Working Paper: NBER ID: w22720
Authors: Mark Pauly; Kyle Myers
Abstract: This paper examines trends in the aggregate productivity of the pharmaceutical sector over the past three decades. We incorporate Ricardo’s insight about demand-driven productivity in settings of variable scarce resources, and estimate the industry’s responsiveness to changes in demand over this timeframe using therapeutic class-specific data. In contrast to many analyses, our empirical estimates indicate that the industry has “met demand” with remarkable consistency since the late-1980s. The growth in total R&D spending, and therefore R&D costs per new drug, appear to have been profitable and productive investments. While we identify a significant increase in the industry’s fixed costs - the intercept of the production function - we find no decline in the marginal productivity of industry investments that might suggest significant supply-side frictions. While we cannot diagnose in detail why average, but not marginal, productivity declined, the data suggests that firms have finally begun to compete down returns from the supranormal levels of decades past.
Keywords: Pharmaceutical R&D; Demand-driven productivity; New molecular entities
JEL Codes: D20; I11; L10; L65; O31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| demographic changes (J11) | potential demand (R22) |
| potential demand (R22) | private R&D investments (O32) |
| private R&D investments (O32) | outputs of NMEs (E01) |
| demographic changes (J11) | private R&D investments (O32) |
| NIH investments (H54) | outputs of NMEs (E01) |
| private R&D investments + NIH investments (O32) | outputs of NMEs (E01) |