Analysis method of internet-related market definition and abuse of dominant market position

      —— Qihu v. Tencent on Abuse of Dominant Market Position

 [Syllabus]

1. In handling anti-monopoly cases, defining the relevant market is an important analytical step. However, clearly defining the relevant market depends on the case’s specific circumstances. In a case of abuse of dominant market position, the definition of relevant market itself is not the purpose, but a tool to assess the business operator’s dominant market position and impact of the alleged monopolistic practice on competition. If the business operator’s market position and market impact of the alleged monopolistic practice can be assessed through direct evidence of elimination or hindrance to competition, it will be unnecessary to clearly and conclusively define the relevant market in every case involving abuse of dominant market position.

2. Hypothetical Monopolist Testing (HMT) is a generally applicable analytical method that defines the relevant market. In practice, it is assumed that HMT can be conducted through methods such as Small but Significant and Non-transitory Increase in Price (SSNIP) or Small but Significant and Non-transitory Decrease in Quality (SSNDQ). The free-of-charge features of Internet-based instant messaging (IM) services, makes users highly sensitive to price. As the SSNIP test method will lead to an excessively broad definition of the relevant market, SSNDQ should be adopted for qualitative analysis.

3. Keeping in mind IM service’s low cost and high coverage in defining the relevant geographical market, a comprehensive assessment can be made based on the actual region where majority of users select the goods, provisions of laws and regulations, status quo of overseas competitors and timely access to relevant geographical markets and other factors.

4. In the Internet sector, market share is a relatively crude and potentially misleading indicator for evaluating dominant market position. Its position and role in determining dominant market position must be determined based on the circumstances of specific cases.

 [Case No.]

Supreme People's Court (2013) MSZZ No.4

 [Cause of Action]

Dispute on misuse of dominant market position

 [Keywords]

Monopoly, abuse of dominant market position, relevant market, market share

 [Relevant Legal Provisions]

Article 17, Article 18 and Article 19 of Anti-Monopoly Law of the People's Republic of China

 [Basic Facts]

This case was filed by Beijing Qihu Technology Co., Ltd. (Qihu) in the High People's Court of Guangdong Province, alleging that Tencent misused its dominant market position with respect to relevant IM software and service market. On November 3^rd, 2010, Tencent released A letter to QQ users requesting them to stop running QQ on computers that had Qihoo 360 software. On November 4^th, after strong intervention of relevant departments of the State, 360 Security Center announced that the current QQ version and 360 software were fully compatible. In September 2010, Tencent QQ Instant Messaging software and QQ Software Management were provided to users for installation as a package. However, users were not prompted that QQ Software Management would be installed when QQ Instant Messaging software was installed. On September 21^st, 2010, Tencent issued a notice that current QQ Software Management and QQ Doctor will be automatically upgraded to QQ Computer Housekeeper. Qihu claimed that Tencent refused to provide related software services to users who had installed Qihoo 360 software and forced users to choose between Tencent QQ and Qihoo 360, thus constituting a restrictive trade practice, which is prohibited under the Anti-Monopoly Law. Tencent’s act of bundling QQ Computer Housekeeper with its Instant Messaging software and installing QQ Doctor under the guise of upgrading QQ Housekeeper constituted a bundled sale, which is prohibited under the Anti-Monopoly Law.

The case was filed in the court of first instance, the High People's Court of Guangdong Province, which held that: 1. On the definition of relevant market, Qihu’s claim that integrated IM service constitutes an independent relevant commodity market and relevant geographical market, which in this case was the Mainland China market, could not be established. The relevant commodity market in this case goes far beyond integrated IM service market and the relevant geographical market should be the global market. However, the court did not clearly define the scope of relevant commodity market in this case. 2. On the dominant market position,  as Qihu misjudged the relevant commodity market in this case, the evidence it provided did not prove that Tencent had a monopolistic position in the relevant commodity market. Qihu’s litigation claims lacked factual and legal basis, and thus could not be established. Therefore, Qihu’s entire claim was rejected in the judgment.

Not accepting the judgment, Qihu filed an appeal. The main points of contention in the appeal are: (1) The first-instance judgment did not determine the relevant commodity market in this case, and the basic facts of the case were not clearly established. (2) Basic method used in the first-instance judgment for the analysis of the relevant commodity market is incorrect. Hypothetical Monopolist Test (i.e. SSNIP Test) should not be directly applicable to free-of-charge products, to define relevant market in this case, and it is incorrect to use the SSNIP Test. Relevant commodity market in this case should be defined as a personal computer-side instant messaging software and service that integrates text, voice and video. (3) Determination of relevant geographical market in the first-instance judgment is obviously incorrect. Relevant geographical market in this case should be Mainland China. (4) The verdict of the first-instance judgment that Tencent did not possess dominant position in the relevant market is incorrect. Tencent’s share in the relevant market is more than half, and thus it should be presumed to possess a dominant market position. (5) Tencent has abused its dominant position in the market and should bear legal liability according to law.

[Holding]

　  On October 8^th, 2014, the Supreme People's Court rendered the Civil Judgment (2013) MSZZ No.4, and rejected the appeal and upheld the original judgment.

[Reasoning]

Based on these grounds of appeal, the Supreme People's Court summarized 22 specific controversial issues in terms of five aspects, and analyzed each of the issues individually. In particular, with respect to the role, purpose and method of defining the relevant market, the Supreme People's Court made innovations and developments in the analysis of traditional Anti-Monopoly Law in the judgment, based on the Internet sector’s unique features and gave a creative answer on the method of defining the relevant market in the global field. For example, on the issue of whether defining relevant market is a unavoidable step in resolution of a monopoly dispute related to abuse of dominant market position, the industry widely recognizes that accurate definition of the relevant market is a prerequisite for determining dominant market position. This refers to the traditional analysis paradigm of "Relevant market - Market power - Competition effects (R-M-C)". The Supreme People's Court's judgment reconsidered the purpose and role of the relevant market based on the Internet’s characteristics, illustrated the tool of defining relevant market and proposed an analysis paradigm of "Market power - Competition effect (MC)" and "Conduct - Competition effect (CC)", both not based on determination of relevant market.

The Supreme People's Court held that: The focus of dispute involved in this case mainly include: first, how to define relevant market in this case; second, whether or not the appellee possesses dominant market position; and third, whether or not the act of the appellee constitutes abuse of dominant market position and other aspects that are prohibited by Anti-Monopoly Law.

　　I. How to define the relevant market in this case.

　　The focal point of the dispute can be further divided into numerous specific issues, which can be summarized as follows:

　　First of all, it is not necessary to define relevant market clearly, in a specific case involving abuse of dominant market position. Competition takes place and is carried out within a certain market scope. Defining relevant market can define the market scope and the competition constraints faced by business operators. In a case of abuse of dominant market position, a reasonable definition of relevant market is of great importance in correctly identifying the business operator’s market position, analyzing the influence of the business operator's behavior on market competition, judging whether the business operator's acts are illegal or not, and determining legal liabilities and other key issues. Therefore, in handling anti-monopoly cases, definition of relevant market is usually an important analytical step. Nevertheless, a clear definition of the relevant market depends on specific case circumstances, particularly on the evidence in the case, availability of relevant data and complexity of competition in the relevant field. In handling cases of abuse of dominant market position, definition of the relevant market is a tool for assessing the business operator’s dominant market position and influence of alleged monopolistic act on the competition, and is not the purpose in itself. Even if the relevant market is not clearly defined, possible market impact of the alleged business operator’s market position and alleged monopolistic act may be assessed through direct evidence of elimination or hindrance to competition. Therefore, it is not necessary to clearly define relevant market in every case of abuse of dominant market position. The court of first instance has actually defined relevant market in this case. As the boundary of relevant market in this case is ambiguous, the court of first instance merely analyzed the possibility of its boundary without arriving at any definite conclusion on the relevant market’s boundary. In view of this, Qihu’s grounds for appeal that the failure of the court of first instance to clearly define relevant commodity market in this case, is a failure to find basic facts of the case, cannot be established.

　　Second, on the issue of whether the method of Hypothetical Monopolist Test is applicable to free commodities, the judgment held that: (1) As an analytical way of defining the relevant market, it is assumed that the Hypothetical Monopolistic Test (HMT) is universally applicable. In practice, there are serveral methods of  hypothetical monopolist test that can be conducted, either by using the SSNIP method or the SSNDQ method. At the same time, as an analysis or thinking method, it is assumed that HMT can be conducted using both qualitative and quantitative analysis when conditions permit, in practice. (2) In practice, choosing the HMT method depends on the specific field of market competition involved in the case and relevant data available. If the homogeneity of commodities in a particular market field is pronounced and price competition is a relatively important form of competition, it is more feasible to adopt the SSNIP method. However, in a field where product differentiation is obvious and where non-price competition factors such as quality, service, innovation, consumer experience etc. become important competitive forms, it will be difficult to adopt the SSNIP method. In particular, using the SSNIP method is difficult when the market equilibrium price of commodities in a particular field is zero. When using the SSNIP method, it is usually necessary to determine the appropriate benchmark price, and make a price increase of between 5% and 10% to determine the consumer's reaction. Where the benchmark price is zero, price increased by 5% and 10% will still be zero. If the price is increased from zero to a smaller positive price, it will be equivalent to an indefinite price increase, which means that the characteristics of commodities or business models have undergone major changes, making it difficult to conduct the SSNIP test. (3) On applicability of HMT in this case, Internet service providers are placing greater emphasis on quality, service and innovation rather than price competition in Internet competition. In circumstances where free internet-based IM service has existed for a long time and has become a pervasive business model, users are highly sensitive to price. Therefore, changing toll-free tactics to charging even a small amount of fees can result in a massive effect on users. Likewise, changing from free-of-charge to charging fees also means a major change in commodity characteristics and business model, i.e. changing from free goods to paid goods, from indirect profitability model to direct profitability model. Under such circumstances, if HMT based on relative increase in price is adopted, it is likely to include commodities without substitutability into the relevant market, resulting in an overly broad definition of the relevant market. Therefore, HMT based on relative increase in price is not entirely suitable for this case. Although it is difficult for HMT based on relative increase in price to be fully applicable in this case, many alternatives to this method are still available, such as HMT based on decrease in quality. As it is difficult to assess quality degradation and obtain relevant data, HMT based on decrease in quality can be used for qualitative analysis rather than quantitative analysis.

　　Thirdly, whether relevant market in this case should be identified as an Internet application platform, the appellant believes that Internet application platform has nothing to do with definition of relevant market in this case. The appellee held that Internet competition is actually a platform competition and scope of relevant market in this case is far beyond the IM service market. In light of the competition feature of the Internet platform, the judgment expounds how to consider the feature and approach of platform competition in defining the relevant market, and held that: (1) Internet competition shows features of platform competition to some extent. When the sued monopolistic act occurred, the Internet platform competition feature becomes obvious. Internet operators enter the Internet field through a specific entry point to play an intermediary role for different types of consumers with different demands, to create value. (2) Judging whether the relevant commodity market in this case should be identified as an Internet application platform. The key issue lies in whether the competition between network platforms for the users and advertisers completely crosses the boundary determined by the characteristics of products or services, and imposes enough competitive constraints on the business operators. The answer to this question ultimately depends on empirical testing. In the absence of definitive empirical data, attention shall be paid to the following aspects at least: Firstly, competition between Internet application platforms for user attention and advertiser competition is based on key core products or services they provide. Secondly, key core products or Internet application platform services differ in terms of attributes, features, functions and usages. Although advertisers may not care about differences in these products or services but only care about prices and effectiveness of advertisements, and from their perspectives, different Internet application platforms may be considered to be alternatives to each other, it is very difficult for majority of free users to consider products or services of different platforms that have completely different functions and usages to be effective alternatives to each other. A user trying to find out the life of a historical figure typically uses a search engine instead of instant messaging, and hardly ever thinks the two can replace each other. Thirdly, differences in characteristics, functions and usages of key core products or services of Internet application platforms determine that there may be differences between major user groups and advertisers for whom they compete. Therefore, there are obvious differences in the mode of obtaining economic benefits, target user groups, and follow-up market products provided. Finally, the case should focus on whether the appellee has taken advantage of their potential dominant market position in the field of instant messaging to eliminate and hinder competition in Internet security software, and extended its dominant market position in the field of instant messaging to the field of security software, and if this competitive process occurs more often for free users. In view of the above grounds, the nature of competition in the Internet platform in the defining the relevant market in this case, is not considered a major factor. (3) The way of considering the competitive feature of the Internet enterprise platform, in this case. The purpose of defining the relevant market is to clarify competition constraints faced by business operators, reasonably determine their market position and correctly judge the impact of their actions on market competition. Even if the Internet platform’s competition feature is not considered at the stage of defining the relevant market, due consideration can still be given to such a feature in recognizing the market position and dominant market position of the business operator. Therefore, in this case, failure to mainly consider the Internet platform’s competition feature while defining the relevant market does not mean ignoring this feature, but rather considering this feature in a more appropriate way.

　　Finally, issues that need to be clarified when defining the relevant geographical market for IM service. The judgment held that definition of the relevant geographical market in this case should begin with the target area of the IM service market in Mainland China. As Internet-based IM services can be delivered at low cost, and can be available to or cover the entire world without additional or noteworthy shipping costs, price costs, or technical hurdles, the actual area where majority of users select the goods, provisions of laws and regulations, status quo of overseas competitors and timeliness of entering the relevant geographical market will be considered in defining the relevant geographical market. As each factor is not decisive, a comprehensive assessment based on the above factors is required. (1) Vast majority of users in Mainland China choose to use IM services provided by business operators in Mainland China. Users in Mainland China do not pay much attention to international IM products. (2) China's Internet-related administrative regulations and rules clearly lay out the requirements and conditions for operating IM services. China implements a system of administrative license for value-added telecommunications services such as instant messaging. Foreign business operators usually cannot directly enter Mainland China, and need to enter by establishing Sino-foreign joint ventures and obtain corresponding administrative licenses. (3) The actual situation of IM service operators located overseas. Prior to filing of this case of alleged monopoly, most international instant messaging operators such as MSN, Yahoo, Skype, Google, etc., entered the Mainland China market through joint ventures. Therefore, when the alleged monopolistic act took place, there were very few major international instant messaging service operators that had not yet entered Mainland China. If the quality of instant messaging service in Mainland China had decreased, there would not have been many overseas IM service operators available for domestic users to choose from. (4) It is quite difficult for overseas instant messaging service operators to enter Mainland China in a relatively short period of time (for example, one year) and to develop enough to restrict scale of domestic operators. Overseas instant messaging service operator need to first establish a joint venture, satisfy a series of licensing conditions and obtain appropriate administrative licenses, which to a certain extent, delays entry of the foreign business operator. In summary, the relevant geographical market in this case should be the Mainland China market.

　　Based on other evidence and actual situation of this case, the relevant market in this case should be defined as the IM market in Mainland China, including both PC-based and mobile-based IM services; both integrated IM services and non-integrated IM services such as text, audio and video.

　　II. Whether the appellee has dominant market position.

　　With respect to the position and role of business operator’s market share in the relevant market in determining its dominant market position, the judgment held that the position and role of market share in determining the dominant market position must be determined according to specific case circumstances. In general, higher the market share and longer its duration, the more likely it is to indicate the existence of dominant market position. However, market share is only a relatively crude and potentially misleading indicator of dominant market position. Under circumstances where the market is relatively easy to enter, or high market share stems from the business operator’s higher market efficiency or provision of better products, or products outside the market impose a strong competitive constraint on business operators, then high market share does not directly infer existence of dominant market position. In particular, competition in the Internet environment is highly dynamic. The boundary of the relevant market is far less clear than those in traditional fields. In this case, the role of market share as an indicator of dominant market position cannot be overestimated. Instead, more attention should be paid to market entry, business operator’s market behavior, impact of competition and other specific facts and evidences that can help determine dominant market position. 

　　Combining the above ideas, the judgment considered and analyzed whether the appellee possesses dominant position in the market based on aspects such as market share, competition condition in relevant market, capacity of the accused business operator controlling price, quantity or other trading conditions of goods or business operator's financial and technical conditions, degree of dependency of other business operators on the business operator with respect to transactions, and degree of difficulty with which other business operators enter the relevant market, especially the fact that when Tencent forced its users to choose between Tencent QQ and Qihoo 360 on Nov. 3, 2010, the monthly users of MSN, one of Tencent's competititors, increased 23 million in that month and many competitors entered, competing for Instant Messaging market. It was finally determined that existing evidence in this case was not sufficient to support the conclusion that the appellee possessed dominant market position.

　   III. Whether the appellee's act constitutes abuse of the dominant market position as prohibited by the Anti-Monopoly Law.

　　 The judgment broke from the traditional "three-step" approach of analyzing the abuse of dominant market position and adopted a more flexible analytical procedure and approach. It considered that: In principle, if the accused business operator does not have a dominant market position, it is not necessary to analyze whether it has abused its dominant market position, and it can be directly determined that its act does not constitute abuse of dominant market position as prohibited by the Anti-Monopoly Law. However, when the relevant market boundary is vague and it is not clear whether or not the accused business operator possessed the dominant market position, the effect of the alleged monopolistic act on competition may be further analyzed to test whether the conclusion of its dominant market position is correct or not. In addition, even if the accused business operator possessed the dominant market position, to judge whether the act constitutes abuse of dominant market position, it is necessary to comprehensively evaluate the possible negative and positive effect that such an act has on consumers and competition, and further judge the legitimacy of the act. This case mainly involves two aspects:

　　First, whether the act of "product incompatibility" (the user has to choose one of two products) conducted by the appellee constitutes a restrictive trade practice as prohibited by the Anti-Monopoly Law. According to provisions of Article 17 of the Anti-Monopoly Law, an act of a business operator with market dominance that requires counterparties to do business only with it or operators designated by it without justified ground, shall constitute abuse of the dominant market position. In this respect, the judgment held that although the act of "product incompatibility" conducted by the appellee caused inconvenience to the user, it did not result in the obvious effect of eliminating or restricting competition. It shows that the act of "product incompatibility" conducted by the appellee does not constitute abuse of market dominant position as prohibited by the Anti-Monopoly Law. Thus, it supported the conclusion that the appellee did not possess the dominant market position.

　　 Second, whether the act of the appellee constituted a bundling as prohibited by the Anti-Monopoly Law. According to the provisions of Article 17 of the Anti-Monopoly Law, the act of a business operator with market dominance conducting a bundling or attaching unreasonable trade conditions without justified grounds shall constitute abuse of dominant market position. In this respect, the judgment held that the appellant's appeal against the appellee's abuse of dominant market position was not well grounded.