Managing Incentives in Drug Development

Abstract

Pharmaceutical innovation faces two persistent challenges: for tropical diseases, commercial incentives are too weak to support drug development, while for advanced and rare-disease treatments, regulators must make approval decisions under severe uncertainty and limited benchmarks. This dissertation adopts a mechanism design framework to analyze these distinct but related problems, aiming to enhance social welfare without undermining firms’ incentives to innovate.The second chapter examines funding mechanisms for pharmaceutical innovation in tropical diseases, where commercial incentives are often too weak to sustain adequate research and development. For instance, although malaria caused over 600,000 deaths in 2021, private investment in drug and vaccine development for malaria and similar diseases remains limited. Governments and nonprofits address these market failures through push (e.g., grants) and pull (e.g., prizes). We propose a third mechanism in which the funder only pays when the firm fails. That is, the funder reimburses the firm a share of its testing costs. This reimbursement for failure is optimal when the commercial market is large enough to reward success, but too small to induce investment. The optimal mechanism addresses adverse selection and moral hazard. For most tropical diseases, including malaria, we recommend pull funding with supplementary push support. Reimbursement is optimal for tuberculosis if testing costs are less than a billion dollars. These findings challenge current practices dominated by push funding and extend to funding innovations in other sectors.The third chapter analyzes regulatory policies for advanced and rare-disease treatments, where few benchmarks exist and confirmatory testing is costly. Regulators often allow firms to sell new drugs based on preliminary efficacy evidence, with final approval contingent on confirmatory testing. We characterize optimal approval policies when firms have private information about testing costs and the regulator’s payoff depends on expected efficacy. The optimal policy may include partial conditional approval for drugs with low expected efficacy, and leniency in granting final approval below conventional standards. Our calibration suggests these tools can generate hundreds of millions of dollars in annual social value.