Modeling Tiered Pricing Frameworks: A Simulation Approach.

Objective: Drug plans take different approaches to determining reimbursement prices for generic drugs. One common approach is to set the maximum reimbursement price as a percentage of the price of the interchangeable branded drug. In many countries this percentage depends on the number of generic entrants, a model we call "tiered pricing." This paper seeks to enhance understanding of how to set the tiers.

Methods: We construct a simple model of tiered pricing and set parameters to match evidence on generic drug costs and the distribution of revenues. Using simulation methods, we then assess different tier structures in terms of total surplus and average drug cost.

Results: We find when tiers are bunched tightly together welfare outcomes are poor. Moreover, there are large welfare gains from increasing the number of tiers from one to two, and only small welfare gains from increasing the number of tiers beyond four.

Conclusions: The choice of tiers has substantial welfare and cost implications. While it is possible to refine the simulation analysis based on specific market characteristics, an optimal tier structure, such as the one we propose in the paper, should have at least two tiers.