Productivity, Innovation, and Entrepreneurship

March 20, 2015
Nicholas Bloom of Stanford University and Josh Lerner of Harvard University, Organizers

Prithwiraj Choudhury and Tarun Khanna, Harvard University

Ex-ante Information Provision and Innovation: Natural Experiment of Herbal Patent Prior Art Adoption at the USPTO and EPO

Choudhury and Khanna study how ex-ante information provision, in the form of codified prior art, affects innovation outcomes. Using a unique dataset of herbal patents filed on the United States Patent Office (USPTO) and European Patent Office (EPO) from 1977 to 2013, the researchers exploit a natural experiment where the USPTO and EPO adopted a codified database of traditional herbal medicine prior art at different points in time. This database (the 'Traditional Knowledge Depository Library' or TKDL) was created by Indian state-owned R&D labs and provided the USPTO and EPO patent examiners with codified, searchable prior art on herbal formulations based on a translation of ancient Indian medicinal texts. The authors establish that the time lag of the USPTO adopting TKDL compared to the EPO was related to idiosyncratic differences in how the agreements were structured and negotiated, not differences in policy toward herbal patents at the USPTO and EPO. The researchers find that the adoption of TKDL affects the level of herbal patent filing and grants. It also shifts the composition of patenting away from pure herbal formulations that are similar to prior art available in the ancient texts towards applications involving both herbs and synthetic compounds, which are more distant from the prior art and arguably less contestable. The authors also use unique data coded from patent image wrappers at the USPTO and validate the smoking gun that prior art codification affects the search strategies of patent examiners.


Ariel Dora Stern, Harvard University

Innovation under Regulatory Uncertainty: Evidence from Medical Technology

This paper explores how the regulatory approval process affects innovation incentives in medical technologies. While prior studies of medical innovation under regulation have found an early mover regulatory advantage for drugs, Stern finds the opposite to be true for medical devices. Using detailed data on over three decades of high-risk medical device approval times in the United States, she shows pioneer entrants spend approximately 34 percent (7.2 months) longer in the approval process than the first follow-on innovator. Back-of-the-envelope calculations suggest that the opportunity cost of capital of a delay of this length is upwards of 7 percent of the total cost of bringing a new high-risk device to market. Stern considers how different types of regulatory uncertainty affect approval times and find that a product's technological novelty is largely unrelated to time spent under review. In contrast, uncertainty about application content and format appears to play a large role: when objective guidelines for evaluation are published, approval times quicken for subsequent entrants. Finally, Stern considers how the regulatory process affects firms' market entry strategies and find that financially constrained firms are less likely to enter new device markets as pioneers.


Lee G. Branstetter, Carnegie Mellon University and NBER; Chirantan Chatterjee, Indian Institute of Management Bangalore; and Matthew Higgins, Georgia Institute of Technology and NBER

Starving or Fattening the Golden Goose? Generic Entry and the Incentives for Pharmaceutical Innovation (NBER Working Paper 20532)

Over the last decade, generic penetration in the U.S. pharmaceutical market has increased substantially, providing significant gains in consumer surplus. What impact has this rise in generic penetration had on the rate and direction of early stage pharmaceutical innovation? Branstetter, Chatterjee, and Higgins explore this question using novel data sources and an empirical framework that models the flow of early-stage pharmaceutical innovations as a function of generic penetration, scientific opportunity, firm innovative capability, and additional controls. While the aggregate level of early-stage drug development activity has increased, their estimates suggest a sizable, robust, negative relationship between generic penetration and early-stage pharmaceutical research activity within therapeutic markets. A 10 percent increase in generic penetration is associated with a 7.9 percent decline in all early-stage innovations in the same therapeutic market. When the researchers restrict their sample to first-in-class pharmaceutical innovations, they find that a 10 percent increase in generic penetration is associated with a 4.6 percent decline in early-stage innovations in the same market. The estimated effects appear to vary across therapeutic classes in sensible ways, reflecting the differing degrees of substitution between generics and branded drugs in treating different diseases. Finally, the researchers are able to document that with increasing generic penetration, firms in their sample are shifting their R&D activity to more biologic-based (large-molecule) products rather than chemical-based (small-molecule) products. The researchers conclude by discussing the potential implications of their results for long-run welfare, policy, and innovation.

David C. Chan, Jr, Stanford University and NBER

The Efficiency of Slacking Off: Evidence from the Emergency Department

Work schedules play an increasingly important role as production becomes more time-sensitive and utilizes workers more interchangeably. Chan finds two types of strategic physician behavior near end of shift (EOS) in emergency department shift work. First, on an extensive margin, physicians accept fewer patients near EOS ("slacking off"). Second, on an intensive margin, physicians distort patient care, incurring higher costs as they spend less time on patients arriving near EOS. Chan demonstrates a tradeoff between these two strategic behaviors, by examining how they change with shift overlap. Accounting for both costs of physician time and patient care, he finds that physicians slack off at approximately second-best optimal levels.


Paul Gompers, Harvard University and NBER; Steven N. Kaplan, University of Chicago and NBER; and Vladimir Mukharlyamov

What Do Private Equity Firms Say They Do?

Gompers, Kaplan, and Mukharlyamov survey 79 private equity (buyout) investors with a total of over $750 billion of assets under management about their practices in firm valuation, capital structure, governance and value creation. Few investors use discounted cash flow or present value techniques to evaluate investments. Rather, they rely on internal rates of return and multiples of invested capital. They also use comparable company multiples to calculate exit values rather than discounted cash flow valuations. Private equity investors typically target a 25 percent internal rate of return on their investments. They also report that their limited partner investors focus more on absolute, not relative performance. Capital structure choice is based equally on optimal trade-off and market timing considerations. Private equity investors anticipate improving the performance of the companies in which they invest, with a greater focus on increasing growth than on reducing costs. They devote meaningful firm resources to do this. The researchers also explore how the actions that private equity managers say they take group into specific firm strategies and how those strategies are related to firm founder characteristics.


Benjamin Pugsley and Aysegul Şahin, Federal Reserve Bank of New York

Grown-up Business Cycles

Pugsley and Şahin document two striking facts about U.S. firm dynamics and interpret their significance for aggregate employment dynamics. The first observation is the steady decline in the firm entry rate over the last thirty years, and the second is the gradual shift of employment from younger to older firms over the same period. Both hold across industries and geography. The researchers show that despite these trends, firms' lifecycle dynamics and their business cycle properties have remained virtually unchanged. Consequently, the reallocation of employment towards older firms results entirely from the cumulative effect of the 30-year decline in firm entry. This "startup deficit" has both an immediate and a delayed (by shifting the age distribution) effect on aggregate employment dynamics. Recognizing this evolving heterogeneity is crucial for understanding shifts in aggregate behavior of employment over the business cycle. With mature firms less responsive to business cycle shocks, the cyclical component of aggregate employment growth diminishes with the increasing share of mature firms. At the same time, the trend decline in firm entry masks the diminishing cyclicality in contractions and reinforces it during expansions, which generates the appearance of jobless recoveries where aggregate employment recovers slowly relative to output.


Antoine Dechezleprêtre and Ralf Martin, London School of Economics, and Myra Mohnen, University College London

Knowledge Spillovers from Clean and Dirty Technologies

How much should governments subsidize the development of new clean technologies? Dechezleprêtre, Martin, and Mohnen use patent citation data to investigate the relative intensity of knowledge spillovers in clean and dirty technologies in two technological fields: energy production and transportation. They introduce a new methodology that takes into account the whole history of patent citations to capture the indirect knowledge spillovers generated by patents. The authors find that, conditional on a wide range of potential confounding factors, clean patents receive on average 43 percent more citations than dirty patents. Knowledge spillovers from clean technologies are comparable in scale to those observed in the IT sector. The radical novelty of clean technologies relative to more incremental dirty inventions seems to account for their superiority. The results can support public support for clean R&D. They also suggest that green policies might be able to boost economic growth through induced knowledge spillovers.