Behavioural mechanisms of cooperation and coordination

This thesis consists of three independent chapters investigating behavioural mechanisms of cooperation and coordination. In particular, chapter 1 analyses a voluntary contribution game and proposes a simple behavioural mechanism to achieve social efficiency. Specifically, in this mechanism, each player can costlessly assign a share of the pie to each of the other players, after observing the contributions, and the final distribution of the pie is determined by these assignments. In a controlled laboratory experiment, I find that participants assign the reward based on others' relative contributions in most cases and that the contribution rates improve substantially and almost immediately with 80 percent of players contribute fully. Chapter 2 studies the effects of costly monitoring and heterogeneous social identities on an equity principle of reward allocation. The investigation is based on the mechanism proposed in chapter 2. I hypothesised that the equity principle may be violated when participants bear a personal cost to monitor others' contributions, or when heterogeneous social identities are present in reward allocations. The experimental results show that almost half of the allocators are willing to sacrifice their own resources to enforce the social norm of equity principle. Likewise, with the presence of heterogeneous social identities, though a few participants give more to their in-group member, the majority of them still follow the equity principle to allocate. Chapter 3 explores the behavioural mechanism of communication and leadership in coordination problems. Specifically, I consider two types of leaders: cheap-talk leaders who suggest an effort level, and first-mover leaders who lead by example. I use experimental methods to show the limits of these two mechanisms in avoiding coordination failure in a challenging minimum effort game, with low benefits of coordination relative to the effort cost. The results suggest that both types of leadership have some ability to increase effort in groups with no history, but are insufficient in groups with a history of low effort