CHINA LEGAL SCIENCE 2018年第1期 | 中国《反垄断法》对垄断行为的法律适用： 标志性案例分析（中）

III. THE PROHIBITION OF ABUSE OF A DOMINANT POSITION

A. Abuse of a Dominant Position under the AML

The abuse of a dominant position to eliminate or restrict competition is prohibited under the AML. Article 17 provides a non-exhaustive list of conduct which is deemed to be abusive: selling at unfairly high prices or buying at unfairly low prices; below cost sales without justification; refusals to deal without justification; exclusive or designated dealing without justification; tying or imposing unreasonable trading conditions without justification; and discriminatory dealing. From the wording of article 17, the opportunity to claim justification is not available for unfair pricing and discriminatory dealing, which raises particular concerns since the freedom for a monopolist to set prices is significantly restricted even though such stipulation is not uncommon in other jurisdictions. It is argued that a cautious approach should be adopted to unfair pricing claims, otherwise the incentives of undertakings to innovate would be reduced.

A dominant market position is defined in paragraph 2 of article 17 as a market position held by a business operator having the capacity to control the price, quantity or other trading conditions of commodities in the relevant market, or to hinder or affect any other business operator to enter the relevant market. In the EU, a slightly different definition of dominant position was laid down in Continental Can Co., and later upheld in United Brands as “a position of economic strength enjoyed by an undertaking which enables it to prevent effective competition being maintained in the relevant market by giving it the power to behave to an appreciable extent independently of its competitors, customers and ultimately of its consumers”.

Article 18 provides a non-exhaustive list of factors for assessing whether a business operator holds a dominant market position: the market share of a business operator and the competitiveness of the relevant market; the capacity of a business operator to control the sales markets or the raw material procurement market; the financial and technical conditions of the business operator; the degree of dependence of other business operators upon the business operator in transactions; and the degree of difficulty for other business operators to enter the relevant market.

Article 19 contains three rebuttable presumptions of dominant position based on the market share hold by the business operators in question. It provides, inter alia, that a business operator can be assumed to have a dominant market position if the market shares of one business operator account for 50 percent or more in the relevant market. Although the EU Competition Law does not have similar presumptions of dominance, the market share of an undertaking is an important factor in the assessment of a dominant market position. In Hoffmann-La Roche, the court held that “the existence of a dominant position may derive from several factors which, taken separately, are not necessarily determinative but among these factors a highly important one is the existence of very large market shares”. “Very large market shares” were defined to be 50 percent or more in AKZO. Nonetheless, the presumption of the existence of a dominant position based on substantial market share can be rebutted since the importance of market share varies in different markets. This is especially true in relation to dynamic and volatile markets like technology markets, in which incumbents can be more easily surpassed by new entrants no matter how powerful they once were, like IBM, Motorola and Nokia. In such markets, competition is always intense and market conditions change significantly over time. The fact that a particular undertaking holds a large share of the market does not necessarily give it the power actually to control prices or output or implement other anti-competitive policies. Considering the fact that the AML is not helpful in clarifying the factors which are deemed relevant in rebutting the rigid presumptions of dominance based on market shares, it is thus important for competition enforcement agencies to rely more on a competition-oriented assessment of dominant position as provided in article 18. The next section will examine how the Supreme People’s Court had analysed market dominance in dynamic technology market without relying on article 19 presumptions.

B. Qihoo 360 v. Tencent 

On October 16, 2014, the Supreme People’s Court delivered its first AML judgement, which concerned an abuse of dominance dispute between Qihoo 360, a provider of free security software known as Koukou Bodyguard, and Tencent, a provider of a free instant-message (IM) application, QQ. In the judgment, the Supreme People’s Court upheld the decision of the court of the first instance, Guangdong High People’s Court, and dismissed claims made by Qihoo on exclusive dealing and illegal tying.

In September 2010, Qihoo released a new security named Koukou Bodyguard, and claimed that this product was able to protect the privacy and internet security of QQ users by preventing QQ from scanning the hard drive of its users. As a counter measure, on November 3, 2010 Tencent announced that it would make QQ incompatible with any of Qihoo’s security software by implementing the “choosing 1 from 2” plan, which forced the users of both products to choose either one to uninstall. However, the compatibility was restored on the next day “under the intervention of relevant government department” after the recall of Koukou Bodyguard.

In April 2012, Qihoo brought a private action before Guangdong High People’s Court claiming that Tencent had abused its dominant position in IM software and service market in mainland China by carrying out exclusionary practice, and by tying the installation of QQ with that of QQ Software Manager, which included Tencent’s security software QQ Doctor. The claims were dismissed by the Court on the ground that Qihoo’s definitions of relevant product and geographic markets were too narrow. Since Tencent did not possess a dominant position, there was no abuse. The case was then appealed by Qihoo to the Supreme People’s Court, where an in-depth economic analysis was conducted focusing on three main aspects.

1. Market Definition. — The Supreme People’s Court stated that an unequivocally and clearly defined relevant market was not necessary for individual abuse of dominance case in which the definition of relevant market was merely a tool to assess the market power of business operator and the effects of the alleged monopoly conduct on the competition. 

In the first instance judgment, the Guangdong High People’s Court applied the “small but significant and non-transitory increase in price” (SSNIP) test to define the product market. Since the test required an increase of the product price by 5 percent to 10 percent, which was not possible with a totally free product, the court alternatively assumed a small absolute increase in price. Although the Supreme People’s Court acknowledged that the Hypothetical Monopoly Test (HMT) was generally applicable in defining the relevant market, it rejected this rigid application of the HMT on the ground that charging a price on a previously free product would result in the fundamental change of business model in a two-sided market and a significant decline in the number of users who might be forced to look for alternative products with no substitutability. It therefore rendered the market definition too broad. The more appropriate test, in the opinion of the Supreme People’s Court, should be the SSNDQ test. 

Instead of applying the HMT, the Supreme People’s Court assessed the demand-side substitutability based on product characteristics, functions, quality and accessibility among integrated and non-integrated IMs, Mobile Instant Messaging (MIM), social networking websites, Weibo, text messaging and email. In conclusion, the Supreme People’s Court rejected the finding of the court of the first instance that social networking websites and Weibo were included in the relevant market, and defined the relevant product market as IM service market, including both PC-based IM and MIM as well as both integrated and non-integrated IM. The Supreme People’s Court also overturned the finding of Guangdong High People’s Court that the relevant geographic market was global on the grounds, inter alia, that most major foreign IM service providers such as MSN, Yahoo, Skype and Google had already been operating in China before the alleged monopoly conduct was carried out. The choice of overseas IM services for Chinese users was considerably limited, and since foreign IM service providers could only operate in China’s IM service market by obtaining necessary government permits and establishing joint ventures with Chinese telecommunications companies, the likelihood that other foreign IM service providers could enter China’s market and possess significant market power in a relatively short period of time (for example, a year) was low. The Supreme People’s Court therefore concluded that the relevant geographic market in this case was mainland China.

2. Market Power. — The Supreme People’s Court determined whether Tencent had a dominant position in the IM service market in China by assessing market share, competition condition in the relevant market, the ability of Tencent to control product prices, outputs and other trading conditions, Tencent’s financial and technological capacity, the reliance of business operators on the transactions with Tencent, and the entry barriers of the relevant market. This analytical approach is in conformity with article 18 of the AML which sets out the factors to be considered in determining a dominant market position.

First, the Supreme People’s Court found that, from 2009 to 2011, Tencent had an average annual market share of over 80 percent in the PC-based IM service market, while the closest competitor during the same period had only 4.2 percent. In the MIM service market, Tencent had an average monthly market share of over 90 percent since August 2012. Under article 19(1) of the AML, when the market share of a business operator accounts for 50 percent or more in the relevant market, it may be assumed that it has a dominant market position. Nonetheless, the Supreme People’s Court stated that high market share did not necessarily equalise a dominant market position especially in the telecommunications industry characterised by dynamic competition.

Second, in the IM service market in China at the time when the alleged monopoly conduct took place, there were dozens of IM products, some of which had more than 100 million users. In addition, there was a trend in the IM service sector to integrate other internet platforms such as advertisements, news, online dating, and microblogs into the IM platform, and as a consequence competition became diversified. Therefore, there was sufficient competition in the IM service market.

Third, since IM services had been conventionally provided to users for free, users would lack the willingness to pay for any IM products that were charged and no IM service providers would have the ability to control the prices of IM services. Moreover, owing to the fact that substitution between IM products would induce no or insignificant financial and technological obstacles, and many IM products were homogenised without major difference in the functions, IM service providers would normally not risk refusing to provide services or changing trading conditions. Therefore, the ability of Tencent to control product quality, quantity or other trading conditions was weak.

Fourth, although Tencent had abundant financial and technological resources, other major competitors in the IM service market such as Alibaba, Baidu, Microsoft and China Mobile all had comparable conditions which were sufficient to impact the leading role of Tencent. Therefore, the influence of the financial and technological conditions of Tencent on its market power was very limited.

Fifth, Qihoo claimed that IM service had significant network effects and customer stickiness: an increase in the number of users would attract more users, and the cost to change IM service was high after the formation of a social network through long-term use. The Supreme People’s Court rejected this claim and stated that it was common for a user to use multiple IM products simultaneously. According to a survey, 90 percent of IM service users used 2 or more IM products, therefore the network effects and customer stickiness were substantially alleviated. Moreover, the development of new technology such as easier export and import of contact details, which could be accessed by IM products, between different mobile phones deepened the alleviation of network effects and customer stickiness. The Supreme People’s Court supported its reasoning with the example of MSN. MSN had a global market share of 40 percent in 2011, however, the number of active users decreased to only 100 million after only one year in 2012. Therefore, customer stickiness is not necessarily equal to a reliance on a particular business operator.

Sixth, the Supreme People’s Court assessed the difficulty to enter the market and expand market share. It was possible for a new entrant to increase its market share dramatically in a short period of time and create an effective competitive constraint to the incumbent. For example, the number of users of Fetion, developed by China Mobile, increased from 0 to 100 million in less than one year. The Supreme People’s Court also found that the increase rates of users of several IM services surpassed the increase rate of the number of internet user during the same period. This evidence indicated that the entry barriers and difficulty of market expansion were low.

In response to Qihoo’s argument that Tencent dared to implement the “choosing 1 from 2” plan only because Tencent believed that most users would choose QQ over Koukou Bodyguard and it indicated Tencent had a dominant position, the Supreme People’s Court stated that the subjective intention of Tencent to implement the plan was irrelevant to the determination of a dominant position. Besides, the implementation of the plan, which lasted merely one day, led to a dramatic increase in user numbers of major competing IM services. In the same month in which “choosing 1 from 2” took place, the user numbers of MSN, which had been declining, increased by 61.93 percent compared with the previous month while Fetion by 9.95 percent and Alibaba by 5.15 percent. Since many IM service providers were able to expand their market due to Tencent’s  short-lived alleged monopoly conduct, it could be proved that Tencent did not have a dominant market position.

3. Abusive Conduct. — Although the Supreme People’s Court could choose to exercise judicial economy after finding that Tencent did not have a dominant position, it nonetheless continued with the analysis of Tencent’s alleged abusive conducts. 

Qihoo claimed that the “choosing 1 from 2” conduct was a violation of article 17(4) of the AML because Tencent forced the users to uninstall their own software. The Supreme People’s Court found that: first, the incompatibility practice targeted only the products and services of Qihoo. On the face of it, Tencent required users to choose between QQ and Koukou bodyguard, which in fact restricted its own operating environments. Although such restriction might cause inconvenience to users, the inconvenience could be mitigated by the fact that there were sufficient substitutions in both the IM service market and security software market; second, the incompatibility practice was a countermeasure in response to Qihoo’s unfair competition practices, i.e. the release and operation of Koukou Bodyguard which specifically targeted QQ, rather than a conduct intended to eliminate or restrict competition in the IM service market; third, as discussed above, the one-day “choosing 1 from 2” conduct in fact generated more vigorous competition. It was therefore justifiable to speculate that if the conduct had lasted longer, the market share of Tencent would have declined more significantly. On the other hand, although Tencent’s conduct had substantial adverse effects on the market share of Qihoo, the focus of the AML was not on protecting individual undertaking but on whether a healthy market competition mechanism was distorted or undermined. After the “choosing 1 from 2” took place, Qihoo’s market share in the security software market dropped from 74.6 percent to 71.3 percent while during the same period, Tencent’s market share in the security software market only increased by 0.57 percent from 3.89 percent to 4.46 percent. It was clear that the effects of Tencent’s conduct on the security software market were negligible, and competition in the security software market was not eliminated or restricted. In conclusion, the “choosing 1 from 2” conduct did not amount to monopolistic conduct for the purpose in the AML.

In relation to whether the tying practice of Tencent was in breach of article 17(5) which prohibited business operators with the dominant market position from tying products without any justification, the Supreme People’s Court found that: first, Qihoo failed to provide sufficient evidence to prove that Tencent had leveraged its competitive advantages in the IM service market into the security software market since Qihoo had a market share of no less than 70 percent in the security software market after both the “choosing 1 from 2” and the tying practice took place while Tencent only had less than 5 percent. Second, it was reasonable to tie the installation of QQ with that of QQ Software Manager since such practice could realise the integration of functions to protect the security of accounts and improve the performance and value of QQ. Third, the coerciveness of Tencent’s tying practice was not obvious. Although there was no notification when QQ and QQ Software Manager were installed, users could freely choose to uninstall any of the software. Therefore, the supply of IM service was not restrained upon the use of QQ Software Manager.

4. Implications of Qihoo v. Tencent. — In upholding the decision of Guangdong High People’s Court, the Supreme People’s Court corrected several mistakes made by the court of the first instance, the most noteworthy being the recommended replacement of the SSNIP test with the SSNDQ test. Although the Supreme People’s Court did not actually apply the SSNDQ test in the decision, it did carefully analyse the scope of the SSNIP test and its limited relevance to a product that was provided free of charge. It identified the special characteristics of the IM product, and examined the standard context in which the SSNIP typically applied. Instead of rigidly applying the SSNIP test, the Supreme People’s Court acknowledged that the SSNDQ test, although it is in a way more difficult to apply owing to the non-quantifiable feature of the quality of products, is more adaptable to fit the special conditions of the internet industry. It displays the growing familiarity of the court with well-established competition doctrines and the application thereof.

The same degree of flexibility can also be found in the determination of market power, where the court commendably ruled that a considerable market share, even as high as 80 percent, did not necessarily render a dominant market position. Normally, the AMEAs rely heavily on market share to determine if a business operator has a dominant position, but when a relatively high market share does lead to the establishment of a dominant position, market share cannot accurately demonstrate the real market power a business operator possesses. The characteristics such as high innovation rate and low dissemination cost of the technology sector in general and PC-based or mobile software in particular dictate that the enforcement experience in traditional sectors should be adjusted accordingly to get the law effectively applied. Therefore, the Supreme People’s Court successfully set a standard for assessing market power in a dynamic and volatile market through quantitative analysis.

Some parts of the analysis can be further elaborated. For example, in relation to the effects of Tencent’s “choosing 1 from 2” conduct on the competition condition of IM service, it would be better to establish more firmly the casual link between Tencent’s conduct and the increase in its competitors’ market share, instead of taking the statistics for granted, by examining if the increase in market share was a result of a newly released IM products or a major upgrade of the previous version. Nevertheless, this landmark judgement provides valuable guidance, for example, an effects-oriented and statistical approach towards assessing alleged abusive conducts for future AML litigation as well as administrative enforcement, especially in the vigorously developing internet sector and also the two-sided markets. Competition issues are derived from economic development, and a myriad of novel competition issues which cannot be foreseen by any legislature will be raised as economic activities continuously develop. It requires both the courts and the AMEAs to apply the AML in a way which preserves the basic norms and principles on which the competition law is founded such as the distinction between protecting competitive process and protecting competitors, and addresses the unique features of a given product and market by giving a sophisticated economic analysis. 

Chapters II and III above discussed the application of the AML respectively to monopoly agreements and abuse of a dominant position. The discussion found that while both courts, the Shanghai High People’s Court and the Supreme People’s Court had done an excellent job in conducting economics-based reasoning and analysis, both NDRC and SAIC fell short of this standard. The next section will analyse MOFCOM’s application of the AML in reviewing concentrations. It intends to show that, compared with the approaches taken in Qihoo v. Tencent, the same degree of inflexibility and opacity presented in the decisions of NDRC and SAIC can also be found in merger control.

长按识别图中二维码，快速关注我们！