Agency Theory: Designing Optimal Incentives in the Insurance Sector

One of the main mechanisms when it comes to influencing autonomous agents is the establishment of incentives, or what is the same, defining rewards (for example, monetary) based on the achievement of certain objectives. The design of those incentives is one of the fields of study grouped on the so called “principal-agent problem”. The principal-agent problem occurs when one entity (the “agent”), is able to make decisions and/or take actions on behalf of another person or entity (the “principal”). The agent earnings are regulated under a contract designed by the principal and under the principal point of view, the main goal while designing that contract and the payment rules (commissions and incentives) incorporated on it is to align the actions made by the agent to his own goals. A poorly designed incentive could lead to economic loss or or be unreachable for the agent. In this paper we will describe a method to check the goodness of those incentives using mixed integer linear optimization.