The Research Of Antitrust Law Regulations Of Algorithms Under The Digital Economy Context

Algorithms might exert functions of improving economic effeciency,enforcing antitrust by competition authorities,promoting monopoly in the digital economy.The two formers are positive effects,while algorithms as the tools of assisting monopoly,are negative effects.Algorithms might serve as In the era of artificial intelligence,markets become increasingly transparent.Through platform-based e-commerce trading,there is a large amount of data that market participants seek to utilise for their bebefit.Algorithm-driven platforms have become key instruments for success in a digital economy.Although they can generate positive effects on consumer welfare and welfare in general,algorithms may foster anticompetitions such as tacit colusion,personal pricing,self-favouring etc,adversely affect consumer choice,even pose a threat to pluralism.Anti-competitive effects through the use of algorithms are considered possibe,particularly in the form of collusion.However abuse of dominate position arising from the algoritms need to definite the relevant markets in the digital markets.Meanwhile traditional definiton approaches of markets on static comoetition are not suitable to definiton of digital markets in dynamic competition,thereby it is measured with the new non-price competition standards,for instance,big data,indirective network effects,platforms,online attention,quality.Yet collusion is typicallyunderstood as a market outcome in which companies achieve higher profits than in competition through forms of coordination,for example,by coordinating prices or quantities.Collusive behaviour is therefore to the detriment of customers and is undesirable from the point of view of society as a whole.In a word,despite algorithms may propel the development of digital economy,they boost platforms to carry out anti-competitive actions most likely.However due to algorithmic particular attribution,anti-competitions arisen from algorighms such as revisiting the concept of agreement,allocation of the algorithmic antitrust’ liabilities rsesult in greater disputes in theoretical and practical circles,which lead to the questions whether the current antitrust laws around the world deal with the anti-competitive and offensive actions by algorithms.Algorithms raise the economics and legal issues in the areas of abuse of domminance,algorithmic pricing and collusion,merger and acquisitions.Algorithms can differentiate marginal customers from inframarginal cuntomers,so cost of predation and exclusion conduct is dramatically reduced because predatory price,the rebate or bundle is offered only to the select few customers or only for spefific transtions.Therefore below-cost pricing can be implemented without a substantial loss,challenging the traditional assumptions of predatory pricing and recoupment.In addtition,self-preferring facilitated by algorithms are similiar with leveraging,discriminate,refusal and tying conducts,but legal plights,such as differences between self-preferring and the former conducts,what conditons platforms perform the obligations not to self-prefer their own products are arisen,too.In data-intensive sectors such as the digital economy,pricing algorithms can facilitate collusion by automating collusive behaviour and thus technically accelerating it.For example,they can stabilize collusion by allowing the collection of information on competitors’ prices and sanctioning deviations from collusive market outcomes more quickly.The use of pricing algorithms can also render explicit anticompetitive agreements or concerted practices dispensable.Lastly,in the case of self-learning algorithms,the decisive business decision is moved to the time of the decision on the pricing algorithm and is not only made during pricing.It is regularly difficult for the competition authorities to detect collusive behaviour,because they don’t understand how algorithms are run,namely“black-box”.Moreover,the reason that blaming who shoud bear the resposibility isn’t rather easy is hard to detect violating responsible subjects and judge their violating intents,This concerns,in particular,the determination of whether a concerted practice is actually taking place.Difficulties tend to increase when the parties use pricing algorithms.This also concerns the proof of a possible price increase.Should algorithms indeed facilitate coordination in markets other-wise not prone to it,we need to explore what tools,if any,can be used to reduce the negative welfare effects of algorithmic-facilitated coordi-nation.In addition to previous market and state regulations,this article focuses on legal remedies.Indeed,algorithms challenge the assumptions on which antitrust law is currently based.lgorithms,unlike humans,can “read the minds” of other algorithms even before they perform any action,thereby transforming the need for an explicit commitment to coordinate or to punish deviations.This new reality requires us to rethink concepts that stand at the basis of our laws,like the meeting of minds,intent,consent,and communication,and possibly create a new taxonomy to fit the algorithmic world.The analysis is timely: competition authorities all over the world are starting to explore such issues in depth,and the legality of algorithmic-facilitated coordination is likely to become a major issue,given rapid advancements in machine learning.Conscious parallelisms result in hot debates of agreement concept on tacit collusion,especially how to judge whether unilateral signalings are illegal actions violating cartels.Due to powerful algorithmic self-learning,predicting and reinforcing,sometimes algorithms transcend the scope of developers and users,so that how to establish the resposibile subjects according to law challenges the legislators.When competition enforcers detect anticompetitions algorithms propel,algorithmic opacity leads to the disputes over commercial secret and intellectual property.When algorithms analyse the influencing factors of questions with big data,they may unfairly classify types of data,ultimately making unfair decisions to consumers and customers.for example,personal pricing to consumers are manifestation of price discrimination to them,which originate from dicriminative,unfair,bias decison of algorithms.Laws should confer them with the explainable rights why they ask developers and users of algorithms to make a decision in order to eradicate bias.Whether competition authortities have the capability to ascertain that algorithms infringe competiton law,consumers protection law etc.is a challenge to academical circles and conpetition enforcers.The paper puts forward auditing systems algorithmic bias,algorithmic risk assessments and algorithmic impact evaluation,where algorithmic risk assessments include internal auditing of platforms or enterprises,external auditing of outside expertises and enforcers,which they respectively have advantages and disadvantages,and should complement each other,to propel algorithmic compliance,,protect fair competition enviroments,maintain consumers’ benefits.At last,the issue of algorithms emerged in merger control.disclosure of a pricing algorithm between competing firms may contravene competition law prohibitionson the sharing of competitively sensitive information.As such,merger parties may face challenges in the due diligenceprocess,as well as difficulty in knowing whether the merger has an anticompetitive effect vis-à-vis increasing the riskof tacit collusion.Based on the above analysis,academic circles,legislators and pratical fields cooperate each other absolutely sincerely to settle the harm to which algotithms give fair competition and consumers.Due to dynamic competition of digital markets,antitrust analysis ways of static competition in the offlinemarkets should modified.With respect to definition of digital markets,establlishing regulatory principles of digial markets,building up the new systems of market domination ect.,antitrust analysis index of traditional markets should be updated.While data bridge market with algorithms,legislators not only exploit potentials and value of data set fully,but protect rights of data subjects maximumly.On the basis of ascertaining and confirming algorithmic infringement,how to accrue to accountability is a difficult problem enforcers face.Afer comparing diversities of ways with which scholars bring out pursuing algorithmic liabilities,the paper sugggests single economic doctrine,which treats algorithms as employees of algorithmic company,whose libilities areattributed to the employing conmpany.Finally,for the harm which algorithmslead to competition and consumers,victims and consumers can maintain their own legal rights through market’self-regulation ways like black box tankering r,increasing likelihood of deviation,through judical remedies of structure remedies like divestitures and behavioral remedies like mandatoryrequirements of transparancy,or through administral remedies like deceleration,price regulation,smart regulation,compliance by design etc.In short,under the notion balancing regulation and innovation,competition policy makers should not neglect the actions that algorithims threaten competition enviroment s and theoretical and pratical prolem they will face,at the sme time they can’t excessively intervene with the innovation and development,so that they delay the opportunities of economic movement,influence the improvement of comprehensive national strength Therefore they escort for createing really free,fair digital markets.