Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
I extend the model of legal standards in Katsoulacos and Ulph (2009) to allow for optimal statistical decision-making by the regulator. I show that an effects-based standard produces weakly lower decision error costs than a per se standard in this setting. I also use the framework to highlight an important implicit assumption underlying the comparison of legal standards in the cited paper.