Moral hazard severity and contract design
Ronald A. Dye, Swaminathan Sridharan
Abstract
Ronald A. Dye, Swaminathan Sridharan
Abstract
In an agency setting where the agent must be compensated both to exert effort to produce a new project and to announce honestly when the new project has been produced, we show that Holmstrom's (1979) well-known informativeness criterion does not, by itself, determine whether a variable is optimally incorporated into the agent's contract. What also matters is how severe the control problem is between the principal and the agent. We further show that the severity of the moral hazard problem also determines whether it is desirable for the principal to have the agent implement the project more often that warranted by first-best implementation considerations.
OpenAlex reports 11 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In an agency setting where the agent must be compensated both to exert effort to produce a new project and to announce honestly when the new project has been produced, we show that Holmstrom's (1979) well-known informativeness criterion does not, by itself, determine whether a variable is optimally incorporated into the agent's contract. What also matters is how severe the control problem is between the principal and the agent. We further show that the severity of the moral hazard problem also determines whether it is desirable for the principal to have the agent implement the project more often that warranted by first-best implementation considerations.
Key concepts: Moral hazard, Agency (philosophy), Principal (computer security), Principal–agent problem, Control (management), Hazard, Business, Variable (mathematics)