Over time a multitude of studies have investigated geographical proximity’s relevance between collaborators, ways in which collaborators agglomerate or ‘cluster’, their means of interaction for knowledge transfer, learning, innovation, and general economic outcomes. In this context, numerous studies found that geographical proximity benefits collaboration, as knowledge ‘spillovers’ are localized (e.g. Camagni 1991; Storper 1995; Malmberg and Maskell 2002). Other studies have challenged this (e.g. Breschi and Lissoni 2001): knowledge does not automatically ‘spill over’ between agents who happen to be co-located because any kind of knowledge transfer results from ‘. . .intended and specific actions of individual agents trying to interact, to exchange information or knowledge, to share common values, or to cooperate’ (Hussler and Rondé 2007: 288). Hence, geographical proximity is not enough to facilitate spillovers, which raises the question of what are the factors that do.