Rate and Direction of Inventive Activity

September 30 - October 2, 2010
NBER Research Associates Josh Lerner of Harvard Business School and Scott Stern of the Sloan School of Management, Organizers

Timothy F. Bresnahan, Stanford University and NBER; Shane Greenstein, Northwestern University and NBER; and Rebecca Henderson, Harvard University and NBER
Schumpeterian Competition and Diseconomies of Scope; Illustrations from the Histories of Microsoft and IBM

Bresnahan, Greenstein, and Henderson address a long-standing question about the causes of creative destruction. Dominant incumbent firms, long successful in an existing technology, are often much less successful in new technological eras. This is puzzling, because a cursory analysis would suggest that incumbent firms have the potential to take advantage of economies of scope across new and old lines of business and, if economies of scope are unavailable, simply to reproduce entrant behavior by creating a "firm within a firm." There are two broad streams of explanation for incumbent failure in these circumstances. One posits that incumbents fear cannibalization in the market place, and so under-invest in the new technology. The second suggests that incumbent firms develop organizational capabilities and cognitive frames that make them slow to "see" new opportunities and that make it difficult to respond effectively once the new opportunity is identified. Bresnahan and co-authors draw on two of the most important historical episodes in the history of the computing industry, the introduction of the PC and of the browser, to develop a third hypothesis. Having been extremely successful in an old technology, both IBM and Microsoft came to have grave difficulties competing in the new, despite some dramatic early success. These authors suggest that the difficulties reflect not only the existence of strategic interdependencies between old and new businesses, and perhaps betweem inherited cognitive frames, but also the presence of diseconomies of scope caused by the presence of assets that are "necessarily shared" across both businesses. Bresnahan and colleagues show that both Microsoft and IBM were initially very successful in creating free standing business units that could compete with entrants on their own terms, but that the interaction between the strategic interdependencies across old and new businesses and the need to share key firm-level assets imposed significant costs on both businesses and created severe organizational conflict. In IBM and Microsoft's case this conflict eventually led to control over the new business being given to the old - a result that is intriguingly consistent with some recent developments in organizational economics - and that in both cases effectively crippled the new business.

Josh Lerner and Peter Tufano, Harvard Business School and NBER
The Consequences of Financial Innovation: A Counterfactual Research Agenda

Financial innovation has been both praised as the engine of growth of society and castigated for being the source of the weakness of the economy. Lerner and Tufano review the literature on financial innovation and highlight the similarities and differences between financial innovation and other forms of innovation. They also propose a research agenda to systematically address the social welfare implications of financial innovation. To complement existing empirical and theoretical methods, they propose that scholars examine case studies of systemic (widely adopted) innovations, explicitly considering counterfactual histories had the innovations never been invented or adopted.

Alexander J. Field, Santa Clara University
The Adversity/Hysteresis Effect: Depression Era Productivity Growth in the U.S. Railroad Sector

Do depressions have a silver lining, in the sense that they boost long-term productivity growth? The 1930s were a decade of double-digit unemployment, but the years 1929-41 were the most technologically progressive, as measured by total factor productivity (TFP) growth in the private nonfarm economy, of the twentieth century. Field argues that the advances were fueled by: strong progress in manufacturing; the result of a maturing privately funded R and D system; spillovers in transportation and distribution related to the build-out of the surface road network; and innovation, particularly in railroads, resulting from an effective response to adversity. He suggests that the first two effects likely would have played out without the depression, but that the third does represent something of a silver lining. Faced with growing competition from trucks and an economy mired in depression, the railroads innovated both technologically and logistically, engineering reductions in employment, rolling stock, and miles operated. And yet output rose. Data on labor productivity growth in individual firms show that advances of the time affected roads both large and small; affected freight rather than passenger carriage; were a national phenomenon, although southern railroads played catch up; and that railroads that reduced their employment more experienced more output growth per worker.

Timothy F. Bresnahan
Generality, Recombination, and Re-Use

Bresnahan studies the economic conditions relevant to the emergence of new general purpose technologies and, more generally, to inventions which will be recombined or otherwise re-used. The decentralized distribution of knowledge about technical and market opportunities ex ante is central to his analysis. Changes in the distribution of knowledge caused by the market presence of early inventions play an equally important role, particularly when they convert entrepreneurial knowledge into market knowledge. Bresnahan applies this analysis to the history of invention of computer technologies for white collar work automation.

Joshua Gans, University of Melbourne, and Fiona E. Murray, MIT
Funding Conditions, the Public-Private Portfolio, and the Disclosure of Scientific Knowledge

Gans and Murray investigate project selection, conditions imposed by public funders (both governmental and non-governmental), and the impact of these two factors on the portfolio of public and private research projects funded, their commercialization, and the level of openness in funded research. The authors begin by reviewing project selection criteria and policies towards disclosure and commercialization (including patent rights), noting significant variability across funders. They then provide a model of how funding conditions that restrict commercialization opportunities affect the projects that accept public funds and the overall level of openness in research.

Ralf R. Meisenzahl, Federal Reserve Board of Governors, and Joel Mokyr, Northwestern University
The Rate and Direction of Invention in the British Industrial Revolution: Incentives and Institutions

During the Industrial Revolution technological progress and innovation became the main drivers of economic growth. But why was Britain the technological leader? Meisenzahl and Mokyr argue that one hitherto little recognized British advantage was the supply of highly skilled, mechanically able craftsmen who were able to adapt, implement, improve, and tweak new technologies and who provided the micro inventions necessary to make macro inventions highly productive and remunerative. Using a sample of 759 of these mechanics and engineers, the researchers study the incentives and institutions that facilitated the high rate of inventive activity during the Industrial Revolution. First, apprenticeship was the dominant form of skill formation. Formal education played only a minor role. Second, many skilled workmen relied on secrecy and first-mover advantages to reap the benefits of their innovations. Over 40 percent of the sample here never took out a patent. Third, skilled workmen in Britain often published their work and engaged in debates over contemporary technological and social questions. In short, they were affected by the Enlightenment culture. Finally, patterns differ for the textile sector; therefore, any inferences from textiles about the whole economy are likely to be misleading.

Petra Moser, Stanford University and NBER, and Paul Rhode, University of Michigan and NBER
Plant Patents and the American Rose

Moser and Rhode use data on plant patents and registrations of new roses to examine whether the Plant Patent Act of 1930 encouraged the development of a domestic U.S.rose breeding industry. Patent data suggest a significant effect: nearly half of all plant patent grants between 1931 and 1970 are for roses, and data on patent assignments indicate that a small number of commercial breeders account for most of these patents. Rose patents and the share of assignments increase until the mid-1950s and decline gradually afterwards. Matching patents with registrations, however, suggests that changes in rose patents were driven by changes in the rate at which new roses were patented, rather than reflecting true changes in innovation. Registrations were more responsive to demand conditions and wars that disrupted the European rose industry than patents, and foreign producer continued to register new roses after patenting declined.

Kevin J. Boudreau, London Business School, and Karim R. Lakhani, Harvard University
The Confederacy of Software Production: Field Experimental Evidence on Heterogeneous Developers, Tastes for Institutions, and Effort

At least as much as other innovative and creative problem-solving sectors of the economy, software development takes place in an extraordinary range of different sorts of organizations, thus forming a patchwork or "confederacy" of institutional forms. Boudreau and Lakhani suggest one explanation that may begin to account for this heterogeneity: workers have different intrinsic tastes or preferences for different institutional regimes. They present field experimental evidence to show that software workers who are sorted into either a cooperative or competitive regime, depending on their tastes, exerted almost twice as much effort as those in a group of randomly-assigned workers (controlling for their skills). These results suggest that sorting to different types of organizations on the basis of institutional tastes and preferences may have large efficiency implications.

Pierre Azoulay, MIT and NBER; Joshua S. Graff Zivin, University of California, San Diego and NBER; and Bhaven Sampat, Columbia University
The Diffusion of Scientific Knowledge Across Time and Space: Evidence from Professional Transitions for the Superstars of Medicine

Are scientific knowledge inflows embodied in individuals, or "in the air"? To answer this question, Azoulay, Graff Zivin, and Sampat measure the effect of labor mobility in a sample of 9,483 elite academic life scientists on the citation trajectories associated with individual articles (resp. patents) published (resp. granted) before the scientist moved to a new institution. They find that article-to-article citations from the scientific community at the superstar's location of origin are barely affected by their departure. In contrast, article-to-patent citations, and especially patent-to-patent citations, decline at the origin location following a star's departure, suggesting that spillovers from academia to industry are not completely disembodied. The authors also find that article-to-article citations at the superstars' destination location markedly increase after they move. These results suggest that, to be realized, knowledge inflows to industry may require more face-to-face interaction than those to academics. Moreover, to the extent that academic scientists do not internalize the effect of their location decisions on the circulation of ideas, these results raise the intriguing possibility that barriers to labor mobility in academic science limit the recombination of individual bits of knowledge, resulting in a suboptimal rate of scientific exploration.

Daron Acemoglu, MIT and NBER
Diversity and Technological Progress

Acemoglu proposes a tractable model to study the equilibrium diversity of technological progress, showing that it may exhibit too little diversity (too much conformity -- in particular, foregoing socially beneficial investments in "alternative" technologies that will be used at some point in the future. The presence of future innovations that will replace current innovations implies that social benefits from innovation are not fully internalized. As a consequence, the market favors technologies that generate current gains relative to those that will bear fruit in the future; current innovations in research lines that will be profitable in the future are discouraged, because current innovations typically are followed by further innovations before they can be profitably marketed. A social planner would choose a more diverse research portfolio and would induce a higher growth rate than the equilibrium allocation. The diversity of researchers is a partial (imperfect) remedy against the misallocation induced by the market. Researchers with different interests, competencies, or ideas may choose non-profit maximizing and thus more diverse research portfolios, indirectly contributing to economic growth.

Daniel Spulber, Northwestern University
How Entrepreneurs Affect the Rate and Direction of Inventive Activity

Entrepreneurship in established industries poses a puzzle. Entrepreneurial entry increases competition, which suggests that innovators will earn greater returns simply by transferring their technology to incumbent firms. Spulber presents a strategic innovation game in which an innovator and an established firm choose whether to cooperate or to compete. The main result of the analysis is that with sufficient product differentiation, the equilibrium outcome is entrepreneurship rather than technology transfer. Product differentiation increases the returns to entrepreneurship relative to the returns from technology transfer. Second, with sufficiently differentiated products, an innovator who can choose between licensing the innovation to the incumbent and becoming an entrepreneur has a greater incentive to invent than an incumbent monopolist. Third, when an independent inventor is an additional player in the strategic game, the market equilibrium always involves entrepreneurial entry. Fourth, with sufficiently differentiated products, an independent innovator has a greater incentive to invent than an incumbent monopolist.

Shulamit Kahn, Boston University, and Megan MacGarvie, Boston University and NBER
The Effects of the Foreign Fulbright Program on Knowledge Creation in Science and Engineering

Kahn and MacGarvie examine the impact of the Foreign Fulbright Program -- which requires foreign students funded by the program to leave the United States after completing their studies -- on the production of research in science and engineering. They track the post-Ph.D. careers of 244 Fulbright scientists and 244 otherwise similar control scientists. They find that, on average, the Fulbright program encourages more international mobility of U.S.-trained Ph.D. scientists. Scientists trained in the United States through the Fulbright program spend more than twice as many of their post-graduation years outside the United States when compared to controls. The effect is particularly large for students from countries with low GDP per capita or less well-developed science bases. Fulbrights from poorer or weak-science home countries also have lower numbers of some kinds of publications and less impact on global knowledge, while Fulbrights from richer/high-science countries with strong science bases have publication and citation records similar to comparable Ph.D.s of foreign origin without return requirements. Fulbrights from rich or high-science countries produce substantially more articles listing home country authors while Fulbrights from poorer or weaker science countries produce substantially fewer articles listing U.S.authors. Finally, for the poorer countries, the Fulbright program does appear to increase U.S. home country collaborations, an explicit goal of the program.