Essays on Asymmetries in Contest

This thesis is concerned with the effects of asymmetries in ability and social preferences in contests and conflict networks. Standard models find that asymmetries monotonically decrease total and individual efforts. I demonstrate that this result does not necessarily hold when players are embedded in complex networks, have preferencesregardingthefairnessofthecontestortheoutcomesofothers, and when real subjects play these games in the lab. Chapter 1 formulates a network of bilateral contests in which locally unique equilibria always exist, and global uniqueness is possible. I find that an increase of one player’s ability can increase her effort and the effort of the entire network. If one player targets a specific opponent, other players follow. Chapter 2 imposes a budget constraint on this model. Most findings are robust to this modelling choice. This allows an investigation of topics like the effects of heterogeneity on team performance and the effect of asymmetries in the number of conflicts a player and her rivals are involved in. Chapter 3 documents that there exists no agreed way for implementing social preferences in contests. I derive four possible versions and critically assess their properties. When costs are considered, the magnitude of predicted overspreading and overbidding is reduced. Mild asymmetry can result in higher effort from the high ability player. In chapter 4, I present a pilot experiment in which social identities, with and without a hierarchy, are induced. We find that identities with such a hierarchy can trigger more aggressive play. To structure these findings, I suggest a foundational model of social preferences that relates them to social identity, where ‘close’ players are treated with altruism and ‘distant’ players are treated with spite