(Source: "China Market Supervision News" 2026-03-19)

Zhang Yi Tian Chuan

Building a unified national market is a major reform task deployed by the Party Central Committee and the State Council. It aims to establish a unified, transparent and predictable market system, break down regional segmentation, industry barriers and rule differences, and promote the free flow and efficient allocation of factor resources across the country. However, with the in-depth development of the platform economy and the widespread application of algorithms, the formation mechanism of market prices, the way competition is carried out, and the boundary shape of the market have all undergone profound changes, posing new topics for the construction of a unified large market.

In-depth understanding of the internal logic of algorithms embedded in market operations

The widespread application of algorithmic technology is reshaping the way traditional markets operate. First, the algorithm is embedded in the price formation mechanism. Platform companies use dynamic pricing and differentiated pricing implemented by algorithms to transform prices that are traditionally formed by game entities into a process driven by algorithmic rules.

Secondly, algorithms reshape the way of market competition. The platform uses algorithms such as search ranking and traffic distribution to determine the competitive position of merchants, and it has also transformed itself from a market "participant" to an "internal designer" of competition rules.

Again, the algorithm dynamically adjusts the market boundaries. By controlling data interfaces, matching logic and interoperability conditions, the platform continuously generates and adjusts market boundaries, affecting the openness and fairness of the market structure.

In the face of these changes, building a national unified market requires not only regulating a single behavior or individual case, but also building an algorithm evaluation mechanism that can penetrate the technical appearance and identify structural impacts under the existing compliance framework, so as to effectively respond to the reconstruction of the market operating mechanism by algorithms from the perspective of maintaining the stability, fairness and sustainability of the market system.

Accurately grasp the characteristics and shortcomings of the existing algorithm regulation system

Currently, our country has initially established a regulatory framework for the operation of platform algorithms and its market system from multiple dimensions such as technology, behavior, and structure. The technical regulation path focuses on the security, fairness and transparency of the algorithm itself. The behavioral regulation path follows traditional legal frameworks such as anti-unfair competition and consumer rights protection to regulate algorithm-driven market behavior. The structural regulation approach focuses on the "gatekeeper" function of large platforms and emphasizes the restriction of their rule-making power.

However, this regulatory framework still has three shortcomings in practice:

First, technical regulations are out of touch with structural regulations. Technical review focuses on model risks, and structural review focuses on market dominance. However, the algorithm itself, as a generating mechanism for market rules, lacks systematic specifications and evaluation standards, resulting in a poor connection between the two.

Second, the compliance evaluation does not touch the core of the algorithm. Existing inspections mostly focus on the fulfillment of procedural obligations and fail to systematically evaluate whether core mechanisms such as algorithm sequencing logic and traffic distribution rules are fair and reasonable.

Third, the threshold for triggering structural regulations is too high. The identification of abuse of market dominance is highly dependent on traditional indicators such as market share and turnover. The triggering threshold is high and identification is difficult. The shaping effect of algorithms on market structure often occurs before market dominance is formed. The review of concentration of business operators is based on turnover, which makes it difficult to prevent “killer mergers” involving data and technology. In addition, algorithms can achieve de facto market integration through API interface control, data sharing agreements, ranking rule adjustments, etc., bypassing traditional centralized review of property rights.

The practice of the EU’s Digital Market Law provides useful reference. It presets "gatekeeper" standards and proactively intervenes and supervises platform behavior based on pre-set behavioral norms and obligation frameworks. This regulatory idea with forward-looking and structural recognition capabilities is worthy of our reference when building a platform algorithm evaluation mechanism.

Basic principles for scientifically constructing platform algorithm evaluation mechanisms

The core of building a platform algorithm evaluation mechanism that meets the needs of a unified national market is to establish a preventive regulatory framework aimed at maintaining the stability, openness and fair competition of the market system. The following principles should be followed during implementation:

The first is to strengthen algorithm responsibility and interpretability. The platform must bear the main responsibility, based on transparency and explainability, prove the fairness of the algorithm in sorting, recommendation, and traffic distribution, protect the data subject's right to know, and prohibit the use of information asymmetry to implement implicit discrimination or unfair traffic tilt.

The second is to implement the pre-positioning of the burden of proof. Before the algorithm is launched, major rule adjustments are made, or core parameters are changed, the platform should proactively explain its potential impact on price fairness, competitive opportunities, and market openness to regulatory authorities, put the responsibility for risk assessment upfront, and turn ex post facto accountability into ex ante prevention.

The third is to practice algorithmic procedural justice and negotiation. Establish a complete manual intervention, feedback and complaint relief mechanism, promote supervision from one-way review to two-way consultation, encourage industry self-discipline and disclosure of algorithm principles, form a social consensus and supervision basis, and enable algorithm rules to be continuously improved during consultation.

The platform algorithm evaluation mechanism is designed to be deeply integrated with existing supervision. Through regular evaluation and dynamic reporting, the operating status of the algorithm can be monitored on a regular basis. Institutionalizing the above principles can not only ensure the innovative vitality of the platform economy, but also prevent algorithms from causing implicit and cumulative distortions in the market structure without clear illegal circumstances, and lay a solid institutional foundation for the long-term stability, openness and fair competition of the national unified market.

(The author is a professor at the w88 casino of Government, w88 casino; a senior engineer at the Beijing Institute of Standardization)