Abstract

Wealth distribution in an economic system is studied by means of an agent model, where agents have a certain spending propensity and they interact over a given network. When the network is random, or scale-free (∼ k - α) with α below 1, approximately, results are equivalent to having all agents allowed to interact with any other agent. However, values of α > 1 affect both the wealth distribution and the behavior at the tail. These results hold both in the absence of spending propensity and when the spending propensity follows a power-law. Results suggest that Pareto's law is a very robust phenomenon with respect to the details of the connectivity of the agents and that the ubiquity of Pareto's law in actual systems may have implications on the topological properties of the underlying networks of interaction.