General introduction

While there is no established definition of e-commerce in China, it is broadly understood to mean the sale of goods and services online.[2] China, being one of the fastest growing economies, has become the largest e-commerce market in the world, representing 40 per cent of e-commerce businesses worldwide.[3] China’s e-commerce transaction volume reached 28.66 trillion yuan in 2017, representing an increase of 20.81 trillion yuan over the past six years.[4] The amount of online payment users in China reached 531 million by the end of 2017, reflecting an increase of 310 million users since 2012.[5]

The rapid growth of the e-commerce sector has contributed significantly to China’s economic development. It has facilitated the emergence and fast development of some well-known Chinese e-commerce companies such as JD, Alibaba (Tmall), Meituan-Dianping and Suning, which have achieved great business growth in recent years. In addition, e-commerce has also brought benefits for businesses and consumers by promoting the improvement of logistics systems and facilitating the development of mobile payment, the internet of things, big data and cloud computing, among others. E-commerce has helped businesses to gain a huge number of new customers through efficient and targeted advertising and has also enabled consumers to find the products and services they want more easily. As the OECD noted in their paper on the implications of e-commerce for competition policy: ‘The growth of e-commerce has the potential to increase competition within retail markets, to greatly enhance consumer choice, and to prompt and facilitate innovation in product distribution.’[6]

However, from another perspective, the e-commerce sector is also characterised by its special business model and some completely brand new commercial practices and relationships, all of which has brought great challenges to the traditional market regulatory system including antimonopoly law enforcement activities.

China’s Antimonopoly Law (AML) has been in effect for more than 10 years, since 1 August 2008. During that time, China’s competition practice, enforced by both the administrative agencies and courts, has seen significant achievements. In early 2018, the State Administration for Market Regulation (SAMR) was established to undertake the unified enforcement of antimonopoly law, which had previously been enforced by the Ministry of Commerce (merger control), the National Development and Reform Commission (price-related monopolistic conduct) and the State Administration for Industry and Commerce (responsible for non-price related monopolistic conduct). That reform was welcome and accepted as the most significant change to China’s antitrust enforcement since the AML became effective. In addition to the enforcement agency reform, the past 10 years have also witnessed various benchmarking improvements and achievements in the regulation of concentration of undertakings, anticompetitive agreements and abuse of dominance, among others. Both the competition agencies and the courts have dealt with many significant cases including Qualcomm’s abuse of market dominance, Tetra Pak’s abuse of dominant market position, Medtronic’s price monopoly case and Qihoo 360 v. Tencent concerning the abuse of dominant market position. Further, a series of competition rules and regulations (and some draft guidelines) have also been enacted by the competition agencies to provide guidance on the interpretation and enforcement of competition law in China, such as the Provisions of the Supreme People’s Court on Several Issues Concerning the Application of Law in the Trial of Civil Dispute Cases Arising from Monopolistic Conduct; the Guiding Opinions on the Declaration of the Concentration of Business Operators; and the draft Antimonopoly Guidelines for Automobile Industry.

Nevertheless, despite all the achievements, China’s competition practice has also faced lots of challenges, especially related to the rapid growth of the e-commerce sector. Although the e-commerce sector has brought numerous benefits to consumers and the economy, the potential adverse effects of e-commerce on competition arising from the change of bargaining power between e-commerce operators and traditional business operators have created tremendous challenges for the competition agencies in China. As the OECD has observed: ‘Competition agencies across the OECD are increasingly involved in both enforcement and advocacy efforts relating to e-commerce markets.’[7]

Two big concerns are big data, as a new dimension for market power assessment, and the two-sided market characteristic of e-commerce business, which leads to questions and difficulties for the definition of the relevant market. In addition, the entirely new business models and commercial practices in the e-commerce sector have triggered new issues in the analysis of horizontal monopoly agreements, vertical monopoly agreements and the abuse of dominant market position.

The development of e-commerce in China and new challenges to the AML

Compared with the traditional model of ‘bricks-and-mortar’ retailers, the booming of e-commerce in China has brought more complicated challenges to the legal regime under the AML, mainly from the following two dimensions:

Big data – market power, conspiracy and a potential essential facility

With the significant development of internet and mobile devices, the amount of data in our daily life has been expanding. Capturing and evaluating large data sets, accumulated in ‘volume, velocity and variety’, becomes critical to provide more insights for business and better commercial strategies. For e-commerce, or more specifically the online retail business, exploring needs and preferences, which customers themselves may not be aware of, by using the big data available, has become the most powerful driving force for market players. In China, the layout and application of big data has existed for a long time. For instance, Alibaba, and have leveraged and grappled with the implications of big data to reduce cost, enhance efficiency, share data and conduct targeted sales in accordance with the result of an analysis of consumers’ demand. There is no doubt that big data has become one of the core competitive factors of e-commerce business.

On the other hand, big data also raises challenges for competition regulators. In recent years, there have been more complaints brought against e-commerce giants, due to the potential anticompetitive effects resulting from their possessing a big data advantage. Data-driven mergers are increasing globally,[8] as an effective solution for those undertakings lacking a big data advantage or as a way of further enhancing such an advantage. Further, conspiracy among competitors via algorithms has been found to be an advanced illegal collaboration in certain jurisdictions and there is also a possibility that big data could be deemed an essential facility for the purposes of analysing potential abuses of dominance.

Nowadays, various competition authorities have been spending more effort in considering the impact of big data on competition.[9] In China, although there is no official document on competition cases dealing with big data issues, a few officials from the competition agency have raised their concerns on the potential competition issues related to big data in China.[10] They have said that data has become an important factor for market competition, especially as a basic strategic resource for data-driven businesses and is a core factor affecting the competitiveness of internet businesses. The officials also commented that big data may also be relevant to dominance assessment. Owning important or a massive volume of data may amount to a dominant position in the market and refusing to deal with such data may amount to an abuse of dominance, raising competition issues, which will also become a big challenge for enforcement agencies in China.

The two-sided market and the SSNIP test[11]

It is almost inevitable for e-commerce players to have a platform as part of their business model, whether their business is B2B or B2C. That makes legal assessment under competition law rather complicated, as the platform model is more or less characterised by two-sided or multi-sided markets.

In economic theory, a ‘two-sided market’[12] refers to an economic platform with two distinct user groups that provide each other with network benefits. The organisation that creates value primarily by enabling direct interactions between two (or more) distinct types of affiliated customers is called a multi-sided platform (MSP). Compared with a one-sided market, the two-sided market is characterised by a cross-group network. For example, in a typical two-sided market involved in an e-commerce platform, the increase of consumer numbers will attract more contractors to join the platform and the increase in the number of contractors will also increase the value of the platform, attracting more consumers.

Two-sided markets raise challenges to the traditional approach to defining the relevant market under competition law. In the traditional one-sided market, undertakings set the price of their product or service based on consumer demand and the SSNIP test is a well-accepted economic approach to define the relevant market. However, the unique features of two-sided markets make it difficult to define the relevant market by virtue of the SSNIP test: one side is usually free for consumers and the other side is monetised based on consumer resources and data (for instance, advertising). Furthermore, the competition constraints on the players are more complex than in the one-sided market. In particular, if we only apply the SSNIP test to one side of the market, the price increase will not only lead to the loss of consumers, but also result in a decline in the value of the platform and the loss of contractors on the other side of the market. In other words, it is impossible to test the price effect on one side only, without considering the reaction from the other side. Due to the complicated cross-group network, the SSNIP test in the two-sided market will amplify the influence of the small price rise. Therefore, the SSNIP test may not be applicable in a two-sided market.

Although there have been few competition cases related to the e-commerce sector in China, the difficulty of market definition in the two-sided market is prominent in those cases. For example, in the case of Renren v. Baidu,[13] the first competition case in China, the court only analysed the substitutability of a product or service from the side related to the consumer and defined the relevant product market as the Chinese search engine market, which triggered criticism from various scholars and lawyers in China.

Interaction between e-commerce and AML

Horizontal monopoly agreements

The AML prohibits illegal conspiracy among competitors in the form of price fixing, allocation of market or customers, output restriction, boycott, etc. With the significant development of e-commerce platforms and big data collection and calculation, we have observed in recent years that price fixing can be achieved by algorithms in the e-commerce sector. In particular, some e-commerce platforms have begun to use automatic pricing algorithms based on big data and deep learning technology. The pricing algorithms dynamically react to changes in markets. Integrating with big data, computer algorithms may respond based on the changing pricing data of competitors or other factors relating to pricing such as the supply and demand situation at the time.

The new technology and practice of pricing has both advantages and disadvantages. On the one hand, computer algorithms enable undertakings to respond to the dynamic competitive situation more quickly and, to some extent, more accurately and farsightedly than a response by individuals. On the other hand, pricing algorithms bring new challenges to competition law as well.

For instance, if the same pricing algorithms are used by competitors and their pricing behaviours turn out to be consistent, it will be difficult for competition authorities to investigate and assess whether the behaviour constitutes price fixing with the subjective conspiracy intention, as both the behaviour and the intention are concealed.

Things become more complicated in real commercial life. In a recent US case,[14] the US Department of Justice dealt with a ‘hub-and-spoke’ conspiracy related to an algorithm. In that case, the plaintiff claimed that drivers contract with Uber and charge fees based on the pricing set by the Uber algorithm. It was argued that this arrangement excluded competition among the Uber drivers and constituted a de facto price-fixing agreement between the drivers, with Uber playing the role of ‘hub’ and the drivers playing the role of ‘spokes’. The court recognised the existence of hub-and-spoke conspiracies, namely an entity at a certain level in the market coordinating multiple competitors at another level in the market to reach agreements. When the parties to the vertical agreement recognise that other market participants have also joined the same agreement (it is based on this understanding that they participate in the vertical agreement), these parties (in a vertical relationship) may be considered to be involved in a horizontal agreement to restrict trade.

In summary, with the development of big data and deep learning technology, more and more undertakings use computer algorithms to improve their pricing models and customer service, and to predict the trend of market development. It is very likely that ‘conspiracy by algorithm’ will be a new focus in various competition jurisdictions, including China.

Vertical monopoly agreements

As far as vertical restraints are concerned, Article 14 of the AML explicitly prohibits fixing the resale price of distributors or their minimum resale price (collectively referred to as RPM). In addition to RPM, the article also contains a catch-all clause that permits the antitrust enforcement agency under the State Council to determine other agreements as illegal vertical monopoly agreements. The draft Antimonopoly Guidelines for the Automotive Industry (Auto Guidelines) published on 23 March 2016 provided detailed guidance on RPM and various other vertical restraints not prescribed in the AML such as territorial and customer restrictions. Although they apply only to the automotive industry, it is generally accepted that it should be used as a reference for the same vertical restraints in other industries.

As mentioned earlier in this article, the e-commerce sector has grown significantly in recent years and the total sales through retail e-commerce platforms worldwide reached US$2.304 trillion in 2017.[15] As for China, it has emerged as the world leader in e-commerce in less than a decade, with more online shoppers than any other country and region.[16]

The rapidly developing e-commerce sector, especially online retail business, brings more challenges to the regulation of vertical restraints under the AML, considering the features of the online sales channel and competition environment.

Resale price control on e-commerce business

There are countless sellers on e-commerce platforms that compete with each other in respect of product price, quality, categories, and the pre-sale and after-sale service in order to win business. The price competition between online businesses is much more fierce than that between offline ‘bricks-and-mortar’ stores. As a result, to prevent online distributors from competing too fiercely on price and thus driving down their profits, suppliers may impose resale price control on their online distributors.

Based on the current prevailing practice, such resale price maintenance control on e-commerce businesses is subject to competition scrutiny without any differentiation from traditional RPMs. In fact, suppliers are prohibited from entering into any agreements that fix the distributors’ resale price or the minimum resale price on e-commerce platforms directly or indirectly. In other words, online distributors shall not be subject to any restrictions on their ability to determine the sales price, which can deviate from the list price or recommended price as provided by the suppliers.

There have been no online RPM cases in China, but other jurisdictions have seen a number of such vertical restraint cases. For instance, in 2017, the UK Competition and Markets Authority (CMA) fined a supplier of domestic light fittings £2.7 million for breaking competition law by restricting resellers’ freedom to set their own prices online.[17] The supplier dictated the minimum prices at which resellers could sell its products online by setting a maximum discount off the recommended resale price that resellers were allowed to offer. The supplier used an internet licence agreement as a way of enforcing the pricing policy, an unwritten condition under which the resellers agreed to the pricing restriction.

Restriction on online sales of distributors – passive sales restriction

Passive sales are conducted by responding to unsolicited requests from individual customers, as opposed to active sales which are conducted by actively approaching customers. Online sales, to a great extent, represent a typical category of passive sales – customers browse the official websites of various brands or of a third parties (for example, Tmall, JD) and contact them to make a purchase.

Based on the Auto Guidelines and well-accepted theory and practice in China and other jurisdictions, a general prohibition on offline distributors from engaging in online sales would trigger competition concerns as a vertical restraint regardless of the market share or market power of the supplier. Accordingly, once a supplier has allowed a distributor into its distribution system, it cannot ban the latter from selling products online. A good example is a case in 2017, in which the golf club manufacturer Ping Europe Ltd (Ping) was fined £1.45 million by the CMA for banning UK retailers from selling its golf clubs online.[18]

Territorial and customer restrictions on e-commerce platforms

Territorial and customer restrictions on e-commerce businesses may refer to any restriction of the online distributors’ freedom in respect of where and to whom they may sell products. Territorial or customer restrictions would be legally risky when combined with exclusive arrangements, as they will result in the reduction of competition in a limited region, especially where cross-territory passive sales are restricted.

Based on the Auto Guidelines and well-accepted competition theory and practice in China and other jurisdictions, the risk of exclusive territory restrictions on active sales being in breach of competition law is low if the market share of the supplier as well as the distributor is below 30 per cent. Online promotions or advertisements specifically addressed to a certain territory or customer group are generally considered to be active sales. For example, territory-based promotional banners on third-party websites targeting consumers outside the designated territory would be regarded as a form of cross-territory active sales and would have a low risk of being in breach of competition law if the market shares of the parties are well below 30 per cent. On the other hand, passive sales across the territory should not be prohibited. As to the feasibility of territorial restrictions on passive online sales, it could be achieved by imposing obligations on distributors to automatically reroute customers located outside their territory to the website of another distributor or to terminate an online sales transaction if the credit card data of the consumer contains an address that is not within the distributor’s territory. Such restraints would not be permissible as they would restrict passive sales to consumers outside the designated territory.

Abuse of dominant position

The AML prohibits a dominant undertaking from abusing its position. In order to assess whether an undertaking has abused its dominant position, in common with all major jurisdictions, there is a three-stage test: firstly, define the relevant market; secondly, determine whether the undertaking concerned has a dominant position in the relevant market; and thirdly assess whether there is abusive behaviour.[19]

Relevant market definition in e-commerce cases

It is well accepted that: ‘Market definition is a tool to identify and define the boundaries of competition between firms.’[20] Its objective is ‘to identify those actual competitors of the undertakings involved that are capable of constraining those undertakings’ behaviour and of preventing them from behaving independently of effective competitive pressure.’[21] Compared to horizontal and vertical monopoly agreements, the definition of the relevant market in assessing abuse of dominant position is crucial.

Pursuant to the relevant guidelines,[22] when defining a relevant market, the demand-side substitutability test is often conducted with a focus on the demand for products’ functions, quality recognition, acceptance on price and availability, etc. If constraints from supply-side substitutability are similar to demand-side substitutability, supply-side substitutability should also be taken into account. If uncertainties arise from a certain case in defining the relevant market, the SSNIP test could also be applied.

However, as explained at the beginning of this chapter, the traditional approach to market definition is applicable to one-sided markets and fares poorly when applied to e-commerce cases that always involve two-sided markets.

In Qihoo v. Tencent (the first competition decision decided by the PRC Supreme Court), concerning an internet-based instant messaging (IM) service with a ‘free’ business model, the Supreme Court considered that the applicability of the traditional SSNIP test based on price was limited.

The case involved a dispute between two internet companies in China. Each company built its own user base via a free core product: antivirus software (Qihoo 360) and the QQ instant messaging (IM) software (Tencent). Both companies earnt their revenue through selling online advertising and offering ‘value-added’ services to users. The case was triggered when Qihoo 360 introduced its 360 Bodyguard software, allowing users to control the number of ads that QQ’s IM software could display. On 3 November 2010, Tencent responded by requiring its users of IM services to choose between Tencent’s QQ or the Qihoo 360 software. Tencent’s users therefore had to use security software provided by undertakings (including Tencent) other than Qihoo 360. On 4 November 2010, due to the intervention of the regulator, the Ministry of Industry and Information Technology, Tencent and Qihoo restored the compatibility of their software.

However, Qihoo 360 brought a lawsuit against Tencent at the High Court of Guangdong Province on 15 November 2011, alleging that Tencent had a dominant position in the provision of IM services in mainland China, and that it had abused its dominant position by requiring its customers to choose between its software and the Qihoo 360 software and by bundling its IM and security software to eliminate and hinder competition in violation of the AML. The High Court ruled in favour of Tencent. Qihoo 360 subsequently appealed against the High Court’s ruling and the Supreme Court dismissed Qihoo 360’s claims, upholding the decision made by the High Court.

In defining the relevant market, as previously indicated, the Supreme Court realised the limited applicability of the traditional SSNIP test to internet-based IM services with a ‘free’ business model. Instead, it focused on demand-side substitution. The Supreme Court emphasised that both qualitative and quantitative methods could be used when defining the relevant market; that the qualitative method is generally the starting point; and that it is unnecessary to conduct the quantitative method if a qualitative method by itself is sufficient to reach a clear conclusion. The court then analysed the substitution relationship between IM and Weibo, SNS, mobile text messaging and email. The court defined the relevant market as the IM service market in mainland China. Unlike the High Court, the Supreme Court excluded Weibo and SNS from the relevant market.

Therefore, in a two-sided market such as those typical in the e-commerce sector, one single product or service is related to two groups of consumers, namely end users and contractors. In that situation, it becomes difficult to identify the source of revenue and that puts the SSNIP test in danger of becoming useless.

Introducing a new dimension to dominance assessment

Once the relevant market has been defined, the next step under the AML is to determine whether the undertaking concerned has a dominant position. Article 18 of the AML stipulates a number of factors to be considered including market share and the competitive landscape; the ability to control the market for raw and semi-finished materials; financial strength and technical conditions; the degree to which other undertakings depend on it; and market entry barriers.

As previously mentioned, big data has become one of the core competitive factors of e-commerce business and could be deemed a new benchmark for market power assessment. Based on the practice in other jurisdictions, big data could be regarded as a market entry barrier. In various merger cases, US antitrust enforcement agencies have found data to be a barrier to entry when a firm has a database that would be difficult, costly or time-consuming for a new firm to match or replicate. For example, in Nielsen-Arbitron (2013), the Federal Trade Commission (FTC) determined that the proprietary data of Nielsen and Arbitron was a key part of offering measurement services of audiences across multiple media platforms. The parties had ‘the most accurate and preferred sources of individual-level demographic data for television and radio audience measurement purposes’.[23] The FTC found access to television audience data with individual-level demographic information to be a significant barrier to entry.

Under EU competition law, barriers to expansion or entry can take various forms.[24] They may take the form of advantages specifically enjoyed by the dominant undertaking, such as privileged access to essential inputs or an established distribution and sales network. In other words, big data may be an essential facility for downstream business.

Therefore, with reference to the law and practice in the US and EU, it is probable that certain data may be deemed a significant barrier to entry or even an essential facility for downstream business. As a result, any denial of access to such data might be deemed a refusal to grant access to an essential facility or a network under certain circumstances, which will be discussed in further detail below.

Abusive behaviours: big data as essential facility

One of the practices related to database use that raises concerns in antitrust circles is denial of access. Although denial of access has so far not been challenged in China as an abuse of dominance, Article 17(3) of the AML lists unjustified refusal to deal by a dominant undertaking as an abuse of dominance, which may serve as the legal basis for such a challenge. As such, it is very likely that the competition agencies and the courts in China would refer to the practice in the EU and US when analysing denial of access to big data in the future.

Currently, US antitrust agencies have resisted the idea of big data as anything like an ‘essential facility’ triggering ‘duty to deal’ concepts and have instead treated it only as an asset within the existing merger review context.[25] The prevalent view in the US is that companies are generally under no antitrust obligation to deal with or make available their assets to another company under Section 2 of the Sherman Act. Despite an earlier Supreme Court decision regarding the duty to allow access to services similar to ‘essential facilities’, in its most recent decision dealing with an alleged refusal to deal, the Supreme Court declined to rule that there is a duty to allow access to a telecommunication network in Trinko.[26]

However, in the EU big data could be subject to abuse of dominance rules under EU competition law. In particular, big data might be deemed as an essential facility and thus refusal to grant access to it might be deemed as refusal to deal. There are several cases concerning data sharing such as Magill,[27] IMS Health[28] and Huawei.[29] The Court of Justice has set the following conditions that need to be fulfilled:

(i) the data is indispensable for the downstream product, (ii) there would not be an effective competition between the upstream and downstream product, (iii) refusal prevents the emergence of the second product, and (iv) there is no objective reason for the refusal.[30]

Bronner[31] is often considered to be the authority establishing those tests. In that case, Mediaprint, a newspaper publisher with the only nationwide home delivery scheme, refused to grant another newspaper publisher access to its delivery scheme. Although the court did not challenge the proposition that Mediaprint may enjoy a dominant position in the nationwide home delivery market, it held that there could be an abuse of dominant position only if the refusal was likely to eliminate all competition in the downstream newspaper market; the refusal was incapable of being objectively justified; and the service was indispensable to carrying on downstream business inasmuch as there is no substitute.[32] The court found no abuse of dominance because other less advantageous methods of distributing newspapers existed and there were no technical, legal or economic obstacles to establishing another nationwide delivery scheme.[33] The Bronner decision is novel in that it restricts the scope of the essential facilities doctrine by setting out a requirement that a facility be indispensable rather than a mere dominant position. It also emphasises the importance of a forward-looking downstream market assessment. The Commission has applied the Bronner standards to information sharing.[34]

In China, similar disputes have already emerged in respect of e-commerce networks. Cainiao ran an internet logistics platform in China, connecting and coordinating with delivery services to facilitate overall logistics efficiency. SF Express was a major delivery service in China that connected to Cainiao’s platform. SF Express ceased uploading its logistics data to Cainiao in order to protect its clients’ privacy in May 2017. In response, Cainiao unilaterally terminated SF Express’s access to its network, claiming that the suspension of uploading data had caused a confusion of information in its network to the harm of certain vendors and consumers. SF Express complained that, without access to Cainiao’s network, its delivery service had been disturbed.

Cainiao was reportedly the only provider of an internet logistics platform service of this kind. Questions might arise as to whether the denial of access to its network constituted the refusal to deal under the essential facility doctrine. Assuming that the EU standards in this regard would be applicable, the following questions would have to be answered: (1) whether Cainiao’s network was indispensible to the downstream delivery service; (2) whether the termination could be justified by SF Express’s refusal to upload its clients’ data; and (3) whether the termination eliminated or foreclosed competition in the delivery service market. It is inevitable that economic as well as technical evidence would be required to solve this potential antitrust claim where data value and privacy issues interplay.

The above dispute between Cainiao and SF Express was reported to have been mediated through the initiative of the State Post Bureau of China. No related competition claim has been actually brought to the courts so far. However, in the context of the prospering and diverse delivery market in China and the increasingly critical role of internet platforms, access to these platforms is likely to be scrunitised under the abuse of dominance rules in the future. Given the EU practice and the importance of big data for competition in the e-commerce sector, we cannot exclude the possibility that there are potential antitrust risks for dominant firms in denying access to their data resources in China in the future. However, as discussed above, the authorities are more likely to consider the test applied in the Bronner case and would act cautiously to avoid any inappropriate encouraging of the free-rider problem, namely when people can benefit from a good or service without paying anything towards it, which could lead to the underprovision of the good or service.

Merger filings

E-commerce as a thriving business has also had an impact on the merger filing review sphere, another important field of antitrust enforcement activities in China and other jurisdictions. For instance, in the acquisition of TTK Express by Suning Commerce Group Co Ltd, a B2C e-commerce platform was defined as a relevant product market[35] and in the acquisition of Shanghai Youyi E-Commerce Co Ltd by Global Top E-Commerce Co Ltd, the relevant product market was defined as the e-commerce business in imported goods.[36] There have also been mergers in the taxi hailing app market that were not notified for merger control review and there have not been any decisions or rulings that the transactions are notifiable, probably because the parties’ revenues in each transaction did not satisfy the filing thresholds. Didi Dache and Kuaidi Dache, two of China’s leading companies in this market, merged and rebranded as Didi Chuxing in 2015.[37] In 2016, Uber China and Didi Chuxing merged as one company to explore new ways to gain profitability and build a sustainable business.[38] Both companies had invested billions of dollars in China without turning a profit.

To conduct competition analysis on a merger filing transaction, the first step is to define the relevant market. The e-commerce business model has brought some challenges in that regard. As stated above in the section on abuse of market dominance, relevant market definition is complicated in the cases related to e-commerce.


With the fast development of the internet, the internet of things and artificial intelligence, it could be expected that the interaction between e-commerce, big data and competition would be more and more significant. The traditional view on market power and competition assessment will be further changed by the dynamic effect of the developing e-commerce economy.


[1] Janet Yung Yung Hui is a partner, and Xuefei Bai and Huting Li are associates, at JunHe LLP.

[2] OECD, ‘Implications of E-commerce for Competition Policy – Note by BIAC’ (28 May 2018), available at

[3] E-commerce data: ‘2017 China E-commerce Development Annual Report’ (13 November 2017), available at

[4] ‘China E-Commerce Research Centre’, 2017 China E-commerce Market Data Monitoring Report (26 June 2018), available at

[5] ‘Analysis on Online Payment Users in 2017: Mobile Payment Users Reached 527 million’ (1 February 2018), available at

[6] OECD, ‘Implications of E-Commerce for Competition Policy—Background Note’ (4 May 2018), available at

[7] id.

[8] It is well accepted that the acquisition of Uber by Didi is a data-driven merger.

[9] For instance, the French Competition Authority launched a sector inquiry into big data in online advertising. The German Federal Cartel Office opened an investigation into a potential abuse of dominance by Facebook arising from the company’s privacy policies.

[11] The SSNIP test requires that a hypothetical profit-maximising firm, not subject to price regulation, that was the only present and future seller of those products (‘hypothetical monopolist’) likely would impose at least a small but significant and non-transitory increase in price (5–10 per cent) on at least one product in the market.

[12] ‘Two-sided markets’ are also referred to in the literature as ‘markets with two-sided platforms.’ See Janusz A Ordover, ‘Comments on Evans and Schmalensee’s “The Industry Organization of Markets with Two-sided Platforms”’ 3 J. COMPET, POL’Y INTL 181, 181 (2007), available at

[13] Tangshan v. Renren.

[14] Meyer v Kalanick, U.S. District Court, Southern District of New York, No. 15-09796.

[15] ‘Worldwide Retail and Ecommerce Sales: eMarketer’s Updated Forecast and New Ecommerce Estimates for 2016–2021’, eMarketer (29 January 2018), available at

[16] ‘This Is Only the Beginning for China’s Explosive E-Commerce Growth’ (5 December 2017), available at

[17] CMA decision on online resale price maintenance in the light fittings sector, available at

[19] Steven D Anderman & John Kallaugher, Technology Transfer and the New EU Competition Rules: Intellectual Property Licensing after Modernisation (2006), Oxford: Oxford University Press, page 269.

[20] See §2 of Commission Notice on the definition of relevant market for the purposes of Community competition law.

[21] See §2 of Commission Notice on the definition of relevant market for the purposes of Community competition law.

[22] The Guidelines of the Antimonopoly Committee of the State Council for the Definition of the Relevant Market.

[24] Guidance of 3 December 2008 on the Commission’ s Enforcement Priorities in Applying Article 82 EC Treaty to Abusive Exclusionary Conduct by Dominant Undertakings OJ [2009] C 45/7. Available at

[25] Paull Lugard and Lee Roach, ‘The Era of ‘Big Data’ and EU/US Divergence for Refusals to Deal’, Antitrust, Vol 31, No. 2, Spring 2017.

[26] Verizon Commc’ns Inc v. Law Offices of Curtis V Trinko, LLP, 540 US 398 (2004).

[27] Case C-241/91P and C-242/91P, Radio Telefis Eireann (RTE) et Independent Television Publications (ITP) v Commission, ECLI:EU:C:1995:98.

[28] Case C-418/01, IMS Health v. NDC Health, ECLI:EU:C:2004:257.

[29] Case C-170/13, Huawei Technologies v. ZTE, ECLI:EU:C:2015:477.

[30] Paul Lugard and Lee Roach, ‘The Era of “Big Data” and EU/US Divergence for Refusals to Deal’, Antitrust, Vol. 31, No. 2, Spring 2017 at pages 62–63.

[31] Bronner, 1998 ECR at I-7791.

[32] id. at para 41.

[33] id. at paras 43, 44.

[34] Commission Staff Working Document of 10 January 2017 on free flow of data and emerging issues of the European Data Economy, SWD (2017) 2, page 22. See also Guidance of 3 December 2008 on the Commission’ s Enforcement Priorities in Applying Article 82 EC Treaty to Abusive Exclusionary Conduct by Dominant Undertakings OJ [2009] C 45/7, paras 75-90. Available at

Unlock unlimited access to all Global Competition Review content