Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We study a multi-unit auction, where bidders differ in their marginal costs, and compare the outcome to the Bertrand (or single-unit auction) analogue.