Dissertation Title:  "Essays on Pharmaceutical and Health Insurance Markets"Pharmaceutical policy must balance two objectives: promoting new drug development and pricing drugs to ensure access. Though pharmaceutical innovation has made significant contributions to health and social welfare, drug costs are rapidly increasing. This combination makes it essential to understand how both regulation and market dynamics shape pharmaceutical innovation, coverage, and prices.

Chapter 1 studies pharmaceutical firm decisions on the timing of follow-on product introductions. Follow-on drugs, termed line extensions, receive a fixed exclusivity period that starts upon approval. Firms can choose to introduce a line extension earlier to attract new consumers, or delay introduction so the line extension’s exclusivity extends beyond that of their original drug product. I show that the firm’s incentive for delayed introduction increases with the share of line extension sales that would cannibalize sales of the original drug. I test this prediction using a novel dataset of over 700 pharmaceuticals approved in the United States from 1985-2016, linked to all subsequent line extensions in that period. Consistent with strategic delay, an original product is almost twice as likely to have a line extension approved in the period leading up to expected generic entry than in the three or more years prior. Using Monte Carlo simulations, I find that line extensions that are more cannibalizing are delayed up to 2.5 years, compared to an average of five months for those that are less cannibalizing. Delays in the introduction of new products can create welfare losses for consumers and payers, and I consider implications for optimal innovation policy.

Chapter 2 (with Thomas McGuire) develops a diagrammatic model of the static and dynamic tradeoff in pharmaceutical markets. Using this framework, we examine the role of different health care financing institutions, specifically, health insurance for drugs and pharmacy benefit management firms (PBMs) on both static and dynamic efficiency. Insurance can help on both margins by decoupling the price paid by the consumer and the price received by the pharmaceutical firm. In what we refer to as a “competitive” PBM, discounts are fully passed on to health plans. In this case, the PBM has no effect on the tradeoff between dynamic and static efficiency but may affect the equilibrium. By contrast, a PBM with market power keeps some or all discounts, degrading the tradeoff. If PBM profits come from pharmaceutical firms, these firms have less incentive to invest in innovation. If PBM profits come at the expense of higher prices to consumers, efficient drug consumption is curtailed. Our framework emphasizes how prescription drug policy reforms can have effects, either intended or unintended, on both innovation and access.

Chapter 3 studies variation in Medicare Part D prescription drug coverage. Part D was designed to increase drug competition by allowing plans and PBMs to selectively contract with drug manufacturers. Though Part D plans must be at least actuarially equivalent to an annual “standard benefit,” PBMs and plans have flexibility on what products they include in their benefit and at what cost sharing levels. To compare plan coverage, I define three measures of class-level drug insurance generosity. Using Part D plan benefit data for calendar year 2016 and novel data on the PBMs each plan was associated with, I describe Part D coverage for ten therapeutic classes, including protected classes, classes with no generics, and classes where products are more likely to be close substitutes. I discuss findings through the lenses of coverage generosity, clinical adequacy, and PBM differences.