Innovation and Labour Mobility: A Case Study of … · Innovation and Labour Mobility: A Case Study of R&D in British Telecom* Matias Ramirez·· Resumen Esie articulo trata acerca

Redefining Firm Competencies,Innovation and Labour Mobility:A Case Study of R&D inBritish Telecom*Matias Ramirez··

ResumenEsie articulo trata acerca de como las insfuucioras del mercaco labora y, mas especmcsmene, los cambios en las pactlras de lamovilidad laboral. moldean la manera en la cua las erpesas esabecldas de tecnologia intensiva reoenren sus cornpeienciastecnoloplcas.A naves de un estudio delallado lIevado a cabo en BT, el argumento central propuestc es que la habilidad de las empresas enimplementar car-cos estrategicos importanles en lyD,involucrando cambios de lIexibilidad latoral. esaao luenerreme inlluenciadapor las estrategias de los propos empleacos en cusuc a lorrraclon de habilidades y esnucnra puesona. Esto. a su vez, esu influidopor el modo especmcc en el cualles insfuuciones del menace letoral se delinen y se regulan. Se presena un marco et cual integrala cuesti6n de menado laboral y la mcvaclcn, especificando las coembuclcres dislintas que tal mercadc lace hacie el procesc deinnovaci6n.

Palabras clavelmovacion, movilidad Iaboral. telecomunicaciones. empresa compaenclas.

I. Introduction

Itis widely acknowledged that over the past decadeand a half, radical new torms of production haveemerged with the pervasive expansion of information

and communication technology (len (Freemanand Perez1988, Best 1990). This has acted as a catalyst for deep-seated changes in the process by which innovations areconceived and have caused important dislocations withthe institutional structures within which innovations weretraditionally generated. Longstanding organisations suchas British Telecom (BT), that roullnely were consideredtechnological leaders and innovators in their industries,have therefore faced enormous challenges to restructure

This is the result of the author's work and research whilsl inEngland. The aUlhor would like to thank Jonathan Michie, FredGuy and Mark Tomlinson tor their helpful comments on earlierversions of this article, Article received in March and approvedin June 2003,

School of Management and Organisalional Psychology, BirkbeckCollege, e-mail: m,ramirez@mbs,bbkac.uk

competencies and generate greater dynamism into theirorganisational routines, This paper contributes to thisdebate around organisational and inslitutional change bylooking at how labour market institutions and morespecitlcally changing practices otlabour mobility In anR&D environment can shape the manner in which largeestablished technology intensive firms redefine theirtechnological competencies. The central argument posedIs that the ability ot firms to implement important strategicchanges In R&D, involving new forms 01 labour Ilexlbllityand mobility amongst knowledge workers, witl be stronglyinfluenced by the employees' own strategies ot skillformation and career structure. This in turn is influencedby the specillc manner In which the labour marketInstitullons are defined and regulaled

The concept 01 the co-evoiution and mutual shapingof firm organisation and labour market institutions isaddressed using a Iramework developed in the paper thaiintenaces elements ot two typologies drawn trom, up untilnow, relatively disparate literatures in Ihe areas 01

183INNOVAR, revists de ciencias administrotivas y socia/es. No. 21, jiJlia . dicil!mbre de 2003

INNOVAR, reeista de ciencies aemnlstrethas y scctaies

organizational change and labour markets. Inherent in theframework's proposition lies the notion that the coherencebetween the innovation strategy of a firm and the labourmarket inslilutions will underpin the ability of firms to ca-rry out successful strategies of innovation.

2. Re-positioning B1 in theIndustry and Labour MobilityThe telecommunications industry has been at the heart ofthe technological and lnstilutlonaluanslormaticns thathave underpinned the ICT paradigm. In terms 01technological change, Ihe so-called intelligenttelecommunications network, whereby software basedcomputerised functions support the electronic processingat digital signals, has been the radical breakthrough thathas transformed how communications is organised(Mansell, 1993) Powertul customer prerruses equipment(CPE) In particular represent a paradigmatic rupture inIhe evolulion 01 the industry, creating the possibilily torusers to shape the process at innovation through designconfigurations based on more open networks (Kavassaliset at. 1996). These developments have laid the basis loran infinite range of diversified services to suit differentcustomer needs.

For firms such as BT, the new technological regimehas posed acute opportunities and chalienges. Thedanger lor incumbent telecoms lirms is that by relying onthe traditionai source at Income, voice telephony and theownership 01 the network, the organisation in effect isreduced to becoming a mere bit transporter for the valueadded services 01 other low cost compelilors. As table1 shows, up 10 1999, BT was still tar more reliant onlts traditional voice business than in value addedservices. Developing technological and organisationalcompetencies in the area of network solutions thereforerepresents a key challenge Iechnoloqically this hasmeant crealing expertise in a whole new series 01areassuch as wireless Inlernet access, direclories and searchengines, internet TV, electronic payment systems,advertising, education and training (Analysis, 1997).

_ BT_3"q_11l8I

Fixed voice service £2.7bnMobile services "'"Data Services £ 597m

Internet sevces £23On

184

The second major development, opening theindustry 10competition, emerged as a consequence of achange in regulation policy by governments, which overmuch at the last century had supported the monopolystatus of service suppliers. The UK government in parti-cular adopled what IS generally known as a "tast track"privatisation programme (Batt and Darbishire, 1997)With the shorter lite cycle at products, BT has beenheavily critclsed for being unable to keep up With thenewer, more agile and more entrepreneurial new entrants.The Economisl for example lamented, "despite BT'sexperience in running networks, it lacks entrepreneurialagility, the speed 01 decision making and the gung-hoculture of investment to match the moves of companiesiike WorldCom" (4/4/1999).

This criticism emerges from the fact that one at thecharacteristics onne giant telecommunications monopolyfirms in the past was that the routines that arose from themonopoly period were shaped by the overriding obiectiveto provide an efficient universal service rather than a fastrollout of dilterentiated services. As a consequence,monopoly public telephone operators (PTOs) developedeltective but highly bureaucratised routines that uphelddetailed operational protocols (Batt, 1995). Hence one atthe main problems tor the large PTDs such as BT hasbeen to overcome the bureaucratic nature of the formerlarge monopolies.

Incumbent high-Iechnology lirms such as BT havetherefore faced two main challenges. The tirst has been toposition themselves strategically in the key areas of busi-ness growth by developing or acquiring competenciesIhal Will generate new streams at revenue through productinnovations. The second chalienge relates to the changesin organisational structures to speed up Ihe rate at whichproducts are rolied out in response to competition andthe taster pace at innovation. Hence, being ahead ofcompetitors is associated not only with devising newsources of innovations, but also having the organisationalstructures and management capability to ensure rapiddeployment and rollout of new products and services.

Labour markets can playa central role in this process01 structural change. In particular, the institutions shapingskililormalion, industrial relations and labour mobility canfacilitate or create obstacles to structural change (Lazonick1998, Bassanni el si, 2002, Saxenian 1996). This po-ses the need for a specific analysis of labour markets inthe telecornrunicatlons industry Yet over the 1990slabour markets in the telecommunications sector alsoexperienced important changes As a consequence of theconvergence of technologies and the fragmentation at

GESTION DE LA PRODUCCION Y TECNOLOGiA

employmenllollowing Ihe entry 01mulliple lirms into theInduslry, skills have become more Iransterable and Ihlshas fomented profound changes in the institutionalstructures regulating recruitment and retention of labour.Various skill types emerging as Increasingly key 10 Iheinduslry, lor example in Ihe area of software development,are also associated with career structures that emphasizeinter firm mobility and more flexible non-standardemployment contacts. as evidenced by the growing useof contractors in research and development environments(CWU, 1997). This may have a crilical impact onknowledge transler in Ihe industry. From the point ofview of the analysis, it also suggests thai institutionalchange in labour markets may shape the Iype 01slruclure firms are able 10 adopt al a given point Intime. For example, one ollhe principal ways In whIChentry barriers have been reduced intelecommunications is throughlabour markets by means of hiringslaff Ihat have accumulatedknowledge 01 how 10 operate thenetwork equipment from specialistsuppliers (Fransman, t 998).

This example can be added to aburgeoning lilerature Ihat suggests Ihallabour mobllily can be a vital meansby Which firms and technologysystems can expand Ihelrcompelencies Ihrough Ihe transler oflacit knowledge (Saxenian 1996.DECO 2001). However, some studieshave also suggesled that policyrecommendations encouraging inler-firm mobility as a means to advanceknowledge flows can be a double- 50uIt'e: O....eloped lrom me UK Labour Force SuMy

edged sword, particularly It Ihis leads10 a very high turnover of specialised labour for firms,Blackler (1995) for example has argued Ihat an inslitulionalstructure is required for occupational labour markets toexisl so Ihat inveslment in training for skills is undertaken.Organisations may be less Willing 10 invest in training forniche skills if they are not guaranteed a return trom theInvestment. Hence, a high degree of mobility foremployees in vocational education and training can lead10 a shortage of inveslmenl In training. II has also beensuggesled that knowledge workers Ihal underlake a greatdeal 01 inter firm mobility may " learn" less Ihan those Ihathave longer lenure (Tomlinson el ai, 1999) This locuseson the tact that Ihe learning base of lirms Is largely madeup of routines Ihal emerge Irom learning-by-doing andlearning-by-interactlng between employees. The diffICulty

at transferring tacil knowledge that is not embedded Innetworks using common protocols and codes ofcommunication is thus emphasized.

The issue of labour mobilily is partiCUlarly relevant forBT While a labour market monopsony, BT had no choicebut to undertake its own investment programme to acquirenew skills. Nevertheless, wilh the enlry of multiple lirmsinlo the market. loday Ihis Is no longer Ihe case andorganisations can hunt within a broader national andinternational labour market for both general and nicheskills. Specitying Ihe degree at turnover amongsl highlyskilled employees in the !elecommunications seclor isdiflicuillo estimate accurately. However, an approximatecalculation using data Irom the UK Labour Force SurveyIs illuslraled in ligure t. This shows Ihat belween 1992and 1998 turnover amongst telecommunicationsknowledge workers grew considerably'.

Turnover Amongst TeJecommunlcatlons "Knowledge Workers"

14 ~--~_----------~

12+------------~ 10 .f---------------tn.il 8+--------e~u~~ 4

6+-------

2

a -1992 1993 1994 1995 1996 1997 1998

The data were collated from the firsl quarter 01 Ihe quarterly UKLabour Force Survey lor each year. The following crena wereused:

J. Includes only individuals that answered that they were Managers,Technical Workers or olhe Professionals at the time when lheywere surveyed

b. Includes only individuals matlnolceteo that they were currentlyworking in a firm In the telecommunications industry.

c. Includes only individuals that answered "NO" when asked if theyhad been working in the same firm one year ago.

d. All missing values were discarded.

e. The sample average of knowledge workers in telecommunicationfor each year was 150. Hence the conclusions are tentative andneed 10be analyzed in coniunction with other inoicalors that showevidence of mobility.

fNNOVAR, revisla de ciencias adminiSlrallvas y sociales. No. 21, enem jlmio de 2003 185

INNOVAR, revista de ciencfas adrnimstretivas y seciaies

3. A framework for changingindustry and labour marketdynamicsThe above discussion has emphasised the need for areinvention of 8T in order to remain competitive. Thisincludes the needto generatenew organisational routinesthai place greater emphasis on product innovation andthe taster roil out at new products and services. From alabour market perspective. emphasis is theretore placedupon bringing in new skills and developing more dynamicmethods of managing employees, covering areas suchas recruitment and knowledge flows. Yet, as has beenmentioned, the institutions defining the nafure labourmarkets are also in a state of flux. In parficular, increasedmobilily 01knowledge workers can be associated bothwith the opportunily to expand firm competenciesrelatively quickly, but also with shorter time someknowledge workers appear to spend within each firmand the growth at non-slandard conlracts within thehigh skill end at the market This suggests furtherresearch is needed to analyse the ditterential impactsmat some recent changes in labour markets can havein terms of the ability of tirms to deveiop newcompetencies. It also suggests the need lor a frameworkat analysis that takes into account the dynamics at aco-evolution of industrial organisation and institutionallabour market structures at the level at Ihe tirm.

This task is undertaken in this paper by intertaclngtypologies drawn from two different sets at lileratures'.The nrst, shown on Ihe leff of figure 2 below, is based ona fypology oftirm competence developed by Lam (2000).Lam's typoiogy draws some Inspiration from the work 01Lundvall and Johnson (1994) and Blackler (1995)among others It is based on Ihe idea that knowiedge inIhe 111mcan be analysed along an epistemologicaldimension i.e. tacit vs explicit knowledge and anontological dimension ie at the level of the individual orIhe collective film. For the purposes of this paper, twoknowledge environments are highlighted. An embodiedknowledge environment is one where tacit, action-orientated skili predominates. This type of knowledgeprovides an environment where there is littlestandardisation 01 knowledge or work practice.Knowledge therefore remains within the individual ratherthan within an organisation and relies to a great extent onmarket know-how and skills that are potentiallytransferable. This can breed instability within anorganisation, as high movement at labour can create

A fuller version of Ihis framework is outlined in Ramirez (2002).

186

problems at knowledgeretention.Organisationsreliant onembodied type knowledge could well characterise somenetwork type forms of organisation such as Silicon ValleyIn the US, wherethere arefew institutional limits to labourmobility and the nature of the work in the area relies onrapid problem solving skills. Inanembedded knowiedgeenvironment on the other hand, firm organisation isunderlined by systemic routines and a strong commonculture. Knowledge is tacit in nature, but It resides inoruanisational capabilities, combining stability andflexibility of team dynamics The stabilify and formalily ofthe structures permits continual and incrementalinnovations.

TacitKnowledge

learning and interaction beyondthe firm, Inter firm movement,

off the job training, OlMs.

1- '"","ooC,"'''''QuaJification

RuleOLMs

Embod'ed(Individual

Knowledge)

Embedded Competence(collective T Rankknowledge) IlMs

Lam (2000) Marsden (1999)

Collective knowledge, firmrcutlree, idiosyncratic learning.

S""""" Developed I,om lhe UK LabooJ' Forco Survey

While Lam's (2000) Iypology focuses predominantlyon the firm as a learning agent, Marsden's (1999)typOlogy, on the right of figure 2 draws on a theory ofmicro labour markel institutions and focuses on the roleof labour market institutions in influencing workorganisation and skill acquisition at the level of the firm.At the core of Marsden's argument lies the notion of theemployment contract as a stable means of aligning jobdemandsfa workercapabilitiesand identifyingcertainrulesthat can be applied in a wide variety of workenvironments'. Forthe purposes of this paper,two distinct

3. Lam (2000) nerseu broadened her typology to encompass inlernaland occupational labour markets.

GESTION DE LA PRODUCCION Y TECNOLOGiA

work environments are of interest. On the one hand thereis an environment where the tasks that employeesundertake are grouped according to technical requirementsarising out at the production needs. Because at theidiosyncratic nature of jobs, qualifications are not a goodguide to assignments, hence jobs are olten decidedaccording to some sort of internal ranking system, suchas seniority, where status depends upon the range anddepth 01 skills at individuals. Hence internal iabour markets(ILMs) play an important role in determining the type attraining employees receive. Similarly career paths are oftenbased on promotions within the firm and associated 10lirrr-specfic skills gained Within the firm. Marsden (1999)calis this a competence rank ruie. On the other hand, in awork environment defined by Marsden as qualificationrules, assignments are set by identifying competenciesrequired for certain kinds 01 work and assigning them onIhe bas.s of recognised qualifications Work is organisedaccording to skill requirements and around functions, Inthis environment, off-the-job training and qualificationsare important, hence training is backed up by developedoccupationai labour markets (aLMs).

The discussion has emphasised that a degree ofcoherence is necessary between firm organisation and therules defining the employment system m the firm. Theimplicafion is that if, as was discussed for Ihe case at BT,the firm wants to fundamentally restructure ils operations,there are iikely to be necessary impilcations for how workrules are defined in terms of job descriptions and howindividuals skili sets are developed. This relationship canbe illustrated by luxtapos.nq Marsden's With Lam'stypology of flrrn competence. Figure 2 places Lam's(2000) and Marsden's (1999) mal rices side by side andthe arrows illustrates the iinks that wouid be ezpecled 10emerge. The foilowing points ouune Ihe baSIS on whichthese iinks are made:

• In organisations reliant on an embodied knowledgeenvironment, skill development needs to be dynamicand highly flexible. Firms in this environment willallen not have the fime to develop in-house skillsin new areas. Hence there is a need for institutionsthat will create ciear guidel ines within which inter-firm movement of labour can take place, such asrecognised qualifications or weli developedinformation peer group networks that can vouch forindividual skills. In Marsden's typology, thisflexibility is aftorded by developed occupationallabour markets as described by the qualitication ruie.

• Organisations that rely on coliective sources ofknowledge and informal non-wrtten tacit type

knowledge structures can be expecled to reiy onsome form of firm specific training structure todevelop common dialogues and understanding offirm routines. This would reflect a competence rankrule with flexible internai labour markets.

There are three important implications that follow fromthe framework described in figure 2. The firsl is that in nuscase we can define two distinctlypes of firm organisationthat have quile ditierent logics in terms of knowledge tlowsand inslitutional structures supporting them. Secondly I aspecific form 01 industrial organisation as described bythe lell hand typology tn figure 2 will be more stableand tram the point of Vl€W of knowledge transfer, moreeffective, when supported by a specific labour marketinstitutlenal structure. Thirdly, the logic of figure 2 isIhat a change in the dominant firm srucune. say fromembedded to embodied, as a consequence 01 changein the technoloqical paradigm, wouid also impiy theneed lor a change in Ihe supporting institutional labourmarket structures. Otherwise a mismatch between firmand institutional structures may emerge. The follOWingsection will deveiop Ihe case study 01 BT 10 shed iighton these propositions.

4. The Case StudyMost of the data presented in this paper is based oninterviews that took place at Adastral Park, BT's researchand development (R&D) laboratories in Martiesham. Teninterviews were carried out, the earliest in 1999, with BTempioyees m the R&O department Five follow~upinterviews were also undertaken. The paper firstly presentsa discussion emerging from interviews undertaken withsenior managers holding strategic decision-makingresponsibilities. These prOVide an overai i Vl€Waffirm levelpoiicy and thinking on strategy and human resourcepolicy (HR). This is foliowed by a discussion of the datacollected from interviews conducted with non-supervisoryR&D employees, principaliy developers, designers andengineers on the impact of company level HR policies ontheir own career prospects and innovation in BT. Materialwas also collected, including the "Investors in Peopie"document from a ieading research department and a se-ries of executive presentation notes from a senior mana-ger describmg aspects 01 BT's research strategy. Datawere aiso collected from The Manager. an internai BTjournal It was agreed to keep the identities ot aliindividuais inte~iewed and department names confidential,so will remain anonymous .

The methods for interrogating the data lollowed Ihe"Coding" process described by Strauss (19B7). This Is a

187fNNOVAR, revisto de cienoes administrolivos y saciales No. 21, esero- junia de 2003

INNOVAR, revista de oencias aemolsuauvas y socteres

methodology whereby codes or subheadings are lormedto link raw data fragments such as paragraphs, phrasesor sentences from the interviews and documentaryevidence to key concepts and themes. The identificationof codes and themes was based on the theoreticalpropositions laid out in the framework presented in figure2. The nexl step Involved outiining a descriplion 01changes in the innovation process, including changes inindustrial organisation over time, using evidence frominterview data with managers and documentary evidence.Subsequentiy delailed information was analysed relatingto the nature of changing work organisation, careerdeveiopmenl pattems and lirm ievel empioyment systemsIn BT Documentary evidence and descriplions 01iabourmarket institutions was also provided

Accelerating product life cyclesand intra-firm mobilityAlthough today it is a private corporation, BT's strengthin the market place still lies to a great extenl on Its tradition01providing an eflicien!, reliable service and the scalingup at lechnology tor its giant communications network,However, as was discussed earlier and as a senior mana-ger in BT's research department beiow points out, theroutines 8T developed to maximise performance in theconlext 01predictable changes in lechnology during much01the 20'" century, have become totally unsuitable tor thenature of tecnnoloqical change today where, as well asrellabliity, speed and diversfy in Ihe rollout at Innovationsare of essence:

"the average time to launch something at theminimum was 1B months to 2 years, It was nonsense,because by the lime you finished launching it, thething was obsolete, So internally there is a view thatwe had to have a radical rethinking in our systems".

BTresponded fa the demand lor laster roll out at newtechnology by adopting a series at organisationalinnovations in an attempt to reduce the bureaucracies andlong drawn out processes that characterised its routinesIn Ihe past As tar as research and development ISconcerned, these Initiatives inciuded locussing on thedeSign rather Ihan the bUilding at goods and intreasnqthesynergresbetweendeveiopersand designersat ditterentstages in the development iitecycle at Ihe product. Thesemeasures mirror initiatives that have been adopted in awide range 01industry environments, thai breakaway tromdeveloping products and services sequentially and adoptan iterative method, with permanent exchanges ofinlormation between the relevant actors, starting from the

188

stage of concepnon'. In this sense, the first significantorganisational innovation that emerged from the interviewsrelates to an attempt by 8T to encourage designers anddevelopers 10follow their ideas out of the confines of thelaboratories via secondments and exposure to otherdepartments of the company. The manager cited abovePOintsout.

"It's one of our policies to push people through thebusiness. If someone has an idea, we would like thatperson to do the research, take that idea intodevelopment, take that idea into field trials anddeployment and then come back perhapsfor the nextgeneration of Ihe product. II this worked they wouldbe so valuable to the business because they knowsomething from beginning to the end, they knowwhat's it's like to try and launch science and supportthings" .

The effort to encourage the spillover of knowledgebetween differenl areas of the firm through intra firmmovement of labour aims to leverage the vision andexperience of researchers throughout the organisation.Individuals tollowing through the development of servicesact as key co-ordhatnq agents of knowledge, sharingthe tacil knowledge of dltterent departments. However,given the lirm specific context in which the disseminationand exchange at knowledge exists, the tacit knowledgegenerated from the exchange of information also improvesexisting firm routines, Using the terminology of Lam(2000), Ihis emphasises lirm specific knowledge and canbe related to an embedded type of knowledgeenvironment.

Broader innovation and Inter-firm mobilityA second significant organisafional change picked up inthe interviews, arose as a direct consequence of the morediverse technological structure at the telecommunicationsindustry. The fragmentation of technologies and theemergence of multiple specialised equipment and soft-ware suppliers have enlarged the lechnological choicesand options for the introduction of the same basic service,The R&D process has in consequence become lesspredictableand the "make or buy" decision more complex,with greater emphasis placed on understanding andtransferring knowledge Irom suppliers rather than

See for example Chapter Three in Kline and Rosenberg (1986) andAoki (1998),

GESTION DE LA PRODUCCION Y TECNOLOGiA

necessarily developing new knowledge in-house(Fransman, 1998)

Firms such as BT that own large networks made upof mixed generation technologies are therefore beingtorced to develop broader competencies 10keep up-to-date with a rapidly accelerating range at technologicaloptions. As a consequence, the labour supply loradvanced telecommunication services draws in a broaderscope at skills trom that traditionally developed withinthe ranks at the core researchers. BT has undertaken aseries of human resource initiatives to develop thesebroader competencies. The tirst at these involves directrecruitment at highly skilled scientists and engineers intothe company. A senior manager in the researchdepartment comments:

"We need as many perspectives on what is possible.Some we can get from within the company, but theyare shaped by company cuiture and thinking. So wegel people in from across the industry and business,short-term fellow schemes 01 people to work withresearchers. This is a relatively new policy. Thereason is because the skills were becoming out 01dale. We believe the half-life of a degree is 3 years",

Similariy, a senior HR manager in the research anddevelopment department comments

"The speed with which the market moves sometimesmeans you cannot re-orientate people sufficiently inthat lime, To suddenly switch that number of peopleis very hard and you don't get advance warning thatil's coming, When you go into a new market areawhere we haven't been in before you may needexpertise in that particular area, that you can buy in,use and once you're in there you decide. Maybe youwill leave there".

BT therelore Is recruiting a greater number 01knowledge workers directly into high grades Irom outsidethe tirm. This is because scientists and experls based ondisciplines beyond those traditionally linked totelecommunications, i.e. physics or chemistry, are beingdrawn into telecommunications. A second lactor is thatoutside recruitment can bring a new "vision" and widerperspective to the firm 01how to develop products and tocombine this with the skills and thinking 01internal people.A third issue is that In some cases the speed of change,in terms at new markets and technologies is so tasl andunpredictable that the organisation does not have the timeto re-train to develop competencies from existingemployees. Signiticantly, the risky nature 01some of thesenew ventures leads the film to lavour a looser relationship

With empioyees, who can easily come all the pay roll II anInvestment proves 10 be short-lived. Finally, there is anIndicallon that the snorter product iife cycle can leaveknowledge and skills out 01 date. Hence greater labourturnover and movement of individuals in and out of thefirm is not as damaging as it might have been in a periodwhere projects were of a longer duration.

The discussion so lar has highlighted Iwo broadchallenges lacing BT. One is 10speed up Ihe pace in theroll out of new products and services, the second todevelop the competencies necessary to generate valuedadded services. Two broad initiatives involving changesin the division of labour have been referred to.Encouraging employees leading innovations to followthrough their new projects between dille rent departmentsin the firm aims to encourage synergies in the initialphases of development As mentioned earlier, this veryclosely describes an embedded knowledge environmentOn Ihe other hand, senior managers have suggested thata greater movement In and out 01 the film at high skitiedemployees may be an important means by which BT canacquire the Ilexibitity and broad competencies to generatenew products and services. This type of structure mirrorsthai described by an embodied knowledge environment

These twin pressures are consistent with the previousdiscussion regarding B1's role in the telecommunicationsIndustry. On the one hand, BT has to cater for a massmarket that focuses on organisational synergies to speedup Ihe rollout at new goods and services, productresilience and system wide process innovations. On Iheother hand, BT is taclng an enormous challenge to shillthe locus at Its R&D operations, Irom product resilienceto customisation of value added services. This can bestbe achieved by introdUCing and continually renewing newsets 01skills and entrepreneuriallaient

Following the logic of the tramework illustrated in fi-gure 3, this almost schizophrenic role at BT. where It hasto adopt dillerent strategies, and as a consequencedifferent structures within the same departments to catertor dillerent types 01 markets, has a number 01 importantrnplicalions. BT appears to be attempting to work withquite dillerent orpanisational structures, thai as Ilgure 2suggests, have contrasting implications for labour marketsIn the lirm. While an embedded knowledge environ men Iis supported by Internal labour markets and tirm speciticknowledge, an embodied knowledge environmenl requiresthe Institulional support at occupational iabour markets toencourage knowledge tlows 01labour between Ilrms. Thequestion Ihat arises is how well can a large firm such asBT accommodate these lensions and support dilterent

189INNOVAR, revista de ciencias adminislrativas y soc/ales. No. 21, enero -junio de 2003

INNOVAR, resista d€ ceneias administrativas t secretes

incentive strcutures, if indeed they do exist? The discussiontherefore turns to the labour market institutions and theexperience of R&D workers themselves.

Combining Incentive Structuresin BTThe scenario 01 lnlra firm labour mobility and temporarysecondment at R&D stall within the Internal BT supplychain was rased as part and parcel of the attempt toincrease the degree of communication and sharing oftacit knowledge between different departments. However,although secondment was identified as a major companyinitiative, when the idea of temporary secondment ofdevelopers and designers was raised with a number ofproject managers in R&D, there appeared to be a greatdeal at scepticism as regards its viability Dne projectmanager in a department rolling out new telemarketingproducts commented:

"Seconcrnen: is a nice theory and there is a lot of talkabout secondment. in practice it rarely works. Typicallya protect will last 6 months, you take somebody outfor 3 months and things have changed a 101.Thingsare moving too quickly so that it doesn't quite work,its such a schism, that there's an awful 101of re-buil-ding that has to go on afterward to synchronise theperson back inlo the team. Plus individuals aren'tkeen to do it unless they are far sightedenough to see it as enhancing their career;it's just a nuisance. I know of very few caseswhere it has been done" (emphasis added)

Secondment is an important concept that has clearlybeen discussed, but appears to face two problems Firstlythere is the short-term nature at projects. The concept 01embedded knowledge emphasised the imparlance ofstable teams, the knowledge at which is Imbued with awider company culture. However, in firms moving rapidlybetween products and projects, although teams playakey role In problem solving, the most important knowledgeagent Is the Individual, that can move rapidly betweenteams, contexts and even workplaces. This provides thefirm the Ilex.bility to move resources into new projects asand when required. In the case 01BT, despite the attemptto provide longer-term support tor products, the relativelyshort nature of each project and the pressure to move on10 new areas makes it difficult to provide longer-termresources for each new products and service.

The second obstacle lies In the apparent lack ofenthusiasm of Individuals to embrace secondment,because intra-firm mobility is not seen as furthering

190

employee's career paths. This suggests that there is ahigh opportunity cost for employees to long-term supportlor products and the development of firm specific skillsthat this Implies. Consequently the discussion tocuseson the career structures of employees and in particular theincentives for progressing through internal labour marketsand building firm specitic skills.

The apparent absence of a career path for R&Demployees willing to develop and follow throughinnovations in BT appears contradictory at first glance,The BT managerial grade structure ISshown in ligure 3'. Itreveals both technical (up to PTG) and supervisory stepsof a hierarchlcal promotion structure through whichmanagement grades can climb as part of their careerdevelopment. This structure is in fact very similar to theinternal labour market inherited from the old monopolycivil service, where direct recruitment to high gradesseldom occurred. OpportUnities still clearly exist tor careeradvancement within the firm. In tact R&D employees withlong tenure in BT interviewed contirmed that they hadworked in many different departments within the company.

PCGC(Senior Board)

tPCGR

tpeGSt

PCGTt

PCGU(Personal comact grade: U)

tPTG(ProfessiOfiailechnical grade)

tNPG4

tNPG2

Nevertheless, the limitations of the BT career path inthe current technological paradigm become clear whenseen in relation to how it impacts individuals developingtechnologies and innovations. The vast majority of

s. All BT permanent R&D employees are on management grades,irrespective of supervisory responsibilities.

6. The STE, the BT manager's union, changed its name to "Connect"after the interview was conducted.

GESTION DE LA PRDDUCCION Y TECNOLDGiA

employees at BT's R&D facilitres Ire on the grades at NPG2and NPG4. These are essentially technical scales. PCGUand above are managerial grades in the sense that theyinvolve supervision over other staff and responsi bility forproiects. This means that the prospects of promotionalong a technical career path are a great deal morerestricted than the scperv.sory path A part of the payaward is determined by performance rating. However,given the promotion structures, most employees with10 years in Adastral Park will have reached the top at thetechnical scale Largely in response to this and to stopemployees with advanced technical computing skillsleaving the company, BT set up the protessionaltechnical grade (PTG) grade in the late 1990s thatIncreased salaries by roughly 20% This was inrecognition at the importance at retaining tecmical skillsin the company. Nonetheless, the more limited careeropportunities tor purely technical grades suggests that itis the supervisory, managerial and organisational skillsthat the firm considers as the key tirm specitic knowledgeagents in contrast to the technical grades.

Numerous comments regarding the limitations of theBT career grade structure were provided by thoseknowledge workers interviewed. These views highllghl theincreasing difficulty BT has had in affracting a layer atyounger employees to commit themselves to a longer runcareer in the organisation at the time of the interviews. Ayoung felemarketing software designer with 5 yearsexperience in BT makes the point

"There is a clearly defined career path, but it'smanagerial. The next step up is to manage morepeople, but not doing any technical work, so if youwant to do technical work there is no career paththere. I think there is just too much emphasis onbeing a manager, there's no career path if you wantto be good technically, YOU'dsoon come to an endon that path. When you have a corporate review ofyour performance, innovation is on there, but irsnot very high. What is really rated is communicationskill, how much can you convince everyone else inthe unit how good you are, If you do that you gel ahigh mark".

Similarly an Internet expert comments:"There is a career ladder in that there are grades thatone can be promoted to, it doesn't in that there areno specific processes that if you go through will getyou promoted We don't have something thai saysonce your performance in these particular areas iscompetent, then you're promoted".

However the appeal of internal labour marketstructures for career development is influenced not onlyby the nature of the internal promotion ladder. It is alsointluenced by the opporlunity costs, I.e the risks andrewards inherent in developing a career outside of the firmthrough inter-firm mobility Particularly during the heightof fhe telecommunications boom in the late 1990s, thisoption became increasingly viable and atlractive for fhebest skilled staff. This can be illustraled Illstly in relation fapay grades amongst peers. Salary ranges for managementgrades' in 8T I" between £18,000 and £38,000'. Yet thisis well below the wages both contractors receive andcompetitors in the software, telecoms and banking industrywere offering in terms of remuneration. In this context 8TWill clearly struggle to lock in ifs younger and besf skilledR&D staff, particularly as skills appear to be highlytranslerable. Underlining the impact that mobility atknowledge workers can have in the firm, a senior mana-ger in the Research department comments:

"We're tending to lose (in research) about 2% a yearthrough turnover. At meta-level, the organisations BTlooks at are old suppliers, Marconi, GEe, bigtechnology companies and here our benchmark payswell But in R&D, our competitors aren't those peopleany more, they are Silicon Graphics, Nortel, Banksand we lose people hand over list to start-upcompanies, Our guys who developed the best in theIP world ailiefl and trebled their pay. Our best securityguys went to a bank, doubled Iheir pay and got a treehouse. BT cannot match thaI You have to accept tworealities, Firstly you have to have a more short-termview, which to some extent is not a problem, hall-lifeof a degree is about three years. On the other handthe movement out does create problems. A lot of theguys that come from university now are very keen ontheir GVs. BT gives training but we are notconverting so much of it into product now"(my emphasis)

A career option for employees interested indeveloping innovations outsns of the normal BT careerpath has also emerged through fhe so-called Bright-starincubator. This initiative was aimed at providing promisingemployees an incentive to develop product innovationsemerging from BT patents through the creation of a start-up firm linked to BT During the course at the interviews itwas reported that BT had some 1200 patents laying on

1 All B1 contracted R&D employees are formally on managemenlgrades, even if they have no supervisory responsibilities,

8 Based on 1999 figures.

191INNOVAR, revist8 de ciencias administrativas r socees. No. 21, eeem . JIJniode 2003

INNOVAR. revista de ciencias adrninistrativas y sociales

the shell not being exploited, suggesting some stifling ofentrepreneurial talent within the organisation. The so-called Brights/ar Incubato~ was sel up by BT as an attemptto overcome this problem by tapping into the skills andinitiatives of R&D employees. Ostensibly the initiativeprovides the opportunity tor groups or individuals withpromising projects to create start-up firms on-site andallract outside venture capital In so doing the normal BTsupply chain process IS detoured and individuals whoare wililng to take the fisk are given the opportunity toreap large rewards by creating a new trrn While In theincubation period, 8T continues to fund the researchers'salaries. However, after the star: up is launched, itbecomes Independent, although BT retarns a stake inthe new company at between 25% and 50%. Brightstartherefore represents an institutional innovation that aims10 encourage a more diverse industrial structure and amore entrepreneurial culture of innovation around thecompany. However this process also ties in to providinga career option for the most talented and promising skillsin BT. As one of the BT employees involved in aBrightstar initiative comments:

"The Brightstar incubator is based on the recognitionthat maybe the lraditional routes 01 trying to get thetechnology downstream weren't working. So this wasa way of avoiding the normal rules of BT productdevelopment. So you have to give up your secureday job with steady pay to become the part owner ofthe new company with a technology you arepassionate about and potentially make millions".

Commenting on his career prospects within BT prior10 the launch 01 Brightstar, the above researcher made thepoint

"Number one priority for my job has to be somethingI get motivated on, something I'm interested in Butalso very much how useful those skills are going tobe 10 me in my career progression. A year ago I wastrying to progress up the technical career ladder. Forme the problem in BT has always been that once youget to PCGU you have to leave the technical side andmove on to the managerial career. Becoming a PTG isslightly higher profile but that's about as far as I cango on the technical grades. I became an NPG2 9 yearsago, and that's typically how long it would takepeople to work their way up to, almost to PCGU"'".

ncebetors of the son set up by 8T have been previously establishedby firms such as Hewlett Packard 10 encourage entrepreneurialskills.

10 See figure 4 lor 8T career grades,

192

Clearly Brightstar offers an alternative career palh forthe brightest individuals. Although this option may involvea long term separation of BTwith some of its best technicalstaff, by maintaining the new spin-offs on site, BT canpotentially tap into the tacit knowledge of its ex-8T stafand raise the profile of Adastral Parkas a high technologycentre, attracting other knowledge workers and innovativeventures. This suggests that the Incubator has a dualrole. One IS to by-pass the normal, longer and morebureaucratic route of product development. Secondly,to encourage the mosltalented researchers to developinnovations that are in some way linked to 8T in thecontext of the limited career opportunities for individualsdeveloping pctentially important technologicalinnovations in BT.

Although this inrtiative can be seen as a means toinject some new radical ideas and innovations into theorganisation, the incubator represents a completelydifferent organisational structure, restricting knowledgespillovers into BT.The following point was made regardingIhe move to the incubator.

"One of the things that changed upon moving to theincubator (was that) the free flow of ideas on site hasbeen restricted, now we don't really want to talk topeople as much as before. There is a vested interestin making sure you keep the ideas to yourself becausein this environment it is the ideas that are the driver,while before the currency was good relations with therest of lhe business, which was based on an inter-flow 01 ideas".

The implications are that the release of entrepreneurialtalent involves creating a differenl structure of organisationwith separate rewards and incentives tor Ihe best talents.This is best mel outside 01rather than within the firm.

GESTION DE LA PRDDUGGION Y TEGNOLOGiA

Widening the use of contractlabour In R&DAs mentioned earlier, changing labour market structuresis also associated with the growing use of contract labourin high skilled environments. Overall the number ofcontract and agency stall 10 BT increased tram 1000 in1980 to 21727 in 1997 (CWU. 1997). BTs policy hasbeen not to use contract labour in the Researchdepartment. However In the Development department theiruse has become more widespread. Contract labour, byIts very nature, would tend to be associated with a relativelynarrow set of clearly detinable skills. The ability atcontractors to otter clearly recognisable and transferablesets at skills determines their relative position and strengthwithin the labour market. Given their relatively narrowset of skill sets, it would be reasonable to assume thatcontractors would be brought IOta BTs R&D area toundertake a set at very specific tecmi cal tasks requiringless Ilrrn specific knowledge, particularly when thereIS surge of demand that cannot be covered bypermanent labour.

Gonventlonal wisdom would also indICate that thenature of the employment conlract creates a dillerent set ofincentives for the contractor than the permanent employee.Permanent statt may well be encouraged to improveproduct quality 10 order to enhance their promotionopportunities and general standing in the firm. Thetemporary nature at the employment contract lor contractstaff might be expected to encourage contractors to meetdeadlines rather than improving product quality and at apersonal level to follow a career strategy based onacquirinq new skills.

It appears however that this is not always the case.The view of the line manager interviewed clearly indicatesthat contract labour does not negatively Impinge on anyaspect at innovation output. It anything, there appears tobe advantage to using contract workers 10carry out eventhose jobs traditionally associated to permanent labour.The line manager comments:

"There are 40 contractors in at the moment in thisarea. The main advantages lie firstly in that some ofthese people are very good, to be honest a lot of thecontractors are better than a lot of our permanent staff,Part of the reason lor this is that, well, they becamecontractors because they had enough "get up andgo" and they must have courage and fortitude. Youalso have to bear in mind that a lot of our contractorshave been with us for quite a long time, lhey tend toget involved in the work for years on end, and they

earn fat salaries. So many are longer term and haveput down roots. I don't have problems with the use ofcontractors, except that there pretty expensive, the ideaof using a resource centre that is systematicallymanaged works line lor me They give me Ilexibilitythat I can hire and fire from a managed pool".

Siqnilicantly, both the formal skill profile and moretacit firm specific knowledge of the contract workerswas not judged to be a great deal dillerent from that atpermanent statl. Many of the contractors had BT speciticskills because they had been on the BT payroll In thepast.

Nevertheless interviews with permanent contract 8Tdesigners working on projects with contractors cast somedoubt on this managerial view. Responding to a Questionon whether in his experience the project teams were affectedby working with contractors, a designer comments:

I've worked in teams where I have been the onlypermanent member of staff, and that has not workedwelt The dillerence is Ihal at the end at the day thecontractor is motivated by the next contract he canget, and that is determined by the skills he's got andso he's more interested in developing skills thanproducts, A lot of contractors, if they're stuck on ajob for six months, they can find themselves pigeonholed and they wouldn't be able to gel a job doinganything else but thaI. So you find if you get a lot ofcontractors en masse, trying to deliver soltware by acertain time to a certain standard can be extremelydifficult. The other factor is that there is no performan-ce review lor contractors, no pressure to say you'vedone a good job, we're going to give you moremoney or promotion. Contractors give the companyflexibility to get rid of people Quickly, but it doesn'tmean you will get better software because there arethese mouvauon problems, where they have aninterest in developing skills that are not necessarilyrelated to the project. because you can put it downon your CV and it will get you that next contract".

These comments highlight the dangers at relYing toogreatly on shari-term labour at this level. Clearly theincentive mechanisms associated with the employmentcontract playa key role, although it is sipniticant that inspite at repeated questioning on the same theme, theproblems signalled by the above employee were notpicked up by the more senior project manager Interviewedearlier In fact It appears that BT is considering the morewidespread use 01 numerically llexiole labour, includingin particular contractors.

193INNOVAR, reaste de ciencias adminislrativas y saciates. No.2 I, enero . junio de 2003

INNOVAR, revisra de ciancias edmlnieuativas V secteles

SummaryThe discussion has highlighted two contrasting trendsapparent in 8Ts organisational structure. The first trendemphasises the development of specitic routines andsynergies to speed up the pace In the roll-out of newproducts and services. In Lam's (2000) typology itemphasisesanembedded knowledge environment.Hencethe importance at coilective or tlrrn specific torms ofknowledge needed to enhance lirm competencies andspeeding up the rollout of new services for a masscustomer base was emphasized. The interviews alsoconfirmed the view from managers that organisationalinnovations that resemble the competence rank rules inMarsden's matrix were being introduced. This stressedintra-tirm mobliity and secondments of employees withinthe tirm to follow through specillc Innovations.

A second distinct and dominant trend was underiinedby a move to an embodied knowledge environment Thiswas associated with rapidiy broadening tirm competenciesand combining firm specitic with market knowledge.Investment in the re-training of researchers in newknowledge areas was accompanied with a policyencouraging outside recruitment This allowedcompetencies to be developed more quickly than internalre-training permitted. More flexible opuons to encourageemployees with marketabie innovations to set up their owntirms outside of, but linked to, 8T were also encouragedthrough the Incubator. Simiiarly, the growing use atcontract labour in innovationprojects underlines the morearms length relationship a growing number at employeesare having with the firm.

The discussion howeveralso errptasseo the difficulty8T was experiencing In deveioping these duaiorganisational structures, partiy because at the contrastingincentives this implies for employees. In particula, theopportunities for inter lirm mobilify at labour amongst thehighest skilled knowledge workers was such, that 8Tinitiatives directed towards reintorclng film specific skiils,such as secondment of researchers to foilow throughinnovations, were not generaily taken up by the mosttalented and skilled staff 8Ts oifficulties in locking in thebest entrepreneurial talents were reinforced by the limitedpromotion opportunities tor technical as opposed tosupervisory management grades. This emerges trom theformal hierarchical grade structure that large firms such as8T inherited trom the past and are required in such largeorganisations.

The discussion therefore underlined the diffiCUlty inexploiting fuily tirm synergies and organisation learningassociated to an embedded system of firm organisation,

194

because at the un-viabiiity of, using Marsden'sterminoiogy, the competence rank system amongst thebest skiiled employees. 8y contrast, the trend towardsestablishing looser and more tlexible ties between theworklorce, as evidenced by the higher turnover and useemployees on non-standard contracts, appears to be Inthe ascendancy. This emphasises the dominance of aquaiification ruie environment in Marsden's typOlogy,linked to an embodied organisational structure.

ConclusionsOverail the case study and lramework emphasised theimportanl roie that fabour markets can play in shapingfirm competencies. In particular, the case study supporteda key proposition that tlowed lrom the lramework, that Inthe current technologicai paradigm, firms embarking onnew strategies to change the nature of their competencies,require changes not only in the division 01 labour, butalso that the labour market institutions within and outside01 the lirm are coherent wilh the type of knowledge transferassociated to particular firm structures. In particular,strategic changes in innovation strategy need to take intoaccount how they implicate employees' own strategies ofski lls and career formation to be eflective.

The need for 8T to maintain a stable and committedworkforce imbued With the knowiedge 01 the organisationand its ways 01 functioning is acknowledged by managersand many existing employees have been re-trained andare highly respected in the new knowledge areas in whichthey work. Nevertheless, it is also apparent that labourmobility isa growing trend, particularly amongst the mostskilled and younger entrants to the labour market This isa phenomenon over which even large IIrms such as 8Thave little control.

The key question that emerges is what impact mightthis trend towards looser and more flexible lies with theworkforce have lor the abil ity offirms to develop differenttypes 01 innovation? As has been discussed, the greatermovement of researchers in and out of firms, far from

GESTION DE LA PRODUCCION Y TECNOLOGiA

debililating Ihe firm's knowledge base, can act as a means10 diversity a lirm's knowledge base, allowing tlrms suchas 8110 be more responsive to rapid changes in the marketIn me conlext 01 Ihe shorler life span of technologies. Inparllcular, if BT is able to atiract more stan up venturesaround Adaslral Park, it may be able 10 create a dynamicnetwork of skills around which II can renovate itscompetencies, much in the same way that for exampleIntel was able 10 do in Massachusetis around route 128In the USA during Ihe 1990s. Hence Ihe iabour marketcan help to redefine firm competencies.

Neverlheless, the Implicalion of Ihe case sludy isalso Ihal BTs abilily to develop essential competenciesrelianl on long-Ierm leam building and lirm specificknowledge may be undermined by Ihe grealer mobilityof labour and shorter lime Ihat skilled staff spends oneach project This Is highly relevant for process-Iypeinnovalions associaled wilh producl support and productresilience thai still play an important role in thetelecommunications service sector.

Thus from a broader policy poinl of view, Ihederegulaled nature of the UK labour market, the absenceof a common apprenticeship programme accepled byemployers and employee bodies and the highlylransferable nature of skills may have some negative long-lerm consequences. This issue has been Ilagged up by anumber of authors such as Prais (1995), Walker (1993)and Steedman and Wagner (1989) as a major weaknessof Ihe UK economic system This emphasises Iheimportance of seeing labour markets as an integral element01 Ihe national system of innovalion and the developingthe policy implicallons thai flow from this.

ReferencesAoki, M. (1988), Information, Incentives and Bargaining Structure

in the Japanese Economy, Cambridge: Cambridge UniversityPress, As cited in

Andreasen, L.E. Doriat, B. den Henoq. F. Kaplinsky, R. (1995),Europe's Next Slep, Organisafionallnnovalion, Competitionand Employment, London, case

Analysys, (1997), TelecomsOperators Mus! Re-Invenl Themselvesto Exploit Po/ential 01 the Internet, Analysis News andFeatures, Analysys Web Sile, www.analysys.Goml.

Batt. R. (1995). Pertormance and Welfare Etfects of WorkplaceRestructuring: Evidence from Telecommunications Services.Ph.D dissertation, Sloan School of Management, M,I.T.

Ball, R. and Darbishire. 0, (1997).lnstilulional Determinants ofDeregulation and Reslruc/uring in Telecommunications:Britain, Germany and the United Stales Compared.International Contribution 10 Labour Studies, pp59-79.

Best, M, (1990). The New CompeWion: Institutions of IndustriaiRestructuring, Cambridge. MA: Harvard University Press,

Blackler. F, (1995), Knowledge, Knowledge Work andOrganizations: An Overview and Interprelalian."Organizational Studies" 16/6, 1021-1046,

Fransman, M. (1998). Analysing the Evolution of Induslry: TheRelevance of the Telecommunications Industry. DRUIDSummer Conference, Campe/ences, Governance andEntrepreneurship. Bornholm, June 9-11.

Freeman, C and Perez, C, Chapter Four in Dnsi. G at al. (1988),TechnicalChange and Economic Theory, Pinter, London, PP38-66

195INNOVAR, resists de ciencias administrativas y satiates. No.2 I, enero . jun;o de 2003

INNOVAA, reveta de ciencias administrativas y snciales

Kavassalis. P.Solomon, R.J. and Benghazi P.J.(1996). TheInternet:a Paradigmatic Rupture in Cumulative Telecom Evo/ulion."Industrial and Corporate Change", Vol 5 N04, pp1097-1127

Kline, S,J. and Rosenberg, N. (1986), An Overview of Innovation,As cited in Andreasen, L,E.Coriat, 8, den Herlog, F. Kaplinsky,R (1995), Europe's Next Step. Organisalionallnnovation,Competition and Employment, London, Casso

Lam. A. (2000). Tacit knowledge, Organisational Learning andSocietal Institutions. An Integraled Framework,Organisational Studies, 21 (3): 487~513.

Lundva!l. B,-A. and Johnson, B, (1994). 'The Learning Economy',(with Johnson, B.) in Journal of Industry Studies, Vol. 1,No.2, Dec., pp. 23-42,

Mansell, R, (1993). The New Telecommunicalions. London, SagePublications Ltd,

Marsden, 0, (1999), A Theory of Employment Systems, Oxford,Oxford University Press.

DECO, (2001), Innovative People: Mobility of Skilled Personnel inNational Innovation Systems: Proceedings from the DECOWorkshop on Science and Technology Labour Markets,DSTI/STP ITI P(99) 2/FI NAl, DEC D, Paris.

Ramirez, M, (2002), Redefining Firm Competencies, Innovationand Labour Mobility: A study of R&D in British Telecom.Working Paper02/04, Department of Management BirkbeckCollege.

Saxenian. A. (1996) Beyond Boundaries: Open Labour Marketsand Learning in Silicon Valley In Arthur, B,M, and Rousseau.D,M. (eds), The Boundaryless Career. A New EmploymentPrinciple for a New Organisational Era, OXford, OxfordUniversity Press,

Steedman, H. and Wagner, K. (1989), Productivity. MachineryandSkills: Clothing Manufacture In BritaIn and Germany, NationalInstitute Economic Review,

Strauss, A. (1987). Qualitative Analysis for Social Scienlists,Cambridge: Cambridge University Press.

Tomlinson, M. (1999), 'The learning economy and embodiedknowledge flows in Great Britain', Journal of EvolutionaryEconomics, Vo1.9, No.4, pp,431-451.

Walker, W, (1993), National Innovation Systems: Britain.ln Nelson,R. (eds), National Innovation Systems. Oxford, OxfordUniversity Press.

196