Learning to Mitigate Externalities: the Coase Theorem with Hindsight Rationality

2024-06-28

AI Summary

This paper addresses social welfare loss due to externalities in multi-agent systems under incomplete information. Motivated by the realistic constraint that agents lack prior knowledge of environmental dynamics and peer preferences, we extend the Coase Theorem—originally formulated for static, fully informed settings—to a stochastic, learning-enabled game-theoretic framework, proposing a prior-free, dynamic property-rights negotiation mechanism grounded in multi-armed bandit learning. Methodologically, we formulate a two-player stochastic bandwidth game and design an online learning strategy coupled with a counterfactual-reward-driven bargaining protocol. We theoretically prove that absence of property-rights allocation inevitably leads to social welfare collapse, whereas our learnable negotiation protocol asymptotically achieves socially optimal welfare. Empirical evaluation demonstrates that the mechanism significantly outperforms non-negotiating baselines. To our knowledge, this is the first Coase-inspired solution for externality governance under uncertainty that is both learnable—adapting endogenously to unknown environments—and verifiably efficient.

Technology Category

• Multiagent Systems: Mechanism Design
• Game Theory and Economic Paradigms: Mechanism Design
• Machine Learning: Online Learning & Bandits

Application Category

• Economics, Online Markets and Human Computation: Social networks and social learning
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In economic theory, the concept of externality refers to any indirect effect resulting from an interaction between players that affects the social welfare. Most of the models within which externality has been studied assume that agents have perfect knowledge of their environment and preferences. This is a major hindrance to the practical implementation of many proposed solutions. To address this issue, we consider a two-player bandit setting where the actions of one of the players affect the other player and we extend the Coase theorem [Coase, 1960]. This result shows that the optimal approach for maximizing the social welfare in the presence of externality is to establish property rights, i.e., enable transfers and bargaining between the players. Our work removes the classical assumption that bargainers possess perfect knowledge of the underlying game. We first demonstrate that in the absence of property rights, the social welfare breaks down. We then design a policy for the players which allows them to learn a bargaining strategy which maximizes the total welfare, recovering the Coase theorem under uncertainty.

Problem

• Incentive Compatibility
• Incomplete Information
• Social Welfare Maximization

Innovation

• Coase Theorem
• Uncertainty Handling
• Negotiation Optimization