Introduction

Firm competencies are discussed in the literature as important sources of firms' competitive advantage. They develop over time and, together with specific routines and communication mechanisms internal to the organization, affect the direction and the outcome of firms' R&D activities (see, for example, Nelson and Winter, 1982, and Dosi et al., 1988).

Recently, however, the economic literature has highlighted the importance of an alternative model for organizing production and innovative activities: the geographical proximity among researchers and institutions in a technological cluster. The contributions on regional clustering and geographically localized knowledge spillovers argue that the cost of transferring knowledge increases with the geographical distance among the parties involved in the exchange. This explains the tendency of innovative activities to locate together (see, for example, Jaffe, 1986; Jaffe, Trajtenberg, and Henderson, 1993; Audretsch and Feldman, 1996; and Swann, Prevezer, and Stout 1998).

So far the literature has analyzed these issues separately. This chapter recombines these two streams of the literature and compares their relative importance in explaining a research output: the value of innovations. It estimates how much of the value of an innovation is affected by the affiliation of the inventors to the same organization, as opposed to spillovers that arise when the inventors are geographically close to each other and to external sources of knowledge. In so doing it expands upon a previous paper by Mariani (2004).