The Incidence of the 340B Program: The Effects of Contract Pharmacies on Part D Premiums and Reimbursements

Empirical evidence has shown in many contexts that taxes levied against producers are passed through to consumers in the form of higher prices. This dissertation asks whether a "tax" on drug manufacturers is passed through to insurers and patients enrolled in their plans. The 340B Program effectively taxes drug manufacturers by requiring them to provide large discounts on drugs purchased by safety-net providers. I assess to what degree insurer premiums are impacted by drug manufacturer pass-through of 340B discounts by exploiting a major expansion in 340B discounts claimed through contract pharmacies. Historically, a 340B provider could claim discounts for prescriptions its patients filled at a single pharmacy with which it contracted-- a so-called contract pharmacy. In 2010, a policy change permitted 340B providers to partner with an unlimited number of pharmacies causing exponential growth in the number of contract pharmacies and a massive expansion in the number of prescriptions eligible for discounts. Between 2010 and 2021, the number of contract pharmacies increased from fewer than 5,000 to 30,000, or one out of every two pharmacies. Using detailed claims and plan-level data for Medicare Part D, I use a two-way fixed effects strategy to estimate the pass-through of 340B discounts to premiums. This strategy leverages changes to the proportion of an insurer's drug expenditures that may be subject to a 340B discount after the 2010 policy change. There is significant geographic variation in contract pharmacy growth across areas over this time period, and insurers are differentially affected by the policy based on their initial enrollment shares across areas. In a second complementary strategy, I use variation in contract pharmacy growth that occurred after the 2014 Medicaid expansions. Both strategies yield evidence that premiums increase in response to expansions in 340B discounts. I find that reducing an insurer's spending that is potentially subject to a 340B discount by 10 percentage points reduces monthly premiums by $3.33 (approximately 6.4%). This finding is consistent with the economic theory of pass-through, but challenges the commonly held presumption by 340B advocates that the 340B Program comes at no cost to taxpayers.