Sunday, January 5, 2014

The more we know, the more we need teams

Several prior studies present evidence that the size of research teams has increased steadily over time (Adams et al, 2005; Wuchy et al, 2007; Jones, 2009). For example, Wuchy et al (2007) show that over the latter half of the twentieth century, team size increased in 170 of 171 fields in science and engineering, 54 of 54 fields in the social sciences, and 24 of 27 fields in the arts and humanities. Furthermore, this increase even occurred in fields traditionally associated with individual-oriented research: “Surprisingly, even mathematics, long thought the domain of the loner scientist and least dependent of the hard sciences on lab scale and capital-intensive equipment, showed a marked increase in the fraction of work done in teams, from 19% to 57%, with mean team size rising from 1.22 to 1.84.” Moreover, they present citation-based evidence that the relative impact of team versus individual output is increasing over time, even after controlling for self-citations. ...

Jones (2009) emphasizes the “knowledge burden” hypothesis in which successive generations of innovators face an increasing education burden due to the advancing knowledge frontier. This advancing frontier, he posits, requires innovators to specialize more and thus necessitates working more collaboratively, which alters the organization of innovative activity towards teamwork. Jones provides descriptive statistics consistent with this theory. For example, he shows that over time: 1) the number of co-authors on academic publications increases, 2) Nobel laureates are older when they perform their great achievement, 3) the number of co-inventors per patent increases, 4) the age at first innovation increases, and 5) the probability of switching fields decreases. ...

We examine whether knowledge accumulation leads to increased collaboration and report evidence consistent with Jones’ (2009) burden of knowledge hypothesis. While we do not rule out other explanations as possible additional drivers of the increasing rate of collaboration, we document that a shock to the knowledge frontier led to increased collaboration and specialization. Specifically, we examine whether the sudden and unexpected increase in knowledge of theoretical mathematics that came with the fall of the Soviet Union led to an increase in collaboration among non-Soviet scholars. Using an identification strategy inspired by Borjas and Doran (2012), we categorize as “Soviet-rich” those subfields of theoretical mathematics where Soviet mathematicians made a high contribution relative to mathematicians from other nations before the collapse of the Soviet Union.

We find that collaboration rose disproportionately in Soviet-rich relative to -poor fields after 1990. ... Moreover, the knowledge shock is followed by a disproportionate increase in researcher team size in Soviet-rich subfields in Japan, a region that did not experience a large influx of Soviet immigrants, suggesting the estimated effect is not caused by an increase in labor market competition.
--Ajay Agrawal, Avi Goldfarb, Florenta Teodoridis, "Does Knowledge Accumulation Increase the Returns to Collaboration?," on the growing importance of working well with others