法学院知识库

随着数字经济时代的到来，市场中的经营者开始利用算法实施共谋行为从而改变传统的合谋市场结构和严重影响到市场竞争秩序。本文主要从算法共谋的定义和特征切入用类型化思维分析四类算法共谋行为，提出规制算法共谋行为面临的法律问题和实际建议。正文内容主要分为五部分，具体内容如下。

文章第一部分是绪论部分，主要以信息时代为社会背景，阐述了研究算法共谋反垄断法规制的理论意义与现实意义，通过对国内外学者的研究成果综述提出本文的研究思路和研究方法，总结了研究的创新点与局限性。

文章第二部分阐述了算法共谋的基础理论。算法共谋行为呈现出智能化、隐蔽化及结构稳定性特征，对其规制的基础理论主要包括经济学维度的勾结寡头垄断理论、社会学维度的风险社会理论及法学维度的垄断协议规制理论。

文章第三部分主要是对算法共谋行为的不同类型进行分析。首先，解释算法共谋行为进行类型化分析的原因在于灵活性和包容性。再者，对信使类算法共谋在内的四种算法共谋行为结合具体案例分析各自的表现形式和构成要件。

文章第四部分主要讨论算法共谋反垄断法规制的法律困境。现行的反垄断法律法规因“协议”概念无法完全规制全部算法共谋行为，存在默示类算法共谋难以规制情形；轴辐类算法共谋场景的出现使传统的垄断协议二分法暴露出过度关注形式与无法应对复杂垄断协议的局限性；自主学习类算法共谋场景中责任承担主体的确定出现争议；反垄断执法机关依靠传统的执法工具很难识别隐蔽性强的算法共谋行为；经营者借助算法工具实施共谋时往往不会存在明确的协议等证明主观意图的证据，这就很难被认定为垄断协议加以处罚。

文章第五部分主要是完善算法共谋反垄断法规制的建议。重新界定合谋中协议的概念，适当延伸合谋协议范围，将默示类共谋纳入规制范围；引入轴辐类协议概念，考虑设置禁止垄断协议制度的一般性条款；明确算法共谋责任主体，秉持人类中心主义视角；在基本价值衡量方面立足消费者权益保护，兼顾效率公平多元化宗旨；加强对算法共谋的执法技术监督，构建完整的算法备案制度；充分利用沟通证据、行为证据在内的间接证据，综合考量行为竞争效果。

With the advent of the digital economy era, operators in the market began to use algorithms implement collusion behaviors and change the traditional collusive market structure, which seriously affected the order of market competition. This article mainly analyzes four types of algorithmic collusion behaviors from the definition and characteristics of algorithmic collusion. The last part of the article puts forward the legal problems and practical suggestions for the collusion behavior of the regulation algorithm.

The first part of the article is the introduction which mainly based on the social background of the information age. This part expounds the theoretical and practical significance of the research algorithm for colluding with anti-monopoly regulations.It also summarizes the research ideas and methods of this article through a summary of the research results of domestic and foreign scholars. The innovation and limitations of the research are discussed.

The second part of the article explains the basic theory of algorithmic collusion. Algorithmic collusion behavior presents the characteristics of intelligence, concealment and structural stability. The basic theories of its regulation mainly include collusion oligopoly theory in economics, risk society theory in sociology and monopoly agreement regulation theory in law.

The third part of the article mainly analyzes the different types of algorithmic collusion behavior. First of all, it explain the reason why analysis of collusion behavior of algorithms. Furthermore, the four types of collusion behaviors of algorithms including messenger algorithm collusion are combined with specific cases to analyze their respective manifestations and constituent elements.

The fourth part of the article mainly discusses the legal dilemma of algorithms colluding with anti-monopoly regulations. The current anti-monopoly laws and regulations cannot completely regulate all algorithmic collusion due to the concept of "agreement" and there are situations in which it is difficult to regulate the implicit algorithmic collusion; the emergence of hub-and-spoke algorithmic collusion scenarios exposes the traditional monopoly agreement dichotomy to excessive Concerns about the form and the inability to cope with the limitations of complex monopoly agreements; the determination of the responsible party in the self-learning algorithm collusion scenario is controversial; the anti-monopoly law enforcement agencies rely on traditional law enforcement tools to find it is difficult to identify the hidden algorithm collusion behavior; when the author uses algorithm tools to implement collusion, there is often no clear agreement and other evidence to prove subjective intent, which is difficult to be regarded as a monopoly agreement and be punished.

The fifth part of the article mainly improve the algorithmic collusion anti-monopoly regulations. Redefine the concept of agreement in collusion, appropriately extend the scope of collusion agreement, and include implied collusion into the scope of regulation; introduce the concept of hub-and-spoke agreement, consider setting up general clauses prohibiting monopoly agreement system; clarify the party responsible for algorithmic collusion; based on the protection of consumer rights in terms of basic value measurement, taking into account the purpose of efficiency, fairness and diversification; strengthening the law enforcement technical supervision of algorithmic collusion, building a complete algorithm filing system; making full use of communication evidence and behavioral evidence indirect evidence, comprehensive consideration of the effect of behavioral competition.