Antitrust Intent in an Age of Algorithmic Nudging
34 Pages Posted: 7 Nov 2018 Last revised: 31 Dec 2018
Date Written: October 15, 2018
This paper revisits the role of intent for the purposes of establishing an abuse of dominance under EU competition law. It does so reviewing cases where the outcome hinged on the existence of anticompetitive intent, and eliciting from that line of decisions a set of guiding principles. The consistency with those principles of the recent European Commission’s Decision in Google Shopping is then discussed. After a critical examination of the definition of abusive conduct identified by the Commission (self- favoring) and the extent to which it satisfies the intent requirement in antitrust, the broader implications for providers of algorithmic intermediation services are explored. Given the absence of a materiality threshold and other limiting principles in the theory of harm used by the Commission, the risks of overreach and unpredictability are substantial. Two recommendations are offered to ameliorate the status quo. First, courts and competition authorities should only take into account a defendant’s subjective state of mind under a “qualified intent” test: a test requiring proof of immediate, substantial and foreseeable anticompetitive effects arising from a purported conduct. Second, to constrain the scope of application of the actionable duty of algorithmic self-scrutiny, regulators should introduce a negligence-based “safe harbor”. This safe harbor, mimicking the specialized regime of intermediary liability for information society service providers, would enable continued investment and innovation in algorithmic services while also promoting adherence to cross-industry best practices in algorithmic design.
Keywords: Antitrust; intent
JEL Classification: JEL9
Suggested Citation: Suggested Citation