Research On Antitrust Regulation Of Algorithm Collusion

The emergence of algorithms has greatly changed the production activities of human beings,and also has a profound impact on people’s daily life.The importance of algorithms in production and life activities should not be underestimated.In the field of commercial activities,the use of algorithms to achieve collusion also occurs,and compared with the traditional collusion,the algorithm collusion may be more harmful,seriously damage to the order of market competition.In view of this,this paper studies the antitrust regulation problem of algorithm collusion.The first part is the basic problem of algorithm and algorithm collusion.This part introduces the basic problem of algorithm and algorithm collusion,including the concept of algorithm,commercial application and the concept and characteristics of algorithm collusion.With data as the core,algorithm is a method,instruction and program to solve problems or achieve tasks.In business activities,it is mainly used by operators for prediction analysis and business process optimization.Algorithm collusion refers to the use of algorithms to implement（price）collusion behavior.It can be divided into narrow sense algorithm collusion and broad sense algorithm collusion.Algorithm collusion has the characteristics of high intelligence,high concealment,high convenience and high stability.The second part is the typing analysis of algorithm collusion.In this section,four different algorithms’ concept and characteristics are expounded respectively,including Monitoring Algorithms to The Computer as Messenger,Parallel Algorithms to Hub and Spoke,Signaling Algorithms to Predictable Agent,Self-learning Algorithms to Autonomous Machine,and combined with the foreign "U.S.v.Topkins","Spencer Meyer v.Uber","AirLine Tariff Publishing" of typical and representative case analysis.And discuss the influence of different types of algorithm collusion on the identification and regulation of illegality.Finally,the paper summarizes the cases of algorithm collusion and four different types of algorithm collusion in the world in recent years.The third part is the antitrust regulation dilemma of algorithm collusion.This part studies the illegal identification especially antitrust law regulation trouble of algorithm collusion,think algorithm collusion to bring antitrust law regulating the plight of the main body elements,namely,how should identify the main body of collusion;the plight of the subjective factor considered,namely,how should prove and identify collusion intentions between enterprises;the plight of the behavior factor considered,namely,how should identify collusion behavior;and the results of elements identified in dilemma,which should take what kind of value concept about collusion results brought by the algorithm.The fourth part is the antitrust regulation suggestions of algorithm collusion.In view of the antitrust law regulation dilemma of algorithm collusion,this part appropriately draws on the experience of foreign countries,and puts forward the antitrust law regulation suggestions of algorithm collusion from the macro-level,meso-level and micro-level of concept,system and practice.On the conceptual level,antitrust law enforcement agencies should pay attention to grasp the "degree" and balance,modest and moderate,uphold the "market competition first,prudential control" law enforcement concept;On the institutional level,the principle of no-fault liability should be established,the concept of "monopoly agreement" should be redefined,and the subject of liability of algorithm collusion should be clarified.On the practical level,practice rules and operation guidelines should be formulated,typical cases should be issued,enterprises should be forced to moderately disclose algorithms within a limited scope,"new algorithms" or other engineering technologies for regulating algorithms should be developed,law enforcement personnel should be expanded,and business training should be strengthened.