Optimal incentive systems for different reward schemes within the Principal-Agent-Theory

In this paper the authors discuss the determination of optimal incentive systems for two types of reward schemes represented by a increasing linear and a piecewise constant function. Contrary to the usual assumption in the Principal-Agent-Theory of a unique realisation of a project they analyse the particular situation that the repeated realisation of the project is possible. It is shown that in the case of a riskaverse agent and a riskneutral principal the piecewise constant function can offer advantages over the increasing linear function.