Algorithmic Collusion: From Smokeroom to Code Lines? How Collusion Can Evolve & Are We Prepared

Abstract

This paper explores whether the current Canadian legal framework is adapted to the challenges that algorithmic pricing may pose. It concludes that creative legal solutions should be envisioned to curtail upcoming challenges, although the most appropriate solutions may come from regulating the design of algorithms. It first provides an overview of the artificial intelligence at hand, and rationales that draw businesses to use them. Secondly, this paper asks whether there is a real threat that pricing algorithms may enable collusion. Through a review of experimental and empirical evidence, it answers that the calls to caution are not unwarranted, although the threats may not yet be at the forefront. Thirdly, after a review of the legal landscape in Canada, this analysis reveals that some actions are needed for regulators to be proactive in the face of upcoming challenges, through a study of the different ways in which algorithmic collusion may occur.