WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE...

27
UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS Introduction & Chapter V International Production System UNITED NATIONS New York and Geneva, 2002

Transcript of WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE...

Page 1: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

UNCTAD/WIR/2002

UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva

WORLD INVESTMENT REPORT, 2002

PART TWO

TNCs & EXPORT COMPETITIVENESS

Introduction &

Chapter V International Production System

UNITED NATIONS New York and Geneva, 2002

Page 2: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

Introduction to Part Two

INTRODUCTION

One of the contributions TNCs canmake to host economies in the developingworld i s to enhance thei r expor tcompetitiveness. Export competitivenesshas many facets, the most obvious implyinghigher exports. But it also means diversifyingthe export basket, sustaining higher ratesof export growth over time, upgrading thetechnological and skill content of exportactivity, and expanding the base of domesticf i rms able to compete g lobal ly ; thus ,competitiveness is sustained and it is generallyaccompanied by rising incomes. TNCs canhelp raise competitiveness in developingcountries in some or all of these ways, buttapping their potential is not easy. Attractingexport-oriented TNC activities is itself anintensely competitive business, and even someof the countries that have succeeded mayfind it difficult to sustain competitivenessas their wages rise and market conditionschange. Coherent and consistent policy supportis essential to ensure that attracting export-oriented TNC activities are embedded in abroader national development strategy. Thisis particularly important as there is a possibletension between the principal objective ofGovernments – which is to maximize nationalwelfare – and the principal objective of TNCs– which is to maximize their global corporatecompetitiveness. Export competitivenessis important and challenging, but it shouldbe seen not as an end in itself but as a meansto an end – which is development.

The link between FDI and trade isnot new but well worth revisiting, especiallyin the light of the growing attention to thechallenge posed by competitiveness,1 butalso in the light of the changing nature ofinternational production systems, the growthof supplier networks and new multilaterald isc ip l ines . In WIR99 , as par t of theexamination of FDI and the challenge ofdevelopment, one chapter was devoted tothe link between FDI and trade. The principalconclusions were that TNCs exert a strong

influence on the patterns of world trade,that much of the international flow of goodsis handled within TNCs in the form of intra-firm trade, and that inward FDI has contributedto boosting the export performance of a numberof developing host countries. This year ’sreport builds on the findings of WIR99 bylooking in greater detail at the role of TNCsin increasing export competitiveness and thecorporate strategies that are driving mostof the recent changes in the pat tern ofin ternat ional t rade . These have to beunders tood in order to achieve a bet terunderstanding of the role that policy canplay to foster export competi t iveness inassociation with TNCs.

Accordingly, Par t Two of WIR02explores the changing nature of expor tcompetitiveness and the role that TNCs playin enhancing it in different countries andactivities. Part Three then deals with thepolicies developing countries (and economiesin transition) might consider to attract export-oriented FDI and benefit from it.

That export competitiveness2 is ofgrowing interest to countries at all levelsof development is clear: the sheer volumeof analysis and benchmarking testifies tothe importance that Governments attach toit. Developed countries regard competitivenessas a prerequisite for maintaining high levelsof income and employment.3 Developingcountries find it essential for development.Improved export competi t iveness al lowscountries to earn more foreign exchange andso to import the products , services andtechnologies they need to raise living standardsand productivity. Greater competitivenessallows developing countries to diversify awayfrom dependence on a few primary-commodityexports and move up the skills and technologyladder, which is essential to sustain risingwages. It also permits the realization of greatereconomies of scale and scope by offeringlarger and more diverse markets.

Page 3: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

New forms of export competitiveness,geared to international systems of production,can al low developing countr ies to entertechnology-intensive activities that they couldnot otherwise undertake. In the process, itallows them to build new productive capacities.Exporting feeds back into the capacities thatunderlie competitiveness: exposure to worldcompetition provides enterprises with greateraccess to information and technology thanexposure to domestic competition alone andleads to more vigorous efforts to acquirenew skills and capacities.4 An export-orientedeconomy also tends to attract more efficientTNC activities, which reinforces the upgrading.

There is ample recent evidence thatexport-oriented development is not only feasiblebut also rewarding. However, raising exportcompetitiveness means more than liberalizingtrade and investment. The most successfulcountries had to make determined effortsto develop new capabilities and to attractforeign capabilities to complement domesticones. All drew heavily on new technologiesfrom TNCs, and many, but not all, reliedon FDI or non-equity forms to spearheadthe process.

This is not to argue that building exportcompetitiveness is a complete strategy foreconomic development. It is only one of anumber of elements needed for development.It has to be complemented by measures toensure that the non-export sectors of aneconomy grow and that the benefits of growthare spread throughout the economy. If theexport sector is de-linked from the rest ofthe economy, it is possible to improve exportcompetitiveness without raising growth ratesor living standards for the population at large.This is, however, rare; a number of exportingeconomies have been able to grow on theback of their export drives, including bymoving up the technological ladder fromlabour-intensive activities towards technologyand skill intensive ones (UNCTAD, 2002a).

TNCs can contribute to the exportcompeti t iveness in host countr ies . Theircontr ibution is important in technology-intensive and internationally branded products,but it goes much further. They have alwaysbeen significant players in primary exportsand the main source of new indust r ia ltechnologies for local exporters (through arm’s-length l icensing and original-equipment-manufacture arrangements). With the spread

of global value chains in many low- andmedium-technology activities, TNCs are nowinvolved in the whole spect rum ofmanufactured exports. In some low-technologysegments, other international players are alsoactive, and TNCs often take the role ofcoordinating local producers in addition toset t ing up their own aff i l ia tes . In manytechnologically complex activities, TNCsare particularly important because a largepart of trade is internal to their internationalproduction systems.

While the growth of internationalproduction systems is well recognized, itis less well known that there is a growingtendency for firms, even large TNCs, tospecialize more narrowly and to contractout more and more functions to independentfirms, spreading them internationally, to takeadvantage of differences in costs and logistics.Some are even opting out of productionaltogether, leaving contract manufacturersto handle it while they focus on innovationand marketing. The main suppliers and contractmanufacturers are themselves often largeTNCs, with global “footprints” matching thoseof the i r pr incipals and wi th the i r ownsubcontractors and suppliers. However, TNCsalso increasingly use national suppliers andcontractors in host economies. Specializationdoes not stop here: leading TNCs are alsoentering into joint innovation arrangementswith other firms – competitors, suppliersor buyers – and with institutions such asresearch laboratories, universities and soon. Thus, the emerging global productionsystem is becoming more multifaceted , butwith tighter coordination by lead playersin each international production system.

What lies behind these trends? Threeforces are driving them. The first is policyliberalization, which opens up national marketsand allows all kinds of FDI and non-equityarrangements . The second is rapidtechnological change, with its rising costsand risks, which makes it imperative forfirms to tap world markets and share thecosts and risks. Technological change – inparticular, falling transport and communicationcosts, the “death” of distance – also makesit economical to integrate distant operationsand ship products and components acrossthe globe in a search for efficiency. Thethird, reflecting both of these, is increasingcompetition, which results in unexpectedforms of relocation to new sites, with new

Page 4: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

Introduction to Part Two

ownership and contractual arrangements, andinvolving new activities.

All the signs are that the export roleof TNCs in host countries, through both FDIand non-equity arrangements, will grow further.I t wi l l cont inue to take new forms andincorporate new locations. The potentialfor generat ing expor ts f rom developingcountries and economies in transition is thushigh. But, as noted, tapping this potentialis not easy.

Part Two of WIR02 deals with thesethemes. It starts with the characteristics ofinternational production systems, exemplifyingthem for a number of firms. It describeschanges in global competitiveness patterns,the direct role of TNCs in exports (theirindirect contribution is difficult to tracequantitatively) and the main country “winners”in export competitiveness. The main messagesof the analysis are as follows.

Trade patterns are changing rapidly,with technology-intensive activities growingconsistently faster than others. As a group,developing countr ies and economies int rans i t ion have done wel l in expor tcompetitiveness, rapidly raising their marketshares and upgrading into advanced activities.

TNCs have played an important rolein the exports of many countries, directlyby establishing in those countries affiliatesincorporated into the TNCs’ internationalproduction systems, and indirectly by enteringinto contractual arrangements, especially withsuppliers linked to the TNCs’ productionsystems. However, such systems are stilllargely concentrated by country, region andactivity. It is possible that the export dynamismseen in the leading “winners” will spreadto other developing countries and economiesin transition as international production gatherspace and increases in scope. But there arerisks and there are opportunities. First therisks:

• The bulk of TNC-related export activityin developing economies and economiesin transition is concentrated in a handfulof economies, mainly in East and South-East Asia and in regions contiguous toNorth America and the European Union,although TNCs are also significant playersin many countries that are not major globalexporters.

• It is unclear whether some large productionsystems can spread further, for technologicalreasons . Once product ion has beenrationalized to serve regional or worldmarkets, first movers tend to build strongcumulative advantages, reaping economiesof scale and scope and drawing uponclusters of suppliers and institutions.

• The entry-level requirements of competitiveproduct ion are r i s ing. Not only i stechnological progress pushing up skillrequirements, TNCs also increasingly needefficient supplier networks that can operateglobally and at much higher levels oftechnological sophistication than before.

• Even “insiders” to international productionsystems face uncer ta in ty about thei rprospects. A number of low-technologysystems such as texti les, clothing andfootwear, on which many countries haverelied, may have peaked in growth. Themore high-technology systems are imposingmore stringent demands on participants.Host countries that cannot muster newcapabilities may lose their competitive edge:even first-mover advantages may not lastwithout upgrading.

• A concentration of exports from developingcountries on a few manufacturers mightlead to oversupply and a subsequentdeterioration of the terms of trade – oftenreferred to as the “fallacy of composition”di lemma (UNCTAD, 2002a, ch . 4) .

• Some trade arrangements that caused aspreading of international production insome low-technology industries will bephased out or else eroded by further tradel i b e r a l i z a t i o n . T h e M u l t i f i b r eArrangement i s an example .

But there are also opportunities:

• International production is spreading andcan be expected to do so as countr iesfurther l iberalize trade and investmentregimes and improve their infrastructureand ski l l s , and as compet i t ion pushesf i rms to spread the i r ac t iv i t i es morewidely to strengthen their competi t iveadvantages .

• Leading TNCs are increasingly drawingi n d e p e n d e n t e n t e r p r i s e s i n t o t h e i rp roduc t ion sys t ems : inpu t supp l i e r s ,service providers and strategic partners.T h i s o f f e r s c o n s i d e r a b l e s c o p e t oenterprises from host economies that buildthe capabi l i t ies to meet TNCs’ needs .

Page 5: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

• TNC suppliers and contract manufacturersa r e g o i n g t r a n s n a t i o n a l t o r e t a i ncompetitiveness and to serve lead firmse ff e c t i v e l y, o p e n i n g n e w s o u r c e s o fe x p o r t - o r i e n t e d F D I f o r d e v e l o p i n gcountr ies .

• Rising costs and congestion may at somestage offset first-mover advantages, andactivities then spread to cheaper and lesscongested areas .

• The increased tradability of services aswell as of specialized service functionsassociated with international productionsystems will open entirely new areas foran in ternat ional d ivis ion of labour.

• Advances in transport technology mayopen up new possibilities for high-valueagricultural and other primary products.

• New global ized act ivi t ies wil l emergeas the economic logic of relocation andn e t w o r k i n g s p r e a d s . T h i s i s a l r e a d yevident in the software industry, wherethe “death” of distance is most evident,but it will also affect other service andmanufactur ing act ivi t ies .

It is too early, of course, to forecastthe net outcome of all these trends. The resultsare very likely to be industry- and context-specific. The forces driving internationalproduction systems are strong, but competitionfor export-oriented TNC activities is alsorising. Achieving and sustaining export-compet i t iveness ca l l s for h igher localcapabilities in all activities and all countries.If developing countries and economies intransition are to strengthen competitiveness(with or without direct TNC participation),they will have to strengthen their capabilities,attract and stimulate activities suited to theirendowments, and upgrade them over time.

Notes

1 For a discussion of the broader question ofTNCs and competit iveness, see WIR95 .

2 “Export competitiveness” is taken here in thebroad sense outlined above. Thus, it does notmean raising world market shares by keepingwages low, but sustaining world market shareswhile raising incomes.

3 For instance, the OECD has this to say:“Competitiveness [should] be understood asthe ability of companies, industries, regions,nations and supranational regions to generate,while being and remaining exposed tointernational competition, relatively high factorincome and factor employment levels on asustainable basis” (OECD, 1994, p. 23). TheGovernment of the United Kingdom’s ThirdWhite Paper on competit iveness starts:“Improving competitiveness is central to raisingthe underlying rate of growth of the economyand enhancing living standards… The needto improve our competitiveness is not imposedby Government, but by changes in the worldeconomy. Improving competitiveness is notabout driving down living standards. It is aboutcreating a high skills, high productivity andtherefore high wage economy where enterprisecan flourish and where we can find opportunitiesrather than threats in changes we cannot avoid”(United Kingdom, Cabinet Office, 1996, p.10). For an analysis of the concept ofcompetit iveness, see Lall , 2001.

4 There is a debate on whether exporting leadsto greater enterprise productivity or vice versa.The relationship between the two is probablyinteractive (Westphal, 2002). The technologycapability literature establishes that export-oriented economies enjoy healthier and morecompetitive capabilities than inward-lookingones, though even the former may need a periodof infant-industry protection when buildingadvanced capabilities in local enterprises (Lall,2001b).

Page 6: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

CHAPTER V

INTERNATIONAL PRODUCTION SYSTEMS

A. Drivers and features

TNC act iv i t ies af fec t the expor tperformance of host countries through arange of equity and non-equity relationships.What i s common to a l l of them is tha tproduction – and, more broadly, the operationsof a firm – is organized under the commongovernance of TNCs. During the past 15years , fa l l ing barr iers to in ternat ionaltransactions have not only invigorated globalmarkets through arm’s-length transactionsbut given rise to elaborate corporate systemsof organizing the production process. Asa result, international production systemshave emerged within which TNCs locatedifferent parts of the production processes,including various services functions, acrossthe globe, to take advantage of fine differencesin costs, resources, logistics and markets(WIR93) . What is distinct about the riseof international production systems as comparedto ear l ier organizat ional s t ructures andstrategies characterizing TNC operationsis, first, the intensity of integration on regionalor global scales and, second, the emphasison the efficiency of the system as a whole(Kaplinsky, 2000, p. 122). In other words,global markets increasingly involve competitionbetween entire production systems, orchestratedby TNCs, rather than between individualfactories or firms.

The rise of such integrated internationalproduction systems reflects the response ofTNCs to dramatic changes in the globaleconomic environment, and, in particular, theirsearch for enhanced competitive advantagethrough an optimal global configuration ofwhere they produce and how they coordinatetheir production activities. Several forces combineto drive this process:

• The rapid decline in barriers to internationaltrade and investment flows. The creationof a corporate international productionsystem is made possible by the freedom

to move goods, services and knowledgeamong l inked corporate ent i t ies wi thminimal impediments, even when theseare situated in widely dispersed locations.

• The greater ease of the international flowsof goods , services and ideas and theresulting drop in the costs of cross-borderbusiness coordination. Advances in transportand communication technologies havecontinued apace in recent years, and TNCshave deployed new systems to enhancetheir internal and external coordination.The advent of the Internet has dramaticallylowered the costs of information exchangeover great distances. More streamlinedand standardized customs administrationand port operations as well as internationaltelecommunication gateways and satelliteand fibre-optic networks further facilitateinternational production, including thecoordinat ion of knowledge- intensivefunct ions such as design and R&D.

• These two forces have, together, led tostronger competition among leading TNCs.In a growing range of industries, majorproducers must contest all of the marketsin which their competitors operate anddraw on sources of competitive advantagewherever they may be located. No longercan a dynamic firm focus only on familiarmarkets, since its competitors may mobilizeprofi ts or resources across the globe.

All of this has led to profound changesin industrial structures. In some industries(e.g. semiconductors and automobiles), theyhave led to consolidation and oligopolisticcompetition; in others (e.g. garments), toa di ffus ion of market power. Corporatestrategies have evolved too, as TNCs havesought to devise new forms of governanceto manage their dispersed activities.

International production systems arethus evolving as corporations respond toeconomic and technological forces. Giventheir growing importance in shaping investment

Page 7: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

and t rade pat terns , unders tanding thei rdynamics is essential for any developingcountry that seeks to enhance i ts exportcompetitiveness through the activities ofTNCs.

Three core elements of internationalproduction systems are critical in this context:governance, global value chains and geographicconfiguration.

The first element is governance , orthe structure of control that determines thegeographic and functional distribution ofbusiness ac t iv i t ies and ensures the i rcoordination. International production systemgovernance occurs in forms ranging fromownership (or equity) linkages that providedirect managerial supervision, to variousnon-equi ty l inkages in which formal lyindependent in termediar ies – suppl iers ,producers and marketers – are linked througha variety of relationships such as franchising,licensing, subcontracting, marketing contracts,common technical standards or stable, trust-based business relationships. Intel has createdan international production systems in whichequity – or ownership – links form the basisfor common governance of the membersof the system (section B.1 below), whileLimited Brands has established a systembased on non-equity relationship (sectionB.2 below). Toyota exemplifies a mixtureof both approaches, by combining close linksamong i t s own ful ly-owned assemblysubsidiaries with a multi-tiered network offormally independent subcontractors (sectionB.3 below). The case s tudy of Ericsson(section B.4 below) traces the transitionin control from a largely equity-based systemto an almost fully non-equity-based system(although Ericsson has retained direct controlover manufacturing, i.e. over product linesrelated to its core focus on innovation anddesign). The continuum of internat ionalproduction system governance thus reflectsthe degree of control over corporate activities.Equity-based governance systems internalizecontrol and allow stronger protection of firm-specific advantages like technology, as inthe case of Intel. Where these advantageslie in brand names and marketing – as inthe case of Limi ted Brands – moreexternalized forms of control may suffice.

A re la ted i ssue i s the degree offunctional hierarchy in the system. In someinternational production systems, each stageof production is assigned to specific corporate

affiliates and deployed globally, with eachunit in the system guided by a headquarterscompany (somet imes refer red to as aninternational production system “flagship”– see Rugman and D’Cruz, 2000). In Intel’shierarchical international production system,individual affiliates specialize in particularstages of innovation or production and areclosely integrated into the parent firm’s globalnetwork. At the other end of the spectrumare systems in which an integrated set ofbusiness functions is decentralized to multiplecentres, often regional headquarters, thatwield considerable autonomy. An exampleis the distribution of “global product mandates”to various affiliates. In this case, internationalproduction system governance takes on amore horizontal character. Toyota presentsan intermediate case, in that affiliates inthe United States and Thailand have beengiven “regional product mandates” for certainproduct l ines . These two aspects ofgovernance (equity or internalized vs. non-equity or externalized, and hierarchical vs.horizontal) are related. In general, equity-based international production systems tendto be organized hierarchica l ly, whi leexternalized systems operate more horizontally.Yet the distinction is important, for the twocriteria are not always perfectly correlated.In the garments industry, for example, theinternational division of labour has, overt ime, become vert ical ly special ized andhierarchically integrated under the leadershipof brand-holding flagship companies likeLimited Brands despite the near-absenceof equity links. On the other hand, someTNCs with global networks of wholly-ownedsubsidiar ies have moved towards morehorizontal forms of coordination.

A striking recent governance trendin many manufacturing and service internationalproduction systems is a shift towards thesystematic outsourcing of a greater rangeof activities, including back-office operationslike customer service and even R&D. Thisreflects TNCs’ efforts to focus on their“core competencies”, i .e. those activitiesin which they can deploy propr ie taryadvantages, wield market power or otherwiseenjoy higher returns. The trend suggestsin particular that technological advancesand competitive pressures have altered thebalance between two opposing firm-specificadvantages sought by TNCs – internalizationversus specialization – in favour of the latter.The outsourcing trend has complex implications

Page 8: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

for global industry structures, creating entirenew industries. In particular, leading TNCsin a range of industries have begun to exitfrom manufacturing altogether. In response,contract manufacturers have emerged tospecialize exclusively in providing turnkeymanufacturing services (see sect ion B.5below). Contract manufacturing differs fromthe earlier system of original equipmentmanufacturers sub-contracting in that thebrand-holding TNC does not simply drawon subcontractors for extra production capacity,but rather outsources the entire manufacturingfunction for individual product lines or, insome cases like Cisco Systems, the entireproduct range.

Despite the outsourcing trend, directequity and managerial control remain keysources of competitive strength in certainindustries. Even when international productionsystems are highly externalized, TNCs typicallyexert powerful authority through their controlof key functions, such as brand managementand product definition, as well as throughthe setting and enforcing of technical, qualityand delivery standards throughout a networkof formally independent producers.

The second element of an internationalproduction system is the organization anddistribution of production activities and otherfunctions in what is commonly known asthe global value chain . I t extends fromtechnology sourcing and development throughproduction to distribution and marketing (figureV.1). The core competitive advantages ofTNCs can reside anywhere along the valuechain, although, in practice, they tend tocluster in one component. Value chains arebecoming fragmented, as business functionsare differentiated into ever more specializedactivities. Functional specificity allows TNCs

to distinguish activities with widely varyinginputs, capacity requirements and financialreturns, even within the same industry orproduction process. As a result , there isa genera l t rend towards funct ionalspecialization, which contrasts with the typeof ver t ica l ly in tegra ted s t ructures tha tcharacterized many TNCs until quite recently.Intel, for example, focuses on the designand manufacture of microprocessors anda few related products (see sect ion B.1below), rather than combining captive in-house chip production with the productionof computer systems, as was long the strategyof IBM, NEC and Samsung. In manyindustries, TNCs have recently tended tofocus more on the knowledge-intensive, lesstangible, functions of the value-chain suchas product def ini t ion, R&D, managerialservices , and market ing and brandmanagement. This has long been true ofthe garment industry, where leading TNCsfocus a lmost exclusively on design andmarketing (see Limited Brands, 2001). Evenin many “high-technology” industries, however,manufactur ing as wel l as logis t ics anddistribution have become standardized andmore easily tradable. In consequence, contractelectronics manufacturers have grown rapidly(section B.5 below).

In general, international productionsystems may be distinguished by the functionalactivity on which a TNC focuses and whichgives it governance authority over the broaderproduction system. Intel’s core strategy isto establish market dominance through theinnovation and design that differentiate itsproducts. Hence the activities that resultin a finished Intel product are those of a“technology-driven” international productionsystem. Toyota’s core competence is i tsability continuously to improve quality and

Figure V.1. The global value chain of product components

Source : UNCTAD.

Page 9: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

productivity in the complex automotive industry.Automotive international production systemsare thus “production-driven”. Finally, inmany consumer-product sectors, the abilityto build brand value and capitalize on changingtastes and styles is the key to competitivestrength, as exemplified by Limited Brands’success in the garments industry. Suchinternational production systems are “marketing-driven”.

These two elements of internationalproduction systems – governance and thevalue chain – can be used to analyse avariety of international production systemsinvolved in international investment and exports.Table V.1 provides a matrix of internationalproduction system systems with illustrationstaken from the case studies presented below.

The third element of internationalproduction systems, which holds particularinterest for developing countries, is theirgeographic configuration in an effort toacquire a portfolio of locational assets thatmaximizes the competitiveness of the corporatesystem as a whole. The past 15 years haveseen great changes in the determinants ofthe optimal location of TNC activities, andhence in the geographic d is t r ibut ion oftechnology, production and marketing activitieswithin international production systems.Production has been internationally dispersedfor decades, but the trend towards integrationover ever larger geographic scales is relativelynew. Supply chains have extended to newareas of the globe and integrated formerlydistinct regional production zones. Contractmanufacturing and the production of keycomponents in electronics and other industrieshave witnessed a consolidation trend in thepast few years. The surviving large-scalesupplier firms increasingly seek a globalpresence, particularly by co-locating nearthe key fac i l i t ies of the i r in ternat ional

production system flagship customers (Sturgeonand Lester, 2002). Related to this is therecent trend towards “postponement”, inwhich components are made or assembledas close to the final point of sale as possiblein order to reduce transport costs, whichremain important for a range of products.This may pose great challenges to those“winners” in the global export trade thathave previously thrived as accommodating,but distant, off-shore production platforms.

Perhaps a more striking trend hasbeen the geographic dispersal of other globalvalue chain functions. The internationalizationof business service and support functionshas progressed rapidly in recent years. Eveninnovation, presumably the function mostf i rmly anchored in home countr ies byspecialized skills and strategic motivations,is increasingly being carried out on a globalstage. Developed countries continue to performmost of the research a imed a t radica lbreakthroughs, but the developing world nowcarries out significant R&D.

Simplified, three sorts of drivers areacting simultaneously on the locational decisionsof TNCs and their partners in internationalproduction systems. Cost differentials remaina fundamental fac tor in the locat ion ofproductive activities. Changing governancemodels and functional differentiation (thefirst two elements of an international productionsystem) have subjected a growing rangeof activities to the logic of cost optimization,including R&D and managerial processesl ike account ing, informat ion sys temsdevelopment, and marketing. In locationaldecision-making, however, production costsare always evaluated relative to the efficiencyand productivity of a location. This pointis of ten over looked in d iscuss ions ofcomparative costs, but it is particularly crucialin that a major focus of TNCs’ geographic

Table V.1. Examples of different international production systems

TNCs’ governance Internalized Mixed Externalizedfunctional focus (Equity-based control) (Equity- and non-equity-based control) (Non-equity-based control)

Technology-driven Case 1: semiconductors – Case 4: telecom equipment –Intel Ericsson/Flextronics

Production-driven Case 3: automotive – Toyota

Marketing-driven Case 2: garments –Limited Brands / Li & Fung

Source: UNCTAD.

Page 10: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

allocation of value-chain activities is to achievesystemic eff ic iencies across their ent i reinternational production systems. A givenlocation, therefore, is judged by how cost-efficiently it performs a given function incoordination with functions located elsewhere,and not merely in isolation.

Asset-seeking motives are also leadingTNCs to tap skills and knowledge in a moresystematic way on the global scale. As notedabove, advances in information processingand telecommunications enhance TNCs’ abilitiesto coordinate complex functions over greatdistances. This not only allows them to deployfunctions in new locations, but also enhancestheir interest in mobilizing a wider rangeof skills and knowledge in a greater varietyof locations.

Finally, clustering has become a keyinfluence on TNCs’ locational decision-making.Earlier studies have described the fortuitousemergence of geographic concentrations ofrelated activities in production, specializedservices, R&D and the like (Schmitz andNadvi, 2000). Increasingly, however, cluster-formation is a global process ref lect ingrecognition by a number of TNCs of thevalue of co-location with suppliers, competitors,service providers , and knowledge-intermediaries. Their intentional efforts tocapture tacit-knowledge spillovers suggeststhat first-mover advantages for nodes ininternational production systems may be durabledespite the increasing mobility of TNC assetsand functions. Alternatively, it might suggestthat countries seeking to enter into internationalproduction systems, or seeking to occupymore complex niches within such systems,might need to reach a critical mass of relatedinvestments before doing so.

These broad structural trends are nowcrucial drivers of the geographic reorganizationof TNC activities. Paradoxically, TNC activitiesof a l l sor ts are becoming increas ingly“footloose”, even as geographic clusters growin importance. While distance might matterless for many transactions, proximity andaccess to tacit or partly tacit knowledgeis increasingly vital for competitive advantage.Nonetheless, it must also be recognized thatthese trends are not the only determinantsof geographic configurations of internationalproduction systems. There are also sectoraland nat ional (or cul tura l ) fac tors . Forexample , compet i t ive s t rength inmicroprocessors depends on design innovation,

and the need to protect technology assetsleads to an internalized, vertically integratedapproach. On the other hand, success inthe garments industry requires knowledgeof fas t -changing market t rends; hence ,externalizing production to a horizontal arrayof independent subcontractors confers costadvantages so long as these are able tomeet design, quality and delivery standards.Then there are the nat ional or ig ins ofinternational production systems (Borrus etal., 2000). For example, electronics TNCsheadquartered in the United States typicallypursue outsourcing strategies while maintainingtight equity control over their verticallyspecialized subsidiaries. Japanese TNCsin the same industry, by contrast, displaya strong preference for integrated production,governed through a mixture of equity controland c lose non-equi ty l inkages wi th keysuppliers (although this is beginning to change).These different structural, sectoral and nationalcharacteristics are combined (as exemplifiedby the case studies provided below) in thecontext of corporate strategies of individualf i rms a iming a t sys temic f i rm-speci f icadvantages.

The dr ivers and fea tures ofinternational production systems signal onepoint of particular interest to developingcountries: Governments that seek export-oriented FDI need to go beyond trade andFDI policies and assess their locat ionaladvantages in the international productionsystem context. More specifically:

“the issue is no longer whether tradeleads to FDI or FDI to trade; whetherFDI substitutes for trade or tradesubstitutes for FDI; or whether theycomplement each other. Rather, it is:how do firms access resources –wherever they are located – in the interestof organizing production as profitablyas possible for the national, regionalor global markets they wish to serve?In other words, the issue becomes: wheredo firms locate their value-added activities?In these circumstances, the decisionwhere to locate is a decision whereto invest and from where to trade. Andit becomes a FDI decision, if a foreignlocation is chosen. It follows that,increasingly, what matters are the factorsthat make particular locationsadvantageous for particular activities,for both, domestic and foreign investors”(WIR96 , p. xxiv).

Page 11: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

Developing countries need to identify theirlocational strengths and weaknesses in relationto their competitors and in relation to thefactors that influence the configuration ofinternational production systems. Only thencan they properly tailor policies to enhancetheir locational advantages. Moreover, inseeking to a t t rac t expor t -or iented TNCactivities or induce existing affiliates (orlocal companies) to upgrade their nicheswithin given international production systems,Governments need to d iscern wheregovernance authority resides within varioussystems so that their efforts are targeted atthe right decision-makers. Both these issuesare taken up further in Part Three of WIR02.

B. Case studies

1. Control through equityrelations in a technology-driven

international productionsystem: Intel

Semiconductors were the most dynamicproducts in world trade during the period 1985-2000, when their exports grew from $26 billionto $235 billion. The demand for semiconductorscomes mainly from the IT industries (computers,telecommunications and consumer electronics).By 2000, they represented 5 per cent of worldtrade, up from 1.5 per cent in 1985, and theyaccounted for 20 per cent of the trade in high-technology non-resource-based manufactures,the most dynamic category of world trade.

In terms of ownership advantages,c o m p e t i t i o n h a s b e e n f i e r c e i n t h esemiconductor industry, and Intel, the marketleader, has se t the pace ( t ab le V.2) . I tadap t ed be t t e r t o t he evo lu t i on o f t heindustry in which there was a long periodof fast growth (1982-1995), fol lowed byconsolidation (1995-1998), followed by ashor t boom and then a sharp fa l l . 1

Intel jumped from seventh to f irstrank by sales in the semiconductor industrybetween 1983 and 2001, and is today byfar the larges t chipmaker in the wor ld .Its sales increased 42 times between 1983and 2000 before declining in 2001 as theIT marke t f e l l . I t a ccoun t s fo r abou to n e - q u a r t e r o f g l o b a l s a l e s i nsemiconductors , most ly to the computerindustry. It also accounts for one quarterof the R&D under taken by the indust ry,and has been the biggest investor duringthe past decade. Intel’s capital expensesin 2001 were $7.3 bill ion. The companyhas developed s igni f icant technologica land product ion advantages : i t has beenable to squeeze more transistors onto siliconwafers for computers than its competitorsand it has developed larger silicon wafers,reducing fabrication costs by about one third.2

Inte l exploi ted i t s manufactur ingprowess by integrating its production systemand establishing identical plants in numerouslocat ions to obta in the opt imal g lobalconfiguration of its production facilities. Thissys tem al lowed the company to locateparticular activities in the sites most suited

Table V.2. The world’s leading semiconductor manufacturers, 2001(Billions of dollars)

Rank 2001 Rank 1983 Company (Home country/region) Sales 1983 Sales 2000 Sales 2001

1 7 Intel (United States) 0.7 29.7 22.72 5 Toshiba (Japan) 0.9 11.0 7.23 3 NEC (Japan) 1.3 10.9 7.04 2 Texas Instruments (United States) 1.6 10.3 6.75 .. STMicroelectronics (EU) .. 7.8 6.36 .. Samsung Electronics (Republic of Korea) .. 10.6 5.17 1 Motorola (United States) 1.6 7.9 5.08 4 Hitachi (Japan) 1.0 7.4 4.79 .. Infineon (EU) .. 6.8 4.6

10 10 Philips (EU) 0.5 6.3 4.6Total top 10 9.4 108.6 73.6Semiconductor industry 17.4 204.4 139.0

Source: UNCTAD, based on IC Insights, cited by Semiconductor Electronics Resource Centre, www.dir-electronics.comand UNCTC, 1986.

Page 12: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

for them. It kept the high-value elementsof the semiconductor cost structure (waferproduction and fabrication) predominantlyin the United States and shifted the morelabour-intensive assembly-and-testing activitiesto lower-cost sites (table V.3). Thus, Intelhas kept its production process internalized.More specifically, it has 13 fabrication plantsand 11 assembly-and- tes t ing s i tes in 7countries. About half of its 86,200 employeeswork in its Technical Manufacturing Group.It has expanded internationally with fabricationfacilities in Israel (1985 and 1999) and Ireland3

(1993 and 1998) and concentrated its labour-intensive operations in Malaysia (1988, 1994,1996 and 1997),4 the Philippines (1979, 1995,1997-1998),5 Barbados (1977, later closed),China (1997, 2001) and Costa Rica (1997and 1999).6 Today, about two-thirds ofIntel’s manufacturing workforce is in theUni ted Sta tes , 11 per cent in Malays ia ,8 per cent in the Philippines, 4 per cent inIreland, 3 per cent in Israel, 2 per cent inCosta Rica and 1 per cent in China. It isthe leading national exporter from Ireland,the Philippines and Costa Rica, and ranks

seventeenth among foreign exporters fromChina.

Many of Intel’s competi tors havereorganized their own international productionsystems, following Intel’s lead and on thebasis of the same overall intra-firm divisionof labour.7 In the process, a number offirms have consolidated their activities. Forexample, Motorola has reduced the numberof its plants from 29 to 14 since 1997, andHitachi reduced its plants from 13 to 8, shiftingproduction from Japan to China and Malaysia.

With regard to transnationalization,In te l ’s opera t ions and i t s in ternat ionalproduction system are designed to distanceitself from competitors by protecting itstechnological advantages inside subsidiariesstrategically located in its home country,or in I re land and Israel . In the case ofassembly and testing facilities, it has expandedinternationally to incorporate a few carefullyselected sites in low-cost locations but alwaysin fully-owned operations. It is an internationalproduct ion sys tem that i s h ierarchica l ,

Table V.3. Intel’s manufacturing sites, January 2002

Country Faci l i ty Funct ion Year built Current process technology Employees

United States Facility 1 Wafer fabrication 1978, 1992, 1996,1999, 2003a 0.13-, 0.25-, 0.35-micron 16 000

Facility 2 Wafer fabrication 1980, 1993, 2002a 0.13- 0.18-, 0.25-, 0.35-micron 5 500

Facility 3 Wafer fabrication 1988 0.13-, 0.18-micron 8 500

Facility 4 Wafer fabrication 1994 0.28-, 0.35-, 0.50-micron 2 700Facility 5 Wafer fabrication,

assembly and testing 1996, 1999, 2001 0.13-, 0.18-micron 10 000Facility 6 Systems

manufacturing 1996 .. 1 400

Facility 7 Wafer fabrication 2001 0.18-micron 1 845

Ireland Facility Wafer fabrication 1993, 1998, 2004a 0.18-, 0.25-micron 3 400

Israel Facility 1 Wafer fabrication 1985 0.35-, 0.50-, 0.75-, 1.0-micron 800Facility 2 Wafer fabrication 1999 0.18-micron 1 500

Malaysia Facility 1 Board manufacturing,assembly and testing 1996, 1997 .. 7 790

Facility 2 Assembly and testing 1988, 1994, 1997 ..

Philippines Facility 1 Assembly and testing 1997, 1998 .. 5 984Facility 2 Assembly and testing 1979, 1995 ..

China Facility Assembly and testing 1997, 2001 .. 1 227

Costa Rica Facility Assembly and testing 1997, 1999 .. 1 845

Source: www.intel.com, January 2002.

a Estimated construction completion.

Page 13: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

integrated (with associated intra-firm trade)and based on tightly controlled subsidiaries.The semiconductor industry is thus a goodexample of a global value chain driven bycarefully protected technological advantages.

Locational advantages also play arole. As technology-intensive stages in thesemiconductor production process can beseparated from labour-intensive ones andthe cost of transport is low relative to thevalue of the output, it is economical to pursuea global production strategy. In the caseof Intel, the principal factors that the firmtakes into account in establishing a newsubsidiary for assembly and testing functionsinclude the avai labi l i ty of a technicalworkforce, construction costs, the qualityof infrastructure, logistics, business costsand supplier capabilities (www.intel.com/pressroom/kits.htm). Host country incentivescan be important as well. The selectionprocess for the plant that was located in

Table V.4. Winners and losers in semiconductor exports,a 1985-2000(Per cent)

Market share increase,Economy 1985 2000 1985-2000c Top 10 TNCs presentb in winner economies

Principal winnersChina 0.14 8.82 8.7 Intel, Toshiba, NEC, Texas Instruments,

ST, Motorola, Hitachi, Infineon, Philips, SamsungTaiwan Province of China 2.72 10.64 7.9 Texas Instruments, Hitachi, Infineon, PhilipsMalaysia 0.36 7.81 7.5 Intel, Toshiba, NEC, Texas Instruments, ST,

Motorola, Hitachi, Infineon PhilipsRepublic of Korea 0.76 4.01 3.2 Texas Instruments, SamsungPhilippines 0.23 3.07 2.8 Intel, Texas InstrumentsThailand 0.46 2.54 2.1 Toshiba, PhilipsCosta Rica - 1.41 1.4 IntelIreland 2.37 3.43 1.1 Intel, Motorola, NECTotal 7.04 41.73 34.7

Principal losersUnited States 29.97 15.40 -14.2Germany 8.76 3.39 -5.4France 6.52 1.71 -4.8Japan 13.83 10.27 -3.6Italy 3.28 1.14 -2.2United Kingdom 6.73 4.88 -1.8Hong Kong, China 3.92 2.11 -1.8Total 73.01 38.90 -33.7

Source: UNCTAD, based on the United Nations’ Comtrade database.

a SITC 7599: parts and accessories of data-processing equipment.b As of January 2002. Developing economies with semiconductor affiliates not mentioned here: Singapore (NEC, Texas Instruments,

ST, Hitachi, Infineon, Philips), Hong Kong, China (Motorola, Philips), Indonesia (NEC); Morocco and Malta (ST).c The concept of “market share increase” is based in the import market shares as calculated by the CAN computer

programme on international competitiveness of UN-ECLAC, which is based on the United Nations’ Comtrade database.The data are classified at 3 or 4 digits of the Standard International Trade Classification (Rev.2). The period ofanalysis is 1985-2000, in which the value of individual years represents 3-year rolling averages (two-year averageof the year 2000) to emphasize the structural aspects of change.

Costa Rica exemplifies the interplay of thesefactors.8 Intel’s strategy of locating labour-intensive activities in low-wage areas, mainlyin Asia, and then moving to yet lower-wagelocations, has also been utilized by manyof its competitors. They too have movedthe more labour-intensive stages of theirproduction to developing countries, oftenthe same ones as Intel , thereby creatinga clustering effect. As semiconductors enjoyunencumbered access to most markets ,market-access factors do not play a rolein these locational decisions.

As a result, a handful of East andSouth-East Asian countries have registeredhigh increases in their export-market sharesin semiconductors, while some developedcountries have experienced large declines(table V.4). During the period 1985-2000,a total of eight winners (mainly from Asia)improved their shares by almost 35 percentagepoints , whi le seven, mainly developed,

Page 14: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

��

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

in distorted trade flows. For that reason,no attempt is made here to match the corporatestrategies of the leaders with trade data. Thissection simply presents one example of thekinds of networking relationships being establishedthrough non-equity forms, illustrating the earlierobservation about the importance of suchrelationships for exports. Still, the trade datado capture the international and regional aspectsof garments markets: China and several Asiancountries compete in most major markets,while the North American and West Europeanmarkets have regional suppliers, mainly asa result of trade restrictions (tables V.5 -V.7).

economies lost a similar percentage. Inother words, a very high proportion of thesetrade gains and losses were accounted forby the relocation of the labour-intensivesegments of semiconductor internationalproduction systems. Another factor was thearrival of newcomers in the industry fromeconomies such as Taiwan Province of Chinaand the Republic of Korea.

2. Control through non-equityrelations in a marketing-driven

international production system:Limited Brands

Clothing remains an important componentof world trade, rising from 2.4 to 3.1 percent of world imports (from $41 to $174 billion)during the period 1985-2000. It accountedfor 20 per cent of low-technology non-resource-based manufactures in 2000 (up from 17 percent). North America ($58 billion in 2000)and Western Europe ($72 billion) are the mostimportant markets for garments.

Barr iers to ent ry are low on theproduction side of garments, in comparisonto complex technology-and-scale-intensiveindustries like electronics and automobiles.However, there are high entry barriers inmarketing in the garment industry. Buyerstherefore occupy an important place in globalvalue chains and dominate the industry. Thereis an ample supply of capable garment makers,and it is relatively easy to create new onesby providing design inputs and some technicalassistance. Thus, the fragmentation of theproduction process is very advanced.

In this situation, the key variable forthe location of manufacturing plants is marketaccess. United States rules favour clothingassembled offshore from United States inputs,and they give special market access toassemblers from Central America, the Caribbeanand Africa (see chapter VII). Within NAFTA,Mexico profits from rules of origin that giveit advantages over competitors, especially fromCentral America and the Caribbean, sinceits domestic inputs count as “NAFTA content”and those of its competitors do not. TheEuropean Union uses a similar mechanism(see chapter VII), also granting special accessto a number of countries. Finally, the quotasof the Multifibre Arrangement have also stronglyinfluenced locational decisions. All of theseschemes and mechanisms – quotas, preferentialmarket access and regional integration – result

The characterist ics of demand forgarments in the United States, Europe andJapan differ greatly, and these differenceshave been important in the choice of suppliercountries (Sturgeon, 2002). United Statesbuyers prefer standardized products at lowprices. Europeans prefer high quality productsof greater diversity and are willing to payhigher prices. Japanese buyers prefer highquality finishing. In practice, this meansthat very few suppliers can meet the needsof all three major markets, and so tend tospecialize in one or two.

Take the Uni ted Sta tes market .Competition increasingly takes place not somuch at the level of f irms as networks.

Table V.5. Winners and losers in NorthAmerican garmenta imports, 1985, 2000

(Percentage)

Market shareincrease

Economy 1985 2000 1985-2000

Principal winnersMexico 1.6 14.0 12.4Honduras 0.2 4.0 3.9China 8.3 11.2 2.9Dominican Republic 1.4 4.0 2.6El Salvador 0.1 2.6 2.5Total 11.6 35.8 24.2

Principal losersHong Kong, China 22.7 8.2 -14.5Taiwan Province of China 15.5 3.3 -12.2Republic of Korea 13.7 3.8 -9.9Total 51.9 15.3 -36.5

Source : UNCTAD, based on the United Nations’ Comtradedatabase.

a SITC 842: men’s outwear non-knit; SITC 843: women’soutwear non-knit; STIC 844: undergarments non-knit;SITC 845: outer garments knit non-elastic; SITC 846:undergarments knitted.

Page 15: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

��

World Investment Report 2002: Transnational Corporations and Export Competitiveness

General retailers like Wal-Mart, Sears Roebuckand J.C. Penney, as well as garment specialtyretailers like Liz Claiborne, The Gap andLimited Brands, both design and marketclothing but do not make the products theysell. These companies are controlling a growingshare of the United States market: in 1977,

the top 50 apparel and accessory retailersheld 28 per cent of the market and the top5 held 9 per cent. By 1992, the share ofthe top 50 had risen to 53 per cent and thatof the top 5 to 18 per cent (Abernathy etal., 1999, p. 76). Continued success forbuyers in the garments industry depends,to a significant degree, on identifying andcontracting the best supplier networks.

An example is Limited Brands. I tis a leading specialty retailer of intimateand other apparel and non-apparel (i.e. beautyand personal care) products. Founded inthe United States in 1963, its net sales doubledbetween 1990 and 1999, from $5 to $10 billion.The dynamism of Limited Brands suffereda bit of a setback thereafter (sales in 2001were $9 billion), but it refocused on “fewerbut better brands” and now seems set fora new expansionary phase (Limited Brands,2001).

Limited Brands influences the globalvalue chain in which it operates throughmarketing, which is based on its two principaladvantages: retail sales outlets and brandmanagement. In 2001 it had 4,614 storesin the United States. The company usesboth in-house and external suppliers in whatit calls its “global network of relationships,resources and support personnel” (http://www.thelimited.com). Its independent division,Mast Industries, is one of the world’s largestcontrac t manufacturers , impor ters anddis t r ibutors of appare l (h t tp : / /www.mastindustries.com). External suppliersinclude a host of firms. Li & Fung has beenone of the most important ones (box V.1).

Mast Industries delivers over 200million garments to Limited Brands per year.It also supplies other retailers. Mast is wellpositioned globally to supply products thatenable its customers to establish and maintaintheir brand identity. It does so through aglobal network of 18 offices in 12 countriesand 400 factories in 37 countries. The latterincludes 53 joint ventures (42 in Asia, mainlySri Lanka) and 600 individual associates(including 400 in Asia and 165 in North America).The key competitive advantage of MastIndustries is “to identify manufacturing partnersin the right place at the right time” (http://www.mastindustries.com). Thus, Mast hasa production migration strategy that constantlysearches for production opportunities. Thisoverarching concern to contain costs istempered by additional factors, such as quality,

Table V.6. Winners and losers in WestEuropean garmenta imports, 1985, 2000

(Percentage)

Market shareincrease

Economy 1985 2000 1985-2000

Principal winnersChina 1.7 9.4 7.7Turkey 2.4 7.1 4.7Bangladesh 0.1 3.2 3.1Indonesia 0.3 2.4 2.1Morocco 1.3 3.2 2.0Romania 1.4 3.3 1.9Tunisia 1.8 3.7 1.9Poland 0.9 2.8 1.8Total 9.9 35.0 25.1

Principal losersItaly 18.5 7.8 -10.6Germany 10.1 5.9 -4.2Hong Kong, China 9.7 6.3 -3.4Republic of Korea 3.6 1.0 -2.5Total 41.8 21.1 -20.7

Source: UNCTAD, based on the United Nations’ Comtradedatabase.

a SITC 842: men’s outwear non-knit; SITC 843: women’soutwear non-knit; STIC 844: undergarments non-knit;SITC 845: outer garments knit non-elastic; SITC 846:undergarments knitted.

Table V.7. Winners and losers inJapanese garmenta imports, 1985, 2000

(Per cent)

Market shareincrease

Economy 1985 2000 1985-2000

Principal winnersChina 25.6 73.2 47.6Viet Nam 0.0 3.1 3.0Total 25.6 86.3 50.7

Principal losersRepublic of Korea 34.2 4.9 -29.3Taiwan Province of China 15.4 0.4 -15.1Hong Kong, China 5.8 0.7 -5.1Total 55.5 6.0 -49.5

Source: UNCTAD, based on the United Nations’ Comtradedatabase.

a SITC 842: men’s outwear non-knit; SITC 843: women’soutwear non-knit; STIC 844: undergarments non-knit;SITC 845: outer garments knit non-elastic; SITC 846:undergarments knitted.

Page 16: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

manufacturing flexibility, capacity and timelydelivery. It coordinates its global networkby way of the Mast Connection, which usesadvanced networking technology and theInternet to create a reliable global networkof Mast associates, manufacturers, customersand shippers.

There are also external “full-packageproviders”. The “full package” involves independentintermediaries supplying products according tothe buyer’s design, assembled by the intermediaries’production network (in which the intermediariesthemselves may have no equity interest). Theproduct carries the name of the buyer and the

Li & Fung Limited is an example of aworldwide trading company that manages a globallogistics chain for its retailer clients and partners.It is a full-package provider that brokers high-volume garments and fashion accessories (Li &Fung Limited, 2001, p. 9). The firm is headquarteredin Hong Kong, China, and is listed on its stockmarket. It has an annual turnover of about $4.2billion, employing about 5,000 people worldwide.In 2001, 72 per cent of the turnover was in thegarments segment; regionally, its orders camemainly from North America (75 per cent) andEurope (21 per cent) (Li & Fung Limited, 2001,p. 6).

The firm’s specialty is supply-chainmanagement within its global supply network.It does not own any production facilities, butmanages the “full package”: product development,product sourcing and product delivery, includingquality control and on-time delivery.

The evolution of Li & Fung into a full-package supplier went through several stages.In the 1960s and 1970s, Li & Fung operated asa regional sourcing agent, with offices in TaiwanProvince of China, the Republic of Korea andSingapore. Activities comprised “assortmentpackaging”: assembling components for a productset from different locations.

In the second stage, during the late 1970s,the business became more sophisticated,functioning, in its own description, as a “managerand deliverer of manufacturing programmes” (http//www.lifung.com). Buyers would indicate the typeof garment they were considering, and Li & Fung,as the agent, would prepare alternative designs.Once a design was agreed upon, the agent wouldsource all components (from yarns and trims topackaging), identify a manufacturing site andarrange the logistics. “Front end” functions(design, engineering and production planning)and “back end” functions (quality control,logistics, testing) remained mostly in Hong Kong,China, while the actual manufacturing wascommissioned from low-cost locations in oneor more countries. During the 1970s and 1980s,manufacturing was located predominantly in China.

The third phase began in the late 1980s:Li & Fung took control of complete garments

programmes for a season for a particular buyer,including the proposing of items and batch mixesand delivery rhythms. The process has sincedeepened. The manufacturing part of the processhas become highly complex and dispersed. Li& Fung “dissect the value chain” (Magretta, 1998).Manufacturing is farmed out to those locationsthat are the most cost-efficient, and the productcomprises components produced in numerouslocations. Out of 15 steps in the manufacturingvalue chain, Li & Fung claims to handle 10(Magretta, 1998).

The firm had about 700 customers in 2000– mainly large retailers in the United States andEurope – and operates through a network of 68offices in 40 countries. Until the mid-1990s, buyerswere mainly from the United States, such as LimitedBrands. In 1995, Li & Fung bought InchcapeBuying Services, a British trader based in HongKong, China, broadening buyer connections intothe EU market and providing itself with anestablished network of offices in Bangladesh,India, Pakistan and Sri Lanka. Li & Fung receivesa commission from the retailer and “the highervalue added lets us charge more for our services”(Magretta, 1998, p. 106). The firm hasmanufacturing contracts with 7,500 suppliers,of whom 2,000 are reportedly active at any giventime. There are estimates that Li & Fung thushas indirect employment links with 1.5 millionworkers (Magretta, 1998, p. 109).

Li & Fung presides over a large networkof contract suppliers in China and other Asiandeveloping countries, notably Bangladesh, India,Pakistan and Sri Lanka, as well as in Egypt,Madagascar, Morocco and South Africa. The firmgenerally takes between 30 to 70 per cent of afactory’s output – less would not give it the cloutto secure orders or reserve production capacityfor incoming orders, and more would make it over-dependent on the supplier (Magretta, 1998, p.106). In sum, the company’s transnationalizationprocess is based not on the possession ofdomestic assets that can be exploited abroad (aswas the case for many conventional TNCs), buton linkages, tapping into the resources of partnersand sharing the risk with them.

Box V.1. Li & Fung: a full-package provider

Source: UNCTAD based on Li & Fung Limited, h t tp : / /www.l i fung.com/about / index.html; Mathews,2001; Magret ta , 1998; Gereff i , 2001.

Page 17: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

intermediary has no influence over its distribution.East Asian firms were the first to become full-package suppliers to foreign buyers (Gereffi,2001), drawing on their ability to coordinatecomplex production, trade and financial networksefficiently (box V.1).

One might say that there are twodifferent worlds in the garments industry:an expanding world and a shrinking one.The first contains buyer-driven value chainsin which brand owners (à la Limited Brands)can move the production systems of thirdparties based on their design and marketingcompetencies. Instead of establishing theirown production facilities, they can contractthem. This can involve the retailer overseeingsome of the individual e lements of theinternational subcontracting process suchas sourcing inputs, assembly, quality controland delivery from distinct providers, as TheMast Industries division of Limited Brandsdoes. Or i t can involve the contractingof full-package providers that, on the basisof the design received, take it upon themselvesto undertake the whole production process,as Li & Fung does. Both Mast and Li &Fung are skilled intermediaries that excelin organizing supplier networks. FDI is notnecessarily involved in the generation ofexports in this first world of garments.

The second world of the garmentsindustry contains brand-name manufacturersthat maintain FDI-based international productionsystems. These manufacturers have beenforced offshore to low-wage s i tes byheightened competition, and rely on preferentialmarket-access schemes that give them anadvantage unti l the quota system of theMultifibre Arrangement is dismantled. TheSara Lee companies involved in garmentproduction are typical examples. They useforeign aff i l iates based in three or fourdifferent production sites in one region (theCaribbean basin, for example) that producesimilar products. That gives them the flexibilityto adapt to the changing competitive situationof each site by adding assembly lines inthe more convenient ones and dropping themin the rest, without the need to close downany site. These FDI-based production systemsare, however, becoming less representativeof the industry as a whole. This kind ofstrategy appears to rely on production aswell as marketing advantages to drive thevalue chain, but buyer-driven chains dominateinternational trade in the industry.

3. Control through equity andnon-equity relations in a

production-driven internationalproduction system: Toyota

The automobile industry maintainedits large share of world trade at 8.8 percent and grew from $149 billion to $486billion between 1985 and 2000. This industryaccounts for about 30 per cent of medium-technology, non-resource-based manufactures.The exis tence of excess capaci ty in theorder of 25 per cent in North America and30 per cent in Western Europe9 has notstopped auto TNCs from continuing to expandproduction capacity both at home and abroad(PricewaterhouseCoopers, 2000). Despitesurface ca lm, a harsh indust ry-wideres t ructur ing has taken place . St rongcompetitors have swallowed weak ones, andnew plants and equipment have replacedold ones. Both factors have influenced theinternat ional expansion of the pr incipalautomobile TNCs.

Unl ike e lec t ronics , automobi leproduction systems tend to be national orregional, rather than global. The high weight-to-price ratio of motor vehicles and strictgovernment policies to protect domesticmarkets and support local production slowthe pace of globalization in this industry.For example, European manufacturers stillproduce a lmost three-quar ters of the i rpassenger cars in the European Union, UnitedStates manufacturers half their passengercars in Nor th America , and Japanesemanufacturers two-thirds of their passengercars in Japan. In the Republic of Korea– a new entrant with a s t rong domest icindustry – national manufacturers suppliedalmost 90 per cent of the total until the Asianfinancial crisis, after which some of thosecompanies sold their assets to foreign TNCs.In other words, home markets are still centralto automobile manufacturers, even thoughmany producers have large stakes overseas.In most developing countries with substantialautomobile industries, however, TNCs aredominant.

A few firms dominate the world motorindustry. In 2000, the top 10 accountedfor three-quarters of global production (tableV.8). Of these, only one company, Ford,had “transnationalized” a good part of itsproduction (defined as over 40 per cent of

Page 18: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

production outside the home country) in 1980.By 2000, all but Toyota had passed thatthreshold. Yet it was Toyota that had stimulatedchange in the industry, including the shiftof production facilities to developing countriesand economies in transition.

Toyota used to supply North Americaand Europe mainly by way of exports fromJapan. It moved abroad only when accessto those markets became restricted becauseof voluntary export restraints and quotas.In order to establish itself successfully inthe highly competitive United States market,it relied on a number of strengths intrinsicto the Toyota Production System:

• Integrated single-piece production flowwith low inventories.

• Small batches made just in t ime.• Defects prevented rather than rectified.• Production “pulled” by customer demand

rather than “pushed” to sui t machineloading. Teamwork with flexible multi-skilled operators.

• Product ion l ine j igs wi th ident ica lspecifications worldwide to add/switchmodels easily.

• Active involvement in root-cause diagnosisto eliminate all non-value-adding steps,interruptions and variability.

• Closer integrat ion of the value chain,from raw materials to finished product,through partnerships with suppliers anddealers.

These strengths enabled Toyota (andother Japanese manufacturers that acquiredthem) to make major inroads in wor ldautomobile markets, first through exportsand later through FDI (Mortimore, 1997).In 1990, Toyota’s production was heavilyconcentrated in Japan, accounting for 86 percent of its sales. By 2000, foreign productionaccounted for 30 per cent. This new capacityis critical to Toyota’s plan to raise its globalmarket share by half (from 10 to 15 per cent).At the end of 2001, Toyota had 12 plantsin Japan and 43 plants in 26 other countries(see annex table A.V.1 for the main ones).Its international production system combinesa set of modern, efficient plants with a setof older, less efficient ones supplying domesticmarkets in such developing economies asArgentina, Brazil, India, Indonesia, Malaysia,the Philippines, South Africa, Taiwan Provinceof China, Venezuela and Viet Nam. Thesmall production scale and export volumesof the latter mean that they do not playa significant role in the global expansionof the Toyota Production System.

Toyota’s success as a major automobileTNC stems primarily from its lean productionsystem, its quality circles, its tiered suppliersand timely procurement. Although it wasa latecomer in establishing its corporatesystem, its superiority allowed it to establishits regional networks in the heart of thehome markets of its principal competitors,

Table V.8. The top 10 automobile manufacturers, ranked by vehicle production, 2000

1980 1994 2000 1980 2000units units units foreign foreign

produced produced produced production productionRank Company Home country (Millions) (Millions) (Millions) (Per cent) (Per cent)

1 General Motors United States 6.7 8.0 8.1 29.2 48.12 Ford United States 4.2 6.5 7.3 54.9 48.13 Toyota Japan 3.8 5.2 6.0 .. 30.34 Volkswagen Germany 2.5 3.2 5.1 35.5 60.75 DaimlerChrysler Germany 1.7a 3.7b 4.7 <20 75.56 PSA France 2.0 2.0 2.9 18.4 40.37 Fiat Italy 1.6 2.4 2.6 14.0 40.18 Nissan Japan 3.1 2.8 2.6 .. 48.69 Renault France 2.1 1.9 2.5 19.8 42.410 Honda Japan 1.0 1.7 2.5 .. 51.1Total top 10 28.7 37.4 44.3Total world 34.9 49.7 58.4

Source: UNCTAD, based on UNCTC, 1983; OECD, 1996; Organisation Internationale des Constructeurs d’Automobiles(www.oica.net/htdocs/main.htm).

a Sum of Chrysler (1.0 mill ion) and Daimler Benz (0.7 mill ion).b Sum of Chrysler (2.8 mill ion) and Daimler Benz (0.9 mill ion).

Page 19: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

driving them to close down less efficientplants and seek out lower-cost productionsites. There are, however, indications thatToyota too will be expanding operations inlower-cost sites – partly as a result of thesuccessful copying of Toyota’s productioninnovations by its competitors.

In the modern and most competitivepart of its international production system,Toyota maintains its ownership advantagein fully-owned assembly plants, with a fewexceptions. Parts are, to a large degree,externalized. Locational criteria relate mainlyto market size in the case of the principalmarkets, and to market access in the older,less compet i t ive par ts of the Toyotainternational production system. Recently,Toyota has indicated that it will extend itsinternational production system around coreregional markets to lower-cost sites, likeMexico for North America, and Turkey andthe Czech Republic (box III.10) for WesternEurope. Production is scheduled to startin 2005.

So far, the developing country thathas gained the most from developments inthe industry is Mexico, which increased itsmarket share by 12 percentage points between1985 and 2000 (table V.9). Volkswagen isone example of how one company reorientedits production in that country (box V.2),Ford is another (box V.3).

Table V.9. Winners and losers in the automobile industrya exports to the NorthAmerican market, 1985, 2000

(Per cent)

Market share increase,Economy 1985 2000 1985-2000 Top 10 TNC presentb

Principal winnersMexico 0.4 12.2 11.8 GM, Ford, DaimlerChrysler, Volkswagen, NissanCanada 23.7 29.0 5.3 GM, Ford, DaimlerChrysler, Toyota, HondaRepublic of Korea 0.6 3.7 3.0Total 24.7 44.9 20.1

Principal losersJapan 41.3 28.1 -13.2 Toyota, Nissan, HondaUnited States 12.0 7.6 -4.4 GM, Ford, DaimlerChrysler, Renault, Toyota,

Nissan, HondaGermany 15.3 12.8 -2.5 GM, Ford, DaimlerChrysler, Volkswagen, FiatTotal 68.7 48.5 -20.2

Source: UNCTAD, based on the United Nations’ Comtrade database.

a SITC 7810: passenger motor cars.b For developing countries with automotive affi l iates not mentioned here, see annex table A.V.2.

4. Control in transition in atechnology-driven international

production system: Ericsson

Telecom equipment was among themost dynamic exports during the period 1985-2000. By 2000, it represented 3.3 per centof world trade (up from 1.3 per cent in 1985)and accounted for 14.3 per cent of high-technology, non-resource-based trade. In2000, the total value of telecom equipmentexports exceeded $173 billion, with the top10 exporting countries accounting for 73per cent of the total .

After the boom of the late 1990s,the industry has faced a sharp downturn,partly because of excessive spending bytelecom operators on licences to operatethird-generation mobile telephony. This hastriggered a dramatic restructuring amongthe leading TNCs, including massive job cuts– at least 500,000 worldwide i f serviceproviders are included (www.FT.com.2 May2002). Over-capacity is driving manufacturersto cut costs and make better use of existingfacilities. At the same time, rapid technologicalchange – including the development of newstandards for mobile telephony and closerintegration of the telecom, consumer electronicsand media industries – is leading to shorterproduct cycles and forcing companies toinvest more in R&D and innovative solutions,precisely at a time when their cash flow

Page 20: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

In 1967, the first Beetle left the assemblylines of Volkswagen de México in Puebla. Duringthe following 20 years, Volkswagen de Méxicofocused its local production almost exclusivelyon the domestic market, achieving only small-scale exports of the classic Beetle model. In1977, production of the first-generation Golfbegan, followed soon after by the Jetta. Theoperation was typical of the import-substitutinginvestments made in protected markets at thetime, characteristically inefficient by internationals tandards .

In 1989, Volkswagen took a strategicdecision to consolidate i ts North Americanoperations. The two North American plants, inPuebla, Mexico, and in Pennsylvania, UnitedStates, were producing at half their capacitydue to the weakness of both markets. Both plantswere producing the same Golf and Jetta models.Mexico was selected as the sole productioncentre for Volkswagen in the regiona and suitablymodernized and upgraded as a result. Of thetotal production of 380,000 vehicles in 2001,300,000 were exported, four-fifths of these toNorth America. These exports from Pueblaaccounted for 60 per cent of the Volkswagensales in North America, with the remaining beingimported from Germany. For the VolkswagenGroup, the Puebla operation is not only the mainproduct source for its most important growthmarket, which is the United States, but, alsoa supplier of engines and other componentsto most Volkswagen factories in the world. TheMexican operations of Volkswagen have beentransformed into a world-class manufacturingplatform.

As NAFTA was implemented in 1994,Volkswagen decided to upgrade its Mexicanfacilities further. It decided to start productionof the prototype called Concept 1, reminiscentof the traditional Beetle silhouette, but on amodern technology platform. A production sitethat guaranteed both highest quality andcompetit ive costs had to be found. Afterreviewing several alternatives for the production

of this car (which had meanwhile been baptizedthe “New Beetle”), Volkswagen finally decidedin 1995 to allocate it to the plant in Puebla,Mexico.b The decision to locate the New Beetleproduction in Mexico meant investing severalhundred million dollars, increasing capacity inthe Volkswagen plant in Puebla and creating1,500 new jobs, bringing more than 20 auto partsuppliers to Mexico and building new productionfacilities.

The New Beetle project is the mostprominent example of how globalization isdetermining the strategy of the automotiveindustry for Mexico. A new trade agreementwith Europe – eventually signed in December2000 – was already on the agenda, and theGovernment decided to open its borders to theimport of assembled cars in advance of the actualnegotiations. Strategically, this allowed Mexicanautomobile assemblers to reduce the productioncomplexity of their Mexican plants to a fewproduct lines, thus increasing scale effects forimproved cost competitiveness. The low volumesof different models to supply the Mexicandomestic market could now be sourced fromother assembly plants all over the world, notonly from the NAFTA region.

In sum, faced with a decision on how toorganize its production for the North Americanmarket, Volkswagen opted to close its UnitedStates operation in favour of its Mexican one.That decision was ratified when Volkswagenassigned to its Mexican operations the worldwideproduction of the new Beetle model. Volkswagenhas been able to improve its competitive situationin the North American market based solely onits upgraded and modernized plants in Mexico.Its subsidiary has become the third biggestmanufacturing operation among foreign firmsin Latin America and the sixth biggest exportoperation of all firms in the region. At the sametime, Volkswagen’s success provided anotherstrong boost to the international competitivenessof the Mexican automobile industry.

Box V.2. Volkswagen’s strategy in Mexico

Source: UNCTAD, based on mater ia l provided by the Department of Communicat ions , GovernmentAffai rs and Corporate Stra tegy of Volkswagen de México, February, 2002.

a Given comparable quality levels, arguments in favour of the Mexican plant were lower production costs, a broadsupplier base in Mexico and the fact that closing the Mexican plant would have meant abandoning the Mexicanmarket, while the United States market could still be supplied from Mexico after the closing of the Pennsylvaniaplant.

b The main reasons for this decision were the excellent quality levels achieved in the Puebla plant after six yearsof export experience to the United States; the very competitive cost levels both of the plant itself and of theMexican supplier base; and NAFTA providing a clear long-term framework with regard to both market access andthe gradual elimination of performance requirements.

Page 21: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

long-term growth of the companies is involved,the immediate threat is to survival. Thatmeans that a good part of their energy isdirected at the need to improve efficiency.

The market for telecom equipmentis dominated by a small number of largeTNCs ( table V.10) wi th ra ther s imi larownership-specific advantages. (Cisco Systems,a router specialist, is the principal exception.)Eight companies in 2000 accounted for about54 per cent of worldwide sales of telecomequipment ($381 billion).10 For the mostpart, they drive their value chains throughtechnological advantages in the sense thatnew technologies have led to a surge inthe use of mobile phones and the Internet,and this has resulted in increased demandfor the installation and upgrading of moderntelecom infrastructure. Of significance forthe configuration of international productionsys tems in th is indust ry were d i f fer ingstandards, the timing of the shift to mobilephones, and the timing of the privatizationof incumbent service providers in manycountries, where national or regional prioritiessometimes came to the fore.

In response to the rapid market growthof the late 1990s, most equipment vendorsexpanded their international production systems

Table V.10. The top telecom equipmentmanufacturers, 2000

(Bill ions of dollars)

Rank Company Home country Sales

1 Ericsson Sweden 31.32 Nortel Networks Canada 30.33 Nokia Finland 27.24 Lucent Technologies United States 25.85 Cisco Systems United States 23.96 Siemens Germany 22.87 Motorola United States 22.88 Alcatel France 21.6

Source: Gartner Dataquest, www.cellular.co.za/stats/top_telecoms_infrastructrue_vendors_2000.htm.

and now have to adapt to the crisis in theindustry. The situation of Ericsson exemplifiesthis in many ways.

During the past decade, Ericsson,11

the wor ld’s la rges t suppl ier of te lecomequipment , reduced the number of i t sproduction plants from about 70 to less than10 worldwide and outsourcing productionto contract manufacturers . Er icsson hasmaintained two kinds of foreign affiliates:p lants needed for the development andmanufactur ing of new products whoseproduction is not standardized enough to

Ford was one of the most globalizedautomobile companies in the early 1990s. Itinvested heavily in foreign markets and lower-cost production sites to rectify its loss of marketshare in the United States. In 1980, over halfof Ford’s production capacity was located outsidethe United States, mainly in Europe. It thensought to compete better in the North Americanmarket by producing in Mexico, where it madesizeable investments in assembly operations,making use of the United States production-sharing market-access mechanism and theMexican maquiladora export scheme (see chapterVII). Beginning in 1994, Ford’s operations therebenefited from the NAFTA rules of origin.

The building of Ford’s new productioncapacity in Mexico resulted in dramatic changesin the quality and type of models produced inMexico. Prior to 1987, its products were sold

only in the Mexican market. There were a greatmany models, and production rarely exceeded20,000 units per year. In 1987, Ford began toexport two models from its new plant. By 2000,the company was producing 193,204 vehiclesand exporting 181,099 of them. Ford’ssubsidiaries in Mexico became the third biggestmanufacturing operation among foreign firmsin Latin America and the eighth biggest exportoperation of all firms in the region.

In short, Ford’s response to the loss ofmarket share in the United States to Japanesefirms was to integrate Mexico into i tsinternational production system and to focusits operations there on two compact-sizedvehicles and one engine with state-of-the-arttechnology, both manufactured for export. Itwas helped in this by the United Statesproduction-sharing mechanism and the NAFTArules of origin.

Box V.3. Ford’s strategy in Mexico

Source: UNCTAD, based on ECLAC, 1998; Mortimore, 1998b; Shaiken and Herzenberg, 1987;Carrillo, 1995; and Shaiken, 1995.

Page 22: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

motivate a shift to low-wage countries; andthe most cost-efficient plants for the morestandardized products. The former producenon-standardized products and need to beclose to the design and development unitsthat can remove “bugs”. The latter engagein high-volume manufacturing concentratedin a few low-cost sites like China, Polandand Estonia. In the case of China, Ericssonexpects exports to grow significantly in thecoming years. Operations in other partsof Asia are l ikely to move there. India,Indonesia and the Russian Federation maybe candidates for future production plants.For site selection, Ericsson takes into accountthe following host-country factors: markets ize , level of bureaucracy, qual i ty ofinfrastructure (including customs clearanceprocedures, the tax system, EPZs), tradepolicies affecting access to internationalmarkets, level of political risk, productioncosts ( inc luding labour cos ts ) , and theavailability of contractors and suppliers.Ericsson protects i ts core technologicaladvantages through i t s fu l ly-control ledsubsidiaries. In choosing locations for itssubsidiaries for equipment manufacture, itfocuses on efficiency factors plus domesticmarket demand, the latter being particularlyrelevant to the telecom industry.

The role of outsourcing has growndramat ica l ly. Much of the none-coreproduction has been externalized to contractelectronic manufacturers for cost-sharingreasons. The firm is disposing of manyplants to contract manufacturers such asFlextronics (see below) and Solectron, whichare willing to purchase them in order tostrengthen their strategic relationships withEricsson. Sourced components are aggregatedat the highest possible assembly and testinglevels, so that a smaller number of strategicpartners or first-tier suppliers are required.In January 2001, Ericsson transferred itscomplete supply chain for mobile phonesto the contract electronics manufacturerFlextronics to improve economies of scaleand volume production flexibility, and to reducecapital exposure as well as risk. Flextronicstook over all related Ericsson facilities inBrazil, Malaysia, Sweden, the United Kingdomand parts of Ericsson’s plant in Lynchburg,Virginia, in the United States. Ericsson nowfocuses on other par ts of the te lecomequipment value chain, such as design, R&D,product development , and sa les andmarketing.12

Ericsson’s competitors have been undersimilar pressures to restructure and adjusttheir international production systems. Anexample is Nokia. It is the world’s leadingmobile phone maker, with a share of 35per cent of the worldwide market of about400 million units in 2001.13 The companyhas production facilities in Finland (the homecountry) and 10 other countries. Telecominfrastructure is produced in China, Finland,Malaysia and the United Kingdom; mobilephone handsets are made in Brazil, China,Finland, Germany, Hungary, Mexico, theRepublic of Korea and the United States.In the case of mobile phones, high-volumeproduct ion of key inputs (engines) i sconcentrated in selected production units,while al l factories are involved in f inalassembly. This type of operation requiresthe ability to adapt quickly to shifting tastesand a very reliable and flexible material-supply system.14 Due to the need for high-volume production, Nokia only adds a newplant if it is clear that there is a marketfor many millions of units per year fromthat plant.15

An analysis of the production systemsof the other main actors in the telecomequipment industry reveals many similaritieswith those of Ericsson and Nokia. Outsidethe EU and the United States, most equipmentmakers have set up production plants inBrazil and China. Other countries that haveattracted telecom manufacturing include somein Central and Eastern Europe (e.g. Estonia,Hungary and Poland), in South-East Asia(Malaysia, Singapore and Thailand), EastAsia (Republ ic of Korea) and Mexico.Siemens has the widest geographical spreadof plants, with a presence also in India,Romania, the Russian Federation, Switzerland,Turkey and Ukraine.

Most of the key players havestreamlined and outsourced considerable partsof their production in reaction to the industrycrisis and the corresponding systemic efficiencyrequirements. Cisco Systems, for example,outsources almost all manufacturing to contractmanufacturers around the globe. Norte lNetworks reduced its overall workforce in2001 from 93,000 to 48,000, mainly bydivesting its non-core activities. Manufacturingand repair operations in Europe, North Americaand Asia were outsourced to the contractelectronics manufacturer, Solectron. Motorolaannounced, in June 2000, a major deal under

Page 23: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

which Flext ronics was to take over asignificant part of Motorola’s mobile phoneproduction. Lucent Technologies unveileda plan in April 2000 to increase the shareof manufacturing by contract partners from20 per cent to 60 per cent over 18 to 24months .16 Lucent has t ransferred twomanufacturing plants in the United Statesto Celestica.17 Through these and othermeasures, Lucent cut its workforce by almost45,000 in 2001. In China, on the other hand,Lucent announced, at the end of 2001, thatit would expand its manufacturing base inShandong Province.18

This analysis suggests the following:

• Most telecom companies are relocatinghigh-volume manufacturing activities fromhigh-cost to low-cost locations (particularlyin a small number of developing countriesand economies in transition).

• Throughout the industry, there is a growingfocus on core competencies and a greaterreliance on contract manufacturers for morestandardized and less sophisticated products.

• The downsizing of high-cost productionsi tes has not af fected the des ign andR&D activities in these sites.

Thus, telecom equipment TNCs drivetheir value chains by way of their technologicaladvantages in core products and throughinternational production systems based onforeign affiliates for the more standardizedand less technologically sophisticated items,and they rely on others, increasing in thismanner their production efficiency. Theownership advantages of their advancedtechnologies are protected in a similar fashionas Intel’s in the semiconductor industry, byinternalizing them. Ericsson’s principaladvantages are in equipment for ground stations(i.e. for transmission and reception). Ini ts deal ings with telecom operators , thecompany prefers to opera te through i t sinternational production systems.

Facing that competitive situation in mobilephones, Ericsson outsources much of its previousin-house production in order to reduce costs.It should be mentioned that, with regard tolocational advantages (as well as the typicalefficiency-seeking considerations), TNCs intelecom equipment prefer that productionlocations include a strong potential for domesticmarket growth as well as a relatively well-developed supply infrastructure.

The geographical shift in the exportsof telecom equipment has been dramatic(table V.11). Over the period 1985-2000,f ive countr ies had market share gainscorresponding to almost 30 percentage pointsin total, with the Republic of Korea, Mexico,China, Sweden and Finland reporting thelargest increases. Ericsson and Nokia arebehind many of these changes, either directlyor via the manufacturers they contract.

Table V.11. Winners and losers intelecommunications equipmenta

exports, 1985, 2000(Per cent)

Market shareincrease, Top 8 TNCs

Economy 1985 2000 1985, 2000 presentb

Principal winnersRepublic of Korea 3.5 11.2 7.8 NokiaMexico 1.0 7.4 6.4 Nortel, Nokia,

MotorolaChina 0.04 5.7 5.7 Ericsson, Nokia,

Siemens,Motorola

Sweden 2.5 8.1 5.6 EricssonFinland 2.0 7.2 5.2 NokiaTotal 9.1 39.7 30.6

Principal losersJapan 29.1 4.6 -24.5United States 23.5 10.9 -12.7Total 52.6 15.4 -37.2

Source: UNCTAD, based on the United Nations’ Comtradedatabase.

a SITC 7643: television, radio and related transmittersand receivers.

b As of January 2002. Developing economies with telecomequipment affiliates not mentioned here include: Brazil(Ericsson, Nortel, Nokia, Siemens, Motorola), Singapore(Motorola), Malaysia (Motorola), Thailand (Siemens),India (Siemens) and Egypt (Siemens).

The case of the technologica lleadership of Er icsson in the te lecomequipment value chain demonstrates someof the principal features of internationalproduct ion sys tems in th is indust ry : atechnological leader facing restructuringpressures dur ing a harsh indust ry-widerecession combines with a fast-growing andefficient production specialist (Flextronics– box V.4) to exploi t the compet i t iveadvantages of both. Ericsson and manyof its principal competitors are focused moreon restructuring than on the expansion oftheir in-house production systems. Theybecome, therefore, more selective about wherethey locate manufactur ing opera t ions ,expanding only in countries like China thatoffer both efficiency-enhancing possibilitiesand strong domestic demand. Nonetheless,

Page 24: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

��

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

the s t ronger expansion of in ternat ionalproduction systems within this value chaincurrently comes from contract manufacturersrather than the technological drivers.

5. Outsourcing becomes moregeneralized: the rise of contract

manufacturers

These case studies have served todemonstrate the diversity of internationalproduction systems and suggest a trend towardsless hierarchical international production systemswith more non-equity elements. While non-equity forms linked to supplier networks havebeen around in the garment and electronicsindustries for a while, other new forms arerapidly appearing. Especially impressive arecontract manufacturers that actually establishtheir own international production systemsthrough which they serve their customers(typically other TNCs). In consequence, theirpresence has an immediate impact on trade.

The growth of contract manufacturingcan be exemplified by the electronics segmentof this market. It is expanding not only inNorth America and Europe, but also in Asia.Between 1998 and 2002, the global marketfor contract manufacturers in electronics hadbeen expected to grow by 140 per cent, from$58 billion to $139 billion. Some estimatessuggest that the share of the total marketfor electronics equipment controlled by contractmanufacturers will increase from 8 per centin 1999 to 18 per cent in 2004(www.solectron.com). The largest four contractmanufacturers each had revenues of over$10 billion in 2002 (table V.12). Outside theEU and the United States, the bulk of theirfacilities are located in Brazil, China, Hungary,

Mexico and Malaysia, followed by Singaporeand Thailand and, recently, Japan (chapterIII).

Contract manufacturers offer manyadvantages. They achieve greater efficiencythrough higher capacity use because theycan simultaneously assemble or manufactureproducts for several or iginal equipmentmanufacturers using the same plant. Bybuilding in-house capabilities, they are ableto develop new process technologies andsometimes even help their principals withproduct innovation. As they evolve, theyundertake a host of manufacturing-relatedfunctions such as logistics and procurement.They go global, both to take advantage oflow-cost locations and to support clientsin all their major production sites aroundthe world, introducing new products andproviding aftersales repair services centres.

Important bonuses for the countriesthat attract leading contract manufacturersinclude the increased scope advantages thatcan accompany such an investment; clusteringeffec ts and the associa ted longer- termupgrading of economic activities; and, ofcourse, exports. For example, Flextronicshas become the sixth largest exporter inHungary and one of the top 15 TNC exportersin China. The effects for host countriesmay become even more important if theprocess of outsourcing leads to a concentrationof production and export activities in a smallnumber of clusters, especially industrial parks;the greater the numbers of plants and themore numerous the local l inkages wi thsuppliers, the less likely will TNCs be tomove to other locations (box V.4).

C. Conclusions

The increasing intensity of competitionin all industries, and especially export-orientedindustries means that leading companies arecontinually challenged by competitors andnewcomers. That competition in itself provokesstrategic reactions. TNC responses to thiscompetitive pressure vary according to theircompetitive advantages at different stagesof the global value chain. In the high-technologysector, competitive advantage lies mainly intechnological capacity and speed of innovation.In the medium-technology sector (characterizedby mature technologies), firms tend to focusmore on efficiency through economies ofscale. In the low-technology sector (where

Table V.12. The five largest contractelectronics manufacturers,

1995 and 2002(Bill ions of dollars)

Revenue

Company Headquarters 1995 2002a

Solectron United States 1.7 16.5Flextronics International Singapore 0.4 13.2SCI Systems/Sanmina United States 3.5 12.1Celestica Canada 0.6 11.3Jabil Circuit United States 3.6 4.9

Source: UNCTAD, based on Sturgeon, 2002, p. 14.

a Est imated.

Page 25: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

��

World Investment Report 2002: Transnational Corporations and Export Competitiveness

A closer look at some aspects of thestrategy of one of the fastest growing companiesin electronic manufacturing services, Flextronics,provides a better understanding of what drivesthe transnationalization strategies of contractmanufacturers. With corporate headquartersin Singapore, Flextronics grew from $448 millionin revenues in 1995 to about $13 billion in 2002.A significant share of Flextronics’ productionis in low-cost economies; in eight locationsin five countries (i.e. Brazil, China, Hungary,Mexico and Poland), Flextronics has developedwhat it calls integrated “Industrial Parks” (boxtable V.4.1). These help the firm to overcomeinfrastructure bottlenecks as they are largeenough to attract suppliers to set up shop closeto them. Each Flextronics’ Industrial Parkprovides the necessary capabilities for thecompany to undertake high-volume productionand provide cost-effective delivery of finishedproducts within a day or two to the productowner’s end-users, greatly reducing the freightcosts of incoming components and outgoingproducts .

The most sophisticated operations,including the manufacture of routers andwireless base stations, are performed in placeslike Silicon Valley (United States) and Sweden,where the right mix of skilled labour is available.The most labour-intensive operations are inDoumen, China, where Flextronics makes, amongother things, PC parts, mouse assemblies andmobile phones.a In Europe, the company hasexpanded rapidly into Central and EasternEurope.

In addition, Flextronics also has regionalmanufacturing operations in multiple locationswithin Brazil, Europe, India, Israel, Malaysiaand North America that complement its Industrial

Box V.4. Flextronics: specializing in the manufacture of others’ products

Parks. In contrast to the latter, which weretypically set up as greenfield investments, mostof the regional manufacturing operations areacquisit ions of existing plants previouslycontrolled by Flextronics’ key customers orcompetitors, and often lack the efficiency-seeking characteristics of its Industrial Parks.Flextronics cements its strategic relationshipswith the former owners in this manner, thenendeavouring to improve efficiency in thoseplants.

Flextronics, l ike most of the moresophisticated contract electronics manufacturers,also provides such services as productintroduction centres and design and engineeringcentres to its strategic partners and buyers.This demonstrates how the boundaries betweencontractee and contractor blur over time asstrategic partners reassign tasks along theglobal value chain.

When investing in the expansion of itsinternational production system, Flextronicsexpects host countries to offer an environmentconducive to manufacturing for the worldmarket. It takes into consideration factorssuch as the productivity of the workforce, thecapabilities of domestic suppliers, the qualityof public utilities and infrastructure, accessto inexpensive and easily accessible land,investment incentives, labour market rules,customs clearance procedures and numerousquality-of-life aspects (Pfaffstaller, 2001).

Since Flextronics specializes in themanufacture of other firms’ products, the valuechain is driven by the owners of the products,not by the contract manufacturer. However,contract manufacturers are obliged toaccompany their clients in different markets

and continually improvetheir process technologyand production efficiency.For both reasons,contract manufacturersmust establish suitableinter-national productionsystems of their own, andhave consequentlybecome leading investorswith a critical influenceon the export-competitiveness of hostcountries.

Box table V.4.1. Flextronics’ selected global facil it ies, 2002

Regional Design andIndustrial manufacturing engineering

Location Park operation centre

North America .. 6 9Western Europe .. 18 14Other developed countries .. 1 2Latin America 2 4 ..East and South-East Asia 2 9 2Central and Eastern Europe 4 2 2Total 8 40 29

Source: http:/ /www.flextronics.com/Globalman.

Source : UNCTAD, based on Pfaffs ta l ler, 2001 and other mater ia ls .a Time, 5 August 2001.

Page 26: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

CHAPTER V INTERNATIONAL PRODUCTION SYSTEMS

barriers to entry are low), cost cutting andmarketing are the most critical. However,despite these general tendencies, it shouldbe noted that firms in the same sector canreact in quite different ways to similar stimuli.In the telecom industry, Nokia focuses moreon in-house production of mobile phones, whileEricsson has completely outsourced it.

Most of the industry leaders reviewedabove are keeping their core competitiveadvantages in-house in their home countries,e i ther in R&D and des ign ( technologydevelopment), or production processes, orsales outlets and brand management. Non-core functions, such as the labour-intensiveparts of the production process, the assemblyof less sophisticated products, or the logisticalorganizat ion of product dis t r ibut ion areoutsourced to low-cost sites. The contractmanufacturing phenomenon epitomizes theseoutsourcing trends as the complete productionprocess is outsourced. It increases the scaleand importance of suppliers’ operations, asglobal value chains are more and more finelys l iced in to specia l ized funct ional andgeographical elements.

The splitting of the global value chainand the multiplication of supplier networksopen up new opportunities for developingcountries and economies in transit ion toparticipate in international production systems.Indeed, TNCs play a critical role in manymanufactured exports. While retaining theircore competencies, TNCs are set t ing upinternational production systems on the basisof corporate strategies that seek to obtainthe optimal configuration of their productionprocess by spreading production to locationsthat offer significant advantages in productioncosts and access to third markets. Thus,labour-intensive activities are moved to siteswith cheap but efficient labour. The slicingof the value chain also means that newopportunities in the export of services openup to countries that can provide these servicesat low cost.

However, g lobal suppl iers mustincreasingly provide independent processdevelopment capabilities and the ability toperform a wide range of value-added functionsassociated with the manufacturing process,including help with product and componentdes ign, component sourc ing, inventorymanagement, testing, packaging and outbound

logistics. The increasing demands put onkey suppliers raise the barriers to marketentry for the smaller and younger suppliersfrom developing countries and economiesin transition.

The spreading of in ternat ionalproduction systems, through either FDI ornon-equity supplier forms, and – equallyimportant – the upgrading of activities offoreign affiliates in specific locations alongthe value chain, depends not only on thestrategies of firms but also on the policiesof host countr ies . The lat ter can play as igni f icant ro le in the conf igura t ion ofinternational production systems if theirgovernments have a clear understanding ofhow they “fit in” with the corporate strategiesthat determine the nature and location ofinternational production systems.

The next chapter looks a t g lobalcompetitiveness patterns.

Notes

1 During the first period, Japanese TNCs mounteda serious challenge to United States producers,winning significant market share (from 32.5 percent to 51.2 per cent during 1982-1988), thenfalling back and reaching a low of 26.4 per centin 1998. The main United States transnationalproducers (Motorola, Texas Instruments, NationalSemiconductor, Intel and AMD) collectivelyreached bottom in 1989 (at 37.3 per cent) beforemaking a strong comeback (peaking with about50 per cent of the global semiconductor marketshare in the late 1990s). While United Statesand Japanese leaders were fighting for marketshares, new producers were making inroads,more than doubling their market shares, from9 per cent to about 20 per cent over the period(based on data from the Semiconductor IndustryAssociat ion) .

2 On Intel’s lead in flash memory designtechnology, lithography and capacity, see http://www.intel .com/intel/f inance/presentations/pdf_files/nichols.pdf.

3 By end 2000, Intel’s investment in Irelandsurpassed $3 billion. A new $2 billion waferfabrication unit was under construction(www. in t e l . com/ in t e l / communi ty / i r e l and /aboutsite.htm).

4 Intel’s investment in Malaysia totalled $1.9 billionin 2000 (www.intel.com/intel/community/malaysia/aboutsite.htm).

5 Intel’s investment in the Philippines surpassed$1 billion by 2000 (www.intel.com/intel/community/philippines/aboutsite.htm).

6 Intel’s investment in Costa Rica was about $450million (www.intel.com/costarica.htm).

7 AMD is Intel’s most direct competitor; however,it is not in the list of the top ten semiconductormakers.

Page 27: WORLD INVESTMENT REPORT, 2002 · 2020. 9. 4. · UNCTAD/WIR/2002 UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT Geneva WORLD INVESTMENT REPORT, 2002 PART TWO TNCs & EXPORT COMPETITIVENESS

���

World Investment Report 2002: Transnational Corporations and Export Competitiveness

8 See Spar, 1998; Rodriguez-Clare, 2001; Shiels,2000; Egloff, 2001b.

9 See “Car manufacturing: incredibly shrinkingplants”, The Economist , 11 April 2002.

10 See http://www3.gartner.com/5_about/press_releases/2002_02/pr20020204b.jsp.

11 This section is based on direct interviews withEricsson executives.

12 In 2001, partly in response to increasedcompetition, a deal was concluded between SonyCorporation and Ericsson to merge their mobilephone businesses worldwide, relying mainlyon contract manufacturers for assembly.

13 See www3.gartner.com/5_about/press_releases/2002_03/pr20020311a.jsp.

14 For example, Nokia’s production of about 140million mobile phones in 2001 implies the handlingof several million components every hour, makingthe efficiency of the overall production networka key competitive factor.

15 For Nokia, every factory needs to be designedto serve a certain well-defined market and tobe adapted accordingly, depending on whetherit will produce engines or assemble mobilehandse t s .

16 http://www.vnunet.com/News/602409.17 http://www.lucent.com/press/0901/010904.coa.html.18 http://www.lucent.com/press/1201/011225.coa.html.