The Detroit of the New Economy: The Changing Workplace, Manufacturing Workers and the Labor Movement in Silicon Valley

Boy Lüthje

Paper presented at panel "Labor in the Silicon Valley Economy"

Annual Convention American Historical Association

San Francisco, California, January 6, 2002

ABSTRACT

The paper explores the new patterns of manufacturing organization and work in the information technology industry of California's Silicon Valley. It provides an account of the profound transformations in the production process through the 1980s and 1990s, characterized as a transition from a regime of "bloody Taylorism" to a sophisticated system of network-centric mass production. From a theoretical approach based on French regulation theory, critical U.S. industrial geography, and German industrial sociology the paper argues that the Silicon Valley of the 1990s has become a place of prime importance for the development of new manufacturing arrangements in the information technology industry. The article discusses the implications for Silicon Valley's manufacturing workforce, which is mainly made up of women and non-white immigrants, and the challenges for the labor movement.

The paper summarize the results of a comprehensive study which is published in German by the Institut für Sozialforschung as "Standort Silicon Valley: Ökonomie und Politik der vernetzten Massenproduktion. Frankfurt/New York: Campus-Verlag. 2001." A publication of the study in English is under consideration with several U.S. publishers. A revised version of this paper, referring to the current crisis in the high-tech industry, will be submitted to the British journal "Historical Materialism".

.01. Introduction

California's "Silicon Valley" has often been portrayed as a role model for the global "information society" (Castells 1996). This perception has been mostly shaped by economists and theorists of industrial policy. For neo-liberals, the region's phenomenal economic success provides a paradigmatic example for the virtues of market-driven technological innovation, based in an environment of start-up entrepreneur-ship (Gilder 1989). Theorists of "flexible specialization" emphasize the cultural and political embeded-ness of Silicon Valley's "networks of innovation" (Saxenian 1994), displaying similar mechanisms of economic coordination as some of the more successful industrial regions in Europe like Italy's Emiglia Romagna or Germany's Baden-Wurttemberg. Adherents of neo-Schumpeterian industrial policy approaches see Silicon Valley computer companies at the center of a new industrial architecture based on tight control over key technology standards and cross-national-production networks - a system of "post-taylorist" and post-Fordist industrial organization for which the brand names of two of its most successful players may offer the name: "Wintelism" (Borrus/Zysman 1997, Borrus 1997).

Absent from the political and the academic debate is the situation of workers, production workers in particular. While the IT-industry and Silicon Valley particular attract all kinds of sociological reflections on the "information society", the sociology of the labor process in this basic industry of modern capital-ism largely remains a "black box". However, the few critical studies on the subject, mostly by theorists of gender (e.g. Hossfeld 1988, 1989, 1995) or industrial geography (Henderson 1989, Storper/Walker 1989, see also Walker/Sayer 1992) provide ample evidence that industrial work in high-tech-centers like Silicon Valley generates only modest wages and living conditions even for the more skilled workers and is heavily typed along ethnic and racial lines (Harrison 1994, Eisenscher 1993, Bacon 1997a). Given the importance of production work in the industry, its global character, and the recent growth of important segments of high-tech-manufacturing in the United States (AEA 1996a and 1997, Sturgeon 1997), it may be worth to examine the emerging patterns of ethnic segmentation and racialization in the context of the new forms of "network-centric-production" in the IT-industry.

The underlying thesis of the following reflections is that Silicon Valley plays a central role in the global production networks of today's IT-industry. As a "hub" for the global R&D- and manufacturing operations of the industry, Silicon Valley is a location where state-of-the art manufacturing know-know in a key industry of its time is generated. Although much of the research literature tends to underrate the significance of Silicon Valley's manufacturing base, the region's strategic role for the creation and coordination of advanced manufacturing know-how may well be compared to places like Detroit in the context of Fordist capitalism earlier in this century. In the radically de-verticalized environment of modern IT-production, however, the recomposition of the working class is taking place in a highly globalized environment, of which the transnational mobility of both "knowledge workers" and industrial workers is an important element.

The emerging mode of industrial organization and control over the labor process can only be understood through a broader analysis of the political economy of industrial restructuring and the sociology of the labor process in the industry. The underlying shift in the "social division of labor" (Sayer/Walker 1992) in the information industry will be discussed in the framework of French "theories of regulation" and current European labor process sociology theories. Referring to the work of authors like Michel Aglietta (1978), Alain Lipietz (1985 and 1987), Benjamin Coriat (1979 and 1990; see also Hübner 1989, Vincent/Negri 1994, Hirsch 1996) as well as to our own research (Esser e.a. 1997, Lüthje 1993;), vertical disintegration and global re-consolidation in the IT-industry will be analyzed as a de- and re-composition of capitalist norms of production and technologies, reflecting the ongoing structural instability of capital accumulation in the industry and beyond. The organization of the labor process in highly "flexibilized" and segmented regional production networks and the corresponding forms of labor policies are to be characterized as a mode of "systemic rationalization" (Bechtle 1993, Schumann e.a. 1994, Aulen-bacher/Siegel 1995) - a concept that refers to the permanent creation of unequal working conditions as epitomized by the concept of "lean production". The new thing, not alone to be learned from Silicon Valley, is that race and ethnicity are constituting elements in the creation of heterogenous conditions of work and control "along the chain" (Sauer/Döhl 1994)

The first section of this paper will briefly explore the historical specifics of Silicon Valley as a "post-Fordist production complex and the role of the large-scale employment of non-white immigrant workers for the emerging regime of production (Burawoy 1985). The second section summarizes the profound economic, technological, and organizational transformations in the IT-industry during the last decade that shaped the industrial structure of Silicon Valley in the 1990s. The third section will take a closer look at the contemporary structure of network-centric-production in the region and discuss the role of ethnic discrimination and non-white immigrant entrepreneurship in this context. The fourth section will trace the shape of the regime of production along the various segments of the local chain of production and discuss the changing patterns of ethnic segmentation in the industrial workforce. The final chapter will touch upon some implications for further research and labor organizing in the IT-industry and the"new U.S. economy" of the 1990s.

.02. Silicon Valley as a post-Fordist production complex

From the vantage point of regulation theory, the Silicon Valley electronics industry represents a para-digmatic example of a post-Fordist system of production. The emergence of the new producers of semiconductors and microcomputers in Santa Clara County during the 1960s and 1970s marked the up-coming of a new norm of production and a rupture with the existing structure of industrial organization in the U.S. electronics industry. The often told story of the rebellion of young engineers and scientists against the established forms of technology development and industry organization, supported by the enormous resources of Northern California's research system, was far more than a cultural phenomenon. In economic terms, it engendered a profound rupture with the prevailing structure of oligopolistic regulation in the U.S. electronics industry and the established patterns of mass production. Under the Fordist model of accumulation, which was centered on the large U.S. domestic market and constantly growing defense spending (Aglietta 1979), the U.S. electronics industry consisted of vertically integrated mass-producers like General Electric, Westinghouse, AT&T/Western Electric, or rapidly growing computer companies like IBM (Lüthje 1993a and b). In the context of the crisis of the post-war Fordist accumulation cycle since the mid-1970s (Davis 1986), the new industry increasingly re-defined the patterns of technology and development in key areas of the electronics sector.

With regard to industry organization, the most visible aspect of this rupture was the character of the new semiconductor companies as "merchant producers" (Ziegler 1991). National Semiconductor, Intel, and others delivered to the"open market" without being tied to captive intra-company markets as prevalent in the older parts of the U.S. electronics industry. This new model of specialization entailed enormous economic risks and instabilities for the new companies only to be matched by continuously successful introduction of new products and the definition of new markets by technological innovations. The highly cyclical character of the industry and its characteristic form of technology competition has its roots in this situation. As microelectronics transformed from "movement to industry" (Siegel 1986), it became apparent that the distinctive feature of the new industry was not the technological character of its products alone. Rather, the industry developed a new model of the regulation of the wage-relation which stood in fundamental contrast to the established patterns of capital accumulation in the industry.

The new regime of production eclipsed the "social equation of Fordist mass production" (Park 1992) which historically had guaranteed relatively stable jobs, wages and benefits in exchange for stability at the shop-floor (Davis 1986). The fundamentals of the "social contract" of the post-war era (Moody 1988) had not only been prevalent in the unionized sectors of the U.S. electronics industry (most pronounced in the regulated telephone industry, Lüthje 1993a and b), but also in the "corporate welfare states" of non-union corporations like IBM. In its early days, the new industry adopted essential parts of this system of regulation, especially the idea of a non-union "corporate community" with "no-lay-off" policies and relatively decent benefits, and combined it with some elements of the new entrepreneurial culture, like the much heralded "open door"/ "no hierarchies" attitudes as developed by the oldest of the native electronics companies, Hewlett-Packard. The "team-spirit" of the new work culture, however, remained largely restricted to engineers and professionals. For production workers, management-labor cooperation did not apply to wages, which were relatively low, and working conditions, characterized by often horrifying conditions of workplace safety and health (especially with regard to the widespread use of toxic materials in semiconductor manufacturing; Eisenscher 1993, La Dou 1984 and 1994).

• While production technologies and"best practices" of manufacturing organization in the new industry remained highly unstable well into the 1980s, the shift away from the "social equation of fordism" was based on three features, each one marking another rupture with established norms of production.
• The widespread employment of women and recent immigrants of Asian and Latin American origin as production workers (Hossfeld 1988). The industry had a policy of hiring workers without previous industry experience, as a means to create a workforce adaptive to the needs of the new industry. Although never admitted, this policy also excluded the hiring of black workers, available from older industries in the San Francisco Bay Area.
• The distinctively global organization of the production process, built upon newly established manufacturing facilities in South East Asia. Other than in the traditional sectors of the U.S. electronics industry with its often very large multinational operations in Europe as well in the Third World, off-shore production (chip assembly and packaging) was not primarily related to local markets but formed an integral part of the production chain of the new industry (Henderson 1989).
• The emergence of a sector of small, low-wage local subcontracting companies (often referred to as "board stuffers"). These companies fulfilled important functions of the local production process, especially related to the emerging PC-industry and other segments of electronic equipment manufacturing located around the new merchant semiconductor complex.

A crucial political precondition for the new regime of production was the scattered structure of unionism in the U.S. post-war electronics industry (Eisenscher 1993). The origins of this situation date back to the post-World-War-II conflicts within the U.S. labor movement. Anti-communism motivated the exclusion of the largest union in the electrical industry, the left-led United Electrical Workers (UE), from the AFL-CIO (Schatz 1983, Rosswurm 1992) and prepared the stage for the subsequent failure of U.S. labor to organize the newly established non-union operations of electronics companies like General Electric or Westinghouse in the U.S. South (Davis 1986, Goldfield 1997). At the same time, most labor organizations were uninterested in organizing immigrant low-wage workers in new high-tech centers like Silicon Valley. Usually justified by the low dues paying potential of these workers, this attitude reflected the racist prejudices against non-white workers in U.S. post-war business unions (Eisenscher 1993, Hossfeld 1995, Bacon 1997a, for a general discussion of the problem: Lüthje/Scherrer 1997).

The absence of industry-wide bargaining structures enabled the electronics industry to develop a system of highly localized regulation of the labor process and labor markets. Whereas wages and compensation policies typically were determined at corporate level, shop-floor practices and recruitment procedures remained local. In this perspective, the regime of production in the newly emerging semiconductor and computer companies in Silicon Valley did not represent a complete break with prevailing industry practices, but rather a variation within the general model of non-union labor relations. For the new merchant chip producers, these conditions provided ample "flexibility" to marry the new practices of production in uncharted technological and economic territory with the prevailing forms of the regulation of the wage relation.

It should be noted, however, that the new regime of production did not remain uncontested. Attempts by the UE in the late 1970s and early 1980s to organize workers for a living wage and better health and safety conditions took root in a number of semiconductor plants. The elimination of the main activists from the plants revealed the anti-union character of the industry. The underlying workplace problems in the new industry remained an issue in the community, particularly with regard to occupational safety and environmental pollution caused by the extensive use of toxics in semiconductor manufacturing. Community activities were an important factor for subsequent changes in the regime of production, forcing the industry to bring under control at least some of its environmentally most damaging manufacturing practices (Eisenscher 1993, Bacon 1997a).

.03. From "bloody Taylorism" to network-centric mass production

The mid-1980s marked a period of structural crisis for the U.S. semiconductor and electronics industry. The world-wide recession in semiconductor production of 1985/86 for the first time showed the limits of growth even of the technologically most advanced product-markets (Drüke 1993) and resulted in heavy job-losses for production workers in Silicon Valley, particularly workers of color (Ong 1990, Sevilla 1992). At the same time, the vulnerability of the merchant semiconductor industry to the competition of vertically integrated foreign multinationals with long-term investment perspectives in new manufacturing equipment and processes became apparent. What was publicly perceived as the "Japanese" and later the"Korean challenge" in microelectronics was to a large extent a crisis of the early post-Fordist model of production in the merchant semiconductor sector, characterized by the broad use of low-wage labor at home and abroad and relatively low and short-term oriented investments in production processes and equipment (Borrus 1988, Florida/Kenney 1990). The subsequent structural changes in the U.S. semi-conductor industry and the IT-sector as a whole entailed a profound transformation of the norms of production.

In theoretical terms, this process, which involved heavy intervention of the U.S. government and the traditional "big players" of the U.S. electronics industry, may be characterized as a transition from the early practices of"bloody taylorism" (Lipietz 1987) towards a model of highly automated network-centric mass-production". The "remaking" of the U.S. microelectronics industry, however, should not be explained in terms of successful corporate and industrial policy strategies alone. It is part of a broad shift in the division of labor and the forms of capital reproduction across the IT-sector, driven by the rapid globalization of norms of technology. In terms of industry structure this process resulted in the creation of new inter-linked industry segments with relatively autonomous norms of production and technology at world-market level.

On the side of the semiconductor industry, the restructuring can be described as a process of respecialization and automation. In the second half of the 1980s, U.S. semiconductor firms abandoned most of the commodity-type business in memory chips, known as DRAMs, and concentrated on the development of more design intensive types of logic chips (microprocessors) and Application Specific Integrated Circuits (ASICs). At the same time, the biggest semiconductor producers heavily automated their design and manufacturing processes through the introduction of new technologies, providing comprehensive control of manufacturing quality. This was accompanied by a profound restructuring of engineering processes, designed to develop several subsequent generations of processors in overlapping time-cycles ("concurrent engineering"). At the shop-floor level, a wide array of quality management practices including teamwork-type work arrangements were introduced (Leachman 1996, Brown 1996). These changes were part of a broader transformation of production arrangements. In the more customer-specific chip-markets so-called "fabless" producers emerged, which contracted the entire manufacturing process to larger or more specialized chip-makers (Saxenian 1989). In a global context, these so-called "foundry agreements" were accompanied by transnational alliances to share the ever-increasing development costs of new chip generations and a refinement of the early "international division of labor" (Henderson 1989, 1994) through a limited technological upgrading of existing off-shore facilities of U.S. corporations in South East Asia. U.S. semiconductor producers also changed the "division of labor" inside its core manufacturing processes, shifting the mass production of chip-wafers to large facilities in other newly emerging U.S. high tech centers in states like Texas, Oregon, Arizona or New Mexico (Angel 1994).

Politics played an important role in this development. The U.S.-Japanese semiconductor trade agreements of the 1980s provided for a certain international stability of semiconductor prices. At the same time, the government and industry-sponsored Sematech program supported a wide array of research initiatives in the field of semiconductor manufacturing technology. The most visible impact of Sematech was the consolidation of the hitherto fragile U.S. semiconductor equipment industry, in which Japanese electronic producers had acquired world-leadership (Stowsky 1987). Through the concertation with large electronics producers U.S. semiconductor equipment firms regained considerable competitive strength and emerged as a new industry segment of its own (Angel 1994, Langlois 1992, Grindley e.a. 1994). Sematech also supports a number of human resource initiatives, related to the "quality revolution" in U.S. semiconductor manufacturing (NOVA 1995). In a broader context, a most important political factor in the restructuring of the U.S. IT-industry was the deregulation of the U.S. telecommunications industry, following the divestiture of AT&T in 1984. It laid the groundwork for the shift towards a radically de-centralized architecture of computer and telecommunications networks and the related hardware products (Bar 1990, Davies 1994), as epitomized in the concept of a "national information infrastructure" pro-posed by the U.S. government since 1992 (Clinton/Gore 1993).

Apart from these strategic efforts, the remarkable resurgence of the U.S. IT-industry in the 1990s has to be explained through the global shifts in the industry, resulting from the crisis of Fordist norms of production, technology and consumption across the electronics sector (Esser e.a. 1997). What has been widely perceived as the "decentralization of information technology" (Ferguson/Morris 1993, Davies 1994) can also be interpreted as the emergence of a whole new structure of the IT industry, characterized by the generalization of vertical disintegration and the commodification of an increasing array of IT products and systems, previously being offered as part of larger computer and communications systems (Ernst/O'Connor 1992). Driven by the accumulation crisis in numerous segments of the IT industry, including the traditional mainframe computer industry (IBM, DEC e.a., Drüke 1992), the recomposition of capital assumes the shape of an ongoing creation of new interrelated industry segments with relatively autonomous norms of production and technology (Lüthje 1997). The so-called "PC-revolution" of the 1980s, in which merchant PC-producers like Apple or Compaq together with Intel and Microsoft in the microprocessor and software field became global industry forces, epitomizes this development. The re-cent explosion of decentralized computer-based mass-communication embedded in the "Internet" (Hart e.a. 1992) engendered the advent of another new industry segment, the so-called "networking equipment industry". Less visible, an increasing number of IT-components is produced on a merchant basis, notably PC disk-drives, today one of the fastest growing segments of the electronics industry.

As the IT industry has almost completed the shift from a reproductive structure based on large-scale vertical integration towards a "confederation" of interrelated layers of hardware products, software systems and the related distributive system, the trend towards increased capital intensity and the resulting problems of capital valorization remains (Esser e.a. 1997). This is particularly true for semiconductor manufacturing, characterized by the enormous increase of investment costs for new plants. The control of the time-cycle of the development and production of new product generations and the related cycle of product prices and equipment depreciation has become the chief problem of manufacturing organization in the industry (most visibly in semiconductors, Borrus/Leachman 1994). As the most successful proponents of the new model of industry organization explain, the interlinkage of technology development and a "high performance organization" of front-line manufacturing activities is crucial for the survival of even the strongest players in the industry (Grove 1996).

This problem is characteristic for all segments of the IT industry today, its regulation beyond the capacities of individual firms or firm-networks. Besides concentration and centralization of capital through mergers and acquisitions within particular industry segments, the permanent reorganization of the"social division of labor" under the "network-centric-production" model has become a dominant mode of regulation for the post-Fordist information industry. As the problem of time-cycle-management has become global in scope, new models of vertical de- and re-integration emerge, characterized as "contract" or "turnkey manufacturing" (Sturgeon 1997). Brand-name producers (OEMs) in the area of computer systems, communications equipment and a wide array of other electronics products increasingly contract out the production of entire systems to specialized "contract manufacturers", in order to reduce cycle-time and investment risks in increasingly expensive production equipment (Curran 1996).

As opposed to traditional forms of subassembly in the electronics industry ("board stuffing"), contract manufacturing is based on capital intensive equipment for printed circuit board production and assembly. In addition, contract manufacturers provide substantial capabilities of design and"supply chain management" (Feitzinger/Lee 1997, Lee 1996) to OEM-customers. With a current growth rate of 25-30% per year and an employment of almost 200.000 (IPC 1997)"Electronics Manufacturing Services" is rapidly evolving into a general manufacturing base for a broad array of electronics-related production, concentrated in net-works of small and mid-sized manufacturing establishments in industrial centers throughout the U.S. with a high proportion of workers of color. Driven simultaneously by the recent large-scale outcontracting of manufacturing activities of traditional electronics producers (also involving defense, aircraft, and auto-mobile electronics) on the one side and specialized "fabless" electronics start-up companies on the other, the convergence of Fordist and post-Fordist norms of production increasingly seems to take place at the level of standardized manufacturing processes.

.04. The transformation of local production networks in the 1980's and 1990's

The development of local production networks in Silicon Valley simultaneously was shaped by and shaped these global trends. The semiconductor crisis of the mid-eighties became the starting point for a profound transformation in the organization of innovation and manufacturing. It brought about a whole new structure of vertical integration and disintegration, which can hardly be described by micro-economic "models" of company networks and external cost economies. The characteristics of this process emerge from its distinctive transnational and global character. As a major center for the de-verticalised IT-industry of the 1990s, Silicon Valley was transformed into a "hub" for global strategies of technology development and manufacturing, embodied in the combination of various vertically disintegrated industry segments with relatively autonomous norms of technology, production, and consumption. As can be stated today, the transformation of local production networks created the base to overcome the chronic weaknesses in moving "from innovation to mass production" (Florida/Kenney 1990). This problem, however, was "solved" neither in a traditional way of vertical integration and consolidation, nor can it alone be explained by the local start-up and innovation culture. Rather, Silicon Valley should be seen as a place where new forms of integration between "innovation" and production emerge, at both firm and inter-firm levels (Voskamp/Wittke 1994).

At the first glance, the transformation of the reproductive structure of capital at the local level may be described as a rapid process of industrial diversification. As the local press puts it, the Silicon Valley of the 1990s is "not just chips and computers anymore" (SJMN 04/22/1996). The downsizing in semiconductor manufacturing has been offset by the growth of new industry segments, each of them global in character. The economic and social structure of the local industry has proven very adaptive to the global processes of vertical de- and re-integration. Like few other "high tech-centers", Silicon Valley benefited from the dramatic shift towards more decentralized architectures of data systems. Silicon Val-ley firms like Intel, Hewlett-Packard, Apple or Sun essentially shaped and control the norms of technology and production in the field of PCs and workstations. From this platform, major players of the industry along with new software-firms like Netscape or Oracle emerged as prime forces in the field of"network computing", currently reshaping the norms of technology across the entire information and communications sector. On this basis, Silicon Valley acquired the position of the largest metropolitan ex-port region in the U.S. (before Detroit and New York), accounting for one-third of all exports of the state of California (SJMN 09/30/1997).

The changes in the industry resulted in a remarkable concentration of advanced manufacturing know-how, embedded in the local linkages and networks between vertically disintegrated industry segments. In the context of the ongoing global recomposition six major segments in the Silicon Valley IT-complex can be identified (see chart):

• Semiconductor manufacturing: Still the largest local industry segment, the leading manufacturers Intel, National Semiconductor, and AMD shifted volume production of wafers to other U.S. locations in Oregon, Texas, Arizona and New Mexico and expanded and automated their assembly and testing facilities in South East Asia. As part of the global reorganization of manufacturing operations, the remaining local manufacturing facilities underwent substantial modernization and automation. As"pilot fabs", these plants perform the extremely complex task of developing the manufacturing process of new generations of chip-products up to the point of volume production, which is handed over in detail to the volume production plants across the U.S. (Leachman 1996). ASIC producers like LSI Logic or VLSI have strategically central manufacturing in more specialized local plants. As global manufacturing activities have become more and more dependent on sophisticated pilot manufacturing and a skilled workforce in this field, new facilities are created in Silicon Valley, resulting in a gradual resurgence in engineering and manufacturing employment in semiconductors in the mid-1990s (NOVA 1995).
• Semiconductor manufacturing equipment: Resulting from the government-supported concertation of major semiconductor manufacturers and their equipment suppliers, local companies like Applied Ma-terials, Varian, Lam Research or KLA/Tencor have emerged as important global players in the field. Close cooperation in the development, installation, and fine-tuning of equipment with the leading semiconductor producers have become crucial to the competitiveness of both equipment and chip producers. Emulating the strategies of independent producers of machine tools for other manufacturing industries, large and medium-sized semiconductor equipment vendors maintain most of their engineering and manufacturing capacities in Silicon Valley but have a world-wide presence in most semiconductor manufacturing centers. The vulnerability of the merchant equipment producers to the global investment cycles of the semiconductor industry still reinforces the necessity for long-term equipment vendor/producer-relationships at the local level (Angel 1994).
• Computer Hardware: Dominated by Hewlett-Packard, Apple and the vendors of workstations and high-performance computers Sun, Silicon Graphics and Tandem the industry has shifted most in-house manufacturing activities away from their Silicon Valley operations. The regional industry has remained an important hub for the global manufacturing organization because of the strategic cooperation with locally based contract manufacturing companies. Pioneered by Sun Microsystems, Silicon Valley computer companies were at the forefront of establishing new cooperative relationships with contract manufacturers (Saxenian 1994), which are involved in important aspects of the engineering and design process of new products and systems (Feitzinger/Hau Lee 1997).
• Disk-drives and storage systems: As part of the global restructuring of the computer hardware sector, a whole new generation of merchant disk-drive producers emerged. All of the major global players in this industry (Seagate/Conner, Maxtor, and Quantum) and their world-leading suppliers of key storage technologies (like ReadRite or Komag) are located in Silicon Valley, commanding about 85% of the world-market. With an employment of 10.000 - 65.000 each, these companies maintain extensive off-shore manufacturing operations, especially in South East Asia (Business Week 07/22/1996, Ernst 1997). Local manufacturing in Silicon Valley has been mostly shifted to these locations, the industry, however, maintains strong links with local contract manufacturers. The only remaining large-scale disk-drive manufacturing operation in Silicon Valley is IBM's storage division facility in San Jose, which after a period of downsizing and subsequent modernization is supposed to become one of the largest disk-drive plants in the world (SJMN 04/12/1996).
• Networking equipment: In the course of the so-called "Internet-revolution", Silicon Valley has become home to the leading players in the world-market of decentralized data communications equipment as routers, modems, network control systems and components like data communication cards. The three major companies in the field Cisco, 3Com, and Bay Networks maintain only minimal in-house manufacturing operations, most manufacturing is done in close cooperation with contract manufacturers. Smaller networking equipment companies use smaller assembly contractors, often on a long-term base with highly personalized relationships.
• Contract manufacturing and components assembly: Driven by the "fabless" manufacturing strategies in the equipment vending sectors, most manufacturing in Silicon Valley (except for semiconductors) today is performed as contract manufacturing. At the top of this segment are rapidly growing, globally acting contract manufacturing companies like Solectron and Flextronics or the local subsidiaries of large contract manufacturers from outside the area like SCI or Hadco. The sector can be divided into electronics systems assembly (often referred to as Electronics Manufacturing Services, EMS) and printed circuit board (PCB) manufacturing (IPC 1997). PCB manufacturing is done by specialized producers ranging in size from a dozen to about 1800 employees, which handle the highly toxic process of etching printed circuits into epoxy laminate boards. In both systems assembly and PCB-production persists a large segment of small and very small companies, which handle work from smaller customers and "start-up" companies or capacity-overflows from larger contract manufacturers. As the traditional polarized manufacturing structure of Silicon Valley today is replicated within the contract manufacturing sector, a stable model of contract manufacturing relationships based on a well-defined hierarchy of "system suppliers" and various levels of subcontractors (as epitomized in the "supply-pyramid" of the auto-industry) has not yet emerged. A "market" for contract manufacturing services still hardly exists. Contract manufacturers rely on a broad variety of often very personalized customer relationships, reinforcing their dependency on local business networks (see also Lüthje 2001b, Lüthje/Schumm/Sproll forthcoming).

Silicon Valley IT-Industry Structure 1995

Industry segment	Major Companies	Employment Global/Local
Semiconductors	Intel, National Sem., AMD, LSI, VLSI	82000/16400
Semiconductor Man. Equipment	Applied Materials, Lam, Varian, KLA/Tencor	26700/14500
Computer Systems	Hewlett-Packard, Apple, Sun, SGI, Tandem	141000/33000
Disk Drives/ Storage	Seagate/Conner, Maxtor, Quantum, IBM Storage Div.	122000/12500
Network Equipment	Cisco, 3Com, Bay Networks	ca. 12000/8000
Contract Manuf./ PCB	Solectron, SCI, Flextronics, Zycon, Sanmina	ca. 27000/ 9000

Employment figures: San Jose Business Journal and Corporate Reports

The local agglomeration of these globally oriented industries makes Silicon Valley unique among high tech-regions in the U.S. and throughout the world. The manufacturing base also consists of a number of industry segments of local character. Of importance here are traditional small- and medium-sized firms in the field of precision metal parts and sheet metal manufacturing, used in electronics systems. More important in size is the vast segment of"business services", which encompasses a number of industries performing vital functions for the local production infrastructure. Among them are building maintenance and janitorial companies, waste disposal and"management" firms, and temporary employment services. As these elements of the production process have been increasingly transformed into con-tract work, the respective sectors - all with extensive internal subcontracting relationships - have been growing rapidly. The term "business services" also covers genuine manufacturing activities. In industry and labor market stat statistics many electronics contract manufacturing firms appear in the category of "engineering services". This is also true for software duplication firms, that handle the physical portions of the production and packaging of software products.

It should be noted that the transformation of the local production structure of the Silicon Valley electronics industry did not come about through the pressure of global market forces alone. It occurs in a system of capitalist regulation which, again, is both local and global in scope. With regard to the overarching industry structure, the framework of financial institutions as embedded in the NASDAQ-technology stock market in New York and the famous venture capital firms in Palo Alto and Menlo Park is of central importance. Although highly speculative in nature, this system should not be described in categories of "casino capitalism" (with regard to Silicon Valley: Florida/Kenney 1990), but of "Finanzkapital" (Hilferding 1910) in the very traditional sense of an organic formation of productive and monetary capital. The local venture capital system is constituting a specific mode of mobilization of productive re-sources which shapes both the system fabless" technology companies and contracted manufacturing. As new start-up firms avoid entering in costly manufacturing activities, it also makes sense to support new "manufacturing services" firms through venture capital. Some Venture capital banks even pretend to mimic the Japanese "keiretsu" model, nurturing "families" of fast growing software, systems, and manufacturing firms (Herhold 1997).

The formal and informal networks and institutions of engineers, management and local and national industry associations, most visibly Joint Venture: Silicon Valley, the Santa Clara Valley Manufacturing Group, or the American Electronics Association, constitute another important regulative structure, that particularly shapes the regime of production at the shop-floor level. Frequently referred to as "networks of innovation" these organizations do not only constitute a distinctive mode of technology development and political organization of the local industry (Saxenian 1981, 1994), but also an important social institutionalization of manufacturing practices. "Post-Fordist" vertically disintegrated manufacturing requires a high degree of political organization, which provides for the "socialization" of the norms of production across the sector and regional industry. A most important vehicle for this is the standardization and normalization of production activities as embodied in systems of quality norms as ISO 9000 and in the various "benchmarking" and "best practice" initiatives in shop-floor management, training, and "workforce development" (AEA 1996). The latter includes the field of public schools and vocational training, in which the industry (under the leadership of some of the more important venture capitalists) has taken the initiative for numerous reforms (NOVA 1995). The institutional networks in this field also are instrumental in developing new models and techniques to manage "flexible employment", downsizing, and outplacement situations (Joint Venture: Silicon Valley, 1996a and b)

An important observation in this context is that the process of capital restructuring and formation within the local production networks has become increasingly shaped along ethnic lines, embedded in the immigration situation, the patterns of transnational elite mobility, and racial and ethnic discrimination in the industry. While the high degree of international labor mobility can be regarded as a central feature of the global character of the Silicon Valley economy, the region provides a prime example of the strong social stratification of the immigration process into major U.S. industrial centers today. Like in New York, Los Angeles, or Chicago, the Latino/Chicano communities in San Jose and other South Bay cities are overwhelmingly working-class, whereas the various Asian nationalities are highly differentiated by class and immigration status. As other students of the "new Asian immigration" have pointed out, immigration is not just an "economic" phenomenon, triggered by the demand for low-wage labor, but essentially "political" in character, structured by immigration laws and regulations, the involvement of political, military, and financial U.S. interests in the countries of emigration, as well as the inability of the U.S. system of higher education to meet the demand for skilled engineers and scientists in high-tech industries (Sassen 1988 and 1991, Ong e.a. 1994, Ong-Hing/Lee 1996). In the case of Silicon Valley, the migration of these elite workers and the related phenomenon of "entrepreneurs immigration" (Keil 1993) is closely bound to the transnational movement of capital in the industry, especially with regard to South East Asia, were most leading Silicon Valley maintain major manufacturing operations (Henderson 1989, Alarcón 1997).

Highly-skilled immigrants, engineers of Asian origin in particular, have played a significant role in the transformation of local industry structures. A number of important companies in the newer segments like semiconductor equipment, networking equipment, storage systems, and contract manufacturing are headed or owned by Asian entrepreneurs and managers, some of them educated in U.S. elite institutions. As the multinationals in the industry began to employ foreign engineers of East Asian, Indian, and also Latin American origin in large numbers (Alarcón 1997), new patterns of ethnic discrimination have emerged, that spurred the growth of smaller "ethnically owned" businesses in the industry. As immigrant engineers widely experience discrimination in payment and status, many start their own companies, often employing immigrant workers. According to a recent study, Asian American entrepreneurs head 300 of 800 high tech firms in Silicon Valley (Park 1996). This phenomenon has been particularly wide-spread at the lower end of contract manufacturing and components production (Ewell 1996), producing highly problematic patterns of exploitation through relationships of intra-ethnic "solidarity" and trust (Park 1992). Immigrant entrepreneurs have also created a number of institutions for mutual support and "networking", as the Asian American Association of Manufacturers.

.05. Production networks and regimes of production in the 1990s

Along with the changes in the reproductive structure of capital, the Silicon Valley IT-industry experienced a profound transformation of the regime of production. This process reflects the transition from the earlier post-Fordist model of "bloody taylorism" to a sophisticated high-performance work organization shaped along the principles of contemporary "Total Quality Management". The process of vertical de- and reintegration and the commodification of IT-products and systems engenders a reorganization of globalized mass-production processes within individual companies and industry segments with a high level of standardization in manufacturing procedures across the internal chain of production. The new face of "systemic rationalization" is characterized by radical outsourcing of most Original Equipment Manufacturing (except for semiconductors), a gradual but strategically important reconcentration of manufacturing activities in Silicon Valley, and increasing inequality of working conditions "along the chain". The network-based regime of production has been shaped through the management practices inherent to the early "post-Fordist" labor policies of the 1970s and the general power relationships be-tween capital and labor in the U.S. It is built upon a highly differentiated pattern of racial and ethnic segmentation of the workforce, developing from the regional immigration situation.

As Silicon Valley transformed into a hub for global manufacturing strategies, the size of the manufacturing workforce has remained relatively stable. According to data provided by the California Employment Development Department, overall employment in the electronics group in Santa Clara County (excluding defense electronics and business services) ranged between 160.000-180.000 throughout the late 1980s and 1990s. Taking into consideration the geographic expansion of Silicon Valley into South-ern Alameda and San Mateo Counties and the high number of manufacturing workers counted in categories as business and engineering services (Joint Venture 1996a and b), manufacturing related employment can be estimated at a minimum of 40.000. What the official labor market data hardly reveal is the shift in the composition of the industry and the workforce. The manufacturing workforce in semiconductors and in-house assembly operations of computer and disk-drive manufacturers has been declining, semiconductor equipment, contract manufacturing, and business services (especially temporary employment services) have been growing. The workforce in contract manufacturing can be estimated at 20.000, temporary help agencies employ about 10.000 production workers (ibid., see also Benner 1996). As the industry has added thousands of jobs in all areas of design, engineering and manufacturing following the crisis of the Californian economy in 1991/92, wages especially in the lower segments have remained at recession level (SJMN 03/17/1997).

Along the position of local manufacturing activities in the global corporate organization and the employment practices prevalent in the particular industry segments, four sectors of the local manufacturing workforce can be distinguished (figures based on industry information and interviews):

• The workforce in the core operations of semiconductor producers and equipment vendors. These workers are involved in the extremely complex processes of "pilot production", both from the manufacturer and the vendor side. With wages of 14-18$ they are the highest paid manufacturing workers. There is also a considerable amount of craft work involved, with even higher pay rates. In contrast to earlier practices, the major semiconductor manufacturers today demand extensive formal training, resulting in a gradual erosion of traditional operator-type job structures and a formalization of hiring procedures. This has also limited the role of kinship relations for recruitment, however, the workforce remains very diverse with a growing proportion of relatively skilled immigrant workers. The industry practices teamwork extensively. However, traditional separations of operative, craft and engineering work remain largely intact, specifically favored by the individualized system of pay determination and the absence of work rules. Semiconductor companies also have a very rigid system of performance evaluation, producing a high level of employment insecurity even for skilled workers. Given the high cost of skill development for clean-room workers today, semiconductor producers use relatively few temporary workers. In contrast, equipment vendors use a high proportion of temporary workers to keep their workforce adjustable to the investment-cycles of their major customers.
• The workforce in high-end local assembly operations, encompassing the remaining in-house assembly operations of the larger OEMs in computers, disk-drives, network equipment, and the new globally operating contract manufacturers. As product markets differ, work structures in these companies are similar: highly automated assembly work with at least some basic computer skills, English language proficiency and communications skills. Entry level wages are 6,50$ to 9,50$, with relatively decent benefits. Like in semiconductors, companies have raised their formal skill and training requirements. As most of the local production activities by OEMs have been eliminated, the larger contract manufacturers are the biggest employers of manufacturing workers in Silicon Valley today. There is also a number of companies under distinctively immigrant or minority management, mostly Asian, that have merged the traditional model of the non-union "corporate community" in the U.S. electronics industry with control strategies from East Asian labor systems (Deyo 1989, Ong 1991). In this sector, the workforce is overwhelmingly Asian with a high proportion of recent immigrants, equal employment opportunity regulations enforce a certain representation of other minorities.
• The lower end of component assembly and PCB manufacturing. Although the labor process is becoming increasingly automated and technologically complex, the work situation is characterized by low wages at or just above minimum wage and poor working conditions, especially with regard to health and safety, even in the larger establishments of this sector. The workforce is almost entirely immigrant, with a broad mix of ethnicities in the larger companies and ethnicity-specific hiring practices in the smaller ones. Some of these companies organize teamwork along ethnic lines, congregating workers of specific nationalities into separate work teams in order to overcome language barriers and create competition at the shop-floor. In many of the smaller establishments we find highly problematic practices of "intra-ethnic" exploitation, especially directed against immigrant women (Park 1992).This sector continues to reproduce the sweatshop-type working conditions of the"board stuffing" firms of the 1970s and 1980s. As the impact of toxics related to new processes of PCB manufacturing have not been seriously researched, the health and safety situation in PCB-production is of particular concern (CRT 1997).
• Production-related business services, as building maintenance or waste management. Wages and working conditions in this sector are shaped by extensive subcontracting relationships, with large con-tractors governing networks of small and very small firms, some of them under minority or immigrant ownership (Zlolniski 1992). As generally true for these "service" industries, wages are at or just above minimum wage, benefits and working conditions are poor. The smaller waste disposal companies working for electronics firms offer the worst working conditions in the Silicon Valley (SCCOSH 1996).

As differences in working conditions between these sectors have become more complex, the regime of production is remarkably uniform across the industry. Driven by industry efforts to institutionalize the norms of flexibilized mass-production, all industry segments with the exception of the very low-end assembly sweatshops and business service firms employ teamwork approaches and demand substantial efforts of skill development from their workers. As employment has become flexibilized in all areas, major variations in the regime of production between companies and industry segments do not primarily occur along varying degrees of employment security (as in traditional concepts of core and peripheral work-force), but along wages, working conditions, and opportunities of skill development. It should be noted that the "teamwork-revolution" in Silicon Valley electronics manufacturing has not changed substantially traditional separations of engineering and shop-floor work as well as the authority of supervisors and engineers at the shop-floor. As this confirms the characterization of the team-model as "management-by-stress" (Parker/Slaughter 1994), its Silicon Valley version is typified by a marriage of employer mandated teamwork with an extremely flexibilized system of workforce recruiting and ethnicity-based company paternalism. Temporary employment has become an almost universal mode of the regulation of the wage relation, as virtually every major company hires manufacturing workers through temp agencies (Benner 1996). A major contradiction of this system is that "team-building" and skill-development is counteracted by the generalized employment insecurity for most sectors of the workforce.

In this context, the patterns of ethnic and racial stratification have changed substantially, too. Today, the South San Francisco Bay is among the regions with the highest concentration of recent immigrants of Latin American and Asian origin in the country (Alarcón 1994, Martínez-Saldana 1996). The character of the area as a suburban metropolis has essentially shaped the formation of immigrant communities, resulting in a high degree of mixed-class neighborhoods, concealing existing poverty, and exposing relatively low levels of urban conflict and crime (Joint Venture 1997). Notwithstanding the largely "classless" face of immigration, recent indicators point to rising segregation, particularly at the workplace (McLaughlin 1997). As it is still true that most production work is performed by immigrants, women in particular, a considerable part of the immigrant workforce has been able to move into higher skilled and supervisory positions of the manufacturing workforce. Apart from the "high tech-braceros" in product and software engineering (Alarcón 1997), there is also a substantial number of locally born and educated engineers of Latino and Asian origins in manufacturing related activities. The traditional pattern of "white (male) managers vs. immigrant (women) workers" (Hossfeld 1989, Siegel 1992) today is mostly prevalent in the lower-end assembly segment of the industry. At the same time, the rapid growth of minority-owned businesses has produced new and highly problematic patterns of intra-ethnic exploitation and interethnic competition. The larger companies under minority management tend to encounter their "ethnic business" image and ethnicity-based solidarization at the shop-floor by promoting explicitly multicultural company values.

Although relatively limited and weak, patterns of industrial contest and workers' solidarity equally develop mostly through immigrant communities. The janitorial industry has seen a remarkable insurgence of mostly Latino immigrant workers, organized through SEIU's "Justice for Janitors" campaign, achieving the unionization of the building maintenance contractors of IT-companies like HP, Apple, or Intel. Resulting from a distinctively industry-based organizing approach, janitorial companies form the only unionized segment of the production networks of the Silicon Valley electronics industry, aiming at state-wide bargaining patterns in the future (Benner 1995, Bacon 1997b). In some of the lower-end assembly firms, labor conflicts over working conditions, dismissals, and payment practices have taken place. As known from other sweatshop-type industries, these conflicts remained largely limited to individual companies and eventually resulted in the closure of these establishments. Nevertheless, these conflicts mobilized considerable community support (Bacon 1997a). In a comparable environment, some waste management firms currently face workers and community mobilization against the miserable health and safety conditions in this industry (SCCOSH 1996). Most important, inside the leading chip-producer of Silicon Valley, immigrant engineers have started to challenge the system of performance evaluation and its inherent mechanisms of radicalized selection (FACE-Intel 1997).

.06. The political context: "network-centric production" and labor in the "New North American Economy"

For all its local peculiarities, the Silicon Valley situation points to the broader context of industrial restructuring and labor relations in high tech industries. As Silicon Valley today depicts much of what has been portrayed as the "New North American Economy" (Moody 1997), our analysis points to the complexity of this picture. De-verticalized high tech production in regional "industrial districts" does not simply entail the proliferation of sweatshop-type subcontracting networks, as in many areas with high concentrations o f "light-manufacturing" for local markets like garment, auto parts, or furniture (Olney 1993). Network-centric production in electronics emerges in the context of the global recomposition of capital and the underlying norms of production and technology, and comprises various levels of working conditions at wages below traditional union-represented Fordist industries. The ability of U.S. firms to quickly combine and recombine these arrangements in the global and in the local arena may be considered a key to their competitiveness in the current period of global restructuring. In this context, globalization of pro-duction is not only characterized by the shift of manufacturing activities to countries with lower wage levels, but through the simultaneous rationalization of the production chain in U.S. and off-shore manufacturing centers. With regard to subcontracting relationships, turnkey manufacturing" indeed marks the emergence of a new form of the manufacturing infrastructure, which seems to be of great significance for the restructuring of norms of production in the electronics industry on a global scale (Sturgeon 1997, Lüthje e.a. forthcoming).

The complexity of this process exposes the analytical weaknesses of mainstream approaches in the cur-rent industrial policy debate. The failure of neo-classical conceptions of "market-driven innovation" be-comes particularly visible vis-à-vis the production process. Not only are network-centric production arrangements embedded in a complex web of industry, community, and government institutions; day-to-day shop-floor practices are based on a distinctively "social" and "political" organization of production relationships. With regard to neo-Schumpeterian conceptions of innovation and manufacturing it has to be stated that the pattern of district-based industrial restructuring shaped by Silicon Valley companies has been relatively successful in linking product and manufacturing innovation without resuscitate traditional forms of vertical integration. As major proponents of the vertically disintegrated IT-industry of the 1990s have become global players in their own right, discussions about the virtues of "locally" vs ."globally" oriented industrial policies seem to loose sense. The term "glocalization" (Ruigkrog/van Tulder 1996) characterizes the prevailing strategic pattern for investment and manufacturing strategies throughout today's IT-industry more appropriately. In this context it becomes more and more apparent, however, that production relations and intra-industry cooperation in regional industry networks are to an important degree shaped along ethnic lines, both with regard to capital formation and the composition of the work-force.

From a more critical perspective, the Silicon Valley situation explains why the analytical juxtaposition of Fordist and post-Fordist manufacturing practices seems more and more outdated today. It may help to understand how a post-Fordist industrial formation is reproduced in the context of extremely volatile global norms of production and technology without a clearly defined hegemonic center of industrial and political power. Silicon Valley as an industrial formation will not be replicated elsewhere, but vertical disintegration of production accompanied by the large-scale outsourcing of manufacturing activities in the more traditional oligopolistic sectors of the IT industry creates comparable effects and conditions for manufacturing organization and the composition of the workforce in other industry segments and locations. Notwithstanding different forms of industry organization and labor relations in the international context, an increasingly important and generalizable feature of post-Fordist regulation of the wage relation seems to be the permanent creation of unequal working conditions both at a global and a local level. This process does not only occur between "core" and "peripheral" industries and countries, but in the very center of advanced capitalist manufacturing. With regard to current debates on regulation theory (Vincent/Negri 1994), this may illustrate why in today's electronics industry the conditions for a post-Fordist capital-labor accord on new technologies and the "modernization" of manufacturing relationships hardly exist.

Before this background, chances for cooperative policy-making in local "networked-based" environments as suggested by theories of flexible specialization (Piore/Sabel 1984) remain very limited. In Silicon Val-ley and most other U.S. high tech centers, the historical and present weakness of labor at the shop-floor precludes any substantial presence of workers and minority communities in the political process. As the "Japanese challenge" for the U.S. IT-industry has vanished, the industrial policy debate has lost momentum (Stowsky 1996). The increasing political significance of Silicon Valley for industrial policy strategies in the national arena, especially under the Clinton administration, has not engendered any significant public debate about the social implications of the manufacturing practices in localized production net-works (U.S. Commission on the Future 1994). In this political context, viable alternatives can hardly be based on public concerns over industrial competitiveness and national security. The small number of community organizations dealing with the impact of the manufacturing practices of the "new" IT-industry have to confront a situation in which existing federal and state regulations over the environmental and human impact of electronics manufacturing are increasingly "streamlined" in the name of international competitiveness of the U.S. economy and individual high tech regions (SWOC 1995, Silicon Valley Toxics Action Fall 1996). The basic policy problem in this scenario seems to be the almost total lack of public control over the norms of production and the permanent recomposition of capital in the industry. In the decentralized IT-industry of the 1990s any viable conception of public control would have essentially to touch the cycle of innovation, product development, and capital investment established by the leading players of the key IT markets (Esser e.a. 1997).

With regard to the labor process, the system of network-centric production in Silicon Valley reveals a number of issues which need closer theoretical and political discussion. It demonstrates that the development of production networks and the underlying inter-company relationships is a highly political process, shaped by relationships of power and trust (Saxenian 1994) rather than micro-economic models of firm-networks. The process of institution-building fostered by the Silicon Valley IT industry in the field of manufacturing practices illustrates that network-based production acquires a high degree of political organization of shop-floor practices in order to guarantee the cohesion of a norm of production and the politics of production in a vertically disintegrated environment. At the same token, it becomes evident that systems of "lean" or "ultra-lean" production are heavily segmented along ethnic and racial lines. As ethnic and racial cleavages seem to be a constituting element of the "Gesamtrationalisierungsprozeß" (Schumann e.a. 1994) in network-based industries, ethnic and racial segmentation along the production chain deserves to be considered not only an accidental" by-product" but a systematic aspect of "systemic rationalization".

The Silicon Valley experience may also show that even in a total absence of labor organizations from the shop-floor and almost undisputed political "hegemony" of local-global business interests, the workplace remains a "contested terrain". Even the most sophisticated models of "outplacement-assistance" and "workforce resilience" demonstrate that the legitimacy of the regime of production is not being taken for granted. Workplace contest in this context mostly develops through immigrant communities. This con-firms the importance of class-based forms of organization and education within immigrant communities (Wong 1996, Grenier 1997). In Silicon Valley, this has become a defining question for the social repre-sentation of the very "core" of the manufacturing working class. As Latino/Chicano workers in Santa Clara County have a history of community-based labor mobilization rooted in the traditions of farm worker unionism (Matthews 1985, Sevilla 1992), most Asian immigrant groups lack such experiences. For a class-based discourse on work, which would also have to counter the dominance of immigrant elites at the workplace and in the communities, issues of quality management, teamwork, and flexibilization seem of central importance. The industry-promoted generalization of teamwork-based forms of labor control along with the common experience of temporary and contingent employment for most sectors of the manufacturing workforce may facilitate a multi-cultural and multi-community approach. The local presence of ongoing conflicts over the "team-model" as epitomized by General Motors' NUMMI plant in Fremont may help to understand experiences from older manufacturing industries (Parker/Slaughter 1994).

In a broader perspective, it remains true that the electronics industry in new high tech centers like Silicon Valley has been notoriously neglected by the U.S. labor movement (Eisenscher 1993, Bacon 1997a). By and large, the"new course" in the AFL-CIO has not changed this situation substantially. Whereas the Sweeney leadership has made the organization of new sectors of immigrant workers a cen-tral issue, the electronics industry does not have priority in this scenario. Even the implications of SEIU's much heralded "Justice for Janitors" campaign for the electronics manufacturing workforce have not been discussed in a broader context. As the electronics industry with 1,9 Mill. employees is by far the largest U.S. manufacturing sector today, current debates on new organizing strategies in the U.S. labor movement have to be rated against this situation. A new look on the industry may be taken as the rapid proliferation of contract manufacturing relationships throughout the U.S. becomes more and more inter-twined with the outsourcing strategies of the electronics-related elements of unionized manufacturing industries as automobile, aircraft, or defense systems. This situation has to be of concern not only for the unions in the electrical and electronics industry like IBEW, IUE or UE, but also for major industrial unions in related fields as the UAW, IAM, and the CWA.

From an international perspective, especially from a European background, the Silicon Valley situation suggests a very different look at the IT industry. As European labor unions tend to perceive the IT industry as a "service" industry with high dues-paying potentials in the technical professions, Silicon Valley shows the importance of the manufacturing workforce in this industry. It also demonstrates, that even an extreme local concentration of world-class research and engineering capacities and its integration with technology intensive manufacturing operations does not in itself guarantee a fair amount of stable and well-paid industrial jobs. As the norms of vertically disintegrated production become generalized throughout the global electronics industry, this experience may become more commonplace in other industrialized countries. The mostly unnoted expansion of contract manufacturing in Europe and of con-tract-type supplier clusters centered around PC- and other electronics manufacturing activities in high-wage countries like Sweden or Germany as well as in Eastern in Europe (Lüthje e.a. forthcoming) point into this direction. In a more political sense, the Silicon Valley of the 1990s may show why previously touted cooperative schemes of labor relations like Japan's company union-based team-model or Northern European and German "models" " of co-determination have lost much of their attractiveness to U.S. policy makers. As is becoming generally true in the "post-cold war" world-economy, the U.S. business establishment can rely on its internal strength in achieving global competitiveness. Globalized high tech manufacturing through localized production networks is definitely a major factor in this scenario.

.07. References

AEA - American Electronics Association (1996a) Cyberstates. A State-by-State Overview of the High-Technology Industry. Washington, D.C.: AEA

AEA - American Electronics Association (1996b) Working in New Ways. An AEA Survey on Changing Skills in the High tech Industry. Santa Clara: AEA.

AEA - American Electronics Association (1997) Cybernation: The Importance of the High-Technology-Industry in the United States Economy. Washington, D.C.: AEA

Aglietta, M. (1979) A Theory of Capitalist Regulation. The U.S. Experience. London: Verso.

Alarcón, R. (1994) 'Labor Migration from Mexico and Free Trade: Lessons From A Transnational Community',

Alarcón, R. (1997) 'From Servants to Engineers: Mexican Immigration and Labor Markets in the San Francisco Bay Area', Chicano Latino Policy Project Working Paper, Berkeley: Institute for the Study of Social Change.

Altvater, E. (1987) Sachzwang Weltmarkt. Verschuldungskrise, blockierte Industrialisierung, Ökologische Gefährdung - der Fall Brasilien. Hamburg: VSA.

Altvater, E. (1987) Sachzwang Weltmarkt. Verschuldungskrise, blockierte Industrialisierung, Ökologische Gefährdung - der Fall Brasilien, Hamburg: VSA.

Altvater, E. (1992) 'Fordist and Post-Fordist International Division of Labor and Monetary Regimes', in Scott, M./Storper, A. (eds.): Pathways to Industrialization and Regional Development. London/New York: Routledge.

Angel, D.P. (1994) Restructuring for Innovation. The Remaking of the U.S. Semiconductor Industry, New York: Guilford.

Aulenbacher, B./Siegel, T. (eds.) (1995) Diese Welt wird vÖllig anders sein - Denkmuster der Rationalisierung. Pfaffen-weiler.

Bacon, D. (1993) Organizing the High tech-Sweatshop, in Third Force, September/October 1993.

Bacon, D. (1997a) Silicon Valley: Gewerkschaften und ImmigrantInnen in der High tech Industrie, in: Lüthje, B./Scherrer, C. (eds.) Zwischen Rassismus und Solidarität. Diskriminierung, Einwanderung und Gewerkschaften in den USA, Mün-ster: Westfälisches Dampfboot.

Bacon, D. (1997b) West Coast Janitors Get Ready To Fight, Z Magazine, April 1997.

Bar, F. (1990) 'Configuring the Telecommunications Infrastructure of the Future: The Economics of Network Control', Ph. D. dissertation, University of California Berkeley.

Bechtle, G. (1993) 'Systemische Rationalisierung als neues Paradigma industriesoziologischer Forschung?' In: Soziale Welt, Sonderheft Umbrüche gesellschaftlicher Arbeit.

Benko, G./Lipietz, L. (1992) Les régions qui gagnent: district et reseaux, les nouveaux paradigmes de la géographie économique, Paris: PUF.

Benner, C. (1995) '"Win the Lottery or Organize": Economic Restructuring and Union Organizing in Silicon Valley', unpublished manuscript, UC Berkeley, Dept. of City and Regional Planning. Berkeley.

Benner, C. (1996) 'Shock Absorbers in the Flexible Economy: The Rise of Contingent Employment in the Silicon Valley' San Jose: Working Partnerships USA.

Bonacich, E. (ed.) (1994) Global Production: The Apparel Industry in the Pacific Rim. Philadelphia.Temple University Press.

Borrus, M. (1988) Competing for Control. America's Stake in Microelectronics. Cambridge, Ma.: Ballinger.

Borrus, M. (1997) 'Left for Dead: Asian Production Networks and the Revival of US Electronics.' BRIE Working Paper 100. Berkeley: BRIE.

Borrus, M./Leachman, R.C. (1994) 'The competitive semiconductor manufacturing survey: first report on results of the main phase', CCC Working Paper No. 94-13, Berkeley.

Borrus, M./Zysman, S. (1997) 'Wintelism and the Changing Terms of Global Competition: Prototype of the Future?' Work-ing Paper 96B. Berkeley:BRIE.

Braverman, H. (1974) Labor and Monopoly Capital, New York: Montly Review Press.

Brown, C. (ed.) (1996) The Competitive Semiconductor Manufacturing Human Resources Project. Second Interim Report. Berkeley: Institute of Industrial Relations/Engineering Systems Research Center.

Burawoy, M. (1985) The Politics of Production. Factory Regimes Under Capitalism and Socialism. London: Verso.

Carnoy, M./Castells, M./Cohen, S./Cardoso, F.H. (1993) The New Global Economy in the Information Age. University Park, Pa.: Pennsylvania State University Press.

Castells, M. (1996) The Rise of the Network Society. Vol. I, Cambridge, Ma./Oxford: Blackwell.

Chicano/Latino Policy Project Working Paper No. 1, Berkeley: Institute for the Study of Social Change.

Clinton, W.J./Gore, A. (1993) 'Technology for America's Economic Growth, A New Direction to Build Economic Strength', Washington, D.C.: GPO.

Coriat, B. (1979) L'ateliér et le chronométre. Paris: PUF.

Coriat, B. (1990) L'atelier et le robot. Essai sur le fordisme et la production de masse à l'age de l'èlectonique. Paris: PUF

CRT - Campaign for Responsible Technology (1997) 'EPA to Report on PWB Technologies'. Connections. Information Bulletin of the Campaign for Responsible Technology, San Jose: Silicon Valley Toxics Coalition.

Curran, L. J. (1996) 'Contract Manufacturers: Binge on Acquisitions', Electronics Business Today, August 1996.

Davies, A. (1994) Telecommunications and Politics. The Decentralised Alternative, London/New York: Pinter.

Davis, M. (1986) Prisoners of the American Dream. Politics and Economy in the History of the American Working Class, London: Verso.

Deyo, Frederic, C. (1989) Beneath the Miracle. Labor Subordination in the New Asian Industrialism, Berkeley/Los Ange-les: University of California Press.

Dosi, G. (1982) 'Technological paradigms and technological trajectories: A suggested interpretation of the determinants and directions of technological change', in: Research Policy, Vol. 11: 147-162.

Drüke, H. (1992) PCs "made in Europe" - ein Auslaufmodell? Berlin: Sigma.

Drüke, H. (1993) 'Restructuring in the PC-Industry: new challenges, new actors, new strategies: a study in labor and indus-trial policy', Berlin: WZB discussion paper.

Eisenscher, M. (1993) Gewerkschaftliche Organisierung in der Computerindustrie: Die Erfahrungen des UE Electronics Or-ganizing Committee im Silicon Valley, in Lüthje, B./Scherrer, C. (eds.): Jenseits des Sozialpaktes - Neue Unternehmens-strategien, Gewerkschaften und Arbeitskämpfe in den USA, Münster: Westfälisches Dampfboot.

Ernst, D. (1997) 'From Partial to Systemic Globalization: International Production Networks in the Electronics Industry'. Berkeley: BRIE Working Paper 98.

Ernst, D./O'Connor, D. (1992) Competing in the Electronics Industry. The Experience of Newly Industrialising Econo-mies. Paris: OECD.

Esser, J. (1993) 'Technologieentwicklung in der Triade. Folgen für die europäische Technologiegemeinschaft' , in Süß, W./Becher, G. (eds.): Politik und Technologieentwicklung in Europa. Berlin: Duncker&Humblodt.

Esser, J./Lüthje, B./Noppe, R. (eds.) (1997) Europäische Telekommunikation im Zeitalter der Deregulierung. Infrastruktur im Umbruch, Münster: Westfälisches Dampfboot.

Ewell, M. (1996) 'Asians rising to the top', San Jose Mercury News, August 11.

FACE Intel - Former And Current Employees of Intel (1997) Is Intel A Great Place To Work? FACE-Intel Website. Folsom.

Ferguson, C.H./Morris, C.R. (1993) Computer Wars. How the West Can Win in a Post-IBM World. New York/

Florida, R./Kenney, M. (1990) The Breakthrough Illusion: Corporate America's Failure to Move from Innovation to Mass Production, New York: Basic Books.

Gereffi, G. (1995) 'Global Production Systems and Third World Development', in: Stallings, B. (ed.): Global Challenge, Regional Response. The New International Context of Development. Cambridge, Ma./New York: Cambridge University Press.

Gilder, G. (1989) Microcosm: The Quantum Revolution in Economics and Technology. New York: Simon&Schuster.

Gingrich, N. (1995) To Renew America, New York: Harper.

Goldfield, M. (1997) 'US-Gewerkschaften zwischen Rassismus und Solidarität', in Lüthje, B./Scherrer, C. (eds.) Zwischen Rassismus und Solidarität. Diskriminierung, Einwanderung und Gewerkschaften in den USA. Münster: Westfälisches Dampfboot.

Grenier, G. (1997) 'Kultur und Solidarität: Latinos in der US-Gewerkschaftsbewegung', in Lüthje, B./Scherrer, C. (eds.) Zwischen Rassismus und Solidarität. Diskriminierung, Einwanderung und Gewerkschaften in den USA, Münster: West-fälisches Dampfboot.

Grindley, P./Mowery, D.C./Silverman, B. (1993) 'Sematech and Collaborative Research: Lessons in the Design of High technology Consortia', Berkeley: CCC Working Paper No 93-21.

Grove, A. S. (1996) Only the Paranoid Survive. How to Exploit the Crisis Points That Challenge Every Company and Career, New York/London e.a.: Currency Doubleday.

Harrison, B. (1994) Lean and Mean: The Changing Landscape of Corporate Power in the Age of Flexibility. New York: Basic Books.

Hart, J./Reed, R.R./Bar, F. (1992) The Building of the Internet, Telecommunications Policy, November 1992.

Henderson, J. (1989) The Globalisation of High Technology Production. London: Routledge

Henderson, J. (1994) Electronics Industries and the Developing World - Uneven Contributions and Uncertain Prospects, in Sklair, L. (ed.) Capitalism and Development. London: Routledge.

Herhold, S. (1997) 'All in the family. Venture capitalists use Japanese business model to help start-ups', San Jose Mercury News, March 16, 1997.

Hilferding, R. (1910) Das Finanzkapital, Re-edition 1969, Frankfurt/KÖln: Europäische Verlagsanstalt.

Hirsch, J. (1993) Internationale Regulation. Bedingungen von Dominanz, Abhängigkeit und Entwicklung im globalen Kapi-talismus, in Das Argument, No. 198.

Hirsch, J. (1996) Der nationale Wettbewerbsstaat. Staat, Demokratie und Politik im globalen Kapitalismus. Ber-lin/Amsterdam: Edition ID-Archiv.

Hossfeld, K. (1988) Divisions of Labor, Divisions of Lives: Immigrant Women Workers in Silicon Valley. Ph. D. disserta-tion UC Santa Cruz.

Hossfeld, K. (1989) 'Their Logic Against Them: Contradictions in Sex, Race, and Class in Silicon Valley', in Ward, K. (ed.) Women Workers and Global Restructuring, Ithaca: ILR Press.

Hossfeld, K. (1995) 'Why Aren`t High tech Workers Organized? Lessons in Gender, Race, and Nationality from Silicon Valley', in Cornford, D. (ed.) Working People of California. Berkeley/Los Angeles/London: University of California Press.

Hübner, K. (1989) Theorie der Regulation. Eine kritische Rekonstruktion eines neuen Ansatzes der Politischen Ökonomie. Berlin: Sigma.

IPC - Institute for Interconnecting and Packaging Integrated Circuits (1997) Snapshot of the EMS Industry, Northbrook, Ill.

Johnson, C./Tyson, L./Zysman, J. (1989) Politics and Productivity. How Japan's Development Strategy Works. New York: Harper.

Joint Venture: Silicon Valley (1996a) Building Workforce Resilience. A Guide to Launching a Process,San Jose: Joint Ven-ture Silicon Valley.

Joint Venture: Silicon Valley (1996b) Index of Silicon Valley. Measuring Progress Towards A 21st Century Community. San Jose.

Joint Venture: Silicon Valley (1997) Index of Silicon Valley. Measuring Progress Towards A 21st Century Community. San Jose.

Keil, R. (1993) Weltstadt - Stadt der Welt. Internationalisierung und lokale Politik in Los Angeles. Münster: Westfälisches Dampfboot.

Kuhn, T.S. (1962) The Structure of Scientific Revolutions, Chicago: University of Chicago Press.

La Dou, J. (1984) The Not-So-Clean Business of Making Chips, Technology Review, May/June.

La Dou, J. (1994) Health Issues in the Global Semiconductor Industry, Annals Academy of Medicine, Vol. 23, No. 5.

Langlois, R.N. (1982) 'Capabilities and vertical disintegration in process technology: the case of semiconductor fabrication equipment', Berkeley: CCC Working Paper No. 92-10.

Leachman, R.C. (ed.) (1996) Competitive Semiconductor Manufacturing Survey. Third Report on the Results of the Main Phase. Berkeley: Report CSM-31.

Lipietz, A. (1987) Mirages et Miracles. The Crisis of Global Fordism, London: Verso.

Lüthje, B. (1993a) Die Neuordnung der Telekommunikationsindustrie in den USA. Krise fordistischer Akkumulation, Deregulierung und Gewerkschaften. Wiesbaden: Deutscher Universitätsverlag.

Lüthje, B. (1993b) On the Political Economy of "Post-Fordist" Telecommunications: The US Experience, Capital and Class, No.51.

Lüthje, B. (1997) Transnationale Dimensionen der "network revolution", in Esser, J./Lüthje, B./Noppe, R. (eds.) Eu-ropäische Telekommunikation im Zeitalter der Deregulierung. Infrastruktur im Umbruch. Münster: Westfälisches Dampfboot.

Lüthje, B. (2001a) Standort Silicon Valley: Ökonomie und Politik der vernetzten Massenproduktion. Frankfurt/New York: Campus.

Lüthje, B. (2001b) Electronics Contract Manufacturing: Transnational Production Networks, the Internet, and Knowledge-Diffusion in Low-Cost Locations in East Asia and Eastern Europe. Research Topics and questions. East-West Center Economics Series, No. 18. Honolulu.

Lüthje, B./Scherrer, C. (eds.) (1997): Zwischen Rassismus und Solidarität. Diskriminierung, Einwanderung und Gewerk-schaften in den USA. Münster: Westfälisches Dampfboot.

Lüthje, B./Scherrer, C. (eds.): Zwischen Rassismus und Solidarität. Diskriminierung, Einwanderung und Gewerkschaften in den USA. Münster: Westfälisches Dampfboot.

Lüthje, B./Schumm, W./Sproll, M. (forthcoming) Contract Manufacturing: Transnationale Produktion und Industriearbeit in der IT-Industrie. Frankfurt/New York: Campus.

Martínez-Saldana, J. (1996) 'The Cardenista Movement in the Silicon Valley, 1987-1994', unpublished manuscript, Santa Clara.

Marx, K. (1885/1963): Das Kapital. Band II. Berlin: Dietz Verlag.

Matthews, G. (1985) 'The Fruit Workers of the Santa Clara Valley: Alternative Paths to Union Organization during the 1930s', Pacific Historical Review, No. 51.

McLaughlin, Ken (1996) 'For Latinos and Asians, Segregation is on the Rise', San Jose Mercury News, October 25, 1996.

Moody, K. (1988) An Injury to All. The Decline of American Unionims, London/New York: Verso.

Moody, K. (1997) 'Pulled Apart, Pushed Together - Die US-Arbeiterklasse in der neuen Ökonomie Nordamerikas', in: NOVA Private Industry Council (1995) Circuits from Sand. Semiconductor Workforce Observations, Sunnyvale, Ca..

Ohmae, K. (1985) Die Macht der Triade. Wiesbaden: Gabler.

Olney, P. (1993) 'The Rising of the Million', Crossroads, July/August.

Ong Hing, B./Lee, R. (1996): The State of Asian Pacific America: Reframing the Immigration Debate. Los Angeles: LEAP Asian Pacific American Public Policy Institute/UCLA Asian American Studies Center.

Ong, A. (1991) 'The Gender and Labor Politics of Post Modernity', Annual Review of Anthropology 1991.

Ong, P./Bonacich, E./Cheng, L. (eds.) (1994) The New Asian Immigration in Los Angeles and Global Restructuring. Phila-delphia: Temple University Press.

Ong, Paul M. (1990) 'Race and Post-Displacement Earnings among High tech Workers', Los Angeles: UCLA Graduate School of Architecture and Urban Planning.

Park, E. J.-W. (1992) 'Asian Americans in the Silicon Valley: Race and Ethnicity in the Postindustrial Economy', UC Ber-keley: Ph. D. dissertation.

Park, E. J.-W. (1996) 'Asians Matter: Asian American Entrepreneurs in the Silicon Valley High Technology Industry', in Ong Hing, B./Lee, R. (eds.), The State of Asian Pacific America. Reframing the Immigration Debate. Los Angeles: LEAP Asian Pacific American Public Policy Institute/UCLA Asian American Studies Center.

Parker, M./Slaughter, S. (1994) Working Smart - A Union Guide to Participation Programs and Reengineering, Detroit: Labor Notes.

Piore, M./Sabel, C. (1984) The Second Industrial Divide, New York: Basic Books.

Poulantzas, N. (1978) Staatstheorie. Politischer überbau, Ideologie, sozialistische Demokratie, Hamburg: VSA.

Rosswurm, S. (ed.) (1992) The CIO`s Left-Led Unions, New Brunswick: Rutgers University Press.

Ruigkrog, W./van Tulder, R. (1995) The Logic of International Restructuring, London: Routledge.

Sassen, S. (1988) The Mobility of Labor and Capital. New York: Cambridge University Press.

Sassen, S. (1991) The Global City: New York, London, Tokyo, Princeton: Princeton University Press.

Sauer, D./DÖhl, V. (1994) Arbeit an der Kette - Systemische Rationalisierung unternehmensübergreifender Produktion, Soziale Welt, 2/1994.

Saxenian, A. (1981) 'Silicon Chips and Spatial Structure: The Industrial Base of Urbanization in Santa Clara County', Ph.D. dissertation, UC Berkeley, Institute of Urban and Regional Development.

Saxenian, A. (1989) 'Regional Networks and the Resurgence of Silicon Valley', Berkeley: IURD Working Paper 508.

Saxenian, A. (1994) Regional Advantage. Culture and Competition in Silicon Valley and Route 128. Cambridge, Mass.: Harvard University Press.

Sayer, A./Walker, R. (1992) The New Social Economy: Reworking the Division of Labor, Cambridge, Ma./Oxford: Black-well.

SCCOSH - Santa Clara Center for Occupational Safety and Health (1996) Struggle & Strength. Stories from theWorker Stories Process, San Jose.

Schatz, R. (1983) The Electrical Workers, Urbana and Chicago: University of Chicago Press.

Schumann, M./Baethge-Kinsky, V./Kuhlmann, M./Kurz, C./Neumann, U. (1994) 'Der Wandel der Produktionsarbeit im Zugriff neuer Produktionskonzepte', Soziale Welt, Sonderband 9.

Scott, A./Angel, D. (1987) The U.S. Semiconductor Industry: A Locational Analysis, Environment and Planning, A 19: 875-912.

Sevilla, R. (1992) 'Employment Practices and Industrial Restructuring: A Case Study of the Semiconductor Industry in the

Siegel, L. (1986) 'Microcomputers: From Movement to Industry', Monthly Review, Vol. 38, July-August.

Siegel, L. (1992) 'Analysis of High tech Employment Patterns in Eight Leading U.S. High-Tech Centers. Survey for the U.S. Equal Employment Opportunity Commission', Global Electronics, Mountain View: Pacific Studies Center.

Silicon Valley 1955-1991', Ph.D. dissertation UC Los Angeles, Dept. of Urban Planning.

Storper, M./Walker, R. (1989) The capitalist imperative. Territory, technology and industrial growth, Oxford/New York: Blackwell.

Stowsky, J. (1987) 'The Weakest Link: Semiconductor Production Equipment, Linkages and the Limits to International Trade', Berkeley: BRIE Working Paper No. 27.

Stowsky, J. (1996) America's Technical Fix. The Pentagon's Dual Use Strategy, TRP, and the Political Economy of Tech-nology Policy in the Clinton Era, Berkeley: BRIE.

Sturgeon, T. (1997) 'Turnkey Production Networks: A New American Model of Industrial Organization?' Unpublished Manuscript, Berkeley/Boston.

SWOP - Southwest Organizing Project (1995), Intel Inside New Mexico - A Case Study of Environmental and Economic Injustice, Albuquerque: SWOP.

U.S. Commission on the Future of Worker-Management Relations (1994) Fact Finding Report, Washington, D.C.: Gov-ernment Printing Office.

U.S. Environmental Protection Agency (1995) Printed Wiring Board Industry And Use Cluster Profile. Design for the Environment Printed Wiring Board Project, Washington, D.C.: G.P.O.

Vincent, J.-M./Negri, T. u.a. (Hg.) (1994) 'école de la régulation et Critique de la raison économique', Futur antérieur, Paris: L'Harmattan.

von Hayek, F.A. (1968) 'Der Wettbewerb als Entdeckungsverfahren', Kieler Vorträge, Neue Folge. Tübingen.

Voskamp, U./Wittke, V. (1994) 'Von Silicon Valley zur "virtuellen Integration". Neue Formen der Organisation von Inno-vation am Beispiel der Halbleiterindustrie'; in Sydow, J./Windeler, A. (eds.) Management interorganisationaler Beziehun-gen, Opladen: Westdeutscher Verlag.

Wallerstein, I.( 1979) The Capitalist World Economy, Cambridge UK/New York: Cambridge University Press.

Wong, K. (1996) Cultural Democracy and the Revitalization of the U.S. Labor Movement, in Darder, A. (ed.) Culture and Difference. Critical Perspective on the Bicultural Experience in the United States, Westport, Ct./London: Bergin&Garvey.

Ziegler, J.N. (1991) Semiconductors. In: Daedalus, Proceedings of the American Academy of Arts and Sciences Vol. 120, No. 4.

Zlolniski, C. (1992) 'The Working Poor in a Post-Industrial Economy: Mexican Immigrants in Silicon Valley's Janitorial Industry', unpublished manuscript, UC Santa Barbara.

Newspapers and Periodicals

Financial Times (FT)

New York Times (NYT)

San Jose Mercury News (SJMN)

Frankfurter Allgemeine Zeitung (FAZ)

Business Week (BW)

Silicon Valley Toxics Action - Quarterly Newsletter of the Silicon Valley Toxics Coalition