On incentive compatible designs of forecasting contracts

In the paper the optimal design of forecasting contracts in principalagent setting is investigated. It is assumed that the principal pays the agent (theforecaster) based on an announced forecast and an event that materializes next.Such a contract is called incentive compatible if the agent maximizes her payoffwhen she announces her true beliefs. This paper relaxes the assumption presentin earlier works on this subject that agent’s beliefs are deterministic by allowingthem to be random (i.e. stemming from estimation). It is shown that for binary ornominal events the principal can learn only expected values of agent’s predictionsin an incentive compatible way independent of agent’s signal space. Additionallyit is proven that incentive compatible payment schemes give the agent a strictlypositive incentive to improve the precision of her estimates