AI-Driven Market Dynamics
Paper Session
Friday, Jan. 3, 2025 10:15 AM - 12:15 PM (PST)
- Chair: Clemens Possnig, University of Waterloo
Robust Identification in Repeated Games: Estimating Algorithmic Collusion
Abstract
Artificial Intelligence (AI) algorithms may set prices or bids that resemble equilibrium strategies in a repeated game, raising concerns over algorithmic collusion. In this paper, we study how researchers can identify primitives of the underlying game in such environments. We make no assumptions on equilibrium selection but maintain that one equilibrium strategy generates the data. We leverage the restrictions on equilibrium payoffs in \cite{AwKr} to construct a law of large numbers that implies bounds on payoff parameters, the discount factor, and monitoring quality. Our method is computationally tractable in generic games. We illustrate the identified sets of parameters using simulated play by Q-learning algorithms. Our results highlight the trade-off between maintaining strong assumptions and the tightness of the identified set.Artificial Intelligence and Spontaneous Collusion
Abstract
We develop a tractable model for studying strategic interactions between learning algorithms. We uncover a mechanism responsible for the emergence of algorithmic collusion. We observe that algorithms periodically coordinate on actions that are more profitable than static Nash equilibria. This novel collusive channel relies on an endogenous statistical linkage in the algorithms’ estimates which we call spontaneous coupling. The model’s parameters predict whether the statistical linkage will appear, and what market structures facilitate algorithmic collusion. We show that spontaneous coupling can sustain collusion in prices and market shares, complementing experimental findings in the literature. Finally, we apply our results to design algorithmic markets.Artificial Intelligence, Algorithmic Advice and Cheap Talk
Abstract
We study a model of communication in which a better-informed sender learns to communicate with a receiver who takes an action that affects the welfare of both. Specifically, we model the sender as a machine-learning-based algorithmic recommendation system and the receiver as a rational, best-responding agent that understands how the algorithm works. The results demonstrate robust communication, which either emerges from scratch (originating from babbling where no common language initially exists) or persists when initialized. We show that the sender's learning hinders communication, limiting the extent of information transmission even when the algorithm's designer's and the receiver's preferences are aligned. We then show that when the two are not aligned, there is a robust pattern where the algorithm plays a cut-off strategy pooling messages when its private information suggests actions in the direction of its preference bias while sending mostly separate signals otherwise.JEL Classifications
- D4 - Market Structure, Pricing, and Design
- L1 - Market Structure, Firm Strategy, and Market Performance