We report experiments designed to test the theoretical possibility, first discovered by Shapley (1964), that in some games learning fails to converge to any equilibrium, either in terms of marginal frequencies or of average play. Subjects played repeatedly in fixed pairings one of two 3 ´ 3...

Local interactions refer to social and economic phenomena where individuals' choices are influenced by the choices of others who are `close' to them socially or geographically. This represents a fairly accurate picture of human experience. Furthermore, since local interactions imply particular...

Suppose an entrepreneur needs funds from a financier to invest in a risky project whose cost is fixed, and whose return may be high or low. Suppose also that the project's realized return is an information that is private to the entrepreneur. If the amount the entrepreneur pays back to the...

This paper studies a principal-agent relationship in a contractual crime setting. Suppose an agent and a principal sign a contract stipulating some transfer of funds from one player (say the agent) to the next (the principal) contingent on the state of the world announced by the first player. In...

The regulation of environmental risks increasingly emphasizes the awareness and empowerment of stakeholders. The success of this approach, however, seems to depend crucially on the quality of environmental disclosures. In this paper we investigate the amount and quality of the information that...

The recent automobile liability insurance crisis in Atlantic Canada has prompted the four provincial legislations (Newfoundland and Labrador, New Brunswick, Nova Scotia and Prince Edward Island) to setup a task force to redesign, if necessary, the personal automobile insurance system. After...

This paper characterizes the optimal insurance contract in an environment where an informed agent can misrepresent the state of the world to a principal who cannot credibly commit to an auditing strategy. Because the principal cannot commit, the optimal strategy of the agent is not to tell the...

We experimentally test the ability of subjects to identify repeated-game strategies from prisoner's dilemma choice data. In the experiments, subjects use a finite state grammar to build models to fit game histories. The histories are designed to distinguish between strategies with empirical and...

We consider the effect of an increase in the risk from pollution. We show that in the case of a flow pollution, when the number of players is sufficiently large, the result of Bramoulle and Treich, showing that a marginal increase of risk in the neighborhood of a risk-free world is...

We experimentally study behavior in a common property renewable resource extraction game with multiple equilibria. In the experiment, pairs of subjects competitively extract and consume a renewable resource in continuous time. We find that play evolves over time into multiple steady states, with...