Evolutionary Dynamics in Data Trading with Risk Compensation

The fact that data can be privately possessed yet contain the attributes of public goods leads to the “Arrow’s Information Paradox” in data trading. If left unchecked, supply-side deception about data quality and demand-side data leakage can seriously undermine the trust between suppliers and demanders. Inspired by the ideas of reputation and punishment, this paper combines a risk compensation mechanism, which is widely available in the financial sector, with data trading. Specifically, we propose a data trading model with risk compensation and study the evolutionary dynamics of the population using evolutionary game theory. We define λ as the demander’s compensation share that the supplier has to bear for data quality cheating and η as the supplier’s compensation share that the demander has to bear for data leakage. Through numerical solution and simulation, we analyze the evolutionary stable states of the population and find that the risk compensation mechanism in some data trading scenarios can limit the supply side from cheating on data quality or the demand side from leaking data. The results show that λ and η act asymmetrically, with λ being able to affect both supply-side and demand-side strategies, while η affects only the demand-side strategy. This work reveals chaos and asymmetry in data trading with risk compensation, and the proposed model and replication dynamic equations may have implications for future research.