UiS Business School Seminar: Popularity and Tolerance in Networks

Abstract

Individuals (political parties) often differ in their identities (ideologies), i.e., the behaviors they consider ideal for themselves and others. They only want to interact (or link) with others whose behavior/action are sufficiently close to their ideals, and they are willing to compromise, i.e, choose costly actions to get closer to others, in order to be accepted by them. Based on the paper by Genicot (2021) on tolerance and compromise we construct a two-stage model in which agents first choose actions and a secondly, a pairwise stable network forms. The novelty of our approach is that tolerance levels depend endogenously on the degree of agents in the network. In particular, the more popular an individual is the more she/he is tolerated by others. We characterize the case with three agents (left, middle and right ideal points) and find that compromising takes quite different forms, which crucially depend on the agents’ beliefs with respect to the network. We consider two extreme scenarios regarding such beliefs: pessimism and optimism. We find that compromise behavior exists in both cases but not necessarily for the same parameter configurations. In particular, in the former case compromise exist for lower compromising costs than in the latter. When considering a larger set of individuals, we find that core-periphery networks as well as polarized societies might coexist.