Goldilocks and the cleaning companies: Collective coordination among contract cleaning companies

Geographical proximity between firms is often believed to favour cooperation, mutual learning, knowledge creation as well as innovation. From Marshall onwards, study after study has demonstrated that both cooperation and successful innovations arise from geographically proximate clusters. Geographers have therefore argued that ‘space matters’. However, all too often, a direct line was drawn from geographical proximity to the assumption of cooperation, or from the existence of cooperation to the assumption of innovation without specifying the links between space and innovation. This inevitably lead to a series of papers pointing out that proximity was not enough, or as Torre and Rallett put it “what does ‘being near’ someone mean” (2005: 48). Therefore space may be a necessary condition but it is not sufficient, ‘something else’ also must play a role (Gilly and Wallet 2001). This was further followed by a series of papers highlighting that it may not be even a necessary condition and other factors count as much as spatial proximity. We seem to have gone full circle, from ignoring the spatial element in economic development, through arguing that it is crucial, to arguing that it is not so important. A recent issue of Regional Studies (January 2005) was devoted to exploring this issue and the role proximity that plays in the cooperation between firms and their subsequent innovation performance. Expanding on the French school, which holds that geographical proximity is only one form of proximity, and that organisational and institutional proximity are important, this issue reopens the proximity/innovation debate. Indeed Boschma (2005) argues that both cognitive and social proximity should be added to the analysis of cooperation as well as innovation. We seek to add to this debate by exploring a geographically proximate cluster which does not cooperate. The cluster is the contract cleaning industry in Dublin which not only fails to innovate but acts in ways that are damaging to their individual and collective interests. Following Boschma (2005) we look at five dimensions of proximity: cognitive, organisational, social, institutional and geographical, to understand why a geographical cluster ‘does not cluster’. The paper draws on an EU funded research project ‘CRITICAL’.