Data Sharing with Endogenous Choices over Differential Privacy Levels

2026-02-10

AI Summary

This study addresses strategic interactions in data sharing, where agents endogenously decide whether to join a coalition and how much differential privacy noise to inject, based on heterogeneous privacy costs. These interdependent decisions generate externalities between participation and privacy choices. To tackle the potential nonexistence of a Nash equilibrium, the paper introduces a novel “robust equilibrium” concept. It systematically characterizes, for the first time, how the structure of privacy costs—whether increasing or decreasing—affects the existence, stability, and efficiency of equilibria. The authors derive upper bounds on the efficiency loss in social welfare and estimation accuracy relative to the social optimum, explicitly revealing their dependence on the number of participants, the form of the cost function, and coalition size.

Technology Category

• Game Theory and Economic Paradigms: Equilibrium
• Multiagent Systems: Mechanism Design
• Machine Learning: Privacy

Application Category

• Security and Privacy: Privacy-enhancing technologies
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

We study coalition formation for data sharing under differential privacy when agents have heterogeneous privacy costs. Each agent holds a sensitive data point and decides whether to participate in a data-sharing coalition and how much noise to add to their data. Privacy choices induce a fundamental trade-off: higher privacy reduces individual data-sharing costs but degrades data utility and statistical accuracy for the coalition. These choices generate externalities across agents, making both participation and privacy levels strategic. Our goal is to understand which coalitions are stable, how privacy choices shape equilibrium outcomes, and how decentralized data sharing compares to a centralized, socially optimal benchmark. We provide a comprehensive equilibrium analysis across a broad range of privacy-cost regimes, from decreasing costs (e.g., privacy amplification from pooling data) to increasing costs (e.g., greater exposure to privacy attacks in larger coalitions). We first characterize Nash equilibrium coalitions with endogenous privacy levels and show that equilibria may fail to exist and can be non-monotonic in problem parameters. We also introduce a weaker equilibrium notion called robust equilibrium (that allows more widespread equilibrium existence by equipping existing players in the coalition with the power to prevent or veto external players from joining) and fully characterize such equilibria. Finally, we analyze, for both Nash and robust equilibria, the efficiency relative to the social optimum in terms of social welfare and estimator accuracy. We derive bounds that depend sharply on the number of players, properties of the cost profile and how privacy costs scale with coalition size.

Problem

• data sharing
• differential privacy
• coalition formation
• privacy cost
• equilibrium

Innovation

• endogenous privacy
• coalition formation
• differential privacy
• robust equilibrium
• privacy externality