Fluctuating Rounds of Inward Investment in ... - Wiley Online Library

Economic Geography 83(1): 51–73, 2007. © 2007 Clark University. http://www.clarku.edu/econgeography

Fluctuating Rounds of Inward Investment in Peripheral Regions: Semiconductors in the North East of England Stuart Dawley Centre for Urban and Regional Development Studies/ School of Geography, Politics and Sociology, Claremont Bridge, University of Newcastle upon Tyne, Newcastle upon Tyne NE1 7RU, United Kingdom [email protected] Abstract: This article extends economic geography research on foreign direct investment episodes by developing a historically grounded understanding of the socioinstitutional relations that shape and constrain different rounds of (dis)investment by multinational enterprises (MNEs) within a host region. Sensitive to the roles of contextuality, path dependency, and contingency, it argues that the temporal and spatial dynamics of volatile MNE (dis)investment are best tackled using a conceptual framework that accords a full and active role to the agency of the firm and its interrelations with the geographically variable socioinstitutional contexts that produce, regulate, and mediate investment decisions. The framework is used to interpret the brief but fluctuating history of the semiconductor fabrication industry in North Tyneside in the old industrial region of North East England. Within each investment episode, the empirical findings reveal the pivotal power and agency of the corporation in shaping and connecting processes across a variety of scales, places, and times. Contrasting corporate strategies illustrate the dynamic and contingent ways in which home and host national institutional contexts matter in mediating and regulating MNE investment decisions. Key words: multinational enterprises, inward investment, historical contexts, socioinstitutional contexts, corporation, home and host nations.

In recent years, a series of high-profile closures of multinational enterprise (MNE) plants and short-lived investment projects across peripheral host-region economies has provided a stark reminder that even prestigious, flagship “performance plant”-type investments remain vulnerable to volatility in investments and corporate rationalization (Dawley forthcoming; MacKinnon and Phelps 2001a; Hudson 2000). Linked to broader questions of the heightened mobility and turnover of multinational capital in time

and space (Jessop 1999), attention is increasingly focusing on the dynamics of “timebased” competition (Schoenberger 1997; van Egeraat and Jacobson 2005) and the shortening life span of individual foreign direct investment (FDI) projects (Phelps and Raines 2003). Driven by disruptive technological change and market preferences, competitive pressures are forcing the reduction in the time-to-market of products and the consequent shortening of the life cycles of products (Stalk and Hout 1990;

Thanks to Bjørn Asheim and the two referees for their helpful comments in the preparation of this article. Thanks are also due to Andy Pike, Nick Henry, Henry Yeung, Nick Phelps, and Andrew Wood for advice on an earlier version of this article, which was presented at “The Political Economies of Inward Investment” session, cosponsored by the Economic Geography Specialty Group and the Political Geography Specialty Group, at the annual meeting of the American Association of Geographers, Denver, Colorado, 5–9 April 2005.

51

52

ECONOMIC GEOGRAPHY

Yeung 2006). According to Phelps and Raines (2003, 3), these competitive pressures expose MNE operations to the “heightened mobility of productive investments as the ‘lives’ of individual production facilities dwindle, their future existence repeatedly and frequently become subject to parent company review.” Compounding the vulnerabilities of host peripheral regions within MNE spatial divisions of labor and accentuated by the recent shift toward deregulated trade and investment policy regimes (Peck and Yeung 2003), the emerging salience of time-based competition and product life-cycle perspectives provides important theoretical insights into the economic geographies of multinational investment activity. However, to date, the dynamics of the volatility of investments, especially corporate abandonment and plant closures, has received little attention in the flourishing studies of global production networks (GPNs), for example (cf. Phelps and Waley 2004; Henderson et al. 2002). This article makes a theoretical contribution to this field by developing a holistic conceptualization of volatile FDI episodes, integrating insights from economic geography’s recent “institutional” and “relational” turns with a geographic political economy approach that seeks to qualify, rather than to deny, structural economic imperatives (Goodwin 2004; Phelps and Waley 2004; Pike 2005). The article contends that research on MNE (dis)investment and the development of peripheral regions can benefit from an economic geography approach that accords a full and active role to the agency of the firm and the multiscalar and territorially variable socioinstitutional contexts that produce, regulate, and mediate investment decisions (Dicken and Malmberg 2001). Therefore, time-based economic and technological imperatives are necessarily mediated across and within a variety of multiscalar socioinstitutional contexts— industrial systems, corporations, and national and local institutional contexts—leading to spatially and historically contingent concrete investment outcomes. Contributing to the “relational turn” within economic geography,

JANUARY 2007

I argue that the temporal and spatial dynamics of volatile FDI (dis)investment are best tackled using a conceptual framework that is sensitive to the roles of contextuality, path dependency, and contingency in shaping economic process and action (Bathelt 2006; Martin 2000). In particular, the approach developed in this article argues for the importance of reconnecting historical contexts with existing economic geography analyses of MNE investment and regional development (Bathelt 2006; Massey 1995). By understanding that economic activity is enabled, constrained, and refracted by the changing contexts of social relations (Martin 2000, 79), one sees that each episode of MNE investment requires articulation as the latest historical “moment” within an unfolding socioinstitutional process that is shaped by its own historical and geographic evolution (Harvey 1996; Pike 2005). Economic imperatives are therefore understood as historically contingent and spatially heterogeneous, with the contingency of corporate strategies and investment decisions molded through a degree of “bounded indeterminacy” (Massey 1995, 316; Bathelt 2006). Some recent studies have developed important insights into the evolutionary contexts of GPNs using the analysis of individual and discrete MNE investment episodes (Hess and Yeung 2006; Phelps and Raines 2003; Phelps and Waley 2004). I extend the existing research base by conceptualizing the conjuncture of causal processes and concrete outcomes both within and across multiple rounds of MNE investment in one locality (Massey 1995). This approach seeks to delineate and subsequently place MNE investment episodes within the complex and geographically situated “institutional ensembles” that regulate, order, and steer economic actions across successive and interrelated historical moments (Wood and Valler 2001). An extended historical focus provides an important analytical lens through which one can gain a better understanding of both the commonalities and the particularities of causal processes, power relations, and institutional configurations of MNE

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

(dis)investment within peripheral host regions (Peck 2000; Sayer 2000). This approach provides the conceptual framework for the analysis of a particular instance of volatile high-technology inward investment in the peripheral host region of North East England. In 1995, German MNE Siemens Semiconductors made an initial inquiry to the Invest in Britain Bureau concerning possible sites within the United Kingdom for a new greenfield semiconductor wafer-fabrication plant. On 23 May 1997, the United Kingdom’s largest ever inward investment project was opened on North Tyneside, representing a £1.13 ($1.9) billion investment by Siemens. For the deindustrialized locality of North Tyneside, situated in the old industrial region of North East England, the creation of 1,200 highskilled jobs by the Siemens “flagship” investment project was suggestive of a “new dawn” for the region’s industrial and economic base. However, after only one year of commercial production, Siemens announced its decision to close the North Tyneside plant. Siemens explained the closure as a “logical sacrifice” (Atkins 1998, 5) in the face of tumultuous conditions within the global semiconductor market, while others suggested that the investment episode illustrated that “the branch plant syndrome was far from dead” (Hudson 2000, 31). In September 2000, just over two years after the announcement of the closure, U.S. semiconductor firm Atmel announced the conclusion of a deal to acquire the redundant Siemens North Tyneside plant as part of a four-year $1.5-billion supply agreement with Siemens (Atmel 2000). For Stephen Byers, the then-secretary of the U.K. Department of Trade and Industry (DTI), Atmel’s decision demonstrated the North East’s ability to “bounce back .|.|. from what proved to be one of the biggest challenges in the region’s economic history” (quoted in DTI 2000). The optimism and enthusiasm demonstrated by the politicians and economic development practitioners who were involved in seducing Atmel to come to the region was buoyed by the company’s pledge of “a commitment on our side not

53

to be a slash and burn operation. We have been through tough times, like everyone in semiconductors, but we have not resorted to plant closures in the past” (Atmel chief financial officer, quoted in Wilson and McKay 2000, 2). The brief but fluctuating history of the semiconductor industry in North Tyneside is important in several respects. First, despite the remarkably persistent cyclical nature of the semiconductor industry, the contrasting investment strategies and responses that Seimens and Atmel adopted illustrate both the importance and variance of corporate agency within economic geography research (Taylor and Ashiem 2001; Schoenberger 1997). In particular, the conceptual framework and empirical results demonstrate the salience of studying the “firm” as a key phenomenon in economic geography that shapes and connects processes across a variety of scales, places, and times (Yeung 2006). Second, the two contrasting corporate strategies illustrate the mutually constitutive relations among firms, territories, and space (Dicken and Malmberg 2001). In particular, the cases of Atmel and Siemens reveal the dynamic and contingent ways in which home and host national institutional contexts matter in mediating and regulating MNE investment decisions (Yeung 1998; Dicken 2000). Third, both cases indicate the path dependency of sociospatial relations within which the trajectories of MNE operations evolve. The altered roles and functions of the North Tyneside plant under successive rounds of investment plants indicate the contingency of individual corporate contexts and the various levels of territorial embeddedness of interplant relations (Phelps and Fuller 2000; Birkinshaw 1999). Fourth, the analysis reveals that the structural political and economic power between the MNE and the host economy remains loaded in favor of capital. In the case of Siemens, the protracted struggle by host national, regional, and local-level institutions initially to attract and subsequently to replace the high-technology investment project reflected the dynamics of the global industry and, more important,

54

ECONOMIC GEOGRAPHY

the asymmetry of power between MNEs and host regions (Phelps and Waley 2004). Even in the aftermath of the closure of the Siemens plant, the region remained relatively powerless to realize the regional assets of the redundant high-technology plant and infrastructure owing to the necessary relationship of ownership and continuing external control (Coe et al. 2004). The remainder of this article is structured into five sections. In the first section, I develop a conceptual approach that articulates MNE investment decisions within a historically grounded, socioinstitutional framework centered on an economic geography analysis of the corporation and the national, regional, and local institutional contexts. The second section introduces the contemporary global semiconductor industry to provide the broader context of global competition and economic imperatives within which each investment episode in the North East region can be situated. The third and fourth sections demonstrate how corporations and (sub)national institutional contexts mediate generalizable economic imperatives to produce spatially and historically contingent investment outcomes. The third section analyzes the evolution of each investment episode in North Tyneside within individual corporate contexts and the intracorporate sociospatial relations of MNE decision making. The fourth section analyzes the complex nexus between firms and the home and host territorial contexts in molding and shaping investment decisions and outcomes. The fifth, concluding, section discusses how this particular story and the historically grounded socioinstitutional understanding developed within this article raises important issues for future research on MNE investment and the development of peripheral regions.

Placing MNE Investment Episodes in an Historically Grounded Multiscalar Socioinstitutional Framework The recent interest in time-based competition and product life cycles necessitates a

JANUARY 2007

brief consideration of former approaches to the temporal and spatial dynamics of MNE investment activity and the development of peripheral regions (Phelps and Raines 2003; Schoenberger 1997). The late 1970s and early 1980s were the heydays of cyclical models of MNE industrial-location behavior, built around the profit-cycle approach (Markusen 1985) and its influential antecedent, the product-cycle model (Vernon 1979; Taylor 1986). Both approaches connected the cyclical evolution of products and sectors to the geographic dispersion and life course of MNE investments, with peripheral regions hosting the most labor-intensive, cost-sensitive, and consequently ephemeral operations. However, with the rise of structuralist realism in economic geography in the mid1980s, cyclical models of industrial change were discredited as essentialist, technologically deterministic, and ultimately disembodied from the concrete phenomena of industrial behavior (Sayer 1985; Walker 1985). Nevertheless, elements of both approaches connected with an emerging concern about the volatility of investments within the classic “branch plant” critique of FDI that conceptualized a geographic division between core and peripheral localities within corporate spatial divisions of labor (Pike, Rodriguez-Pose, and Tomaney 2006). Here, cyclical notions of investment vulnerability were replaced with a nontemporally determined exposure to corporate abandonment through spatial structures of production within broader social relations of production (Massey 1995; Yeung 2005). However, during the 1990s, attention shifted to the rise of so-called performance plant projects embodying a qualitative transformation in the profile and durability of inward investment through heightened levels of plant autonomy, more complex functionality, specialized markets, heightened product and process technologies, and more qualified workforces (Amin et al. 1994). Yet continued skepticism surrounding the nature and scale of any qualitative transformation has been recently reinforced by a series of performance plant closures, rationalizations, and

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

postponed investments (Dawley forthcoming; van Egeraat and Jacobson 2004). Most recently, in the guise of the literature on GPNs, MNE activities and investment strategies have come to be understood as being situated and produced within particular socioinstitutional contexts that are nested across geographic scales and territories (Coe et al. 2004; Smith et al. 2002). Almost 30 years from the heyday of the product-cycle approach, notions of temporal determinism have been replaced with path dependency and an openness to contingency and context (Bathelt 2006). Put another way, how firms respond to economic imperatives and altered external environments reflects their permeability to broader socioinstitutional contexts that envelop, regulate, and mediate corporate strategies and investment decisions (Dicken and Malmberg 2001). However, unraveling causality within diverse socioinstitutional contexts, both as formal organizations and as informal ordering structures, remains an ongoing research agenda for research on GPNs and the articulation of MNE investment episodes more broadly (Hess and Yeung 2006). This article develops a holistic institutional framework, connecting hard and soft institutions, multiple scales, and territorial and historical contexts. Figure 1 presents a schematic multiscalar conceptualization for FDI of what North (1990) and Martin (2000, 80) termed the “institutional regime,” namely, the mutually constitutive “institutional environment” and “institutional arrangements” (see also Barnes 1999). The institutional environment refers to the broad context of both formal and informal conventions, ranging from, for example, the habitual forms of corporate behavior and culture to legally enforced forms of trade, labor, and competition regulation—with both spheres providing important, often territorially varied, influences and controls over socioeconomic behavior. In terms of institutional arrangements, Martin (2000) provided examples of particular organizational forms that are situated and governed by the institutional environment, including markets, firms, and national and local economic development agencies.

55

From this viewpoint, institutional arrangements and environments that shape MNE investment decisions are multiply scaled and mediate economic action simultaneously across and between geographic levels (Wood and Valler 2001). In this sense, although power remains a “multi-scalar process, with varying combinations of actors cooperating or playing off one against the other,” its realization is still contingent—nested at a variety of interrelated scales (Coe et al. 2004, 481; see also Amin 1998; Peck 2000, 2002). At the same time, scalar power relations remain influenced and differentiated by degrees of territorial embeddedness, reflecting a long-standing preoccupation within economic geography with the ways in which MNEs encourage international economic integration but do not undermine the importance of place (Phelps and Waley 2004, citing Jessop 1999). However, research continues to underplay the historical contexts through which multiscalar power relations and territorial embeddedness are produced and played out. This article extends economic geography research on FDI investment episodes by developing a historically grounded understanding of the socioinstitutional relations that shape and constrain economic actions during successive “moments” of MNE (dis)investment in host regions (Bathelt 2006; Pike 2005). Recent contributions to research on GPNs and the “relational turn,” more broadly, have increasingly focused on the historical and spatial development of corporations and their embeddedness and interplay with broader institutional contexts over time, across space, and in place (Coe et al. 2004; Phelps and Waley 2004; Hess and Yeung 2006). Welcome though these accounts are for an understanding of the evolution of economic action, the analytical focus remains centered on the historical contexts of individual and discrete investment episodes. By applying a socioinstitutional framework across and between two successive rounds of investment in one locality, this article develops two important historical perspectives. First, the approach responds to Peck’s (2000, 76) call to go beyond the cata-

ECONOMIC GEOGRAPHY

Figure 1. A framework for identifying institutional environments and arrangements for FDI (developed from Martin 2000).

56 JANUARY 2007

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

loguing and identification of institutions to probing institutional “liabilities, tendencies and contradictions under different contextual conditions.” Unraveling the “causal effects and liabilities” of institutional ensembles (e.g., the firm, national state, markets) across successive investment episodes moves the analysis beyond contextual description toward the identification of common aspects of causal mechanisms (Peck 2000, 68; Yeung 2003; Sayer 2000). Second, resonating with Massey’s (1995) spatial divisions of labor approach, the conceptual framework explores the degree to which previous rounds of investment not only provide the context for, but also shape, subsequent rounds of MNE investment. This line of analysis adds important conceptual and empirical contributions to recent GPN studies that have explored the strategic coupling of “regional assets” that are embedded within localities and the locational requirements of global firms (Coe et al. 2004). In particular, few studies have examined the relative power of host regions to retain, reinvigorate, or recycle the hard and soft infrastructural legacies of disinvesting “focal firms.” In the context of this article, the remainder of this section illustrates how the holistic institutional framework provides the conceptual basis for a more sophisticated articulation of investment episodes across a necessarily selective series of interrelated multiscalar analytical strands: MNE corporate contexts and strategies and national and subnational institutional contexts. MNE Corporate Contexts and Strategies Yeung (2006, 2) lamented the extent to which recent research in economic geography has “sidestepped the issues of researching how business firms perform as the movers and shapers of the capitalist economy” (see also O’Neill 2003). As a purely economic agent, relative to other agents, such as governments, the firm continues to possess structural power (Phelps and Fuller 2000). Moreover, consti-

57

tutive of the institutional contexts that they connect, firms increasingly exhibit political power, for example, using mobility to promote convergence across national regulatory environments (Phelps and Waley 2004). The perspective developed here examines the agency of the corporation within a more “pluralistic industrial geography,” according an active role to the socioinstitutional and cultural regulation of investment behavior (Yeung 2001, 293). In this sense, although the capitalist “firm” continues to respond to economic imperatives, its precise strategies and actions evolve in response to specific social, cultural, and political contexts—both internal and external to the corporation (Schoenberger 2000; Yeung 2000). While much of the research on GPNs has thus far been preoccupied with inter- and extrafirm institutional relations, I open up the black box of intracorporate activity during investment episodes. When the firm is conceptualized as a complex sociospatial and territorial construction, the behavior of corporate actors and strategists need not be confined to a singular logic of profit maximization, but instead reflects the influence of competing discourses, cultures, and politics within the firm (Dicken and Malmberg 2001; Yeung 2001; Schoenberger 1997). In particular, I explore two dimensions in this article. First, in seeking to reconnect historical contexts with economic geography analyses, I contend that investment decisions and corporate strategies are shaped by the evolution and path dependency of the institutional architecture of capital (Clark 1994). Path dependency can reflect the strategic disposition of corporate activities (Heenan and Perlmutter 1979), together with the territorial embeddedness of firms and corporate cultures (Dicken and Malmberg 2001). Second, evidence of timebased competition and the shortening life span of individual FDI projects suggests that elements of the product life-cycle approach still provide some practical bearing on the functional and spatial organization of affiliates within the MNE networks (Phelps and Fuller 2000). In this way, the status and roles of MNE affiliates are connected to the

58

ECONOMIC GEOGRAPHY

life cycle of each plant’s product and process technologies. Intracorporate spatial divisions of labor emerge and are contested as MNE operations compete for the parent company’s investments to obtain first-mover advantages and avoid being locked into declining product markets (Birkinshaw 1999; Phelps and Fuller 2000). Connected to Massey’s (1995) work, and while not situating plants within rigid spatial structures, managerial hierarchies (including economic ownership) and technical divisions of labor continue to contribute to our understanding of sociospatial power relations within corporate networks and GPNs (Phelps and Fuller 2000; Birkinshaw 1999). However, to gain a better understanding of the dynamics of intracorporate divisions of labor, I draw on Yeung’s (2005) review of the relational turn and argue for a more expansive and inclusive appreciation of sociospatial relations within corporate networks by situating plants and their connections within broader and diverse socioinstitutional contexts (Sunley 1991). For example, Schoenberger (1997) illustrated the variations in corporate subcultures and the transmission of knowledge across MNE networks, while Birkinshaw and Hood (1998) focused on the agency of entrepreneurial affiliates to alter their position within corporate sociospatial hierarchies. National and Subnational Institutional Contexts A long-standing literature illustrates that the nation-state and national-level institutional environments and arrangements remain a crucial analytical element in understanding MNE investment and disinvestments (Dicken 2003; Hudson 2001; Yeung 1998). Thus, Dicken (2000, 284) suggested that the interfirm, intrafirm, firm-place, and place-place connections that developed through processes of international economic integration are fundamentally embedded within asymmetrical, multi-scalar power structures.|.|.|. Each of these sets of relationships is embedded within and

JANUARY 2007

across national/state political and regulatory systems which helps to determine the parameters within which firms and place interact.

Considerable attention has focused on the continuing roles and capacities of the nationstate as an institutional arrangement in terms of developing political and policy structures to promote and regulate the spatial strategies of the investment activities of MNEs (Phelps and Waley 2004; Yeung 1998). Moreover, the “complex, dynamic interactions between states and firms” create the context for “regulatory arbitrage” within which MNEs play one state (and the communities within them) off against another (Dicken 1998, 10). However, although states engage in regulatory arbitrage in the attraction of inward investment, they are relatively powerless to prevent companies that are attracted by national regulatory environments that emphasize deregulation and labor market flexibility from taking advantage of low exit costs to close or rationalize operations within host regions (MacKinnon and Phelps 2001a). In parallel, the embeddedness of an MNE within its home nation has been proved to play an important role in influencing and enmeshing corporate strategies and behaviors within territorially distinctive assemblages of institutions and practices (Dicken 2000). Insights are derived from studies that revealed the distinctiveness of national business systems and varieties of capitalism that suggest “although firms do respond and react to (or anticipate) changing competitive conditions .|.|. the strategy they choose is most strongly shaped by the national legacy of their home country” (Gertler 2001, 14; see also Bathelt and Gertler 2005). Through the processes of MNE-driven international economic integration, the roles of home and host political, cultural, and regulatory institutional environments serve to embed the firm-territory nexus and contextualize intracorporate sociospatial relations during (dis)investment episodes (Phelps and Fuller 2000). Finally, the multiscalar institutional analytical framework that is presented here also

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

provides an opportunity to articulate the agency of regional and local institutions within the MNE investment and disinvestments episodes. Coe et al. (2004) restated the importance of local and regional institutional agency and capacity in promoting and negotiating issues of power and control with focal firms in GPNs. Within the global context of geoeconomic deregulation and mobile capital, “regional institutions” (including nonlocal institutions with regional influence) have an important role to play in promoting and coupling “regional assets” with the strategic needs of focal firms within GPNs (Coe et al. 2004; Amin and Thrift 1994). Considerable attention has focused on the roles and efficacy of local and regional institutions in both seducing “flagship” FDI projects and embedding MNE investment within host economies (MacKinnon and Phelps 2001a; Phelps and Fuller 2000; Amin et al. 1994). However, Phelps, MacKinnon, Stone, and Braidford’s (2003) comprehensive analysis of several U.K. host-region economies questioned both the extent of MNE embeddedness and any causal link between local institutional capacity and the embedding process. These points contribute to the emerging research interest in the degree to which localized points of opportunity and resistance (local affiliates, institutions, and communities) are able to moderate the powers and investment decisions of MNEs within host-region economies (Phelps and Fuller 2000; Phelps and Waley 2004; Pike 2005). The empirical findings presented here were based on an intensive research method that used the case studies of the Siemens and Atmel inward investment projects in North Tyneside between 1995 and 2004 (Sayer 2000). In the study, I drew on secondary materials in the form of corporate literature (e.g., press releases, annual reports, and strategy documents), local and national newspaper articles, the specialist semiconductor industry literature, trade journals, governmental reports, and press releases and on in-depth semistructured interviews with representatives of Atmel, Siemens, and a range of local and regional

59

economic development organizations. Finally, I used participant observation undertaken within the region’s Semiconductor Industry Forum (SIF), comprising representatives from Siemens and latterly Atmel, together with a variety of local and regional institutions, whose aim is to support the skills, infrastructure, and business-support needs of the industry (Dawley 2003).1

The Global Semiconductor Fabrication Industry The global semiconductor industry has been described as “the dominant technology of the last three decades, extending its transformative influence into all branches of the economy and into many aspects of society at large” (Dicken 1998, 353). A number of key characteristics of the industry can be identified. First, the spectacular growth and commercialization of the semiconductor industry has been characterized by its continual capacity for technological change in the domains of both product and production technology (Angel 1994; Morgan and Sayer 1988; Sayer 1986). Second, the evolution of the semiconductor industry has been truly global in terms of both production and markets (Dicken 1998). Third, despite longterm growth of 17 percent per year, in reality the semiconductor industry market is highly cyclical and has rarely come within 7 percentage points of the long-term average (Leckie 2001; McClean, Lam, and Kanz 1998). More specifically, since 1970, the modern integrated circuit industry has experienced six industry cycles, characterized by “chip gluts” and “chip famines” (McLean

1 The participant-observation strand of the research was derived from access that was provided through a collaborative doctoral research project funded by the Economic and Social Research Council and the North of England Microelectronics Institute (NEMI). NEMI coordinated the SIF and facilitated my position as “observer-as-participant” (Burgess 1984) in a sphere of research that is beyond public access.

60

ECONOMIC GEOGRAPHY

2001; Morgan and Sayer 1988, 45). While each cycle reflects a contingent combination of forces and events, it is possible to delineate two interrelated drivers of volatility: derived demand and sticky supply. Driven by a reciprocal relationship between semiconductor innovation and derived demand within electronic applications, producers must predict future demand to accommodate the lead times of the construction of plants or the development of process technology. However, owing to the lead times to production (possibly two to three years), by the time firms are able to enter into production, the nature and character of the market demand and hence price may have changed. Even so, large-scale investments necessitate that firms continue to supply a saturated market to achieve capacity use and to compete in a market that is driven by initially high prices as producers reap supernormal profits from new technology that diminish rapidly through fierce price competition (Morgan and Sayer 1988). During the early 1990s, the global semiconductor industry was experiencing its fifth cyclical upturn since the early 1970s. The arrival of Microsoft’s software package Windows 95, for example, was heralded as creating a new demand by consumers and businesses for personal computers (PCs) and hence a boom in the derived demand for semiconductors. In particular, the global semiconductor industry grew by 29 percent in 1993, 32 percent in 1994, and 42 percent in 1995 (McClean, Lam, and Kanz 1998). By 1998, however, the DRAM (dynamic random access memory) industry was experiencing a chip glut running at 18 percent overcapacity (Manners 1998b; Morgan and Sayer 1988, 45). In light of underinvestment in the wake of the previous industry cycle, the strong market growth of 1993 to 1995 led to belated aggressive capacity spending by companies like Siemens to capture the wave of market growth. Yet the large-scale investments in additional capacity merely accentuated the market’s collapse, compounded by the failure of Microsoft Windows 95 to drive anticipated growth within the PC market

JANUARY 2007

and the debilitating implications of the Asian recession (Jones, Jones, Jowit, and Wighton 1998; Robertson 1998a). In the aftermath of the DRAM collapse, the market subsequently entered its sixth cyclical upturn, driven by a derived demand from the telecommunications industry for the more sophisticated flash memory market. In turn, by 2001, the flash memory market experienced its first downturn as a result of the “perfect storm” of a market collapse, which was triggered simultaneously by the global recession, surplus inventory, overcapacity, and a decline in the sales of electronic systems (Romanelli 2002). The following sections connect these broader industry-level processes to the concrete investment episodes in North Tyneside through the analysis of corporate strategies and investment decisions and national and regional institutional contexts.

MNE Corporate Contexts, Strategies, and Investment Decisions The case studies of Siemens and Atmel reveal the important ways in which corporate agency serves to mediate and translate similar broader structural forces into contingent and varied outcomes over time, across space, and in place (Yeung 2006; Dicken and Malmberg 2001; Dicken and Thrift 1992). To understand the nature of corporate agency, the analysis situates Siemens’s and Atmel’s investment decisions as moments within the respective longer-term historical, institutional, and geographic evolutions of both corporations (Pike 2005; Clark 1994). In particular, by opening the black box of the corporation to political, cultural, and territorial contexts, the analysis develops a historicized economic geography understanding of corporate strategies, actors, and investment decisions (Yeung 2000; Schoenberger 1997). The historical analysis then “places” the North Tyneside plant within distinctive sets of intracorporate sociospatial relations that unfolded across the two rounds of investment (Massey 1995).

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

The approach highlights the dialectical relations between corporate contexts and the path dependency of managerial and technical divisions of labor within each investment episode (Clark 1994; Phelps and Fuller 2000). However, contrasting the Siemens and Atmel investment episodes reveals the importance of situating intracorporate social relations within the contingent interplay of socioinstitutional settings and incentives that permeate the organizational boundaries of firms (Dicken and Malmberg 2001; Yeung 2005). Siemens Semiconductors: Corporate Contexts and the Fifth Global Semiconductor Cycle Siemens decided to invest in North Tyneside at a time when it was increasingly looking to accelerate the globalization of its activities to reduce its dependence on the German domestic market and to enhance Anglo-American shareholder values within the corporation (Bathelt and Gertler 2005; Loewendahl 2000; Siemens 1999). According to Loewendahl (2000, 330), the investment decision represented a key stage within the company’s struggle to “break away from its ‘German’ history, driven by technology and engineering rather than profit.” Siemens Semiconductors—one of 16 corporate groups—developed historically as a “captive” semiconductor producer serving internal demand, but in the early 1990s, it switched to a “merchant” producer to capture the burgeoning external chip market (Siemens 1997b). The semiconductor division became a priority within the broader corporate objectives of growth, globalization, and market orientation. Consequently, an ambitious four-year DM 8 ($4.8) billion investment strategy was developed to treble the company’s existing chip sales with the aim of becoming one of the world’s top 10 semiconductor companies (Siemens 1997a; Loewendahl 2000, 343). Four new largescale fabrication projects were at the hub of this new capacity-expansion plan, including the £1.13 ($1.9) billion North Tyneside plant.

61

However, while Siemens’s investment program initially mirrored three years of high growth, by 1998, the industry had plummeted to 65 percent below the market’s historical 17 percent average annual growth rate. The market crisis enforced an annual loss of DM 1.2 billion ($722 million) on Siemens Semiconductors—a stark contrast to the DM 109 ($65) million profit the year before (Siemens Chief Executive Officer (CEO) 1998; “Infineon Reduces Deficit” 1999). Although one-third of the groups within the broader conglomerate experienced negative business trends during 1997–1998, the semiconductor division’s absorption of $1.1 billion of capital expenditure created a “hailstorm” effect for the ailing corporation, which was already lagging behind the German share market by 30 percent since 1997 (“Horrendous Profits Warning” 1998; Siemens CEO 1998). In response, a radical corporate restructuring program was announced that included preparations for the semiconductor division to be spun-off and publicly floated.2 To stabilize the semiconductor division prior to flotation, “a very, very quick decision was made following a review of the semiconductor business” at the Siemens divisional headquarters in Munich to close the North Tyneside facility to reduce the division’s overproduction by 20 percent (Manners 1998a; see also Jones 1998). Despite the cost of the North Tyneside disinvestment process, which accounted for 45 percent of the division’s projected annual losses of $565 million, the closure decision was perceived as “essential .|.|. to move towards restoring our profitability and securing the future of 2 The semiconductor spinoff company—named Infineon—shifted away from the exposure created by large-scale capital investments in solely owned fabrication plants (of which North Tyneside was the most recent), instead increasingly looking toward foundries as suppliers of chips and a more extensive use of contracted-out assembly in an attempt to develop a more riskaverse attack on the DRAM market (Robertson 1998b).

62

ECONOMIC GEOGRAPHY

the group” (Siemens Semiconductor CEO, cited in Osborne 1998; see also Atkins 1998). Placing North Tyneside Within Siemens’s Corporate Strategy From Siemens’s Munich headquarters, the CEO of the semiconductor division explained the North Tyneside closure as the logical sacrifice of the “weakest link in the chain” (quoted in Cane and Jones 1998, 1). Yet, in addition to the closure of an aging and rapidly obsolescent production line in Munich with the loss of 450 jobs, Siemens’s North Tyneside plant was the only major fabrication casualty within the division. However, the whole story of the Siemens closure requires North Tyneside to be placed within the richer context of the unfolding political-economic dynamics of corporate strategies and sociospatial relations (Phelps and Waley 2004). The North Tyneside plant was Siemens’s only wholly owned fabrication plant outside Germany and Austria. The Corbeil-Essonnes (France), Whiteoak (United States), and ProMOS (Taiwan) plants were all forms of joint-venture ownership, licensing, and research agreements. To begin with, the complex nature of joint-venture arrangements within the semiconductor industry makes it much more complicated to exit from jointly owned facilities than from facilities under sole ownership (Osborne 1998). Furthermore, indicative of broader shifts in the industry, a key objective of the Siemens spinoff (Infineon) was to move away from the scale of investment and associated risks of wholly owned fabrication plants (Angel 1994). Thus, as Siemens’s U.K. CEO commented, “the reality is that the North Tyneside plant could not compete on the world stage without a joint-venture partner” (quoted in Linford 1999, 7). In terms of technical divisions of labor, although the North Tyneside plant was little more than a year into commercial production, it was positioned weakly in terms of economies of scope and scale within the intracorporate network. Following Siemens’s

JANUARY 2007

decision initially to equip North Tyneside to capture the dwindling returns of the aging 16Mbit market niche, the plant became path dependent on the fortunes of a collapsing product market. Despite emerging demands in new product markets, the facility became the only project within the division not to be ramped in either the next generation of DRAM or the more stable logic chip markets. Consequently, at the time of closure, the North Tyneside plant was using only one-third of its production potential, meaning “in volume terms we were unable to compete with our other factories, which were running at full capacity .|.|. the plant was therefore financially unviable” (Siemens’s U.K. CEO 1998). The fortunes of the North Tyneside plant also appear to reflect the changing nature of intracorporate politics and power relations that are enmeshed within home-host contexts and settings (Dicken 2000). Prior to Siemens’s decision to invest in North Tyneside, a strong lobby developed within the corporation against the decision to locate a wholly owned fabrication plant outside Austria and Germany: We did hit some quite significant politics in Siemens, and that was a group of people who were very, very keen for this project to go to Villach (Austria). The Austrian contingent was both inside the project team and inside Siemens at a senior management level. (Former Northern Development Company chief executive, interview 2002)

At the same time, the determination of elements of the Siemens management to make an investment that would identify the company as a global company with “Anglo Saxon” market values suggests that the decision in favor of North Tyneside “was never an entirely commercial decision; .|.|. sentiment and company politics had their say” (Manners 1998a). Indeed, soon after the North Tyneside investment was secured, Siemens Semiconductors appointed a new CEO to address the division’s worsening performance. Following the announcement of the

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

closure in 1998, the new CEO was identified by local North Tyneside management as one of a group of individuals who had previously opposed the North Tyneside plant in favor of investment in either Dresden or Austria, leading a senior local manager to posit: Decisions, even on a major £1.1 billion project, are made by people. When different people are changed, then you get a change in attitude .|.|. and that was quite clear in the case of Siemens.|.|.|. We had a shift from pro-British, not to anti-British, but to pro-German. (Former Siemens North Tyneside senior manager, interview 2000).

Atmel: Corporate Contexts and the Sixth Semiconductor Industry Cycle In stark contrast to the long and rich history of Siemens, Atmel was founded in 1984 and represents one of the most successful companies in the history of the Silicon Valley phenomenon, achieving revenues in excess of $1 billion in its first 15 years and establishing itself as a “cash-rich Wall Street darling” (Manners 1996). Atmel developed as a specialist semiconductor firm operating as both a merchant and foundry producer. In contrast to Siemens, it has steered clear of the mature, cyclical, and mass-commodity DRAM market and instead evolved around leading-edge specialized semiconductor products, ranging from advanced logic products to flash memory and radio frequency integrated circuits. In doing so, Atmel invested heavily in developing intellectual, process, and product capacity across a diverse range of leadingedge niches within the semiconductor market. From 1995 onward, it embarked on a steep trajectory of intellectual property, design, and manufacturing-capacity acquisitions, guided by a philosophy whereby “if we are missing something we will buy a company or a small group” (Atmel CEO, quoted in Manners 1999). Likened to a “takeover tiger prowling chip prey in the fierce pursuit of chip markets” (Manners

63

1996), sustained only by the addition of “a fab every two years” (Atmel CEO, quoted in Manners 1997b), the company has been able to capture particular product and process technologies, place-specific expertise, and significant cost and time savings in comparison to the construction of new fabrication facilities (Atmel CEO, cited in “Amtel to Buy 8 Inch Wafer Fab .|.|.” 1999). Atmel’s investment in the then-mothballed North Tyneside plant can be understood as a further episode of its “growth by acquisition” strategy (Manners 2000a). Whereas Siemens responded to similar economic and technological imperatives during the first round of investment by building new fabrication facilities, Atmel’s investment strategy targeted the acquisition of suitable mothballed semiconductor plants. Placing North Tyneside Within Atmel’s Corporate Strategy Following Atmel’s acquisition, the North Tyneside facility began to occupy a more sophisticated and powerful position within both corporate and industry sociospatial relations. In contrast to Siemens, which focused solely on mass-commodity DRAM, Atmel planned to invest $800 million in equipping the plant to make the company’s full range of products, starting with microcontrollers and subsequently shifting to flash memory to serve an array of markets: “If you want to make DRAM, you should be living in Taiwan. We can make many things at North Tyneside, and that is the best way of guaranteeing the future of the site” (Atmel’s chief financial officer, quoted in Manners 2000c). Despite this initial optimism, however, by 2001 the global semiconductor industry was witnessing its first-ever decline in the flash memory market. Faced with a £31 ($45) million loss and a 23-percent decline in revenues, Atmel announced a global restructuring plan with the loss of 2,500 jobs—26 percent of its workforce (“Atmel Stalls Tyneside Fab as Jobs Go” 2001). For a company with no history of plant closures,

64

ECONOMIC GEOGRAPHY

“the most difficult business climate our industry has ever faced” (Atmel CEO, quoted in Manners 2001), led to the cessation of volume production at Nantes, Grenoble, and Colorado Springs, together with a postponement of manufacturing at Atmel’s Texas plant (Atmel North Tyneside director of human resources, interview 2001). In contrast to the position of the North Tyneside facility during the corporate restructuring of Siemens, despite the scaling back of initial recruitment projections, the North Tyneside plant’s position within the corporate network appeared to strengthen. For Atmel, the plant represented its biggest and “newest” production facility by “far” and subsequently benefited from intracorporate technological transfers from smaller plants that were acquired for proprietary technological capture (Atmel North Tyneside senior manager, interview 2001). Situated within a corporate strategy whereby “the biggest and most cost-effective factories stay open,” both the economies of scale at the plant and the low start-up costs of acquisition privileged a position of relative power within Atmel’s network of sociospatial relations (Atmel North Tyneside senior manager, interview 2000). Unlike Atmel’s older fabrication facilities that were embedded within specific technological trajectories, which would incur heavy costs to revise, the North Tyneside fabrication “shell” has been retooled as a foundry offering a diverse array of product and process technologies and therefore “not placing all our eggs in one basket” (Atmel North Tyneside senior manager, interview 2001). With the development of an on-site research and development capacity and a local management structure, the plant holds a more advanced position in managerial and technical divisions of labor than under the ownership of Siemens. As such, Atmel’s historical willingness to reduce relative capacity and investment in its home economy, at an equivalent or greater rate than overseas, indicates an internal corporate political structure that is less susceptible to the socioinstitutional embeddedness of national capital that was illustrated by Siemens (Yeung 2001).

JANUARY 2007

National and Regional Institutional Contexts This section explores in more detail the constitutive relations among firms, territories, and space in explaining how home and host national institutional contexts matter in mediating MNE investment episodes (Dicken and Malmberg 2001). The analysis begins by looking across the two rounds of investment and identifying the dynamic interactions, both economic and political, that shape the “increasingly murky firm-state nexus at the national and local scales” (Phelps and Fuller 2000, 241). The analysis reveals the important ways in which Siemens’s (dis)investment decisions were enmeshed within home and host political, regulatory, and cultural settings and incentives (Dicken 2000; Yeung 1998). In contrast, Atmel’s investment appears to have been shaped less by territorial embeddedness within home-based contexts than by the place-specific assets provided by the legacy of the Siemens investment. However, the final section of empirical analysis reveals the limited powers of host regions relative to MNEs in generating replacement rounds of investment in the aftermath of closure (Coe et al. 2004; McLeod 2001). Placing MNE (Dis)Investment Decisions in Home and Host-Nation Settings The example of Siemens Semiconductors in the North East of England illustrates the significance of the dynamic relations between firms and states in regulating, shaping, and molding patterns of (dis)investment (MacKinnon and Phelps 2001b; Dicken 2000, 1998). The findings illustrate the importance of the British state’s distinctive national regulatory regime and neoliberal approach to FDI as a prime factor in the development of the Siemens investment (Dicken 2000; Hudson 2000). In particular, the United Kingdom’s competitive labor and social costs, low corporate taxation, low cost of utilities, favorable exchange rate, and deregulated environment were

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

compared favorably to the highly regulated practices and structures within the company’s home nation:3 “over the last 10 to 15 years, Britain has become a more pro-business environment (and Siemens) has capitalised on the opportunities now available for foreign companies who set up in Britain” (Siemens U.K. chief executive, cited in Mainwaring 1996). At the same time, Siemens’s investment was used by the ailing U.K. Conservative government to legitimate its long-standing liberalization of regulatory environments as a form of industrial policy. As the deputy prime minister stated: The chief executive of Siemens has praised the pro-business environment that exists here. From every quarter of the globe, the world’s greatest companies are pouring money into the British economy .|.|. to bolster tomorrow’s industries. Siemens has invested £1.1 ($1.9) billion in the North East. (Hansard 1996)

Consequently, Siemens illustrated the power of MNEs to cultivate inward investment “locational tournaments” (Mytelka 2000) and to converge national regulatory regimes. Despite suggestions that governmental support packages amounted only to “petty cash” within the investment decision (former Siemens U.K. chief executive, cited in Munchau, Taylor, and Tighe 1995, 1), Siemens initially rejected the United Kingdom’s first offer of financial assistance, using comparisons with other potential host nations to garner a U.K. package in parity with the Austrian bid of £120 ($173) million (Loewendahl 2000).4 Even so, the 3 For example, “UK labour costs are 50 percent cheaper than in Germany, and direct employment costs represent five to seven percent of the fab’s running costs and indirect employment about 15 to 18 percent” (Siemens Semiconductors senior director, quoted in Manners 1997a). 4 The official aid package totaled £64.26 ($93) million worth of U.K. assistance—to be delivered at the national, regional, and local levels. However, reports suggested that “other benefits” would eventually deliver a package worth around £120 ($173) million to Siemens (Loewendahl 2000).

65

last-minute intensive lobbying activities of both the United Kingdom’s prime minister and the deputy prime minister were necessary to sway the investment away from Austria, indicating that “a high degree of political persuasion at very senior levels encouraged Siemens to set up in the North East” (former Siemens North Tyneside personnel and training manager, interview 2000; see also Munchau, Taylor, and Tighe 1995). Even though Siemens’s investment in the United Kingdom allegedly “symbolised a prime example of the company’s globalisation of manufacturing” (Siemens 1997a), the tensions between capital mobility and territoriality would emerge in the context of closure. In this regard, the comparative developmental trajectories of Siemens’s Dresden fabrication facility and the North Tyneside plant—the only two wholly owned large-scale fabrication plants at the time of closure—appear instructive. Siemens’s Dresden investment was heavily influenced by the availability of large-scale funding and assistance in the aftermath of German reunification and represented the outcome of a long-term understanding between Siemens and the German government (director of the Dresden Council Economic Development, interview 1999; Osborne 1998). From the point of view of the North Tyneside management: The grant thing seemed to be absolutely predominant for making business decisions in favour of Dresden, we [North Tyneside] outperformed Dresden on a number of manufacturing parameters.|.|.|. We had established that we were at least as good as Dresden. (Former personnel director of Siemens North Tyneside, interview 2000)

In particular, the close professional and social networks between Siemens management and the German political elite further entangled territory and investment flows (Loewendahl 2000). Therefore, high rates of national unemployment and an impending general election suggested that any moves to close the Dresden plant would have deep political ramifications. Indeed, Tomaney

66

ECONOMIC GEOGRAPHY

(1998, 8) claimed that the rapid disinvestment of the North Tyneside plant “would simply not have been possible in ‘over-regulated’ Germany. And the domestic and political costs would have been incalculable.|.|.|. German over-regulation has saved jobs in Dresden and cost them in North Tyneside.” The asymmetrical relations of power between mobile capital and deregulated FDI-led institutional arrangements was further illustrated by the political furor that surrounded Siemens’s apparent unwillingness to repay “recoverable” public-sector grants and assistance in light of its short-lived investment (Wighton 1998). Tensions surrounded the government’s legitimate reclamation of a £18 ($28) million Regional Selective Assistance (RSA) grant to Siemens in 1995 and its hesitance to “[say] to foreign investors [that] we are going to give you a different deal from what you get in other countries” (Prime Minister Tony Blair, quoted in Wighton 1998, 10). However, following Siemens’s decision to invest £155 ($251) million at its Corbeil-Essonnes facility (France), the government realized that Siemens was no longer meaningfully marketing the dormant North Tyneside plant and legitimated the reclamation of the RSA grant in its entirety (Hansard 1999). In the case of Atmel, while the U.K.’s probusiness institutional and regulatory environment was an important factor of location, it was less fundamental than for Siemens. Although Atmel recognized the importance of the U.K. government’s £27.8 ($40.3) million RSA grant to assist in capital expenditures at the North Tyneside plant, this grant nevertheless was “icing on the cake” in terms of the investment decision (Atmel North Tyneside director of human resources, interview 2000). In this sense, the nested influences of factors that were drawn from the supranational and local scales appear to have been important. At the supranational scale, no other Silicon Valley-based company has expanded in Europe as rapidly as Atmel. By 2000, over 50 percent of the company’s workforce (3,500) was European based, contributing 60 percent of the company’s manufacturing

JANUARY 2007

capacity (Manners 2000c; Lineback 1999). Atmel identified four influential factors of location within the European institutional environment: market base, highly skilled engineering labor, financial incentives, and low-cost production base (Atmel’s marketing vice president, cited in Lineback 1999). Moreover, following Atmel’s recent pursuit of a listing on the Paris stock exchange to provide the company with a “more European feel” (Atmel spokesperson, cited in Manners 2000b), the U.S. company is developing a form of territorial embeddedness at the European scale. At the local scale, Atmel’s strategy of targeting specific plants and infrastructures for acquisition creates a “business need” for subnational and place-specific factors of location relative to the importance of national-level regulatory regimes (Atmel North Tyneside senior manager, interview 2000). Realizing “Regional Assets”? (Dis)Investing MNEs and the Host Region The research findings on the two rounds of semiconductor inward investment on North Tyneside lend considerable support to the ongoing claims to situate local and regional forms of institutional intervention within a multiscalar and multiagency political and economic context (MacKinnon and Phelps 2001b). Despite the regional institutional infrastructure of the North East possessing a strong track record in capturing inward investment during the 1980s and 1990s, the institutional response to Siemens’s investment inquiry was structured through a “top-down” site-selection framework that was controlled centrally at the national level by English Partnerships, a centrally administered institution dealing with real estate, site selection, reclamation, and development. Nevertheless, drawing upon the regional economic development agency’s (then the Northern Development Company, NDC’s) previous experience in managing large-scale inward investment projects, the host region was able to coordinate a timely process of site development that was partic-

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

ularly suited to the temporal pressures of investment in the semiconductor industry: Of course, Siemens were in a hurry; there was a window of opportunity, which they wanted to capitalise on, and they were very, very keen to move as fast as they possibly could. The speed at which we were able to deal with the site was quite astonishing. I think Siemens recognised that. (Former NDC chief executive, interview 2002)

Even so, the activities of the NDC and local agencies remained nested within a multiscalar institutional response, relying heavily on the commitment of the central government, particularly through financial assistance, tax incentives, and the subversion of local planning procedures (Tewdwr-Jones and Phelps 2000, 434). In the aftermath of the closure of the Siemens plant, North Tyneside possessed a series of infrastructural “regional assets,” through which local and regional institutions could attempt to attract new rounds of investment (Coe et al. 2004). Of fundamental importance was the interaction between the “place-specific” legacy of plant and infrastructures within the North East and the emerging production strategy in the semiconductor industry of targeting the acquisition of specialized infrastructures and mothballed fabrication capacity (Massey 1995). The scale, process technologies, pristine condition, and significantly reduced costs that were associated with the mothballed plant proved to be important “place”specific attractions for Atmel: The fact the North Tyneside plant was available provided us with a great opportunity. The company is going for growth of 10 per cent and needs more capacity. By moving to a capacity that is already built will give us a 2 year advantage than if we had to build from scratch. (Atmel CEO, quoted in Wilson and McKay 2000, 2)

However, the power and agency of local and regional institutions to realize such “regional assets” was shaped by the nationstate and, most important, the disinvesting capital—Siemens. On the day of the

67

announcement of the closure of the Siemens plant, the central government instigated a multilevel and multiagency task force that would do “everything we can to find an alternative owner [and] a solution that will guarantee the long term future of this world class facility” (DTI 1998). Reflecting the political sensitivity of the Siemens closure— enhanced by the plant’s location in the secretary for trade and industry’s local constituency—the regionally based task force relied upon central government actors and resources to provide “invaluable support” in the efforts leading to the eventual sale of the facility (CEO of Siemens U.K., cited in DTI 2000; Pike 2002).5 However, irrespective of the national and local political and institutional pressure within the host economy, the activities of the task force faced “obstacles occurring elsewhere in the system,” bounded within the necessary relations of Siemens’s ownership of the plant and corporate strategies that emerged from Munich (former NDC chief executive, interview 1999). Illustrated by the company’s unexplained refusal to consider a proposed local management buyout, Siemens was seen to adopt “a very curious approach; .|.|. we were never quite sure whether they wanted us to find a buyer or not” (former Siemens North Tyneside personnel director, interview 2000). Local management believed that the company wanted to avoid selling the plant to any potential competitor of the newly formed spinoff Infineon: Siemens .|.|. [was] placing too many restrictions on the potential buyers.|.|.|. There were quite a number of interested parties that got to a reasonably detailed level of negotiation. Then, as rumour has it, these fell through because Siemens couldn’t be satisfied one way 5 For example, the central government was integral in promoting an ambitious, albeit unsuccessful, £380 ($550) million local management buyout (“Hopes Fade” 1999; former Siemens North Tyneside personnel director, interview 2000).

68

ECONOMIC GEOGRAPHY or another. (Former Siemens training and personnel manager, interview 2000)

Despite decommissioning and removal of the process technology within the plant, Siemens maintained the facility in pristine condition during the closure period and retained an experienced and skilled local management team to assist in the marketing and subsequent handover (Atmel North Tyneside senior manager, interview 2000). At considerable cost to Siemens, this strategy was clearly important in Atmel’s decision to acquire the place-based facility, “wheel its equipment through the door,” and therefore reduce the lead time to production (Wilson 2000, 2). Of further significance, Atmel’s acquisition was part of a larger lucrative framework purchasing agreement with Siemens, leading to Siemens’s purchase of $1.5 billion of Atmel’s products over four years (Atmel 2000). This agreement helped Siemens “secure future availability of flash memory chips” (Siemens 2000) and provided Atmel with a guaranteed medium-term supply deal, helping the company to “grow our design and manufacturing presence within Europe” (Atmel CEO, quoted in Atmel 2000). Whether Atmel and Siemens would have agreed to a flash-memory supply agreement in the absence of the North Tyneside plant’s acquisition, or vice versa, is open to conjecture. These findings therefore further contribute to the apparent asymmetry of power between inward investors and regional institutional support structures, especially in shaping, mediating, and contesting investment and disinvestment decisions (MacKinnon and Phelps 2001a).

Conclusion This article has extended economic geography research on FDI investment episodes by developing a historically grounded understanding of the socioinstitutional relations that shape and constrain economic actions during successive “moments” of MNE (dis)investment in host regions (Hess and Yeung 2006). By integrating socioinstitutional analyses across different rounds of

JANUARY 2007

investment, the article has advanced recent contributions to studies of GPN that have focused on the historical contexts that shape individual and discrete MNE investment episodes and networks. Connecting with the relational turn in economic geography, the conceptual framework and empirical evidence have demonstrated the complex ways in which MNE investment episodes are produced through economic-political processes that are sensitive to contextuality, path dependency, and contingency (Bathelt 2006; Phelps and Waley 2004). In particular, the case studies of Siemens and Atmel in the North East of England provide a compelling example of the ways in which an historically grounded, multiscalar, and socioinstitutional understanding of investment episodes reveals how seemingly temporal— even cyclical—economic imperatives are mediated and woven across a diverse array of institutional contexts and settings to create varied investment outcomes over time, across space, and in place (Hess and Yeung 2006; Martin 2000). Centered on the interrelations between corporations and the contexts of home and host economies, the theoretical perspectives presented here offer a series of findings for the conceptualization of MNE investment episodes and the development of peripheral regions. First, the cases of Siemens and Atmel restate the pivotal agency of the corporation as an actor in connecting, mediating, and producing processes of economic and political power over time, across space, and in place (Phelps and Whaley 2004; Taylor and Ashiem 2001; O’Neill 2003). This article has emphasized the need for economic geography analyses to open up the black box of the firm and explore the complex ways in which corporations produce multiple, dynamic, and contested rationalities in response to similar economic imperatives and competitive pressures. Using an historically grounded approach, the findings explain Siemens’s and Atmel’s different responses to the dynamics of the semiconductor industry as moments that were molded within the longer-term path-dependent institutional and geographic evolutions

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

of both corporations (Pike 2005; Clark 1994). More specifically, contributing to a more “pluralistic industrial geography,” the case studies revealed how each corporate investment decision evolved in response to particular multiscalar institutional settings and incentives and specific social, cultural, and political contexts both within and external to the corporation (Yeung 2001, 293; see also Schoenberger 2000; O’Neill 2003). Second, the article has illustrated the importance of “placing firms” and “firming places” (Dicken 2000) within economic geography analyses of MNE (dis)investment episodes. In terms of placing firms and corporate strategies, Siemens’ embeddedness within the distinctive assemblages of institutions, politics, and regulations in its home nation helped shape the nature of internal corporate politics, place-based identities, and power relations that directed the path-dependent character of the North Tyneside investment (Schoenberger 1997; Pauly and Reich 1997; Yeung 1998). In contrast, Atmel’s historical willingness to reduce relative capacity and investment in its home economy at an equivalent or greater rate than overseas reflects a pro-acquisition growth strategy that was shaped more by U.S. capital markets and less by the socioinstitutional embeddedness of national capital (Yeung 2001). In terms of firming places, both Siemens and Atmel illustrated the complex and dynamic ways in which MNEs use space and place. Integrating the investment episodes revealed the sustained asymmetry of power between MNEs and host economies that ceded political leverage through the development of “low cost to enter–low cost to exit” regulatory environments (MacKinnon and Phelps 2001b; Phelps and Raines 2003). The case of Siemens highlights how the power geometries within intracorporate managerial, technical, and divisions of labor both structure and are structured by MNEs’ limited functional and political integration between home and host institutional environments (Yeung 2005; Massey 1995; Phelps and Fuller 2000). However, North Tyneside’s more advanced position within both corpo-

69

rate and industry divisions of labor under Atmel serves to restate the contingency of corporate agency when peripheral host regions are inserted in the historic and geographic-economic evolution of firms (Bathelt 2006; Birkinshaw and Hood 1998). Third, if peripheral regions—already weakly positioned within intracorporate spatial divisions of labor—are to respond to the shortening of the life cycles of FDI projects, then host economies require responsive policies to identify and integrate “strategic couplings” between regional assets that are created by one round of investment and the strategic needs of future rounds of investment (Massey 1995; Coe et al. 2004). The case of Siemens represents a specific instance in which the power of regional institutions to realize the legacy of “regional assets” was constrained by the ongoing ownership and external control of plant and infrastructure in the postclosure period. These findings provide further conceptual and policy-related challenges for research that explores the potential agency of localized points of resistance (local affiliates, institutions, and communities) in moderating the powers and investment decisions of MNEs within host-region economies (Phelps and Waley 2004; Coe et al. 2004).

References Amin, A. 1998. Globalisation and regional development: A relational perspective. Competition and Change 3:145–65. Amin, A., and Thrift, N. 1994. Living in the global. In Globalization, institutions and regional development, ed. A. Amin and N. Thrift, 1–22. Oxford, U.K.: Oxford University Press. Amin, A.; Bradley, D.; Gentle, C.; Howells, J.; and Tomaney, J. 1994. Regional incentives and the quality of mobile investment in the less favoured regions of the EC. Progress in Planning 41:1–112. Angel, D. P. 1994. Restructuring for innovation: The remaking of the U.S. semiconductor industry. New York: Guilford Press. Atkins R. 1998. Munich office sees closure as logical sacrifice. Financial Times, 1 August, 5.

70

ECONOMIC GEOGRAPHY

Atmel. 2000. Atmel signs one and half billion dollar, four year supply agreement. Atmel press release, San Jose, Calif., 19 September. Atmel stalls Tyneside fab as jobs go. 2001. Electronics Weekly, 1 October. Available online: http://www.electronicsweekly.com/ Articles/2001/10/01/22526/Atmel+stalls+ Tyneside+fab+as+jobs+go.htm Atmel to buy 8 inch wafer fab from Hitachi in Texas. 1999. Semiconductor Business News, 1 November. Available online: http://www. eetimes.com/news/semi/showArticle.jhtml? articleID=2905880 Barnes, T. J. 1999. Industrial geography, institutional economics and Innis. In The new industrial geography: Regions, regulation and institutions, ed. T. J. Barnes and M. Gertler, 1–22. London: Routledge. Bathelt, H. 2006. Geographies of production: Growth regimes in spatial perspective 3: Toward a relational view of economic action and policy. Progress in Human Geography 30:223–36. Bathelt, H., and Gertler, M. S. 2005. The German variety of capitalism: Forces and dynamics of evolutionary change. Economic Geography 81:1–9. Birkinshaw, J. 1999. Multinational corporate strategy and organisation: An internal market perspective. In The globalisation of multinational enterprise activity and economic development, ed. N. Hood and S. Young, 55–79. Oxford, U.K.: Oxford University Press. Birkinshaw, J., and Hood, N. 1998. Multinational subsidiary evolution: Capability and charter change in foreign-owned subsidiary companies. Academy of Management Review 23:773–95. Burgess, R. 1984. In the field: An introduction to field research. London: Allen and Unwin. Cane, A., and Jones, S. 1998. Siemens to close North Tyneside plant. Financial Times, 1 August, 1. Clark, G. L. 1994. Strategy and structure: Corporate restructuring and the scope and characteristics of sunk costs. Environment and Planning A 26:9–32. Coe, N. M.; Hess, M.; Yeung, H.W.-c.; Dicken, P.; and Henderson, J. 2004. “Globalising” regional development: A global production networks perspective. Transactions of Institute of British Geographers, NS 29:468–84. Dawley, S. 2003 Rounds of investment and peripheral region development: The case of the semiconductor fabrication industry in the North East region of England 1989–2003. Unpublished Ph.D. thesis, Department of

JANUARY 2007

Geography, University of Newcastle upon Tyne, Newcastle, England. ———. Forthcoming. Making labour market geographies: Volatile “flagship” inward investment and peripheral regions. Environment and Planning A. Department of Trade and Industry (DTI). 1998. Joint task force on future of Siemens plant announced. Department of Trade and Industry, London. Press Release P/98/625, 31 July. ———. 2000. Byers announces £27.8 million grant to return microchip production to North Tyneside. Department of Trade and Industry, London. Press release P/2000/631, 18 September. Dicken, P. 1998. Global Shift, 3rd ed. London: Paul Chapman. ———. 2000. Places and flows: Situating international investment. In The Oxford handbook of economic geography, ed. G. L. Clark, M. P. Feldman, and M. S. Gertler, 275–92. Oxford, U.K.: Oxford University Press. ———. 2003. “Placing” firms: Grounding the debate on the “global” corporation. In Remaking the global economy: Economicgeographical perspective, ed. J. Peck and H. W.-c. Yeung, 27–45. London: Sage Publications. Dicken, P., and Malmberg, A. 2001. Firms in territories: A relational perspective. Economic Geography 77:345–63. Dicken, P., and Thrift, N. 1992. The organisation of production and the production of organisation: Why business enterprises matter in the study of geographical industrialisation. Transactions of the Institute of British Geographers 17:279–91. Gertler, M. 2001. Best practice? Geography, learning and the institutional limits to strong convergence. Journal of Economic Geography, 1:5–26. Goodwin, M. 2004. Recovering the future: A postdisciplinary perspective on geography and political economy. In Envisioning human geographies, ed. P. Cloke, P. Crang, and M. Goodwin, 65–80. London: Arnold. Hansard (House of Commons Debates). 1996. Column 820, 3 December. ———. 1999. Column 599, 29 April. Harvey, D. 1996. Justice, nature and the geography of difference. Oxford, U.K.: Blackwell. Heenan, D., and Perlmutter, H. 1979. Multinational organisation development. Reading, Mass.: Addison-Wesley. Henderson, J.; Dicken, P.; Hess, M.; Coe, N.; and Yeung. H. W.-c. 2002. Global produc-

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

tion networks and the analysis of economic development. Review of International Political Economy 9:36–64. Hess, M., and Yeung, H. W-c. 2006. Whither global production networks in economic geography? Past, present and future. Environment and Planning A 38:1193–204. Hopes fade. 1999. The Journal, 16 March, 4. Horrendous profits warning. 1998. Financial Times, 17 July, 44. Hudson, R. 2000. Production, places and environment: Changing perspectives in economic geography. Harlow, U.K.: Prentice Hall. ———. 2001. Producing places. New York: Guilford Press. Infineon reduces deficit.1999. Semiconductor Business News, 17 May. Jessop, B. 1999. Reflections on globalization and its (il)logics. In Globalisation and the Asia Pacific: Contested territories, ed. K. Olds, P. F. Kelley, P. Dicken, and H. W.-c. Yeung, 19–38. London: Routledge. Jones, J. 1998. Now we are back to square one. Northern Echo, 1 August, 4. Jones, C.; Jones, S.; Jowit, J.; and Wighton, D. 1998. Fujitsu blames UK closure on glut in microchips. Financial Times, 5 September, 2–3. Leckie, R. 2001. The semiconductor cycle—Here we go again. Infrastructure. Special report sponsored by Scottish Enterprise. Infrastructure, Irving, Texas. Lineback, J. R. 1999. Europe may not be building fabs, but industry expansion is still healthy. Semiconductor Business News, 15 December. Available online: http://www.eetimes.com/ news/semi/showArticle.jhtml?articleID= 10817460 Linford, P. 1999. Siemens drops bombshell for prime minister: Policies “a factor” in facility closure. The Journal, 1 May, 7. Loewendahl, H. B. 2000. Bargaining with multinationals: The investment of Siemens and Nissan in North-East England. Unpublished Ph.D. thesis, University of Birmingham, Birmingham, U.K. MacKinnon, D., and Phelps, N. A. 2001a. Devolution and the territorial politics of foreign direct investment. Political Geography 20:353–79. ———. 2001b. Regional governance and foreign direct investment: The dynamics of institutional change in Wales and North East England. Geoforum 32:255–69. MacLeod, G. 2001. Beyond soft institutionalism: Accumulation, regulation and their geograph-

71

ical fixes. Environment and Planning A 33:1145–67. Mainwaring, J. 1996. Innovation is the way. Electronics Weekly, 10 July. Available online: http://www.electronicsweekly.com/Articles/ 1996/07/10/1528/Innovation+is+the+way.htm Manners, D. 1996. Takeover tigers prowl chip prey. Electronics Weekly, 3 July. Available online: http://www.electronicsweekly.com/ Articles/1996/07/03/643/Takeover+tigers+ prowl+chip+prey.htm ———. 1997a. Siemens put SMIF chips into Tyneside. Electronics Weekly, 5 March. Available online: http://www. electronicsweekly.com/Articles/1997/03/05/304 9/Siemens+puts+SMIF+chips+into+Tyneside. htm ———. 1997b. Head for heights. Electronics Weekly, 22 October. Available online: http://www.electronicsweekly.com/Articles/ 1997/10/22/5157/Head+for+heights.htm ———. 1998a. I’m very disappointed by the actions of the EU.|.|.|. Electronics Weekly, 12 August. Available online: http://www. electronicsweekly.com/articles/ 1998/08/12/8358/145Im+very+disappointed+ by+the+actions+of+the+EU.htm ———. 1998b. DRAM: Korea cuts supply. Electronics Weekly, 5 October. Available online: http://www.electronicsweekly.com/ articles/1998/06/10/7652/DRAM+Korea+cuts +supply+Texas+Instruments+in+talks.htm ———. 1999. A family affair. Electronics Weekly, 4 August. Available online: http:// www.electronicsweekly.com/Articles/1999/ 08/04/12430/A+family+affair.htm ———. 2000a. Intel buys old fab shell to avert flash shortage. Electronics Weekly, 9 February. Available online: http://www.electronicsweekly. com/Articles/2000/02/09/14682/Intel+buys+ old+fab+shell+to+avert+flash+shortage.htm ———. 2000b. Atmel feeling European with French listing and expansion. Electronics Weekly, 29 March. Available online: http://www.electronicsweekly.com/Articles/ 2000/03/29/15472/Atmel+feeling+European+ with+French+listing+and+expansion.htm ———. 2000c. A tale of two fabs reborn in the UK. Electronics Weekly, 27 September. Available online: http://www. electronicsweekly.com/Articles/2000/09/27/178 60/A+tale+of+two+fabs+reborn+in+the+UK. htm ———. 2001. Battling DRAM suppliers refuse to damp down output. Electronics Weekly, 17 October. Available online: http://www. electronicsweekly.com/Articles/2001/10/17/227

72

ECONOMIC GEOGRAPHY

18/Battling+DRAM+suppliers+refuse+to+ damp+down+output.htm Markusen, A. 1985. Profit cycles, oligopoly, and regional development. Cambridge, Mass.: MIT Press. Martin, R. 2000. Institutional approaches in economic geography. In Companion to economic geography, ed. T. Barnes and E. Sheppard, 77–94. Oxford, U.K.: Blackwell. Massey, D. 1995. Spatial divisions of labour: Social structures and the geography of production, 2d ed. Basingstoke, U.K.: Macmillan Press. McClean, B. 2001. 2001 IC industry at the crossroads. Semiconductor International, 1 January. Available online: http://www.reedelectronics.com/semiconductor/article/CA5779 3?pubdate=1%2F1%2F2001 McClean, B.; Lam, D.; and Kanz, J. 1998. The semiconductor market upturn: Surprise, surprise, surprise. Fabtech, August. Morgan, K., and Sayer, A. 1988. Microcircuits of capital: “Sunrise” industry and uneven development. Cambridge, U.K.: Polity Press. Munchau, W.; Taylor, P.; and Tighe, C. 1995. Siemens to build £1bn microchip plant in Britain. Financial Times, 5 August, 1. Mytelka, L. K. 2000. Locational tournaments for FDI: Inward investment into Europe in a global world. In The globalization of multinational enterprise activity and economic development, ed. N. Hood and S. Young, 278–302. Basingstoke, U.K.: Macmillan Press. North, D. 1990. Institutions, institutional change and economic performance. Cambridge, U.K.: Cambridge University Press. O’Neill, P. 2003. Where is the corporation in the geographical world? Progress in Human Geography 27:677–80. Osborne, M. 1998. Thanks for the memory: The closure of a semiconductor fab. Fabtech, August. Pauly, L. W., and Reich, S. 1997. National structures and multinational corporate behaviour: Enduring differences in the age of globalisation. International Organisation 51:1–30. Peck, J. 2000. Doing regulation. In The Oxford handbook of economic geography, ed. G. Clark, M. Feldman, and M. Gertler, 61–80. Oxford, U.K.: Oxford University Press. ———. 2002. Political economies of scale: Fast policy, interscalar relations, and neoliberal workfare. Economic Geography 78:331–60. Peck, J., and Yeung, H. W.-c., eds. 2003 Remaking the global economy: Economicgeographical perspectives. London: Sage.

JANUARY 2007

Phelps, N. A., and Fuller, C. 2000. Multinationals, intracorporate competition, and regional development. Economic Geography 76:224–43. Phelps, N. A.; MacKinnon, D.; Stone, I.; and Braidford, P. 2003. Embedding the multinationals? Institutions and the development of overseas manufacturing affiliates in Wales and North East England. Regional Studies 37:27–40. Phelps, N. A., and Raines, P. 2003. Introduction. In The new competition for inward investment: Companies, institutions and territorial development, ed. N. A. Phelps and P. Raines, 1–9. Cheltenham, U.K.: Edward Elgar. Phelps, N. A., and Waley, P. 2004. Capital versus the districts: A tale of one multinational company’s attempt to disembed itself. Economic Geography 80:191–215. Pike, A. 2002. Task forces and the organisation of economic development: The case of the North East region of England. Environment and Planning C 20:717–39. ———. 2005. Building a political economy of closure: The case of R&DCo in North East England. Antipode 37:93–115. Pike, A.; Rodriguez-Pose, A.; and Tomaney, J. 2006. Local and regional development. London: Routledge. Robertson, J. 1998a. Siemens’ big gamble in DRAMs backfires: Remedial action sought. Semiconductor Business News, 23 July. ———. 1998b. How Siemens IC unit will reinvent itself. Semiconductor Business News, 15 November. Romanelli, A. 2002. 2001: What went wrong? Electronic News, 24 January. Available online: http://www.edn.com/article/CA192899.html? text=what+went+wrong Sayer, A. 1985. Industry and space: A sympathetic critique of radical research. Society and Space 3:3–30. ———. 1986. Industrial location on a world scale: The case of the semiconductor industry. In Production, work and territory, ed. A. J. Scott and M. Storper, 107–24. Boston: Allen and Unwin. ———. 2000. Realism and social science. London: Sage. Schoenberger, E. 1997. The cultural crisis of the firm. Oxford, U.K.: Blackwell. ———. 2000. The management of time and space. In The Oxford handbook of economic geography, ed. G. L. Clark, M. P. Feldman, and M. S. Gertler, 317–33. Oxford: U.K.: Oxford University Press.

VOL. 83 NO. 1

SEMICONDUCTORS IN THE NORTH EAST OF ENGLAND

Siemens. 1997a. North Tyneside wafer fab to play a major role in logic offensive by Siemens. Siemens press release, 25 February. ———. 1997b. Products and systems. Siemens corporate literature. Munich. ———. 1999. Facts and figures 1999. Siemens corporate literature. Munich. ———. 2000. Atmel corporation takes over Siemens’ North Tyneside fab. Siemens press release, 18 September. Siemens Chief Executive Officer. 1998. Press conference, 4 November. Siemens U.K. Chief Executive Officer. 1998. Annual press luncheon, 9 December. Smith, A.; Rainnie, A.; Dunford, M.; Hardy, J.; Hudson, R.; and Sadler, D. 2002. Networks of value, commodities and regions: Reworking divisions of labour in macro-regional economies. Progress in Human Geography 26:41–63. Stalk, G. R., and Hout, T. 1990. Competing against time: How time-based competition is reshaping global markets. New York: Free Press. Sunley, P. 1991. Game theories? Spatial divisions of labour and rational choice Marxism. Antipode 23:229–39. Taylor, M. 1986. The product-cycle model: A critique. Environment and Planning A 18:751–61. Taylor, P., and Asheim, B. 2001. The concept of the firm in economic geography. Economic Geography 77:315–28. Tewdwr-Jones, M., and Phelps, N. A. 2000. Levelling the uneven playing field: Inward investment, interregional rivalry and the planning system. Regional Studies 34:429–40. Tomaney, J. 1998. When the chips are down. The Journal, 12 August, 10. van Egeraat, C., and Jacobson, D. 2004. The rise and demise of the Irish and Scottish computer hardware industry. European Planning Studies 12:809–34. ———. 2005. Geography of production linkages in the Irish and Scottish microcomputer

73

industry: The role of logistics. Economic Geography 81:283–304. Vernon, R. 1979. The product cycle in a new international environment. Oxford Bulletin of Economics and Statistics 41:255–67. Walker, R. 1985. Technological determination and determinism: Industrial growth and location. In High technology, space, and society, ed. M. Castells, 226–84. Beverly Hills, Calif.: Sage. Wighton, D. 1998. Renewed attack on Blair over failed inward investments. Financial Times, 11 September, 10. Wilson, B. 2000. US giant set to sign deal for Siemens. The Journal, 16 September, 2. Wilson, B., and McKay, N. 2000. New dawn for Ex-Siemens plant. The Journal, 19 September, 2. Wood, A., and Valler, D. 2001. Turn again? Rethinking institutions and the governance of local and regional economies. Environment and Planning A 33:1139–45. Yeung, H. W.-c. 1998. Capital, state and space: Contesting the borderless world. Transactions of the Institute of British Geographers 23:291–309. ———. 2000. Organising “the firm” in industrial geography I: Networks, institutions and regional development. Progress in Human Geography 24:301–15. ———. 2001. Regulating “the firm” and sociocultural practices in industrial geography II. Progress in Human Geography 25:293–302. ———. 2003. Practicing new economic geographies: A methodological examination. Annals of the Association of American Geographers 93:442–62. ———. 2005. Rethinking relational economic geography. Transactions of the Institute of British Geographers, NS 30:37–51. ———. 2006. From followers to market leaders: Asian electronics firms in the global economy. Asia Pacific Viewpoint Lecture presented at the International Geographical Union Regional Congress, Brisbane, Australia, 3–7 July.