Working Paper: NBER ID: w20677
Authors: Joshua Gans; Fiona Murray
Abstract: Formal attribution provides a means of recognizing scientific contributions as well as allocating scientific credit. This paper examines the processes by which attribution arises and its interaction with market assessments of the relative contributions of members of scientific teams and communities – a topic of interest organizational economics of science and in understanding scientific labor markets. We demonstrate that a pioneer or senior scientist’s decision to co-author with a follower or junior scientist depends critically on market attributions as well as the timing of the co-authoring decision. This results in multiple equilibrium outcomes each with different implications for expected quality of research projects. However, we demonstrate that the Pareto efficient organisational regime is for the follower researcher to be granted co-authorship contingent on their own performance without any earlier pre-commitment to formal attribution. We then compare this with the alternative for the pioneer of publishing their contribution and being rewarded through citations to back to it. While in some equilibria (especially where co-authoring commitments are possible) there is no advantage to interim publication, in others this can increase expected research quality.
Keywords: scientific attribution; coauthorship; market assessments; research quality
JEL Codes: O31; O34
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Pioneer decision to coauthor (D70) | Market attributions (D49) |
| Pioneer decision to coauthor (D70) | Timing of decision (C41) |
| Commitment to coauthoring based on follower's contributions (D70) | Higher expected quality of research outputs (D29) |
| Awareness of contributions (D64) | Higher effort by both scientists (D29) |
| Higher effort by both scientists (D29) | Better quality research (C90) |
| Inability to commit to coauthorship (D86) | Inefficiencies and lower expected quality outcomes (L15) |
| Ex post coauthorship (D70) | Optimal outcomes (L21) |
| Pioneer publishes interim results without coauthoring (A19) | Lower expected quality of research (L15) |