Strategic Framing in Contracts

Abstract

We provide a hidden-action principal-agent model where the agent has referencedependent preferences. The loss-averse agent considers the base wage as reference point, and bonuses and/or penalties as gains and losses, respectively. When choosing optimal payments, the principal strategically sets the base wage, knowing that this determines the agent's reference point. We consider two variants of the model. In a first variant, the agent's reservation utility is not reference-dependent. We show that it is always optimal in this case for the principal to employ bonuses. In a second variant, the reservation utility is reference-dependent and the principal may use penalties.