Network and Egocentric Uncertainty: Relationships Among Art Galleries in the Contemporary Art System

Abstract

Our paper addresses the macro-issue of the relationship between uncertainty and inter-organizational relations. We take the Podolny's theoretical framework which relates different types of market uncertainty to network structures of individual actors. In the sub-network composed by a focal actor and his network contacts (ego-network), structural holes exist when these contacts are (indirectly) connected through the focal actor, due to the absence of direct ties among them. According to Podolny's framework, high levels of egocentric uncertainty of focal actors increases the value of structural holes in their ego-networks. The brokerage ties of focal actors foster the identification of market opportunities and of information useful to address those opportunities. In contemporary art markets egocentric uncertainty is high for all the involved actors: artists, art galleries, museum curators, etc. The intrinsic heterogeneity of art works, the lack of objective criteria of evaluation, the difficult assessment of market needs, make the decision process of focal actors difficult for the lack of critical information. Uncertainty exists (i) for the artist, in the choice of the art work that he creates; (ii) for the art gallery, in the choice of the artists to be traded, their genre and style; (iii) for the museum curator in the choice of the artists and art works to be presented in a temporary exhibition or added to the museum's collection. Network relations with other actors (artists, curators, art galleries, etc) could be useful to access information about market opportunities and information about how to fill those opportunities. Podolny's framework suggests that in contemporary art markets structural holes are more useful to focal actors than in modern art markets, where egocentric uncertainty is significantly lower. We should expect a larger presence of structural holes in the former than in the latter. The empirical context of this research are the editions 2002, 2003 and 2004 of the international art fair Arte Fiera, annually held in Bologna. We observed the identity of the galleries that took part at each edition, and who were the artists whose works each gallery was presenting at the fair. We push further our earlier (mainly explorative) analyses of this network, and model its evolution across the observed period by using Tom Snijders' stochastic models of network change, implemented in his software SIENA. The latest implementation of the models make it possible to assess the co-evolution of the network with actor level variables (attributes of the actors). We test whether a tendency exists for galleries to cluster in cohesive network subgroups (clustering) or, on the opposite, to connect to other galleries not otherwise connected to each other (brokerage). Following our discussion of the implications of the greater uncertainty involved in trading contemporary art works, we test and find evidence of a distinctive tendency toward brokerage of contemporary art galleries.