This paper uses panel data on 339 economists to evaluate the relationship b
etween co-authorship and output. It is shown that for a given individual, m
ore co-authorship is associated with higher quality, greater length, and gr
eater frequency of publications. However, the net relationship between co-a
uthorship and output attributable to the individual is negative after disco
unting for the number of authors. The results of this paper suggest that un
iversities and granting agencies which preferentially reward research colla
boration may be undermining their goal of maximizing research output. (C) 2
001 Elsevier Science B.V. All rights reserved.