Whenever a group of agents interact a leader-follower relationship almost always emerges. To understand, therefore, how leadership emerges is fundamental in analysing strategic behaviour and interaction within groups. It can provide insights, for example, on fashions and trends in consumer purchasing, strategies of financial traders, the success and failure of business leaders, pricing in non-cooperative oligopoly or the evolution of a cartel. Our understanding, however, of the processes by which leaders emerge is still incomplete. The proposed research has three aims. We will develop a theoretical model of leadership, test it experimentally, and apply it to various economic situations. The model will be one of sequential decision making where agents choose when to act. A leader is someone who chooses to choose first while a follower is someone who chooses to 'wait and see'. Leading is potentially risky because an agent who leads is only successful if he attracts followers. Incentives to lead will therefore depend on both the benefits to successful leadership and a judgement of the probability of becoming a successful leader. The model will predict an 'optimal waiting time' before an agent should lead, if no one has yet done so. This will provide clear testable predictions.

Laboratory economic experiments.