Carolina Mattsson (CentAI)

Informal meetings on complexity field

Network Growth Under Opportunistic Attachment

Abstract: Growing network models can potentially be a useful tool in developing economic theory. This talk introduces an “opportunistic attachment” mechanism where incoming nodes, in deciding where to join a network, consider features of the entry points available to them. For example, an entrepreneur looking to start a thriving business might consider the projected revenue of many hypothetical businesses. Here we explore this mechanism, in isolation, via a minimal model where PageRank serves to score the opportunities available to incoming nodes. The model gives rise to rich node dynamics, path-dependence, and an unexpected degenerate structure. We go on to suggest that this model might serve as a maximally stylised model of entrepreneurial growth. Central to our argument is an alternative notion of economic equilibrium introduced in Leontief and Brody (1993) based on the steady state of a random walk. To the extent this argument holds, simulations of our minimal model suggest that opportunistic attachment would be an interesting mechanism for relating the structure of an economic system to its future growth. Moreover, entrepreneurs in the generalized version of the model would face a shifting “opportunity space” where the number of potential business opportunities is effectively unbounded, a scenario amenable to evolutionary selection mechanisms.