The value of the information in the Moral Hazard setting
Ishak Hajjej,Caroline Hillairet,Mohamed Mnif
2023-04-07
Abstract:This article studies the problem of evaluating the information that a Principal lacks when establishing an incentive contract with an Agent whose effort is not observable. The Principal ("she") pays a continuous rent to the Agent ("he"), while the latter gives a best response characterized by his effort, until a terminal date decided by the Principal when she stops the contract and gives compensation to the Agent. The output process of the project is a diffusion process driven by a Brownian motion whose drift is impacted by the Agent's effort. The first part of the paper investigates the optimal stochastic control problem when the Principal and the Agent share the same information. This situation, known as the first-best case, is solved by tackling the Lagrangian problem. In the second part, the Principal observes the output process but she may not observe the drift and the Brownian motion separately. This situation is known as the second-best case. We derive the best response of the Agent, then we solve the mixed optimal stopping/stochastic control problem of the Principal under a fixed probability and on the filtration generated by the Brownian motion, which is larger than the one generated by the output process (that corresponds to the information available for the Principal). Under some regularity conditions, the Principal value function is characterized by solving the associated Hamilton Jacobi Bellman Variational Inequality. At the optimum, we prove that the two filtrations coincide. Finally, we compute the value of the information for the Principal provided by the observation of the Agent's effort. It is defined as the difference between the principal value function in the first-best and second-best cases.
Optimization and Control,Probability