This paper examines how patents affect the direction of pharmaceutical R&D. By capping the period of profits earned, patents create a wedge between private and long-term social value of R&D. This paper shows that this wedge is heterogeneous across diseases, with larger discrepancies for longer duration diseases. I introduce a novel method for quantifying the social and patent-protected private value of R&D based on changes in expected lifetime utility, which I estimate for an exhaustive panel of medical conditions affecting the US population. Employing a shift-share instrumental variable (IV) approach, I estimate the elasticity of R&D with respect to demand in the US from 1990 to 2014. I find that the private-social wedge dampens the responsiveness of R&D to demand changes. The consequence of this is an underinvestment in chronic diseases, which are increasingly prevalent in the US.

This paper investigates the link between technological progress and economic growth in the United States in post-World War II data. To do so, we employ a novel, technology field-specific measure of technological progress constructed using patent text and work task descriptions. Using a VAR framework, we estimate the effects of technology on GDP and other macro variables. The empirical model provides a technological explanation for three salient episodes in US GDP growth: the slowdown starting in the 1970s, the acceleration of the 1990s-2000s, and the post-2000 slowdown. Engineering technology is the main driver of growth overall, followed by IT, which took over as the main source of growth in the 1990s. We also show that engineering technologies disproportionately benefit the income of the bottom 90%, while IT technologies benefit the top 10%, explaining part of the post-1990 increase in US income inequality.