The Competitiveness of Nations in a Global Knowledge-Based Economy

Peter Dicken *

Global-Local Tensions: Firms and States in the Global Space-Economy

Economic Geography, 70 (2)

Apr. 1994, 101-128.


Abstract [Web 1]


An Analytical Point of Entry: Production Chains and Relational Networks

Firm-Firm Competition: The Changing World of the Transnational Corporation

State-State Interaction: The Emergence of the “Competition State” [Web 2]

Firm-State Interactions: Dynamic Bargaining Relationships

Conclusion: The Global-Local, Debate Revisited [Web 3]




A key issue facing researchers of economic, political, social, and cultural change is the dialectical tension between globalizing and localizing processes.  From an economic geography perspective, a major question concerns the relationship between the globalizing tendencies of many business firms and the prospects for a genuinely local economic development, particularly in light of the organizational and technological changes associated with the alleged transition to a post-Fordist world.  This paper addresses a specific aspect of the “global-local nexus” that has not been well developed in the geographic literature: the relationships between transnational corporations and nation-states.  Each can be conceptualized as highly embedded interacting networks.  Firms and states are locked in competitive struggles.  The competitive strategies they employ are both diverse and the outcome of contested power relations, internal and external.  Increasingly, too, they involve various forms of collaborative relationship.  Concrete spatial outcomes, therefore, reflect complex competitive and bargaining relationships between and within firms, between and within states, and between firms and states. 


These are exciting and challenging times in economic geography, as we grapple with the problem of understanding the major transformations in the way economic activities are organized and reorganized.  Not only does the pace of change pose difficulties, but the complex interplay among processes operating at different, but related, geographic scales has become increasingly intricate as well.  In an earlier Roepke lecture, Brian Berry argued that “economic geography is in crisis.  New intellectual vigor will come only if there is creative reconstruction of the conceptual core” (1989, 1).  In these days of theoretical plurality, it is doubtful if we should be concerned with a single conceptual core.  Indeed, one of the most stimulating features of today’s economic geography is this plurality of approaches and the healthy interchange of ideas that such diversity stimulates.

Berry’s search for such a core led him to explore issues of cultural differentiation at a global scale.  Berry located nation-states in a multidimensional cultural-geographic space, which, he suggested, constituted a “differentiating checkerboard of culture” on which “in an increasingly tightly knit global economy, multi-national corporations play their locational games interactively with nation-states” (1989, 1).  Berry did not, however, explore in any depth the actions either of

* School of Geography, University of Manchester, Manchester, M13 9PL, UK

The Roepke Lecture in Economic Geography was established to honor the late Professor Howard G. Roepke, who served on the faculty of the University of Illinois at Urbana-Champaign from 1952 until 1985.  The lecturer is chosen by the geography faculty of that institution in consultation with the editors of this journal.  The paper was read at the Annual Meeting of the Association of American Geographers, Atlanta, 1993.

The author is grateful to Amy Glasmeier, James Huff, Anne Huff, Jamie Peck, Adam Tickell, Liisa Cormode, Henry Yeung, and two anonymous referees for their constructive comments.


multinational corporations or of nation-states.  Rather, he devoted his attention to the dimensions of global cultural differentiation.

In contrast, it is the actions - and, especially, the interactions  - of international firms and nation-states that constitute the focus of this paper.  I share Berry’s view of the centrality of international firms and nation-states in shaping the changing geography of the global economy (Dicken 1986, 1992a).  I want to frame my argument, however, in the specific context of what has come to be seen as a major contemporary issue: the “global-local nexus.”  This is far more than merely a question of the geographic scale at which economic processes occur.  More fundamentally, it is a question of where power lies, and it is a central problematic facing both firms and states.

Analysis of these processes is made more difficult by the uncertainty surrounding the interpretation of the current structure of the world economy.  Making sense of the present is, of course, always more difficult than making sense of the past, because we do not really know where we are.  Throughout history, most observers of the contemporary scene have been tempted to argue for the “specialness” of their own times; the present day is no exception.  But, by definition, we cannot know which of the current developments and changes are permanent and which are purely ephemeral.  The current debates over the alleged new era of flexible accumulation, the transition from Fordism to post-Fordism, the replacement of the “Old Competition” by the “New Competition” (Best 1990) are cases in point.  Gertler (1992) has recently drawn up a very useful balance sheet of views, in which he identifies “areas of consensus, continuing disagreement, and important but still neglected questions” (p. 259).  Considerable disagreement also exists over the interpretation of current developments in the organization of economic activities at the global scale and their implications for firms and states.  The crux of the debate is the extent to which we are moving toward (or already live in) a globalized, rather than an inter-national, economy.  Robert Reich’s view is that

We are living through a transformation that will rearrange the politics and economics of the coming century.  There will be no national products or technologies, no national corporations, no national industries.  There will no longer be national economies, at least as we have come to understand that concept. ... As almost every factor of production - money, technology, factories, and equipment - moves effortlessly across borders, the very idea of an American economy is becoming meaningless, as are the notions of an American corporation, American capital, American products, and American technology.  A similar transformation is affecting every other nation, some faster and more profoundly than others; witness Europe, hurtling toward economic union. (R. Reich 1991, 3, 8)

(The current hiatus within the European Community makes Reich’s use of the term “hurtling” seem rather exaggerated!)

While Reich appears, at least implicitly, to support the view that we are now in a globalizing, if not a globalized, economy, Hirst and Thompson (1992) are more skeptical.  On the basis of a comparison between two ideal types of structure - a globalized international economy and a worldwide international economy - and from an analysis of empirical trends in trade, investment, and political restructuring (notably toward the development of supranational trading and economic blocs), they conclude, “we do not have a fully globalized economy; we do have an international economy and national policy responses to it” (Hirst and Thompson 1992, 394).  Both Reich and Hirst and Thompson agree, in their different ways, that the relationships between firms and states have been drastically changed.  Stopford and Strange (1991) express this in terms of mutual dependence, which, they argue, can be explored in terms of a triangular nexus of interactions comprising firm-firm, state-state, and firm-state relationships.

The aim of this paper is to explore some


of these complex interactions between transnational corporations (TNCs) and states in the context of the global-local debate through a review of a wide variety of literature drawn from different disciplines.  The paper is divided into five sections.  In the first, I make some basic observations about the nature of the production system and its changing organizational structure.  The second section focuses specifically upon the TNC and examines aspects of changing competitive strategies.  Section three is concerned with the competitive rivalry between states.  Section four explores the interactions between TNCs and states.  Finally, in section five, I make some observations on the implications of the discussion for what we might call the “seriously local.”  It is important to emphasize that the terms “global” and “local” are not fixed scales; rather, they represent the extreme points of a dialectical continuum of complex mutual interactions.


An Analytical Point of Entry: Production Chains and Relational Networks

At the crux of the interrelationship between TNCs and nation-states is control of the production system and the relative, distribution of the costs and benefits associated with its operation.  This suggests that we need, first, to understand the production system itself.  One useful point of entry is via the concept of the production chain.  Although the idea itself is far from new, the notion of a chain or filière is central to several recent attempts to explain how production is organized.  In the economic geography literature, for example, Walker (1988, 380) incorporates the concept of the filière—“the connecting filament among technologically related activities” -in his exploration of the geographic organization of production systems.  Storper (1992) uses the concept of the commodity chain as the basis for his analysis of technology districts in a global context, and the basic notion of the chain (although the term is not used) forms part of Storper and Harrison’s (1991) paper on flexibility, hierarchy, and regional development.  Within the world systems literature, Hopkins and Wallerstein (1986) employ a specific interpretation of the commodity chain to exemplify the historical evolution of a capitalist world economy, while Gereffi and Korzeniewicz (1990) provide one of the very few attempts to apply the concept to a detailed empirical study (of the global footwear industry).  The business literature is a third area in which the chain concept has been used.  Porter (1985, 1986) uses the term value chain; others, such as Johnston and Lawrence (1988), use the more conventional term, value-added chain.  Clearly, there is no universally agreed upon term for what is essentially the same concept.  I prefer the term production chain.

The basic structure - grossly oversimplified - of a production chain is shown in Figure la.  It consists of a transactionally linked sequence of functions in which each stage in the sequence adds value to the process of production, whether of goods or services.  Both Hopkins and Wallerstein (1986) and Gereffi and Korzeniewicz (1990) restrict themselves to this basic chain, although they produce more elaborate versions.  They do not incorporate the two additional components shown in Figures lb and lc, yet both are fundamental.  Common to the entire chain of transactionally linked functions are the technological processes involved in production itself and in the physical movement of the constituent elements (Fig. lb); the coordinative, regulatory, and financial functions (Fig. lc); and the whole complex of labor processes that are embedded in each component part of the chain.  An entire production system, therefore, can be thought of as a complex input-output system of linked production chains with vertical, horizontal, and diagonal links (Fig. ld).

There are, of course, many important issues relating to the precise functioning of each of the stages in the production



Figure 1

The Production Chain

Source: based on Dickens, 1992a, Fig. 7.1



chain.  Most attention, however, has been focused on only a very narrow segment of the chain, on production per se, while far less attention is paid to the other functions.  An obvious area of neglect by economic geographers is marketing.  Such marketing geography as does exist is concerned overwhelmingly with retailing and, to a lesser extent, wholesaling.  Yet markets and marketing are just as central to the operation of the production chain as “production” itself (Wells and Cooke 1991).  More generally, overconcentration on production has obscured the importance of services in production.  Increasingly, “services have become a major source of value-added.  Downstream services, in particular, are both a factor contributing to competitive strength and a source of value added” (UNCTAD 1988, 178).  But the value of services to the competitiveness of products is more general than this and applies throughout the production chain (Britton 1990).

For my present purposes, the most significant questions relate to the way in which the production chain as a whole is coordinated and regulated and how it is geographically configured.  As Walker observes, “the pieces of complex production-systems are not just divided up and sent their independent ways on the assumption that their technological input-output imperatives will keep the necessary relations intact.  They must also be actively integrated so that production may proceed to completion.  That is, production-systems must be physically linked up; their labor processes must be coordinated; and the flows of materials, labor, and information between them must be regulated” (1988, 381-82, emphasis added).

The issue of the organization - or governance - of the production chain has been the subject of considerable debate (see Dicken and Thrift 1992 for a summary of views).  It is now widely accepted that the market-hierarchy dichotomy is untenable, as is the notion that the determining influence on the degree of internalization or externalization is transactions costs (Best 1990; Cowling and Sugden 1987; Walker 1989).  What we have, in reality, is a variety of developmental trajectories and a spectrum of different forms of governance, that are best captured by the notion of networked interrelationships structured by different degrees and forms of power and influence.  Such networks invariably consist of a mix of intrafirm and interfirm structures.

Network relationships are central to the concept of the social embeddedness of economic action developed in the literatures of economic sociology and socioeconomics (see, for example, Etzioni and Lawrence 1991; Granovetter and Swedberg 1992; Perrow 1986; Thompson et al. 1991; Zukin and DiMaggio 1990).  Much of the initiative in the business economics literature has been taken by Scandinavian researchers, who have long argued that production systems as a whole, as well as the firms within them, should be interpreted as networks (see, for example, Christensen et al. 1990; Forsgren and Johanson 1992; Johanson 1989; Johanson and Mattsson 1987; Hakanson 1989; Hagg and Johanson 1992).

Networks are relational structures and, as such, take time to evolve, not least because they involve the building of trust.  The nature of the relationships, however, depends a good deal upon the particular power structure.  Several writers have attempted to conceptualize this phenomenon. Christensen et al. (1990), drawing upon the ideas of Reve (1988), identify two broad groups of network organization, one centered around a dominant corporate group and the other based on interface skills.  Storper and Harrison (1991) take a similar stance.  They use a core-ring representation of network power structures to identify four types of network governance structures: (1) all ring, no core; (2) core-ring, with coordinating firm; (3) core-ring, with lead firm; (4) all core, no ring.  Miles and Snow (1992) identify three major types: (1) stable networks, (2) dynamic net­works, and (3) internal networks.

Whatever the precise form taken by networks, business firms clearly remain central to their operation (Dicken and Thrift 1992).  The network conceptualization im­


plies, however, that the boundaries of the firm need to be redefined.  Badaracco (1991) explores this question in his attempt to draw a “new map” of the firm that “displays no sharp dividing line separating the inside of the firm from the outside.  Rather, it shows the firm as a dense network at the center of a web of relationships” (p. 314).  All of this suggests that not all parts of a production network are equal; key roles are played by certain firms.  These are the firms that, in Cowling and Sugden’s definition, possess “the means of coordinating production from one centre of strategic decision making” (1987, 60).  These are the kinds of firm with which I am most concerned in this paper.

Network structures present firms with a considerable degree of flexibility in the extent to which they can internalize or externalize specific functions within the production chain.  Although there may be a general tendency for a particular pattern to develop between firms in a particular industry, there is no a priori reason why this should be so.  Similarly, there is no a priori way of determining the particular ways in which production chains are configured geographically.  Different types of production chain may be configured geographically in very different ways, but the same production chain may be configured differently by different firms.  Certainly, the patterns change over time.  As Porter (1986) has pointed out, the question of geographic configuration/location has to be resolved for each separate function within the chain.  These problems of coordination and configuration of production chain functions face firms of all kinds, but they take on a specific - and more complex - form at the international and global levels and in the context of transnational corporations and their interactions with nation-states.


Firm-Firm Competition: The Changing World of the Transnational Corporation

TNCs today are restructuring their activities in ways that involve: (1) reorganizing the coordination of production chain functions in a complex realignment of internalized and externalized network relationships; (2) reorganizing the geography of their production chains internationally and, in some cases, globally; [1] (3) redefining their core activities and repositioning themselves along the production chain, with a particular emphasis on downstream, service functions.  These developments reflect the nature of TNCs as highly embedded interacting networks involved in competitive struggles in which a diversity of competitive strategies is used.  Such strategies are, themselves, the outcome of contested power relations both inside the firm and, externally, with the constellation of institutions (including the state) with which TNCs interact.

This kind of perspective necessitates a reconsideration both of what we actually mean by a “TNC” and of our conceptualization of how TNCs have evolved over time.  Despite important shifts in the past decade, the literature retains a distinction between those who continue to regard ownership of equity in overseas oper­tions as the key diagnostic feature of a TNC and those who adopt a broader definition that recognizes the imprecision of firm boundaries, emphasizes the role of coordination rather than ownership, and accepts that such coordination extends beyond internalized functions to encompass externalized relationships with independent or quasi-independent firms.  Notwithstanding the problems created for the acquisition of adequate empirical data, my own view is that Cowling and Sugden have devised the most conceptually satisfying definition of a TNC: “a transnational is the means of coordinating production

1. The terms “international” and “global” are not used synonymously.  Internationalization refers simply to the extension of activities across national boundaries; globalization involves more than this and is qualitatively different.  It implies a degree of purposive functional integration among geographically dispersed activities.


from one centre of strategic decision making when this coordination takes a firm across national boundaries” (Cowling and Sugden 1987, 60).  Apart from adopting a less restrictive definition than has often prevailed (and still does in some quarters), we also need to rid ourselves of the stereotypic view of the temporal process of TNC development.

Until relatively recently, much of the business and management literature projected a simplistic sequential path of TNC development.  In fact, there is - and always has been - considerable diversity in TNC strategies and in the kinds of organizational coordination and geographic configuration employed to implement them (see, for example, Perlmutter 1969; Bartlett and Ghoshal 1989; Martinez and Jarillo 1989).  Just as Perimutter (1969), somewhat ahead of his time, was sufficiently sensitive to point to substantial diversity among the “multinational” firms of the 1960s, so, too, it can be argued today that the term “global corporation” obscures a rich variety of actual forms.

Bartlett and Ghoshal’s (1989) typology provides a useful perspective on this issue.  They identify three different organizational models, each with distinctive structural, administrative, and management characteristics.  The major characteristics of each ideal type are summarized in Table 1.  Their multinational organization model is characterized by a decentralized federation of activities and simple financial control systems overlain on informal personal coordination.  The company’s worldwide operations are organized as a portfolio of national businesses.  Each of the firm’s national units has a considerable degree of autonomy; each has a predominantly “local” orientation.  Their international organization model involves far more formal coordination and control by the corporate headquarters over the overseas subsidiaries.  Whereas multinational organizations are, in effect, portfolios of quasi-independent businesses, international organizations clearly regard their overseas operations as appendages to the controlling domestic corporation.  Thus the international firm’s subsidiaries are more dependent on the center for the transfer of knowledge and information, and the parent company makes greater use of formal systems to control the subsidiaries.  Their “classic” global organization model is based on a centralization of assets and responsibilities.  The role of the local units is to assemble and sell products and to implement plans and policies developed at headquarters.  Thus,


Table 1

Basic Characteristics of Different Types of Transnational Business Organization

Source: based on Bartlett and Ghoshal (1989,  Fig. 3.1, 3.2 , 3.3)


overseas subsidiaries have far less freedom to create new products or strategies or even to modify existing ones.  Although each of these three ideal-type models developed during specific historical periods, there is no suggestion that one was sequentially replaced by another.  Each form has tended to persist, to a greater or lesser extent, producing a diverse population of transnational corporations in the contemporary world economy.

Although we can continue to identify these forms of TNC, new forms of transnational organization are also emerging that may - although not inevitably - replace some of the existing forms.  Bartlett and Ghoshal argue that each of their three ideal types of organization possesses specific strengths, but each also has severe contradictions and tensions.  Thus, the global company capitalizes on the achievement of scale economies in its activities and on centralized knowledge and expertise.  But this implies that local market conditions tend to be ignored and the possibility of local learning is precluded.  The more locally oriented multinational organization is able to respond to local needs, but its very fragmentation imposes penalties for efficiency and for the internal flow of knowledge and learning.  The international company “is better able to leverage the knowledge and capabilities of the parent company.  But its resource configuration and operating systems make it less efficient than the global company, and less responsive than the multinational company” (Bartlett and Ghoshal 1989, 58-59).

As Bartlett and Ghoshal point out, the dilemma facing firms - especially large firms - in turbulent competitive environments is that to succeed on a global scale they must possess three capabilities simultaneously.  They need to be globally efficient, multinationally flexible, and capable of capturing the benefits of worldwide learning all at the same time.  Rather confusingly, Bartlett and Ghoshal use the term transnational to describe such an organization and write of the “transnational solution.”  Because I prefer to use the term “transnational” in a generic sense, it is perhaps better to use the term complex global firm for the newly emerging organizational form.  Again, however, this does not imply an inevitable sequential development, but rather that some firms are moving toward such a complex global structure.  A key diagnostic feature of such organizations is their integrated network configuration and their capacity to develop flexible coordinating processes.  Such capabilities apply both inside the firm (the network of intrafirm relationships which, it is argued, is displacing hierarchical governance relationships) and outside the firm (the complex network of interfirm relationships).

Such a conceptualization of the contemporary TNC is consistent with a number of other interpretations of current tendencies in the nature and organization of international business activity.  Best (1990) sees the “entrepreneurial firm” as a key institution in the “New Competition” and emphasizes its strategic, innovation-based, flexible nature and its dependence on learning.  These ideas are echoed in Storper’s (1992) focus on production-based technological learning as a diagnostic characteristic of new forms of production organization and also, indirectly, in Buckley and Casson’s (1992) prognosis for “multinational enterprise in the 21st century.”  A related issue is the apparently increasing emphasis on time as a competitive weapon in business strategy (Best 1990; Stalk 1988; Stalk and Hout 1990; Schoenberger 1991), “the next source of competitive advantage” and one that pervades the entire production process, from initial conceptualization to final marketing and involves the compression of product development cycles.

A central feature of these models of contemporary production organization is that they place particular emphasis on the rich variety of external relationships within production networks.  At the international scale, most attention has been devoted to strategic alliances, which, as Kindleberger (1988) has observed, are by no means a new phenomenon.  But, as


Anderson (1992) has recently pointed out, there has been much ambiguity and confusion of terminology in the literature, and it seems better to use the more general term “collaborative venture” to encompass what is, in fact, a bewildering variety of organizational relationships.  Such relationships are frequently multilateral rather than bilateral, polygamous rather than monogamous.  Collaborative ventures are a long-established form of international business organization that, in the past, involved specific relationships between conventionally organized, hierarchically structured firms as just one element in their competitive strategies.  The argument now is rather different: not only have collaborative ventures in the more traditional sense moved to the center of firms’ strategies, but also - and more controversially - new forms of collaboration are embedded within a much looser network structure, or webs of enterprise (R. Reich 1991).  This view, however, is very much a Western perspective.  Japanese business organizations have long been embedded in a complex structure of interorganizational alliances, epitomized by the horizontal and vertical keiretsu (Gerlach 1992; Eli 1990; Helou 1991).

Seductive as these various ideas about new forms of international business organization undoubtedly are, we face a major problem in distinguishing what is actually happening (and may happen in the future) from the hype and the rhetoric.  When we turn to the hard world of empirical reality we meet up with the intractability - and even total absence - of comprehensive data.  The only variable on which we have (reasonably) comprehensive data is that of foreign direct investment (FDI), which, unfortunately, exclude coverage of most nonequity relationships and activities, the very ones that are becoming especially important.  Within their limitations, however, FDI data provide an important, though partial, indicator of changing levels and patterns of TNC activity at a global scale. [2]  This is not the place to present a detailed analysis of FDI (see Dicken 1992a; Dunning 1993; Julius 1990; Thomsen 1992; UNCTC 1988, 1991, 1992), but it is important for our present purposes to summarize the major contemporary trends.

The recent temporal pattern of FDI growth displays a major upsurge during the 1980s, on a scale exceeding even that of the 1960s.  Julius (1990, 6) estimates that “whereas in the 1960s FDI grew at twice the rate of GNP, in the 1980s it has grown more than four times as fast as GNP.”  The most recent United Nations Center on Transnational Corporations (UNCTC) figures reveal a slackening in FDI growth in 1990-91, associated with a recession in most industrialized countries, but previous experience suggests that FDI growth will resume when the recession eases.  Not only has FDI been growing faster than GNP, but it also has been growing at a much faster rate than world exports, particularly since 1985 (UNCTC 1991).  This figure alone suggests that FDI has become a more significant integrating force in the global economy than the traditional indicator of such integration, trade (Julius 1990).  Indeed,

2.  Dunning observes that “in practice, statistical data on MNEs and their activities, though markedly better than they were even a decade ago, are fragmentary, variable in quantity and quality, and rarely comparable between countries, industries, and firms or over time … while some governments ... compile a fairly comprehensive range of statistics on inward and outward MNE activity, most countries limit their data gathering to the foreign direct investment stake (usually compiled irregularly on a survey or sample basis) and direct investment flows (obtained largely from balance of payments statistics)” (1993, 7-8).  He lists several important specific deficiencies in FDI data, including: variations in coverage, both temporal and spatial; problems associated with using inadequately adjusted book values; variations in treatment of exchange rate differences; the fact that some countries publish data relating to investment intentions while others refer to actual investments.


because TNCs are themselves responsible for a large proportion of international trade (much of this as intrafirm transactions), their global significance becomes even more marked.  Another notable trend has been the massive internationalization of services, much of which has been driven by FDI (Dicken 1992a).

Spatially, a number of important FDI trends can be identified:

(1) Origins have diversified geographically, not only within the industrialized countries, but also from several of the newly industrializing economies (NIEs), notably in East Asia but including some Latin American countries.

(2) Although the United States still accounts for the largest share of the world’s stock of FDI, Japan is now the world’s leading foreign direct investor on an annual flow basis.  The recent collapse of Japan’s “bubble economy,” however, has caused a substantial contraction (and geographic reorientation) of Japanese FDI.

(3) There has been a major intensification of cross-investment between the industrialized economies, an increasingly high level of TNC interpenetration of national economies (Julius 1990).  In this respect, the position of the United States as a home and host country for FDI has shifted dramatically.  In 1975, the U.S. ratio of outward FDI to inward FDI was approximately 11:1; by the end of the 1980s, the ratio had fallen to just below 1:1.  In other words, it is now at a level commensurate with that of the major European economies in which FDI outflows have long been relatively symmetrical.  Overall, then, the relative importance of TNCs (both foreign and domestic) in each major economy has greatly increased, while the global pattern of FDI has become strongly concentrated in the triad regions.

(4) The intensified concentration of FDI in the industrialized economies means that the share going to developing economies remains low: some 18 percent of the world total compared with more than 60 percent on the eve of World War II.  Within the developing countries, the distribution of FDI is extremely uneven: a mere ten countries, primarily the Asian NIEs and “proto-NIEs” and some Latin American countries (OECD 1993), account for 75 percent of the FDI inflow.

(5) The recent political developments in the former Soviet Union and in Eastern Europe have unexpectedly opened up investment opportunities for foreign TNCs.  Although the number of recorded direct investments in those regions in 1992 was 34,422 (UNCTC 1992), their geographic distribution was extremely uneven (almost one-third of the total was in Hungary).

When we turn from FDI to other modes of TNC activity - particularly the kinds of collaborative and network ventures that, it is claimed, have become increasingly important - we face a serious paucity of empirical data and a reliance on case studies and anecdotal evidence at either a sectorally specific or company-specific level.  Particularly useful studies of collaborative ventures are provided by Anderson (1992) (aerospace, automobiles, pharmaceuticals); Cooke (1988), Cooke et al. (1992), Wells and Cooke (1991) (information technology, computers, telecommunications); National Research Council (1992) (semiconductors); Ohmae (1985) (aerospace, automobiles, biotechnology, computers, robotics, semiconductors).  Strategic alliances are particularly important in those sectors that tend to be “typified by high entry costs, globalization, scale economies, rapidly changing technologies, and/or substantial operating risks” (Morris and Hergert 1987, 18; Hagedoorn and Schakenraad 1990).

Most strikingly, the majority of strategic alliances are among competitors.  Of the 839 agreements analyzed by Morris and Hergert (1987) between 1975 and 1986, no less than 71 percent were between two firms in the same market.  In addition, although alliances are certainly not confined to particular sizes or types of firms, they are undoubtedly especially


common between large TNCs with extensive international operations.  Geographically, most alliances in the Morris and Hergert study were between firms from European Community (EC) countries (31 percent) or between EC and U.S. firms (26 percent).  A further 10 percent of the alliances were between EC and Japanese firms and 8 percent between United States and Japanese firms.  Anderson (1992) shows that a substantial proportion of collaborative ventures in the aerospace, automobile, and pharmaceutical industries are “cross-triad” collaborations.

Empirical evidence on the existence and nature of the kinds of dynamic network that are becoming especially important in the global economy is even sparser than that on the more conventional collaborative ventures.  Here, we are limited to a small number of cases, the most notable being the Italian clothing company, Benetton (Elson 1989), and the U.S. athletic shoe company, Nike (Donaghu and Barff 1990; Far Eastern Economic Review 1992).  Clark (1992), however, has recently produced a most interesting and detailed case study of an anonymous small company, headquartered in the Western Pacific region, that appears to be a paradigmatic dynamic network coordinator in the branded textile and apparel industries.  He shows how a small company (a mere 140 direct employees) may, nevertheless, be a global corporation because of its ability to orchestrate the activities of “more than a thousand people involved in designing, producing, distributing and selling its products … [with] … worldwide sales of about US$40 million” (Clark 1992, 2).  Clark places particular emphasis on the significance to such global network firms of access to global communications systems.

The clear message is that transnational reality is one of a spectrum of forms of TNC organization, a diversity of developmental trajectories in which consciously planned global operations exist side-by- side with firms that have internationalized in an unplanned, often adventitious, way.  Across the spectrum, complex restructuring is occurring at all geographic scales, from the global to the local, as strategic decisions have to be made regarding the organizational coordination and geographic configuration of production chain functions.  Decisions to centralize or decentralize decision-making powers, or to cluster or to disperse some or all of the firm’s functions in particular ways are, however, contested decisions (Stopford and Strange 1991).  They are the outcome of power struggles within firms, both within their headquarters and between headquarters and affiliates, and they reflect differences in goals and objectives. How they are resolved depends very much upon the location and nature of the dominant coalition.  Such decisions also have to be seen within the context of the fundamental tension facing all firms that operate at a global scale: whether to globalize fully or to respond to local differentiation.

Anecdotal evidence points to substantial internal reorganization within global corporations, involving (1) allocating greater autonomy to affiliates; (2) responding to major external changes (e.g., regional integration) by spatial rationalization of production, particularly by placing greater emphasis on the market-serving functions; and (3) transforming relationships with supplier firms.  Thus, for example, there has been an outburst of corporate federalism in companies like IBM, Coca Cola, ABB, and BP (Financial Times, 18 December 1992).  In some cases, this also involves a global shift in major divisional headquarters, as in the cases of Hewlett Packard, IBM, Monsanto, and Nestlé, among others.  Spatial rationalization is a continuing process in all TNCs, but, from time to time, specific events generate pressure for wholesale change across the corporate spectrum.  Such, for example, has been the effect of the Single European Market (see, for example, Amin 1992; Amin, Charles, and Howells 1992; Bachtler and Clement 1990; Cantwell 1987, 1992; UNCTC 1990), although even here it is still too early to be sure of the


precise impact amid the flurry of rhetoric and hyperbole.

next page