Speaker(s):  Lee Fleming (HBS)


Title:        Why the Valley Went First: Agglomeration and Emergence in Regional Inventor Networks
    
Author(s):  Lee Fleming, Lyra Colfer, Alexandra Marin, Jonathan McPhie

Abstract
Are the inventor networks of Silicon Valley more densely connected than those of the Boston Route 128 corridor?  The evidence remains mostly historical and controversial to date.  We develop an analysis of all the patented inventors in both regions since 1975 and find that the networks of Silicon Valley are simultaneously more connected and less robust than those of Boston.  Of greatest interest, Silicon Valley demonstrates a dramatic agglomeration of its inventors, such that half of them can trace an indirect path to one another through co-authors by 1999.  Boston, despite a very similar number of patents, inventors, technologies, firms, and overall density of ties, agglomerates later and even today lags Silicon Valley.  This process of emergence of a "giant component" occurs through the linking of a region's larger components.  Based upon interviews with inventors who did and did not create linking ties across a region's components, we identify a variety of similarities and differences in the agglomeration and non-agglomeration processes of co-authoring networks across the two regions.  While our limited sample found more reports of information flow across firms in the Valley, inventors reported very similar experiences and attitudes in the two regions.  Ultimately, we find an institutional explanation for a large portion of the Valley's advantage: a single post-doctoral fellowship program at IBM's Almaden Valley Labs was responsible for up to 30% of the region's initial agglomeration process.