Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We extend Armstrong’s (1996) result on exclusion in multi-dimensional screening models providing support for the view that the result holds true in a large class of models. We first relax some of the strong technical assumptions he imposed and provide alternative sufficient conditions for exclusion not relying on any form of convexity. We then proceed to show that exclusion obtains generically. We illustrate the results with examples and applications.