How Do Health Systems Capitalize on Public Programs? Side Effects of the 340B Drug Pricing Program

Many government programs are designed to transfer resources to disadvantaged people and organizations that provide public services, but these programs often inadvertently create incentives for agents to exploit their provisions. Assessing how agents differentially game programs is essential to understand their incidence and correct market distortions. In this paper, I study how hospitals heterogeneously gamed the 340B Drug Pricing Program — a federal program intended to aid providers that treat low-income patients by requiring drug makers to sell drugs to participants at steep discounts. I focus on the role of health systems, which coordinate the business functions of numerous providers and may thereby facilitate passing 340B discounts on to drugs administered outside hospital walls. Using a staggered adoption design, I find that 340B increased hospitals' Medicare spending on cancer drugs by an average of $200,000 per year. Remarkably, this increase was entirely driven by health system-affiliated hospitals, which increased infusions by 72 percent. System hospitals increased medical oncologist employment only modestly, indicating that 340B did not lead hospitals to forge many new relationships with physicians through practice acquisitions. System hospitals also did not increase cancer screening or adopt new non-medical cancer treatments, indicating little effort to attract new patients. Instead, my analysis suggests that health systems necessarily have advantages in gaming programs like 340B, but resulting distortions may be substantially mitigated by regulation of billing practices.