Algorithmic and human collusion

Arbeitspapier

As self-learning pricing algorithms become popular, there are growing concerns among academics and regulators that algorithms could learn to collude tacitly on non-competitive prices and thereby harm competition. I study popular reinforcement learning algorithms and show that they develop collusive behavior in a simulated market environment. To derive a counterfactual that resembles traditional tacit collusion, I conduct market experiments with human participants in the same environment. Across different treatments, I vary the market size and the number of firms that use a self-learned pricing algorithm. I provide evidence that oligopoly markets can become more collusive if algorithms make pricing decisions instead of humans. In two-firm markets, market prices are weakly increasing in the number of algorithms in the market. In three-firm markets, algorithms weaken competition if most firms use an algorithm and human sellers are inexperienced.

Object Details

• ISBN: 978-3-86304-371-1
• Language: Englisch

References and Relationships

• Bibliographic citation: Series: DICE Discussion Paper ; No. 372

Classification and Topics

• Classification: Wirtschaft ; Design of Experiments: General ; Search; Learning; Information and Knowledge; Communication; Belief; Unawareness ; Oligopoly and Other Imperfect Markets ; Monopolization; Horizontal Anticompetitive Practices
• Subject: Artificial Intelligence ; Collusion ; Experiment ; Human-Machine Interaction

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Werner, Tobias
• Event: Veröffentlichung
• (who): Heinrich Heine University Düsseldorf, Düsseldorf Institute for Competition Economics (DICE)
• (where): Düsseldorf
• (when): 2021

Further information

• Handle: http://hdl.handle.net/10419/246229
• Last update: 10.03.2025, 11:43 AM CET