From Behavior-Based Regulation to Limited Rights Confirmation: An Economic Efficiency Study of Enterprise Data Property Rights Delineation

Abstract

In the digital era, the ambiguous delineation of enterprise data property rights has triggered a dual dilemma of the "tragedy of the
commons" (insufficient investment incentives) and the "tragedy of the anti-commons" (circulation obstacles), becoming an institutional bottleneck constraining the release of data element value. This paper employs the Coase Theorem and transaction cost theory as analytical tools to
construct a three-dimensional evaluation framework of "legal definitioneconomic efficiency" (investment incentives, circulation efficiency,
innovation dynamics), systematically examining the current behavior-based regulatory model dominated by the Anti-Unfair Competition Law.
The study finds that while the behavior-based regulatory model offers case-by-case flexibility, it generates threefold efficiency losses due to
uncertain expectations, subjective discretion, and fragmented protection: high institutional transaction costs, distorted innovation incentives,
and obstructed data circulation. Comparative economic efficiency analysis of different property rights models demonstrates that the "limited
property rights" model achieves a suboptimal equilibrium between incentivizing investment and promoting circulation through institutional
design combining rights allocation with statutory exceptions, outperforming both the full ownership model and the pure behavior-based regulatory model. This paper proposes an institutional design scheme of "layered rights confirmation + supporting mechanisms": differentially allocating rights to three categories of objectsoriginal datasets, derivative data products, and public data derivativeswhile establishing supporting mechanisms such as registration and publicity, compulsory licensing, and standardized contracts to reduce transaction costs, thereby
providing theoretical support from an economic efficiency perspective for the implementation of the "Data Twenty Articles."