Algorithmic Tacit Collusion

Sarah Kaa & Amalie Toft Bentsen

Student thesis: Master thesis


The purpose of this thesis is to conduct an examination of pricing algorithms under the Predictable Agent scenario. Specifically, it examines how pricing algorithms can facilitate tacit collusion in the digital retail market, as well as how algorithmic pricing differs from normal (human) pricing used by firms. In addition, the thesis also considers how the parallel conduct of pricing algorithms should be investigated under article 101 of The Treaty of the Functioning of the European Union (TFEU).
The thesis is divided into three main parts. The first part of which conducts an economic analysis of tacit collusion in an oligopolistic market, with a presentation of the characteristics that strengthen the facilitation of tacit collusion. It is analyzed through the Structure-Conduct-Performance-paradigm how the implications of pricing algorithms influence the market. Finally, it is demonstrated how the algorithms has a negative impact on consumer welfare.
Part two conducts the legal analysis. It starts with a presentation of the prohibition to form an agreement that could disrupt free competition in the internal market, as prescribed in article 101 TFEU. Furthermore, as the distinction made in EU competition law between concerted practice and tacit collusion/parallel conduct has been stated as a grey zone in the interpretation, it is evaluated how to analyze pricing algorithms regarding the prohibition under article 101 TFEU. Here, the regulation forbids conduct which has as its object or effect, a preventions, restrictions or distortions competition. In this regard, the thesis seeks to demonstrate how parallel pricing algorithms can distort competition the same way as exchange of information, as explored in relation to concerted practice.
Part three consists of an interdisciplinary analysis of the problem. The results from the economic part and the legal part are set up to complement each other in the study of, how the regulation is today and what the impact of this is. Moreover, it is discussed how pricing algorithms under the Predictable Agent scenario can pose a problem for the competition in the digital retail market. In contrast to this statement, it is then analyzed how the algorithm can add efficiency and innovation to the market. There will be both benefits and disadvantages to the use of algorithms, and with this knowledge, four different ways to regulate the parallel pricing algorithms are discussed. In the future, more sophisticated algorithms are expected to be applied to the internal market. This case will be briefly evaluated.
Finally, the conclusion of the thesis is presented. In sum, it is found that pricing algorithms can facilitate tacit collusion better than seen before, due to the increase in, among other things, transparency and frequency. This will have an impact on competition, and therefore also the consumer welfare. It will be necessary to regulate parallel pricing algorithms because of the restriction on competition by effect.

EducationsMSc in Commercial Law, (Graduate Programme) Final Thesis
Publication date2020
Number of pages136
SupervisorsMaria Jose Schmidt-Kessen & Kenneth Thomas Baltzer