The Effect of On-net/Off-net Differentiation and Heterogeneous Consumers on Network Size in Mobile Telecommunications An Agent-based Approach

I explore the effects of on-net / off-net differentiation on network sizes in mobile telecommunications when both rational and non-rational consumers coexist in the market. In particular, three different types of consumers are modeled: (1) fully informed rational (FIR) consumers who are perfectly informed about the true market shares of all networks and choose the network with the lowest expected cost of a call; (2) partly informed rational (PIR) consumers who only observe market shares within a circular sensing field and choose the network with the lowest expected cost of a call based on these observed market shares; and (3) non-rational (NR) consumers who choose the network with the highest market share among their immediate neighbors. Using an agent-based simulation approach and by systematical variation of four key parameters of the model, three key results emerge. First, if the share of FIR consumers is too high, all consumers will eventually join the initially larger network A. Second, if their share in the population is sufficiently large, NR consumers can prevent the growth of clusters of consumers subscribed to network B. Third, if the share of PIR consumers is high, clusters of consumers subscribed to network B can grow, thereby increasing network B’s market share, provided that the radius of their circular sensing field is small enough for the cluster size. JEL codes: C63, K23, L14, L96