Knowledge spillovers: The role of geography, technology and intra -firm linkages in the global semiconductor manufacturing industry.

Abstract

Geographic collocation of firms facilitates flow of knowledge spillovers between industry peers. Yet, why firms vary in their ability to use existing geographic locations to acquire external knowledge is not well understood. In an attempt to examine the process, this dissertation relaxes two simplifying assumptions of prior work. First, I allow for the possibility that firms may be multi-unit which introduces complexity to the process. The complexity arises from the simultaneous existence of multi-unit firms in different geographic and different technology domains. Multiplicity of units brings not only opportunities but also challenges in the absorption of knowledge spillovers. Multi-unit firms are able to overcome the challenges of geographic dispersion only when integrative mechanisms, or intra-firm linkages, that allow knowledge sharing across locations, exist within the firm. Second, since the flow of knowledge spillovers is caused by the interaction of firms, hence there is a need to model both the role of the recipient (focal) and source (spillor) firms. Characteristics of the spillor determine where the focal firm is most likely to acquire knowledge spillovers. Finally, characteristics of the dyad of the focal and the spillor firms also contribute to the process. I test the hypotheses in the context of the global semiconductor manufacturing industry. The sample consists of 100 firms over a period spanning 1990--1997. I find strong support for my hypotheses to indicate that geographic concentration of units enhances firm's ability to acquire knowledge spillovers. However, for firms that are geographically dispersed, intra-firm linkages are crucial in the process of absorption. Similarly, intra-firm linkages of the spillor enhance the distributiveness of proprietary knowledge to the focal firm. Additionally, the impact of intra-firm linkages of the focal firm is significantly greater than the impact of intra-firm linkages of the spillor firm, in the flow of spillovers in firm dyads. Thus, firms are benefited overall by such intra-firm linkages. This research enhances our understanding of the role of firm boundaries on flows of knowledge spillovers. The sharing of distributed knowledge within the firm has implications for firms' ability to leverage, not just local but also geographically distant resources for maximum advantage.