Contracting with a naïve time-inconsistent agent: To exploit or not to exploit?

• Repeated principal–agent problem with moral hazard over three periods.
• The agent is time-inconsistent with βδβδ-preferences and is naïve.
• There are no information rents for the principal from naïveté of the agent.
• Principal is indifferent between facing a naïve or a sophisticated agent.

In a repeated principal–agent model with moral hazard, in which the agent has βδβδ-preferences, we analyze the case where the agent is naïve in the sense that he is not fully aware of his inconsistent discounting. We consider the possibility of principal manipulating the naïveté of the agent. Surprisingly, when the principal wants to implement the high effort, there are no gains to the principal from the naïveté of the agent and the principal does not choose to deceive the agent. The principal’s maximum utility is the same from a sophisticated agent and from a naïve agent.