Industrial alliances and firm location behaviour: some evidence from the US semiconductor industry

Abstract

This study uses a log-linear model in order to analyse data on interfirm strategic alliances within the US semiconductor industry. The findings suggest that although the finding that the intensity of information transaction between firms is directly associated with geographical proximity, this effect is much less localized than would be usually expected. The results cast doubt on much of the existing qualitative literature on the nature of agglomeration externalities.

Clusters and regional development: Some cautionary observations from the semiconductor industry

“…These arguments have also been applied to other sectors (Best, 1990). While the empirical validity of some of these arguments has been questioned (Arita and McCann, 2000;Audrestch and Feldman, 1996;Suarez-Villa and Walrod, 1997), in many quarters the Silicon Valley example of industrial clusters, with its many co-located flexible small firms competing in an environment of both competition and trust, is perceived to be the ideal spatial and organisational arrangement for 21st century innovative industries (Castells and Hall, 1995;The Economist, 1997). In cases where such conditions do not exist, proponents of these theories argue normatively that they should exist (Saxenian, 1994).…”

The Location of Technological Innovations within the Japanese Semiconductor Industry

“…Moreover, such organizational issues are determined within a multi-product environment, in which corporate control is exercised with respect to the production of a range of different outputs for either intermediate or final consumption. These resulting spatial patterns of plants are therefore primarily the result of the need for coordination between the plants within a well-defined system of formal inter-plant relations (Arita and McCann 2000), and are not the result of informal information networks of the type often discussed in relation to the electronics industry (Saxenian 1994). This is also the case for the higher level establishments.…”

“…In the case of multiplant multinational firms (Cantwell and Iammarino 2000), any such agglomeration effects may even operate over spatial scales larger than individual countries. These empirical observations therefore cast doubt on the assumed importance of specifically local inter-firm information spillovers as a source of competitive advantage (Porter 1990(Porter , 1998 within the electronics industry, and point rather more to the role of labour market hysteresis as a possible rationale for industrial clustering (Angel 1991;Arita and McCann 2000). More importantly, however, these observations also cast doubt on the whole hypothesis that small firm clusters represent something of an ideal spatial and organizational arrangement ensuring the maximization of innovation, either for the semiconductor industry or any other innovative industry facing short product life-cycles.…”

“…Urban clustering is therefore advantageous for industries which exhibit very short product life-cycles (Vernon 1966(Vernon , 1979). Yet, recent research within the electronics and semiconductor industry (Suarez-Villa and Rama 1996;Suarez-Villa and Karlsson 1996;Wever and Stam 1988) suggests that agglomeration linkages, and the formal outcomes of any informal information spillovers (Audretsch and Feldman 1996;Suarez-Villa and Walrod 1997;Arita and McCann 2000) extend over much larger spatial scales than that of the individual urban metropolitan area. In the case of multiplant multinational firms (Cantwell and Iammarino 2000), any such agglomeration effects may even operate over spatial scales larger than individual countries.…”

Social and Spatial Structures of Innovation in the Irish Animation Industry

“…Clearly, innovations in communication technology have reduced the need for face-to-face contact in the exchange of knowledge, even in the context of detailed technical design issues (Mckinnon, 1997;Torre & Rallet, 2005;Van Egeraat & Jacobson, 2006). In addition, where face-to-face contact is required, the number of contact occasions can be very limited and/or involve short periods that can often be satisfied via frequent long-distance travel and the seconding of research staff for extended periods of time (Arita & McCann, 2000;Torre & Rallet, 2005;Van Egeraat & Jacobson, 2006). 2 In addition, various forms of relational proximity can act as substitutes for geographical proximity and facilitate knowledge flow and distanciated learning in distributed teams (Boschma, 2005;Gertler, 2008;Moodysson, 2007;Torre & Gilly, 2000;Torre & Rallet, 2005).…”