When Can Communication Lead to Efficiency?

2025-10-14

AI Summary

This paper investigates how communication affects efficiency in incomplete-information games. Using game-theoretic and mechanism-design approaches, it systematically analyzes the Pareto-efficiency boundaries of two communication paradigms—cheap talk and Bayesian persuasion—under both state-dependent and state-independent utility structures. Theoretical contributions include: (1) excessive randomization as a fundamental driver of equilibrium inefficiency; (2) a necessary structural condition for efficient communication—namely, the total number of actions must be strictly less than the sum of the number of players and the number of states; (3) cheap talk achieves Pareto optimality only in pure-strategy equilibria where the seller’s preferred action is deterministically implemented; and (4) Bayesian persuasion generally fails to improve efficiency. Collectively, these results identify precise structural constraints governing communication effectiveness, providing formal theoretical criteria for designing efficient information-transmission mechanisms in strategic settings.

Technology Category

• Game Theory and Economic Paradigms: Mechanism Design
• Multiagent Systems: Mechanism Design
• Reasoning under Uncertainty: Decision/Utility Theory

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Security and Privacy: Privacy-enhancing technologies

Abstract

We study games with incomplete information and characterize when a feasible outcome is Pareto efficient. We show that any outcome with excessive randomization over actions is inefficient. Generically, efficiency requires that the total number of actions taken across states be strictly less than the sum of the number of players and states. We then examine the efficiency of equilibrium outcomes in communication models. Generically, a cheap talk outcome is efficient only if it is pure. When the sender's payoff is state-independent, it is efficient if and only if the sender's most preferred action is chosen with certainty. In natural buyer-seller settings, Bayesian persuasion outcomes are inefficient across a wide range of priors and preferences. Finally, we show that our results apply to mechanism design problems with many players.

Problem

• Characterizing Pareto efficiency conditions in incomplete information games
• Analyzing communication model efficiency in cheap talk and persuasion
• Extending efficiency results to multi-player mechanism design problems

Innovation

• Characterizes Pareto efficiency in incomplete information games
• Identifies generic efficiency conditions for communication models
• Applies efficiency analysis to mechanism design problems