Clustering and firm performance in project-based industries: the case of the global video game industry, 1972-2007

Abstract

Explanations of spatial clustering based on localization externalities are being
\nquestioned by recent empirical evidence showing that firms in clusters do not
\noutperform firms outside clusters. We propose that these findings may be driven by
\nthe particularities of the industrial settings chosen in these studies. We argue that
\nin project-based industries, negative localization externalities associated with
\ncompetition grow proportionally with cluster size, while positive localization
\nexternalities increase more than proportionally with cluster size. By studying the
\nsurvival patterns of 4607 firms and 1229 subsidiaries in the global video game
\nindustry, we find that the net effect of clustering becomes positive after a cluster
\nreaches a critical size. We further unravel the subtleties of the video game industry
\nby differentiating between exits by failure and exit by acquisition and conclude
\nthat being acquired is best considered as a sign of success rather than as a business
\nfailure.