Big data and algorithm-driven abuse of dominance: analysis and limiting principles under Article 102 TFEU

Thesis

The rise of big data and algorithmic systems has revolutionised the digital world, introduced novel dynamics and opened new avenues for dominant undertakings to abuse their market power. This necessitated extending the reach of antitrust in general (and Article 102 TFEU in particular), which has simultaneously created both ambiguity and a significant degree of controversy in the applicability of various legal tests in response to novel practices. The thesis investigates whether and under what conditions the gathering and use of big data, and algorithm-driven conduct could amount to abuse under Article 102 TFEU, and what limiting principles should guide their assessment. In answering this question, it focuses on five categories of behaviour as potential abuse under Article 102 TFEU: [1] privacy degrading data collection; [2] personalisation; [3] refusal to grant access to data; [4] automated bias against business users; and [5] self-preferencing. The selection of conduct spans both exploitative and exclusionary theories of harm and reflects different perspectives in the debate over novel forms of abuse. This thesis adopts a nuanced approach: it argues that expanding Article 102 TFEU to address these emerging strategies is both necessary and possible, while at the same time contending that this expansion must be disciplined by the limits of enforcement and oversight of the consistency of Article 102 TFEU. It highlights the importance of clear applicable principles, a structured approach, and the consistent development of case law.

Authors: BAN, E

• DOI: 10.5287/ora-8nawxdz9b
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: abuse of dominance; competition law; competition law and policy in digital markets; big data and algorithm-driven abuse
• Subjects: antitrust law; competition law