesthat

becausethe knowledge people possessis closely linked to the physical and cognitive
activities they undertake, and is embedded in the social context in which such activity
occurs, the knowledge base of all organizations will be fragmented into specialist sub-
communities. From this perspectives,one of the key and most difficult tasksof manage-
ment is to link together and coordinate the organizational knowledge base (Brown and
Duguid 1998; Grant 1996; Kogut and Zander 1992;Tsoukas1996).While the objectivist
perspectivehas a different conceptualization of knowledge, and typically assumesthat it
can be more easily codified, those analysesof knowledge utilizing such a perspective

8 Siteaccessed
16June2003.
 SI V U L T I N A T I O N A L S
K N O W L E D GPER O C E S S EI N

also acknowledge the complexities of sharing and integrating knowledge within

multinational corporations (Gupta and Govindanjan200l'; Szulanski 1996;Tsai2O01).
In global multinational corporations such as those outlined in the introduction to this
chapter, which may employ tens of thousands of workers, operating from possibly
hundreds of different sites,dispersedacrossthe globe, the task of coordinating and inte-
grating organizational knowledge is non-trivial. However, the flip-side of having such a
diversified knowledge base is that it createsenormous potential for qmergy to be devel-
oped through intra-organizational learning and interaction (Macharzina et al. 2OO1;
Morosini 1998; Soderberg and Holden 20O1.;Yan Maanen and Laurent L993). Thus
possessinga large, fragmented knowledge basehas both potential benefits and problems.
One way of supporting and facilitating intra-organizational knowledge processesis
through the structuring of organizations. However, the diversity of ways in which MNCs
are structured in practice suggeststhat there is no consensuson the best way to facilitate
intra-organizational knowledge-sharing.The rest of this section considersthe knowledge-
sharing implications of utilizing two particular structural forms: a centralized,hierarchical-
basedstructure and a decentralizednetwork structure.

The centralized hierarchicalstructure

This means of structuring a multinational corporation assumesthat the home baseof the
corporation, the country out of which a multinational originates,provides a platform and
a foundation from which global advantagecan be achieved (Macharzina et al' 2001').In
such organizations global expansion occurslargely through taking advantageof the home
bases,capabilities,which are developedfrom, basedin, and exploit national and regional
systemsof innovation. Porter (1990), for example, suggeststhat such a logic is highly
prevalent. Laurent (1983, 1986) also supports such a perspective,and arguesthat all multi-
nationals to some extent bear the stamp of the country from which they originate.
Basedon this model the corporate centre, which will be basedin the home country is
typically largeand where not only the vast majority of strategicdecisionsaremade,but where
researchand developmenttype knowledge-creatingactivitiesarealsolocated(seeFigure13.1).
Within such organizations knowledge flows unidirectionally, ftom the corporate centre out
to the organizations businessunits, which arelargely responsiblefor applying this knowledge
to their local market context. Finally, another characteristic of this model is that there are few
independent interconnections between different businessunits. The importance of such a
structural logic is also reinforced by those wdters who suggestthat the extent to which multi-
nationals aretruly intemational hasbeen exaggerated,and that the majority of multination-
als are still home or region centred(Hirst and Thompson 1999;Rugman2000).

To what extentarecontemporary corporations

multinational independent of the countriesin which
lf
they originated? you corporations
multinational
compare that originatedin the USA,UK,France,
Germany,Russia,China,Japan,etc, canyou discerndifferences in the way they operatethat are
relatedto theircountrYof origin?
@ O R G A N I Z A T I O N ACLO N T E X T S

CORPORAT CEE N T RIEN ' H O M E 'C O U N T R Y

Responsibilities:
o strategicdecision-making,
. knowledgecreation

knowledge
Unidirectional f lows:
Centre>> Periphery

A U T O N O M OB
UUSS I N E S S A U T O N O M OB
UUSS I N E S S
. Knowledge application . Knowledge application

NOinter-unit
knowledge-sharing

A U T O N O M OB
UUSS I N E S S
. Knowledge application

F i g . 1 3 . 1 . A c e n t r a l i z e dh, i e r a r c h i c asl t r u c t u r ef o r m u l t i n a t i o n a l s

Dell: codification strategy and centralizedstructure

Hansen et al. (1999),as discussedin Chapter9, suggestedthere were two broad knowledge
managementstrategiesthat companiescould pursue:a codificationor a personalization
strategy.
Dell,the computermanufacturingand retailingcompany,was one companythey describedwhich
strategy.Such a strategyis lT-based,and involvesthe codification
followed a codification-based
of knowledgeinto searchablerepositories(seeTable9.1).With such an approach,the knowledge
in the repositorycan easilybe reusedby anyone.Dell combinesthis type of knowledgemanage-
ment strategy,with a centralized,hierarchicalstructure.Dell utilizesa knowledgerepositoryto
sell computersdirect to their customers,who define the specificationof their machines(either
on the web or via a telephonecall with a customer sales assistant)through selectingcompo-
nents from the knowledge repository.Dell, which has over 34,000 employees worldwide,
spread across thirtv four different countries, utilizes a centralizedcorporate structure.Thus
national/regional
offices,whose main responsibilities are for sellingcomputers,or providingafter-
salessupportand servicingto customers,do not havemuch of a role in strategicdecision-making,
and are more concernedwith administrationand knowledge applicationthan with knowledge
creation(suchas designingor managingthe lT-basedknowledgerepository).

ls therea riskwith sucha strategythat Dellwill be lesssensitiveto the particular

demandsof local
marketsthan if it useda moredecentralized strateov?
 K N O W L E D GPER O C E S S E
I NSM U L T I N A T I O N A L S

The decentralizednetwork structure

As Chapter 12 shows,the network logic for the structuringof multinationals is currently

extremely influential, with the path-breaking work of Ghoshal and Bartlett doing much
to initiate this way of conceptualizingthe internal structureof multinationals (Ghoshal
and Bartlett 1990;Bartlett and Ghoshal 1993).
With a network-based structure, in contrast to centralized hierarchical structures,
knowledgecreationis not the soleresponsibilityof the corporatecentre,with there being
multiple centres of knowledge creation (seeFigure 13.2). Secondly,knowledge can flow
equally in both directionsbetweenthe corporatecentreand businessunits. Thirdly, there
are many interconnectionsbetween interdependentbusinessunits, with a diversity of
mechanismsbeing used (such as staff transferralsbetweenunits, matrix structures,etc.)
to facilitate such interactions.Finally, these are typically complex organizationalstruc-
tures that don't have a clearhierarchy.Hedlund (1986,1994)usedthe term'heterarchy'
to describethis structuralform. Primarily,within a network structure/businessunits are
not controlled in a top-down way by the corporatecentre.
This structuralform, asillustratedin Chapter 12,has a number of advantagesover hier-
archical structuresin terms of knowledge processes.Primarily, the network structure more
effectively facilitates the sharing of knowledge between businessunits (Tregaskis2003;
van Wijk and van den Bosch 2000). Grant (1996), in his development of the knowledge-
basedtheory of the firm, also suggeststhat hierarchicalcoordination is bad for sharing

BUSINESS BUSINESS
. Knowledge
creation o Knowledqe
creation
. Knowledge-sharing . Knowledde-sharing

CORPORATE CENTRE
Responsibilities:
o Facilitate
knowledge-
sharing
. Someknowledge

lnterbusiness
interaction
and
Interbusiness
interaction
and knowledge-sharing
knowledge-sharing

BUSINESS
. Knowledge
creation
r Knowledge-sharing

Fig.13.2.A networkstructurefor multinationals

O R G A N I Z A T I O NCAOLN T E X T S

and integrating knowledge.Grant, in what amounts to a knowledge-basedjustification

for organizational delayering argues that the disadvantage of hierarchical structures is
that they are ineffective at sharing tacit knowledge, as they primarily utilize systemsof
ruies and regulations to coordinate activity and integrate knowledge, which are poor for
sharing tacit knowledge. The most effective means of sharing such knowledge is through
processesof direct interaction between people, where there are minimal levels of
hierarchy, such as in the network form.

N. V. Philips:the network structuring of a multinational

G h o s h aaln dB a r t e t t( 1 9 9 0u) s et h e D u t c he l e c t r l c ag lo o d sc o m p a n ay s a n e x e m p l aorf h o w a

multinational canbe conceptualized as a network.Philips canbe considered to be a multinational
company as it hasoperating unitsin oversixtycountries worldwide. Whilethe company's corpo-
r a t eb a s ei n H o l l a nids u n d o u b t e dtlhye s i n g l em o s ti m p o r t a nhtu bi n t h eo r g a n i z a t i o n ea tl w o r k ,
Philips's structureis closerto a networkthana hierarchy. Thus,manyof its businessunitsare
extremely large,constituting someof the largestorganizations inthe counties theyarelocatedin.
Therearealsodifferentcentresfor research anddevelopment. Thus,manyof Philips's business
unitsarenotsimplyresponsible for the application of knowledge createdat thecorporate centre,
but haveknowledge creationresponsibilities as well.Finally, thereis alsoa diversity of intercon-
nectionsbetweenbusiness units,facilitating the sharingof knowledge betweenthem.Thusfor
examplethe business unitsin regionssuchasAfrica,Europe, theAmericas, andAsia-Australasia
arelinkedtogetherin regional networks.

A contingencyperspectiveon structure

Birkinshawet al. (2OO2), in an interestingarticle that deservesto be widely read,provide

an analysisthat challengesthe logic that network forms of organization representthe
most effective way of organizing multinational companies in everysituation. Overall their
analysistakesa contingency-basedperspectiveto organizationaldesign,and concludes
that the design of an organization'sstructure should account for the characterof its
knowledgebase.Their analysisconsideredhow the level of observabilityand the degree
of system embeddednessof an organization'sknowledge were linked to the degreeof
autonomy and integration betweenbusinessunits. Observabilityrefersto the easewith
which an activity can be understood by simply looking at an organizational process,or its
products, whereassystem embeddednessrefersto the extent to which knowledge is a
function of the systemor context in which it is developedand used.
While their analysiswas based on researchinto the R&D activities in a handful of
Swedish multinationals, making it difficult to generalize widely, they found a strong
relationship to exist between organizationalstructureand the degreeof systemembed-
dednessof organizationalknowledge.Specif,callythey found that the degreeof system
embeddednessof organizationalknowledge was inversely proportional to the level of
 K N O W L E D G EP R O C E S S E ISN M U L T I N A T I O N A L S @
High

TRANSPARENT ISOLATED

Knowledge
observability

INTEGRATED OPAQUE

Low
Low High
System-embeddedness
of knowledge

knowledgebases(from Birkinshawet al.20021

Fig.13.3.A typologyof organizational

inter-unit integration. Therefore, when the knowledge in an organization is highly

system embedded, the level of inter-unit integration is likely to be low, due to the
difficulties and problems involved in sharing such knowledge.
Basedon the two dimensionsof knowledgethey utilized they developeda typology, char-
acterizingthe knowledgein R&D units into four generictypes (seeFigure 13.3).Birkinshaw
et al. (2OOZ) suggestthat different structuresarethus likely to be appropriatefor eachtlpe of
knowledge base. Extrapolating from this framework it could be argued that hierarchical
structuresaremost appropriatewhen organizationalknowledgeis'transparent' (i.e.when it
has a high-level observability and a low level of embeddedness)as in such circumstances
knowledge can be relatively easilycodifled and shared.Further,network forms of organiza-
tion may be most appropriate when organizational knowledge is 'integrated' (i.e. when
knowledgehas a low ievel of observability,and a low level of embeddedness), asthe effective
sharing of such knowledgerequiresextensiveand direct socialinteraction between people.
Overall therefore this section has outlined two different ways in which multinational
companiescan be structured, and shown how the knowledge dlmamics within them vary
substantially.In general,network structuresaremore conducive to processesof knowledge-
sharing/searching than hierarchical structures. Further, drawing on Birkinshaw et al.'s
(2OOZ)analysisit was concluded that the most appropriate structure for a multinational
corporation to adopt may depend on the dominant characteristicsof their knowledgebase.
This analysis has significant managerial implications, as it suggeststhat in the develop-
ment of businessand knowledge strategies,as well as the design of organizational struc-
tures, attention requiresto be paid to the characterof the organizational knowledge base.

Organizationalsize and knowledge processes

As far back as L987, Whitley suggestedthat organizational sizerequired to be taken more

seriously as a variable of analysis in the study of organizational behaviour. However, in
general terms, his call has gone unheeded. The literature on knowledge management is
 I Z A T I O N ACLO N T E X T S
ORGAN

no exception in this respect,as the relationship between organizational size and the
dyramics of knowledge processeshas in general terms been neglected (exceptions
include Fenton and Pettigrew2OOOb; van Wijk and van den Bosch2000; Becker2OOl;and
Forsgren1997).
As has been discussedextensively elsewherein the book, the typically fragmented,
specialized,and dispersednature of the knowledgebasein most organizationsmeansthat
one of the key tasksfor management is to coordinate and integrate organizational knowl-
edge. In general,as organizational size increases,the more complicated the processof
knowledge coordination becomes,as the organizationalknowledgebasebecomesmore
and more fragmentedand dispersed.Drawing on Brown and Duguid's (1991) metaphor
'community of communities', the more organizational (sub)
of an organization as a
communities that exist, the more likely it is that processof coordinating and facilitating
their interactions will increasein complexity.

Rabobank:the knowledge dynamics in an expanding network

VanWilk and van den Bosch(2000)studiedthe evolutionin the structuring of Rabobank, the
Dutch-base bd a n k i nagn df i n a n c i as le r v i c e . y t h e l a t e1 9 9 0 s
c o m p a nby e t w e e n1 9 B Ba n d1 9 9 7 B
it employed 44,000workerswith operations in over100countries. Duringthe time studied,due
to a varietyof externaland internaldrivers,it evolvedits internalorganizational structureaway
from a hierarchical one towardsa network-based structure.Partof the catalystunderlying this
evolution was thatthe mergersandacquisitions undertaken by Rabobank increased the sizeof
the organization suchthat businessunitswere increasingly at arm'slengthfrom the corporate
centre,andalsoincreasingly didn'tknowwhererelevant knowledge was located.
Theutilization of a networkstructure, it wasfelt,wouldhelpaddress theseproblems. However,
the largesize of the organization was found to make difficult the development of a single organ-
izational network.Therelationship betweenorganization/network sizeandthedynamics know- of
e n a s m a l l esrc a l ei,no n eb u s i n e susn i tS P E C T R U w
l e d g ep r o c e s s ewsa sa l s ov i s i b l o Mh, i c hh a d
beenin the vanguard of developing andimplementing the network-based structure. Duringthe
time that this divisionwas studied(approximately six years),it grew from havingonly thirty
employees to having350.The expanding sizeof the SPECTRUM divisionsignificantly affected
patterns of horizontal communication betweenstaffworkingin its differentproductareas,which
is oneof the characteristic elementsof a networkstructure. In general, the increasing sizeof the
organization inhibitedhorizontal communlcation. Thus,when the division had been relatively
small,suchcommunication was widespread, but as the divisiongrew it becameincreasingly
uncommon, with eachproductgroupbecoming moreandmorecompartmentalized.

What implications of networkstructuresto large

do theseflndingshavefor the relevance
organizations?Do they meanthat the knowledge+elated benefitsof usingnetwork-based structures
diminishwith increasing size?
organizational
 INMULTINATIONALS
K N O W L E D GPER O C E S S E S

pos-
of interactionsthat are
Howevet,incleasingolganizationalsizedoesnotsimplymaketheprocessofcoordination
it can funiamentally alter the type and character of
more complex, character
dir.";;;;" chapten 3-6, the tacit and context-dependent
sible. As has been knowledgerequiresextensive
-.un, ,t at effectivelysharing
most organizationatt<now*teig" peopleto be willing to
in1 context where enough trust existsfor
socialinteractions to occur in chapter 6' this is
particularly
in such . nr"."rr. As was extensivelydisLssed
participate
thecasewhenknowledgehastobesharedbetweenpeoplewhoarenotmembersoftheSame
community,astheymayhavedifferentvaluesystemsandlimitedcommonknowledge. necessaryfor such knowledge
the type or 'otiui
Developing u"o "rutions well as Gargiulo and
"""u"tlg Thus Hansen(Iggg), as
processes to be effectFJrriiir"-.""*ming.
Benassi(2000),alguethatsustainingstlongsocialrelationships'requirescontinuous
pelson can
relationships that any
interactionsbetweenpeopleandasustainedrec-ipro.ut.^.t'uns;.ottlowledgeandinfor-
Thus there is a limit to the numb", oi,,,.h researched by
mation. company
tlme' For example' in Kappa' the global R&D
sustain at any one effective
done by staff to sustain
ttl' 'n" ulnot"" or iJ*lliig
orlikowski f'"" ct'upit' organizational size
to tt problem of burn-out' Thus as
social relations cont;t"J "
incleases,sodoesthepotentialproblemsanddifficultiesofsustainingrelationshipswith
allthepeoplewhomayhaverelevanttnowreogeandexperience(seeFigurel3.4fora
of this process)'
graphical representation

network
5mall,cohesive
(a)
Strongtie
Weaktie
O Person

holes
(b) Largenetworkwith structural

'o

of different size
relations within networks
Fig. 13.4. Typical social
@ O R G A N I Z A T I O N ACLO N T E X T S

Table 13.1. Knowledge-relatedbenefitsand dlsadvantages

of cohesivenetworksand
networkswith structuralholes

Cohesivenetworks Network with structuralholes

Characteristics Tghtlyknit networks,where long- Networkswhere interpersonal

establishedsocialrelationsexist,strong connections are loose,limited
normshavedeveloped, and highlevels normsexist,and interpersonal
of interpersonal
trustexist trust is limited
Advantages Createsan environmentconducive
to Providespeoplewith accessto a
knowledgesharingand cooperation
within wide rangeof knowledgeand
tne network whichmakespeople
information
opento changeanda diversityoI
vrewpoints
Disadvantages Createsa potentlalrigidity,due to the Knowledge-sharing and social
effortrequiredto sustainnetwork(sustain interaction
inhibitedand slowed
norms,reciprocate whereexpected), which down by a lackof cohesiveness
may hinderpeople'sabilityto adaptthrough andestablished socialnorms
limitingthe rangeof knowledgeand
information they utilize

Becker(2001)referredto this as the problem of 'largenumbers'. Beckerarguesthat the

typically dispersedcharacterof an organization'sknowledge base createsthree funda-
mental problems/issues for organizationalmanagement,one of which is the problem of
largenumbers.This problem stemsto two factors.Firstly,there is the issueof opaqueness,
or intransparency,which refersto the difficulties of developing an overview when knowl-
edgeis fragmentedand dispersed,which is a problem that increasesasthe level of disper-
sal or number of fragments increases.Secondly,is the issue of resourcerequirements
involved in bringing together the fragmentsof a dispersedknowledge base,which is a
problem that again increasesproportionally with organizational size.Thus, for Becker,
there is a direct relationship between organizational size and the difficulty of managing
and integrating an organization's knowledge base.

Dueto the amountof work involvedin sustaining them, is therea limitto the numberof strongties
that peoplecan have?lf so, what is the approximate
sizeof this limit-5 , 10,20,50,more?

Connecting theseinsights to the work of Gargiulo and Benassi(2000)it can be argued

that the type of network relations that people can have will vary with organizational size.
Gargiulo and Benassicontrast the advantagesand disadvantagesin terms of knowledge
searching and acquisition of cohesive networks compared to networks with structural
holes (seeTable 13.1). In Gargiulo and Benassi'sanalysisthe type of network that any
individual possesses is determinedby personalchoice. However,the difficulties outlined
aboveof trying to support a largenumber of strong,closesocialrelationshipsmeansthat
the larger an organization becomes,the more difficult it will be for people to sustain
 K N O W L E D G EP R O C E S S E ISN M U L T I N A T I O N A L S

cohesive networks with all relevant people, and the more people's social networks will
becomefilled with structuralholes (seeFigure13.4).
Thus the larger an organization, the more people's social networks will have structural
holes, and the smaller an organization, the more easyit will be for people to develop, pos-
SeSs,and sustain cohesivenetworks. As a consequence,the knowledge dynamics within
large and small companies are likely to be quite different. As suggestedby Table 13.1 and
Figure 13.4, this does not mean that large multinationals are less effective at sharing,
searching for, or integrating knowledge than in small companies, simply that their
knowledge dynamics will be different.

Pharma-co:organizationalsize and cohesivenetworks

Pharma-co,as alreadydiscussed in previousexamplesin Chapters 3 and8 is a UK-based phar-

maceuticalcompany, which in the late 1990sbeganattemptingto implementan information
management systemthat wouldimproveintra-organizational communication and cooperation.
However, at the sametime,it was involvedin two mergers which its
trebled turnover,anddou-
bledits numberof employees (employing approximately 10,000staffworldwideby the end of
WithinPharma-co
the 1990s). therehadtraditionally beenlittlecommunication andknowledge-
sharingacrossbusiness units.Instead, staffin eachof Pharma-co'sbusinessunitshadrelatively
cohesivelocalized networks,andeachunitwas narrowlyfocusedon producing theirown prod-
uctsfor theirown customers. Thislackof communication wassummedup by one
andinteraction
'the of the European
manageras follows: thingthat is perceivedto haveimpededintegration
operation is an absenceof anyconnectivity between the groups.
manufacturing . . . Thereis no
dialogue of anyexperience
betweenthemat anylevelin Europe. . . thereis no exchange or infor-
mationor knowledge at all.'
Following themerger, thispatternof businessunitautonomy continued. Whilethemergerssig-
increased
nificantly the sizeof the company, theyalsoincreased the potentialbenefitsf rominter-
due to the increased
unit interaction, overlapbetweenthe differentbusinessunits.But, the
cultureof compartmentalism whichfacilitated
andisolation thecreation of localandcohesive net-
worksthat had existed prior to the mergers,became following
intensified the mergers due to
job
and losses,andactedto
fearsof rationalization prevent suchinteractions occurring,Thus,par-
adoxically,while mergersimprovedthe potentialbenefitsof knowledge-sharing,the increased
combined
sizeof theorganization, withthecultureof autonomy andclimateof anxietyandmistrust
which emergedfollowingthe merger,combined to make the possibility
of suchcollaboration
occurring peopleevenf urtherthanhadbeentraditional
throughentrenching
unlikely, withintheir
localnetworks.

Widerevrdencesuggeststhat the post-merger situationof fearand mistrustthat occurredin Pharma-

co is not untypical(seeEmpson2001)-What can management to developtrust,
do in suchsituations
processes?
knowledge-sharing
reducef ears,andfacilitate
O R G A N I Z A T I O NCAOLN T E X T S

In conclusion, this section has shown how organizational/network size can signific-
antly affect the dynamics of knowledge processes.In general, as organizational size
increases,not only doesthe complexity of managing knowledgeprocesses increase,but
the characterof the network of socialrelationsbetweenpeople,which crucially underpin
knowledgeprocesses, will also change.

Knowledgesharing acrosssocioculturalboundaries
and businesssystems

Chapter6 examinedin detail the dynamicsand complexitiesof knowledgeprocesses that

involve interactions betlveen people from different communities. The speciflc focus here
is on the dynamics of knowledge processesthat involve the spanning of sociocultural
boundariesaswell asdistinctive and quite different businesssystems.As has been shown
alreadyin this chapter,one characteristicof multinational corporationsis the need for
workers from different countries to cooperate.Thus the dlmamics of such interactions are
an important aspectof knowledge processeswithin multinationals. What are not exam-
ined here are the methods by which such boundaries can be surmounted to make
processes of knowledgeprocesses more effective.Suchissuesare dealt with in Chapter 6.
The focus here is on what impact sociocultural and institutional systems have on
processes of knowledge-sharing,integration, and knowledgeproduction.
The socioculturalvaluesthat peoplepossess, and the characterof the businesssystems
that exist, are closely interrelated, as business systems are created and reproduced by
people in possessionof particular socioculturalvalues,while simultaneouslythe socio-
cultural valuespeoplehave are shapedby the characterof the businesssystemsthey work
in. For analytical clarity, however, these topics are examined separatelyhere. In general,
aswith the issuesof organizationalsize,neither topic has receivedmuch attention in the
knowledgemanagementliterature.Thus, the illustrative examplesutilized are not taken
from the knowledgemanagementliterature.Nevertheless,both examplespresenteduse-
fully illustrate the relationshipbetweensocioculturalvaluesand businesssystems,to the
dynamicsof knowledgeprocesses.

Organizational knowledge processesand bridging sociocultural differences

Sociocultural values and beliefs refer to the systemsof values,knowledge, and beliefs that
individual people possess.Such values are shapedby an enormous diversity of social and
cultural factors including social class,the countries in which people are born and live,
educational expedence, family and parental influences, religion, experiencesof work,
professionalcodes of behaviour and ethics, etc. Some, most notably Hofstede (1980,
2001), arguethat distinctively national cultural characteristicscan be identified in differ-
ent countries. But, while this perspectivehas been highly influential, it has simultaneously
been subjectto significantcriticism (McSweeney2002; Soderbergand Holden 2002).
Having said that, numerous examples can be given of differencesin sociocultural
values that exist, and their impact on organizational processes.In the knowledge
 K N O W L E D GPER O C E S S EI N
SN / U L T I N A T I O N A L S

management literature the greatest,if not sole focus, is on differencesbetween Japan and
Europe and the USA.Nonaka and Takeuchi (1995) suggestthat there are quite distinctive
differencesbetween Japan and the Western world (Europeand the Americas)with regard
to the way knowledge is conceptualized and used in organizations. While this could be
criticized as crude cultural stereotyping, there is some evidence that there are distinctive
differencesbetween Asian and European values and attitudes. For example, Pauleen and
Yoong (2001) found there to be a greater degree of respect for authority and a higher
degree of formality in businessrelations in Japaneseand Chinese cultures than in
European and Australian cultures. Such differences were also shown to make misinter-
pretation and distortion possiblein communication processes.
One explanation for the existenceof the differencesin sociocultural values that people
across the globe possessis that they are shaped by the system of cultural values that
people areborn, educated,socialized,and work within. The most well-known advocateof
such a perspectiveis Hofstede,whose influence is visible in the work of some of those who
write about multinational companies, thus, Machalzina et al. (2001) talk about how
knowledgeis deeply culturally bound, while Van Maanen and Laurent (1993, 275) talk
about how values and behaviour are shapedby'underlying codesof meaning'.
Such differences have been shown to have a profound influence on knowledge
processes. Firstly,such differences,as was discussedin Chapter 6, make the sharing and
integration of knowledge between people with different systemsof sociocultural values
extremely complex and difficult. The lack of common knowledge, shared system of
values, or overlapping senseof identity that can exist in such situations is the primary
explanations for these difficulties. Secondly,the sociocultural values that people possess
importantly shapethe way knowledge is produced, meaning is made, and, using the lan-
guage of the practice-basedperspectiveon knowledge, how processesof perspective
making and taking occur. Thus people actively use their sociocultural values to produce
meaning and create knowledge, and two people may construct quite different meanings
from the sameevents,basedon their different value systems'
The example immediately below provides an illustration of such a process.Further, an
acknowledgement of the role played by sociocultural values in shaping the way people
createmeaning and produce knowledge challengesthe idea embeddedin the transmitter-
receivermodel of knowledge-sharingutilized by the objectivist perspectiveon knowledge
(seeChapter 2). Thus, knowledge cannot simply be diffused and transferred, unaltered,
between people with different cultural values.

Disneylandin Japan and the USA: sociocultural influenceson processes

of perspectivemaking

VanMaanenand Laurent(1993)providean analysis valuesaffectthe way

of how sociocultural
visitorsmakesenseof the Disneyland Adventureparksin Tokyoand the USA.At first glance,
Disneyland American
of Disney's
Tokyolooksto be a perfectreplica themeparks.Thusit appears
to containthe same codes
cultural and messages, which are andreceived
interpreted in a similar
@ OR G A N
I Z A TOI N A LC O N T E X T S

way by equallyenthusiastic Japaneseand Americanvisitors.Thus,since Disnevland Tokvo

openedit provedto be justas,if not moresuccessful thanthe AmericanDisneyland parks, and
hasbeenvisitedby enormousnumbers.However,in subtleways Disneyland Tokyohasbeen
modifiedto accountfor differentsociocultural values.ThusDisneyland Tokyohasfeweroutdoor
foodretailers andhasmoresit-downrestaurants thanthe American parks.Therearesomenew,
specificridesthatdescribe anddefendJapanese waysof life.lt haspicnicareascloseto the park
that go againstDisney'svaluesof not allowingfood to be broughtto its parks.Finally,in
Disneyland Tokyo,but notin the American parks,whiteglovesarewornby vehicledrivers, while
secondnamesratherthanfirstnamesareusedon worker'snamebadges.
V a nM a a n e na n d L a u r e nat l s oa r g u et h a tw h i l et h e s a m ev a l u e se x i s ti n t h e J a p a n e saen d
Americanparks,the way theyareinterpreted, andmadesenseof by theirdifferentaudiences is
f u n d a m e n t adl liyf f e r e nTt .h ed o m i n a nvta l u e si n D i s n e y t' sh e m ep a r k si n b o t ht h eU S Aa n dJ a p a n
are of order,safety,and cleanliness. However, whilethis is arguedto appealto the American
visitorsfor the contrastandescapeit provldes to theirtypicallifeexperiences, the samevalues
appealto itsJapanese visitorsbecause theyreinforce andreflect,ratherthancontrastwith,their
dominantculturalvaluesand lifeexperiences. ThusJapanese visitorsare recontextualizing the
valuesof Disneyland's parksthroughthe lensof theirown sociocultural valuesystems,andthe
perspectives theymakearethustotallydifferentf romthoseof American visitorsto similarparks
a conilnenlaway.

Thiscasesuggeststhat culturalvaluesin Japanandthe USAaresignificantly

different.Do such
significant
differences
existbetweenothercountries?

Organizationalknowledgeprocessesand the spanningof

different businesssystems

Lam (1,997),asdiscussedin Chapters6 and 11 (see,p. 78 and p. 1,62),identified significant

differencesbetween the UK and Japanesecompanies she researchedboth in terms of the
characterof their knowledge baseand the dyramics of their innovation processes.Other
studiesby Lam (1994, 1996)also show significant differencesbetweenJapanand the UK in
terms of how technical and knowledge-basedwork is organized. Lam (1997) suggestedthat
the differencesbetween theJapaneseand UK companiesshe examined could be explained
primarily by the different businesssystemsthat exist and operatein Japan and the UK.
Lam's flndings flt within a broader stream of analysisthat considershow the character-
istics of business systems, which vary signiflcantly across the globe, shape the work
practicesand strategiesutilized by the companiesthat operatewithin them (Hall and
Soskice2002;Whitley 1990,1999).As was discussedin Chapter 11, one specificsubtheme
within this areafocuseson the existenceand characterof national systemsof innovation.
Another broad strand within this broad perspectiveexamines how the character of
businesssystemsaffectsthe characterand role of HRM functions in organizations(Ferner
1997;Ferneret al. 2001; Varul and Ferner2000).
 K N O W L E D G EP R O C E S S E ISN M U L T I N A T I O N A L S @
Table 13.2. Key institutionaldimensionsshapingthe characterof businesssystems

lnstitutional dimensions of Examples

business systems

of market
The degreeandcharacter
regulation
The extentof governmentownership
in industry
The roleof tradeunionsin business
decision-making andtheirrelations
with businessmanagement
The roleof banksandfinancial
in industry
institutions
Thetype of financial
systemandthe
economicperformance demandsthey
placeon organizations
ln the USA,labourmarketshavemuchweaker
protectingworkers'rightsthanin other
legislation
countries
In France,comparedto otherWesternEuropean
countriessuchas the UK,the governmentstillhas
levelsof ownershipin a numberof business
significant
sectors
In Germanytradeunionshavea significantrolein
throughbeinggiven
businessdecision-making
significant powerenshrinedin law
bargaining
In Japanbankshavea powerfulroleat the headof
groups,and havecloselinkswith large
largeindustrial
businessorganizations
In the UKthe financialsystemplacespressureon
businessesto focus on relativelyshort-term
economicgoalssuchas sha.eprice

The term 'businesssystemr,as utilized by these writers, refers to the structure of social,
political, and economic institutions that constitute and shape the environment within
which businessorganizations operate. Key institutions in these structures include gov-
ernments and financial institutions. Researchshows that these institutional structures
vary significantly between different countries and regions, with Whitley (1999) develop-
ing a ty?ology of six distinctive types of businesssystemsmade up from sigrrificantly dif-
ferent institutional structures.Some of the key aspectsof the institutional structure that
characterize business systems are outlined in Table 13.2, and include the nature and
degreeof legal regulation, as well as the character of the financial system.

The'Japanization'of UK industry: the role of institutional factors

Following the globaldiffusionof Japanesebusinesspractices and philosophies suchas lean

production,therehavebeendebatesregarding the extentto whichsuch practices havebeen
customized to localconditions.Muchevidence suggeststhatin the UKtheseJapanese working
'select-
practiceshavebeensignificantly customized. Forexample,Taylor et al.(1994)referto the
ive and uneven'adoption of Japanesepracticesin a detailedstudyof two UK organizations.
Morriset al. (2000)in a studyexamining twenty-three companies in the UK foundthereto be
'considerable divergence' from the idealof Japanesepractices. Theyfoundfor examplethat
compatibilityexistedin terms of the careappliedto selection,recruitment, and socialization,
OR G A N
I Z A TOI N A LC O N T E X T S

but that the investment in trainingand 'hightrust'culturestypicalof Japanese practices were

absent.
Thiscustomization of Japanese workingpractices canbeexplained bythecharacteristics of the
UK'sbusinesssystem.Thus,Morriset al. (2000)arguethat the differences they foundcould
l a r g e lb r n dS m i t h( 1 9 9 41, 2 1 )a l s os u g g e stth a tt h e
y e e x p l a i n ebdy i n s t i t u t i o nfa lc t o r sE. l g e a
economicshort-termism prevalent in the UK and the generalunderfunding of trainingthis
produces, hasbeena significant contextual factor,constraining theabilityof UKmanagers to fully
implement Japanese methodsunaltered. Finally, Scarbrough andTeny(1998), in a studyof two
c a rp l a n t si n t h e U KM i d l a n d sf o, u n dt h a tt r a d eu n i o n sh a da s i g n i f i c a rnot l ef o l l o w i ntgh e i m p l e -
mentation of Japanese workingpractices, whichwas anomalous with theirgeneralphilosophy,
whichcouldbe explained by the differenthistorical rolesplayedby tradeunionsin Japanese and
UKbusiness systems.
Thusoverall, dueto the constraints andpressures imposedby the specificinstitutional charac-
teristicsof the UK'sbusiness system,Japanese workingpractice andknowledge havenot been
implemented andtransferred unaltered, but insteadhavebeensignificantly customized.

Knowledge processesthat span different businesssystems,as shown by Lam, can prove

complex, due to the effect they have on the character and structuring of organizational
knowledge. As the above example also shows, the sharing of knowledge across such
boundaries can also result in it being changed and reconfigured. However, the general
lack of attention to such issuesin the contemporary knowledge literature means that the
relationship between business systemsand organizational knowledge processesis rela-
tively uncharted.

Conclusion

The fragmented and dispersed character of the knowledge base within multinationals
means that there are potentially significant beneflts from effectively managing it. Thus
the potential synergy that could be created from bringing together elements of this dis-
persed knowledge is enormous. This helps to explain why multinationals corporations
have been some of the most enthusiastic adoptersof knowledge management initiatives.
However, paradoxically, these same characteristicsof the knowledge basemake its man-
agementan extremely complex and difficult task.This is due to both the sizeof the knowl-
edgebasein these organizations, which means the knowledge baseis highly fragmented,
combined with the fact that this knowledge is dispersedamong communities which can
have different sociocultural values and which operatewithin distinctive businesssystems.
One way in which multinationals can managetheir knowledge baseis through the way
business is structured, with the chapter showing how hierarchical and network-based
structures produce very different knowledge-sharing d;,namics. However, Birkinshaw
et al.'s (2002) contingency perspectivesuggeststhat the dominant logic that suggeststhat
network structures are inherently better for knowledge-sharing compared to hierarchical
stluctures,in all situations,was challenged.
 K N O W L E D GPER O C E S S EI N
SI \ 4 U L T I N A T I O N A L S

The chapter also consideredhow organizational size,a relatively neglectedtopic, affects

the characterof knowledge processes.It was concluded that not only is organizational size
directly related to the complexity of knowledge processes,but that organizational sizecan
also fundamentally alter the characterof knowledge dynamics, through shaping the type
of networks that people can develop and sustain.
Finally, the chapter also considered the complexity of sharing knowledge between
communities that are located in different and distinctive businesssystemsand where
people possessdifferent sociocultural values. The sharing of knowledge across such
boundariesis not a simple, direct transfeqasthe socioculturalvaluesthat peoplepossess
shapethe way they interpret and understandthe knowledgeof others.Thus knowledge-
sharing in this context involves an active process of perspective-making whereby the
knowledgeof others is understoodin relation to a person'sexisting values.Equally,the
sharing of knowledge between people and communities who operate within different
business systems was also not found to be straightforward, and typically involves
the transformation and customizationof any sharedknowledge.

Hofstede(1988,2001)arguesthatdistinctnational
culturescanbe identified.
Towhatextent
doesyourown personal experienceconfirmor challenge
this?Further,
do suchcultural
differences hinderprocesses
significantly of knowledge-sharing?
FordandChan(2003), in oneof the few studiesto examinethe effectof cultural differences
on organizational knowledge processes, foundthatlanguage competences significantly
affectedsuchprocesses. In general, informalknowledge flowsweremostlikelywithin
cultural groups,while{ormalbusiness-related communication was morelikelybetween
c u l t u r aglr o u p sW
. h a td o s u c hf i n d i n gssa ya b o u t h e i m p o r t a n coef p r o v i d i nl g
anguage
training as a way of dealing with the difficulties of cross-cultural knowledge processes?
exampleof Dell{seep. 200)showedthatit hada centralized
Theillustrative hierarchical
structure,
andutilized
a codification-based
knowledge management strategy.Towhatextent
aresuchknowledge management strategies
compatiblewith hierarchical
structures?
Further,
wouldsucha knowledge management strategybe compatiblewith a network-based
structu
re?

'Managing
M. Becker(2001). DispersedKnowledge: Organizational
Problems,Managerial
Strategiesand their Effectiveness',Journalof ManagementStudies,3\ll:1037-51 .
Examineshow organizationalsize affects the characterof organizationalknowledge bases, as well
as the most appropriatestrategies for managingknowledge.

J. Birkinshaw, (2002).'Knowledge
R. Nobel,andJ. Ridderstale as a Contingency
Variable:
Do the
of KnowledgePredictOrganizational
Characteristics Structure?'Organization
Science,1313274-89
Providesan analysiswhich suggests that organizationalstructure needs to be sensitive to the
characterof an organization'sknowledge base.
@ ORGAN
I Z A T I O N ACLO N T E X T S

A. Guptaand V Govindarajan (2000).'Knowledgeflows within MultinationalCorporations',

Strategic M anagement J ournal, 21: 473-96.
Presents an objectivist perspective on the complexities of knowledge sharing in multindtional
corporations.

A-M. Soderbergand N. Holden(2002).'Rethinking CrossCulturalManagementin a Globalizing

BusinessWorld', lnternationalJournal of Cross CulturalManagement,2li: 103-21.
Drscussesthe challengesfor multinationalcorporationsof managingtheir knowledge basesin the
contemporary bus/nessenvironme nt.