History-Dependent Demand and Intermediaries: Explaining Prescription Drug Pricing Dynamics

I show that US prescription drug prices can be better understood by accounting for history-dependent demand and industry pricing structure, which has important implications for potential government negotiation. First, I confirm the presence of inertia and history-dependent demand for generic drugs through a novel quasiexperiment. Next, I capture pricing structure in a dynamic pricing game involving drug companies and an intermediary that designs drug insurance, and estimate it using novel net price data from the anti-cholesterol market. Counterfactuals show that intermediaries reduce expenditures by over 20% through their ability to exclude drugs from insurance, but capture a significant fraction of the savings. Government negotiation would need to have a similar threat to curb the inflationary pressure from history-dependent demand. JEL Codes: L11, L65, I12, H51 ∗I would like to thank my advisers Ariel Pakes, David Cutler, and Josh Lerner. In addition, I would also like to acknowledge Ernst Berndt, Lauren Cohen, Edward Glaeser, Robin Lee, Alex Mackay, Luca Maini, Anders Munk-Nielsen, and Heidi Williams for providing detailed comments and suggestions throughout the research process. Next, I would like to thank various industry contacts for providing valuable background information and data: Ryan Baum and SSR Health for providing data and suggestions on net price construction; Bob Nease and Edmund Pezalla for insights into PBM formulary decisionmaking. Finally, the research also benefited from comments from participants in the Harvard Industrial Organization Lunch, the Harvard Labor/Public Lunch, the Harvard Innovation and Entrepreneurship Brownbag, and the NBER Productivity Seminar.