« Back to Results
Marriott Marquis, Grand Ballroom 11
Hosted By:
American Economic Association
To answer these questions, we use a novel longitudinal dataset containing information about scientists at over thirty U.S. universities who received research funding. This dataset was linked to information about their careers and productivity—specifically to employment and earnings records—and to information about patents. We also use a recently developed, theoretically grounded measure of transformativeness that captures the degree to which new ideas make existing work obsolete. Our analytical approach uses new data and methods to (1) describe the broader organizational context—the formal (i.e., departments, physical space) and informal (i.e., collaborations) structure—within which research takes place, (2) characterize transformational research, and (3) describe the link between the two. To uncover the causal effects of whether one’s network position affects the transformativeness of research, we instrument for network position by exploring credible sources of variation which include: (1) differences in campus organizational structures, (2) spatial proximity of researchers across campus, and (3) distance and frequency of conferences.
We test this hypothesis using a new dataset we have compiled containing the age of approximately 80% of the U.S. inventors on U.S. patents 1976-2018, collected by scraping multiple websites based on the name and address of the inventors as reported on the patents. We examine the importance, originality and radicalness of approximately 1.6 million patents produced by teams of U.S. inventors for which we have identified the age of all inventors on the patent. Because of the unknown accuracy of the age information, we estimate the association of age composition and patent attributes using a multiple over-imputation method for missing and mis-measured data.
Younger workers may benefit from the experience that older workers bring to creative teams. At the same time, older workers may benefit from teaming with younger workers, such that declines in fluid abilities associated with cognitive aging could be mitigated. Age-related cognitive changes, and the possibilities for age-related complementarities in inventive teams, have important implications for the overall inventiveness and productivity of the economy, at a time when serious concerns have been raised about growing life course delays in research success and productivity, and possible productivity declines associated with overall aging of the workforce.
Life on the Edge: Collaboration and the Production of Ideas at the Scientific Frontier
Paper Session
Friday, Jan. 3, 2020 10:15 AM - 12:15 PM (PDT)
- Chair: Bruce Weinburg, Ohio State University
Inventor Teams, Invention Quality and Occupational Contexts
Abstract
The role of teams has received significant attention in the literature on knowledge production over the last decades. Empirical studies in economics and management have largely focused on how individuals contribute to teams and what role team composition plays for team productivity. This project explores how a negative shock to an inventor’s team capital affects productivity and career trajectories. To this end, we exploit a rich dataset on more than 150,000 German inventors that combines employer-employee labor market biography data with patent register data covering the period from 1980-2014. These data enable us to track each inventor’s employment status over time and firm with high precision. Using data on about 700 registered premature deaths of inventors and the same number of matched inventors, we study how their co-inventors were affected by the death of their respective peer. Using a difference-in-differences and an event study design, we investigate differences in the co-inventors’ patenting activities, career advancement and in new team formation. We seek to establish novel results that help researchers and practitioners better understand how inventors adjust their career paths after a negative shock to their team. We do so by tracking the careers of the individuals across organizations, occupations, industries and regional labor markets and discuss our results in the context of industrial innovation in the German labor market.Money for Something: The Link between Research Funding, Collaboration Networks, and Innovation
Abstract
A major goal of research funding is to support transformative science. A challenge to achieving this goal, however, is determining what kinds of research funding will lead to such transformations. Should funding agencies encourage large teams that are connected across a variety of universities and disciplines? Or should they support individual researchers who can generate ideas independent of academic group-think? Are junior researchers more prone to conducting incremental research (possibly in pursuit of tenure) than more secure senior researchers? Do women choose less transformative research paths? If so, is it because they are placed in organizational structures that inhibit the riskiness of their research? To what degree is the capacity for transformative research driven by the organizational context within which research takes place, and to what degree does research funding act to promote transformation by creating and sustaining supportive contexts?To answer these questions, we use a novel longitudinal dataset containing information about scientists at over thirty U.S. universities who received research funding. This dataset was linked to information about their careers and productivity—specifically to employment and earnings records—and to information about patents. We also use a recently developed, theoretically grounded measure of transformativeness that captures the degree to which new ideas make existing work obsolete. Our analytical approach uses new data and methods to (1) describe the broader organizational context—the formal (i.e., departments, physical space) and informal (i.e., collaborations) structure—within which research takes place, (2) characterize transformational research, and (3) describe the link between the two. To uncover the causal effects of whether one’s network position affects the transformativeness of research, we instrument for network position by exploring credible sources of variation which include: (1) differences in campus organizational structures, (2) spatial proximity of researchers across campus, and (3) distance and frequency of conferences.
Multi-Disciplinary Scientists: Field Experimental Evidence from a Call for Grant Proposals
Abstract
Who chooses to participate in multi-disciplinary scientific research? This paper reports on a field experiment in which an open call was made for new scientific research proposals for seed grants. The call went to all researchers in all disciplines at a large research-intensive US university. Researchers were randomized as to whether the call explicitly mentioned a requirement for the research to be multi-disciplinary. The multi-disciplinary treatment cut interest in the grant by one-third relative to the control group. Differences in the multi-disciplinary treatment vary more “across” than “within” research disciplines. For example, Computer Science and Engineering—areas of applied science—were least negatively affected by the multi-disciplinary treatment. Within given research disciplines, propensity to pursue multi-disciplinary research was higher (i.e., less low) among senior tenured professors with a long tenure at the school, non-tenure track researchers, those already with a history of publishing multi-disciplinary papers, and less-than-top-quality researchers. The findings are consistent with multi-disciplinary research often simply being highly applied in its nature and otherwise the non-paradigmatic multi-disciplinary research conflicts with incentives provided by the institutions of science.Fluid Intelligence and Experience in Invention: Complementarity in Age-Heterogeneous Teams
Abstract
Previous research suggests creative ability peaks with age in the 30s and early 40s, and declines thereafter, with some variation across fields. Such patterns are consistent with research from the cognitive aging literature. Cognitive processes show aging-related increases in experience-based knowledge (pragmatics or crystallized abilities) and age-related decreases in the ability to process new knowledge and information quickly and efficiently (mechanics or fluid abilities). On this basis, we hypothesize that age-heterogeneous teams have performance advantages derived from being able to combine the experience of older workers with the fluid intelligence of younger workers.We test this hypothesis using a new dataset we have compiled containing the age of approximately 80% of the U.S. inventors on U.S. patents 1976-2018, collected by scraping multiple websites based on the name and address of the inventors as reported on the patents. We examine the importance, originality and radicalness of approximately 1.6 million patents produced by teams of U.S. inventors for which we have identified the age of all inventors on the patent. Because of the unknown accuracy of the age information, we estimate the association of age composition and patent attributes using a multiple over-imputation method for missing and mis-measured data.
Younger workers may benefit from the experience that older workers bring to creative teams. At the same time, older workers may benefit from teaming with younger workers, such that declines in fluid abilities associated with cognitive aging could be mitigated. Age-related cognitive changes, and the possibilities for age-related complementarities in inventive teams, have important implications for the overall inventiveness and productivity of the economy, at a time when serious concerns have been raised about growing life course delays in research success and productivity, and possible productivity declines associated with overall aging of the workforce.
Discussant(s)
Pierre Azoulay
,
Massachusetts Institute of Technology
Julia Lane
,
New York University
Adam B. Jaffe
,
Brandeis University, Queensland University of Technology, and NBER
Russell Funk
,
University of Minnesota
Ina Ganguli
,
University of Massachusetts-Amherst
JEL Classifications
- O3 - Innovation; Research and Development; Technological Change; Intellectual Property Rights