PAYMENT STRUCTURES IN THE MEDICAL COMMUNITY: AN EXPERIMENTAL STUDY

Purpose: In this study we focus on the dual principal-agent problem in which agents have other-regarding preferences. In the dual principal-agent problem an agent has responsibilities to two different principals. For instance, a physician is an agent of their patients and an agent of their health care employer. A critical feature of this study is that there is an interior solution for the number of services provided to the client (the patient).  While the agent is assumed to know this critical value, the employer only knows what the average value should be over all downstream principals, and the downstream principal, or client, has very little information.

Method: Our study uses experimental methods to analyze the relative performance of a variety of compensation contracts. This study creates an environment in which individuals are paid via common payment structures employed in the medical community (Fee For Service, Capitation, Salary, etc.) to study the effect that different incentive structures have on agent behavior.  We specifically address the effect that other-regarding behavior has on decision making in the physician-patient relationship.

Result: Our results suggest that the existence of other-regarding behavior substantially affects choices made by agents and shows that some compensation contracts outperformed others on a variety of measures. These results also suggest that classic contract theories were misleading. In this study, the payment structure that resulted in the highest quality service was the Salary contract; where in classic theory a salaried agent would under provide services relative to the other compensation contracts.

Conclusion: Standard contract theory emphasizes the need for a direct link between the final product provided by the agent and their compensation for services. However, this study brings to light that this may not be necessary to achieve a high quality of service in the dual-principal agent problem. When taking into account the significant effect that other-regarding behavior has on an individual’s utility, agents do not need to be directly incentivized to provide high quality services.