Incentivizing Truth Exploration and Honest Reporting: A Contract Design Approach

In this paper, we study a Principal-Agent problem in which a principal incentivizes an agent by establishing a payment contract that encourages the agent to exert costly effort in exploring the true state of the environment, which is of interest to the principal, and then report the findings. We consider two feedback setups: (1) the true state is ultimately observable by the principal, and (2) only some noisy feedback related to the true state is observable. In the first setup, we demonstrate that the optimal contract is the one that pays the agent only when the report matches the ground truth, and we derive an efficient algorithm to compute this optimal contract. In the second setup, we design a BDD contract and show its approximate optimality with respect to the optimal honest-reporting incentivizing contract, both theoretically and empirically. Furthermore, we introduce a sufficient condition under which the optimal contract encourages honest reporting.