Developing novel drugs

We provide evidence that risk aversion leads pharmaceutical firms to underinvest in radical andnovel innovation. We do so by introducing a new measure of novelty: a drug candidate is novelif it is molecularly distinct from prior candidates. Using our measure, we show that firms face arisk-reward tradeoff when investing in novel drugs: while novel drug candidates are less likely tobe approved by the FDA, they are based on patents with higher indicators of value. Consistentwith a simple model of costly external finance, a plausibly exogenous positive shock to firms' networth leads the development of novel (and riskier) drug candidates. This pattern suggests thateven large public firms behave as though they are risk averse, reducing their willingness to investin potentially valuable radical innovation