back
Dissertation
Immaterialgüter- und Wettbewerbsrecht

Algorithmic Personalised Price Discrimination (APPD) in the Data Economy. Rethinking the EU Competition Policy vis-à-vis Distributive Considerations.

APPD's harmful effects on efficiency and distribution are not fully captured by the EU competition toolkit. From a liberal-egalitarian approach, the toolkit is interpreted in light of digital market realities, allowing to address theories of harm, expose legal system gaps, and make policy proposals.

Last Update: 01.03.22

APPD might become a common business practice in data-driven markets due to machine learning profiling techniques, anti-arbitrage strategies, and the economic power attained by some firms through their market partitioning strategies and/or market failures inherent to the digital economy. In this context, the three main reactions to price discrimination —competition, consumer retaliation, lack of trust in the market— are less likely to take place. While APPD may improve efficiency and be part of procompetitive strategies (e.g. to attract new consumers and thus gain sufficient network effects to challenge an incumbent), it may also lead to detrimental effects.

To assess these effects, four scenarios are distinguished based on a taxonomy of economic power relations in the data economy: Product-based monopoly power; informational power, as a consequence of information asymmetries and bounded rationality; relational power, which derives from the economic dependence created between economic players in long-lasting commercial relationships; structural power, which stems from firms’ intermediary position and strategic role in platform and ecosystem contexts. In all cases, theoretical economic models show ambiguous allocative effects. From a distributional perspective, concerns are related to the transfer of surplus from consumers to producers; the greater exposure of vulnerable groups in society to higher prices; and competition disadvantages that a distorted playing field might create in up/downstream markets.

To address possible legal-based solutions the analysis focuses on two fields: the law of competition unilateral abuses and fairness-related regulations. For that purpose, a theoretical framework based on Dworkin’s distributive justice theory and Atkinson’s concept of competitive equality of opportunities is set up, followed by an analysis of  EU competition goals vis-à-vis efficiency and distribution, and examination of Art. 102 TFEU from a historical perspective. It is argued that unilateral abuses  can be understood in relation to two benchmarks: perfect competition and equal treatment. While the former became a key for competition law due to Chicago School influence, equal treatment has rarely been considered. Reasons are ideological and factual (its benchmark role is less relevant in pre-digital market structures).

Against this backdrop, a study of the EU case law on price discrimination finds that there is almost no room to enforce Art. 102 TFEU against APPD harmful effects. EU law on unfair commercial practices may achieve tackling some concerns, as it sets price transparency rules. However, it does not solve the B2B dependency nor consumer vulnerability issues.

Lastly, the project proposes a regulation of pricing algorithms, a new legal category to address APPD relational power issues, and an interpretation of Art. 102 TFEU that captures economic power phenomena in network economics contexts and recognizes equal treatment as a key principle.

Persons

Doctoral Student

Germán Johannsen

Main Areas of Research

II.3 Vernetzte Datenwirtschaft