Exploring Tradeoffs in the Organization of Scientific Work: Collaboration and Scientific Reward (with Fiona Murray and Joshua Gans, forthcoming in Management Science)
When do scientists and other knowledge workers organize into collaborative teams and why do they do so for some projects and not others? At the core of this important organizational choice is, we argue, a tradeoff between the productive efficiency of collaboration and the credit allocation that arises after the completion of collaborative work. In this paper, we explore this tradeoff by developing a model to structure our understanding of the factors shaping researcher collaborative choices in particular the implicit allocation of credit among participants in scientific projects. We then use the annual research activity of 661 faculty scientists at one institution over a 30-year period to explore the tradeoff between collaboration and reward at the individual faculty level and to infer critical parameters in the organization of scientific work.
Nearly half a century after Merton’s description of simultaneous discoveries “as a strategic research site” (Merton 1963), they are hardly ever used by social scientists. Although their promise as a research tool has been established, operationalization issues have hindered their use. Disagreement exists over the appropriate amount of scientific and technological similarity necessary to consider two or more discoveries a multiple. This paper intends to unleash the potential of simultaneous discovery as a research tool. First, we review prior work on the topic and describe the vast theoretical potential of this approach. Second, we propose a new method that uses the literature on the social construction of science to go beyond the old debate of scientific and technological similarity. Third, we present a new algorithm that generated a large dataset of scientific multiples in an automated and systematic fashion. Finally, we describe the resulting sample of 578 recent discoveries made by 1,246 teams of scientists working in a variety of settings around the world.
Though scientific knowledge published by universities is increasingly important for the production of new technologies, the obstacles to invention based on academic science remain little understood. This question is difficult to examine empirically because the effect of the academic environment is confounded by the fact that university science tends to be fundamental and therefore little amenable to technology development. This paper addresses this identification challenge by using simultaneous discoveries operationalized as “paper twins.” Analysis of follow-on citation of a sample of paper twins in the patent literature indicates that inventors cite academic papers less than those from industry. Importantly, this difference is not purely explained by the fundamental character of academic research. Not only do academic scientists produce fewer patents based on their discoveries than industry ones; we also find suggestive evidence that issues of awareness, trust, and approach to discovery might hinder a broader use of academic science by inventors.
Geographic Localization of Knowledge Spillovers
I investigate current debates about the extent to which knowledge spillovers are localized. Using citations of 275 twin papers in the patent literature, I can identify the extent to which inventors are more likely to draw on scientific knowledge that emerged within closer geographic proximity—while keeping the discovery constant. Early results indicate that knowledge spillovers are localized not only at the country level, but also very strongly at the metropolitan area level.
In the Shadow of Uncertainty: Entrepreneurial Strategy and the Selection of New Projects
I explore how entrepreneurs exploit uncertainty to compete against incumbents in pharmaceutical R&D. Using instances in which the same discovery is made simultaneously in an entrepreneurial venture and at a large firm, my preliminary results indicate that entrepreneurs tend to disengage from projects involving too little uncertainty for fear of competition with companies that have much greater resources. On the other hand, larger firms tend to reject ideas with high uncertainty, providing space for young firms to grow, “sheltered from competition” by this very uncertainty.