SPC Guiding Case No. 78: Beijing Qihoo Technology Co., Ltd. v. Tencent Technology (Shenzhen) Co., Ltd. and Shenzhen Tencent Computer Systems Co., Ltd. (Abuse of Market Dominance Dispute)

SPC Guiding Case No. 78

Discussed and adopted by the Judicial Committee of the Supreme People’s Court; issued on March 6, 2017

Keywords

Civil; abuse of market dominance; monopoly; relevant market

Holding

1. In the trial of anti-monopoly cases, defining the relevant market is usually an important analytical step. However, whether the relevant market can be clearly defined depends on the specific circumstances of the case. In cases of abuse of market dominance, defining the relevant market is a tool for assessing the market power of an undertaking and the impact of the alleged monopolistic conduct on competition; it is not itself an end. Where the market position of an undertaking and the market impact of the alleged monopolistic conduct can be assessed through direct evidence of the exclusion or hindrance of competition, it is unnecessary to clearly and explicitly define the relevant market in every case of abuse of market dominance.

2. The hypothetical monopolist test (HMT) is a widely applicable analytical approach for defining the relevant market. In practice, the hypothetical monopolist test may be conducted through methods such as a price increase (SSNIP) or a quality decrease (SSNDQ). Because the free-of-charge nature of internet instant messaging services makes users highly price-sensitive, employing the price-increase test would result in the relevant market being defined too broadly; instead, a qualitative analysis should be conducted using a hypothetical monopolist test based on quality decrease.

3. Given the low-cost and wide-coverage characteristics of internet instant messaging services, when defining the relevant geographic market, a comprehensive assessment should be made based on factors such as the actual region in which the majority of demanders choose goods, the provisions of laws and regulations, the current situation of foreign competitors, and the timeliness of their entry into the relevant geographic market.

4. In the internet field, market share is only a relatively rough and potentially misleading indicator for determining market dominance, and its position and role in determining market power must be determined according to the specific circumstances of the case.

Basic Facts

Beijing Qihoo Technology Co., Ltd. (hereinafter “Qihoo Company”) and Qizhi Software (Beijing) Co., Ltd. released the “Koukou Guard” software on October 29, 2010. On November 3, 2010, Tencent Technology (Shenzhen) Co., Ltd. (hereinafter “Tencent Company”) issued “A Letter to the Vast Number of QQ Users,” stopping the operation of the QQ software on computers on which 360 software was installed. On November 4, Qihoo Company announced the recall of the “Koukou Guard” software. On the same day, the 360 Security Center also announced that, under the strong intervention of the relevant state departments, QQ and 360 software had by then achieved full compatibility. In September 2010, Tencent’s QQ instant messaging software was packaged and installed together with QQ Software Manager, and the installation process did not prompt users that QQ Software Manager would be installed at the same time. On September 21, 2010, Tencent Company issued an announcement stating that the QQ Software Manager and QQ Doctor then in use would be automatically upgraded to QQ Computer Manager. Qihoo Company sued in the Guangdong Higher People’s Court, alleging that Tencent Company abused its market dominance in the relevant market for instant messaging software and services. Qihoo Company claimed that Tencent Company and Shenzhen Tencent Computer Systems Co., Ltd. (hereinafter “Tencent Computer Company”) held market dominance in the relevant market for instant messaging software and services, that the two companies expressly prohibited their users from using Qihoo Company’s 360 software and otherwise stopped the QQ software service, refused to provide relevant software services to users who had 360 software installed and forced users to delete the 360 software, and adopted technical means to prevent users who had installed the 360 browser from accessing Qzone, and that these acts constituted restrictive trading; and that Tencent Company and Tencent Computer Company bundled QQ Software Manager with the instant messaging software and installed QQ Doctor in the name of upgrading QQ Software Manager, which constituted tying. It requested an order that Tencent Company and Tencent Computer Company immediately cease the monopolistic conduct of abusing market dominance and jointly compensate Qihoo Company for economic losses of RMB 150 million.

Judgment

On March 20, 2013, the Guangdong Higher People’s Court rendered Civil Judgment No. (2011) Yue Gao Fa Min San Chu Zi No. 2, rejecting the claims of Beijing Qihoo Technology Co., Ltd. Beijing Qihoo Technology Co., Ltd., dissatisfied, appealed. On October 8, 2014, the Supreme People’s Court rendered Civil Judgment No. (2013) Min San Zhong Zi No. 4, rejecting the appeal and affirming the original judgment.

Reasoning

The court in its effective judgment held that the focal issues of this case mainly included: first, how to define the relevant market in this case; second, whether the appellees held market dominance; and third, whether the appellees constituted abuse of market dominance prohibited by the Anti-Monopoly Law.

I. How to define the relevant market in this case

This focal issue may be further refined into several specific questions, summarized as follows.

First, the relevant market need not be clearly and explicitly defined in every case of abuse of market dominance. Competitive conduct arises and unfolds within a certain market scope, and defining the relevant market can clarify the market scope of competition among undertakings and the competitive constraints they face. In cases of abuse of market dominance, reasonably defining the relevant market is of great significance to such key issues as correctly determining the market position of an undertaking, analyzing the impact of the undertaking’s conduct on market competition, judging whether the undertaking’s conduct is unlawful, and the legal liability to be borne in the event of unlawfulness. Therefore, defining the relevant market is usually an important analytical step in the trial of anti-monopoly cases. Nevertheless, whether the relevant market can be clearly defined depends on the specific circumstances of the case, especially the evidence in the case, the availability of relevant data, and the complexity of competition in the relevant field. In the trial of cases of abuse of market dominance, defining the relevant market is a tool for assessing the market power of an undertaking and the impact of the alleged monopolistic conduct on competition; it is not itself an end. Even without clearly defining the relevant market, the market position of the alleged undertaking and the possible market impact of the alleged monopolistic conduct can be assessed through direct evidence of the exclusion or hindrance of competition. Therefore, the relevant market need not be clearly and explicitly defined in every case of abuse of market dominance. In fact, the court of first instance had already defined the relevant market in this case, except that, because the boundaries of the relevant market were ambiguous, it merely analyzed the possibilities of the boundaries without giving a clear conclusion. In view of this, Qihoo Company’s ground that the court of first instance had failed to clearly define the relevant product market, which constituted unclear determination of the basic facts of this case, could not be established.

Second, on the issue of whether the “hypothetical monopolist test” method can be applied to the field of free goods. The court in its effective judgment held: (1) As an analytical approach for defining the relevant market, the hypothetical monopolist test (HMT) has universal applicability. In practice, there are multiple analytical methods for the hypothetical monopolist test, which may be conducted either through a small but significant and non-transitory increase in price (SSNIP) or through a small but significant and non-transitory decrease in quality (SSNDQ). At the same time, as an analytical approach or method of thinking, the hypothetical monopolist test may be conducted through qualitative analysis in practice, or through quantitative analysis where conditions permit. (2) In practice, which method to choose for the hypothetical monopolist test depends on the market competition field involved in the case and the specific circumstances of the relevant data available. If the homogeneity of goods in a particular market field is relatively obvious and price competition is a relatively important form of competition, it is more feasible to employ the SSNIP method. However, in fields where product differentiation is very obvious and non-price competition such as quality, service, innovation, and consumer experience has become an important form of competition, it is more difficult to employ the SSNIP method. In particular, when the market equilibrium price of goods in a particular field is zero, employing the SSNIP method is especially difficult. When employing the SSNIP method, it is usually necessary to determine an appropriate benchmark price, implement a price increase of 5 to 10 percent, and then determine the reaction of demanders. Where the benchmark price is zero, if a price increase of 5 to 10 percent is implemented, the price after the increase is still zero; if the price is raised from zero to a small positive price, this is equivalent to an infinite increase in the price increase range, implying a major change in the characteristics of the goods or the business model, making the SSNIP test difficult to conduct. (3) On the applicability of the hypothetical monopolist test in this case. Internet service providers attach greater importance to competition in quality, service, innovation, and other aspects than to price competition in internet competition. Where free internet basic instant messaging services have existed for a long time and have become a prevailing business model, users have extremely high price sensitivity, and changing the free strategy to charging even a relatively small amount may lead to the loss of a large number of users. At the same time, changing the price from free to charged also implies a major change in the characteristics of the goods and the business model, namely from free goods to charged goods, and from an indirect profit model to a direct profit model. Under such circumstances, if a hypothetical monopolist test based on relative price increase were adopted, goods without a substitution relationship would likely be brought into the relevant market, resulting in the relevant market being defined too broadly. Therefore, a hypothetical monopolist test based on relative price increase is not entirely suitable for application in this case. Although a hypothetical monopolist test based on relative price increase is difficult to apply fully in this case, a modified form of the method may still be adopted, such as a hypothetical monopolist test based on quality decrease. Because the degree of quality decrease is difficult to assess and the relevant data are difficult to obtain, a qualitative analysis rather than a quantitative analysis may be conducted using the hypothetical monopolist test based on quality decrease.

Third, on the issue of whether the relevant market in this case should be determined as internet application platforms. The appellant argued that internet application platforms were irrelevant to the definition of the relevant market in this case; the appellees argued that internet competition is essentially platform competition and that the scope of the relevant market in this case far exceeded the instant messaging service market. With respect to the characteristics of platform competition in the internet field, the court in its effective judgment explained how the characteristics of platform competition should be considered and handled when defining the relevant market, holding: (1) Internet competition to a certain extent exhibits the characteristics of platform competition. When the alleged monopolistic conduct occurred, the platform competition characteristics of the internet were already relatively obvious. Internet undertakings enter the internet field through specific entry points and play an intermediary role among consumers of different types and demands, thereby creating value. (2) The key issue in determining whether the relevant product market in this case should be determined as internet application platforms is whether the mutual competition among online platforms for user attention and advertisers completely crosses the boundaries determined by the characteristics of products or services and imposes sufficiently strong competitive constraints on undertakings. The answer to this question ultimately depends on empirical testing. In the absence of precise empirical data, at least the following aspects should be noted: first, the competition among internet application platforms for user attention and advertisers is based on the key core products or services they provide; second, the key core products or services of internet application platforms differ considerably in attributes, characteristics, functions, and uses. Although advertisers may not care about the differences between these products or services and only care about the price and effect of advertising, and may therefore regard different internet application platforms as substitutable for one another, it is difficult for the vast number of users on the free end to regard products or services with completely different functions and uses provided by different platforms as effectively substitutable for one another. A user trying to look up the life of a historical figure would usually choose a search engine rather than instant messaging, and would hardly consider the two to be mutually substitutable. Third, the differences in the characteristics, functions, and uses of the key core products or services of internet application platforms determine that the main user groups and advertisers they compete for may differ, and there are thus considerable differences in the model for obtaining economic benefits, the target user groups, and the subsequent market products provided. Finally, what should be focused on in this case is whether the appellees used their possible market dominance in the instant messaging field to exclude or restrict competition in the internet security software field and extended their possible market dominance in the instant messaging field to the security software field; this competitive process occurred more on the free user end. In view of the above, the platform competition characteristics of the internet were not a primary consideration at the stage of defining the relevant market in this case. (3) The manner of considering the platform competition characteristics of internet enterprises in this case. The purpose of defining the relevant market is to clarify the competitive constraints faced by undertakings, reasonably determine the market position of undertakings, and correctly judge the impact of their conduct on market competition. Even if the platform competition characteristics of the internet are not primarily considered at the stage of defining the relevant market, in order to correctly determine the market position of an undertaking, they may still be appropriately considered when identifying the market position and market control power of the undertaking. Therefore, for this case, not primarily considering the platform competition characteristics of the internet at the stage of defining the relevant market does not mean ignoring those characteristics, but rather considering them in a more appropriate manner.

Finally, on the issues to be noted in defining the relevant geographic market for instant messaging services. The court in its effective judgment held: the definition of the relevant geographic market in this case should start from the target region of the instant messaging service market in mainland China. Because internet-based instant messaging services can reach or cover the globe at low cost and with little expense, without additional noteworthy transportation costs, price costs, or technical barriers, when defining the relevant geographic market, primary consideration will be given to factors such as the actual region in which the majority of demanders choose goods, the provisions of laws and regulations, the current situation of foreign competitors, and the timeliness of their entry into the relevant geographic market. Since none of these factors is decisive, a comprehensive assessment must be made based on the above factors. First, the vast majority of users within mainland China choose instant messaging services provided by undertakings within mainland China. Users within mainland China do not pay high attention to international instant messaging products. Second, China’s administrative regulations and rules concerning the internet provide clear requirements and conditions for operating instant messaging services. China implements an administrative licensing system for value-added telecommunications services such as instant messaging, and foreign undertakings generally cannot directly operate within mainland China, but must enter by means of a Sino-foreign joint venture and obtain the corresponding administrative license. Third, the actual situation of instant messaging service undertakings located outside China. Before the alleged monopolistic conduct in this case occurred, most major international instant messaging undertakings, such as MSN, Yahoo, Skype, and Google, had already entered the mainland China market through joint ventures. Therefore, when the alleged monopolistic conduct occurred, there were already very few major international instant messaging service undertakings that had not yet entered mainland China. If the quality of instant messaging services within mainland China were to decrease slightly, there would be few foreign instant messaging service undertakings for domestic users to choose. Finally, it is difficult for foreign instant messaging service undertakings to enter mainland China in a timely manner within a relatively short period (such as one year) and develop to a scale sufficient to constrain domestic undertakings. Foreign instant messaging service undertakings first need to establish enterprises through joint ventures, satisfy a series of licensing conditions, and obtain the corresponding administrative licenses, which considerably delays their entry. In summary, the relevant geographic market in this case should be the mainland China market.

Based on the other evidence and actual circumstances of this case, the relevant market in this case should be defined as the instant messaging service market in mainland China, including both instant messaging services on the personal computer end and instant messaging services on the mobile end, and including both comprehensive instant messaging services and non-comprehensive instant messaging services such as text, audio, and video.

II. Whether the appellees held market dominance

With respect to the position and role of an undertaking’s market share in the relevant market in determining its market power, the court in its effective judgment held: the position and role of market share in determining market power must be determined according to the specific circumstances of the case. Generally, the higher the market share and the longer it lasts, the more likely it is to indicate the existence of market dominance. Nevertheless, market share is only a relatively rough and potentially misleading indicator for determining market dominance. Where market entry is relatively easy, or a high market share results from the undertaking’s higher market efficiency or provision of superior products, or goods outside the market impose strong competitive constraints on the undertaking, a high market share cannot directly infer the existence of market dominance. In particular, competition in the internet environment is highly dynamic, and the boundaries of the relevant market are far less clear than in traditional fields. Under such circumstances, the indicative role of market share should not be overestimated; instead, more attention should be paid to specific facts and evidence helpful to determining market dominance, such as market entry, the undertaking’s market conduct, and the impact on competition.

Combining the above approach, the court in its effective judgment considered and analyzed whether the appellees held market dominance from such aspects as market share, the competitive situation of the relevant market, the alleged undertaking’s ability to control the prices, quantities, or other transaction conditions of goods, the undertaking’s financial resources and technical conditions, the degree of dependence of other undertakings on the undertaking in transactions, and the difficulty for other undertakings to enter the relevant market. It ultimately determined that the existing evidence in this case was insufficient to support the conclusion that the appellees held market dominance.

III. Whether the appellees constituted abuse of market dominance prohibited by the Anti-Monopoly Law

The court in its effective judgment broke away from the traditional “three-step method” of analyzing abuse of market dominance and adopted more flexible analytical steps and methods, holding: in principle, if the alleged undertaking does not hold market dominance, there is no need to analyze whether it abused market dominance, and it may be directly determined that it did not constitute abuse of market dominance prohibited by the Anti-Monopoly Law. However, where the boundaries of the relevant market are relatively ambiguous and it is unclear whether the alleged undertaking holds market dominance, the impact of the alleged monopolistic conduct on competition may be further analyzed to verify whether the conclusion on market dominance is correct. In addition, even if the alleged undertaking holds market dominance, judging whether it constitutes abuse of market dominance also requires a comprehensive assessment of the negative effects and possible positive effects of the conduct on consumers and competition, and then a judgment on the lawfulness of the conduct. This case mainly involved two issues:

First, whether the “product incompatibility” conduct implemented by the appellees (the “choose one of two” for users) constituted restrictive trading prohibited by the Anti-Monopoly Law. Under Article 17 of the Anti-Monopoly Law, an undertaking with market dominance that, without legitimate reasons, restricts its trading counterparties to trading only with itself or only with undertakings designated by it, constitutes abuse of market dominance. The appellant claimed that the appellees, without legitimate reasons, forced users to stop using and uninstall the appellant’s software, constituting restrictive trading that is abuse of market dominance prohibited by the Anti-Monopoly Law. In this regard, the court in its effective judgment held that although the “product incompatibility” conduct implemented by the appellees caused inconvenience to users, it did not produce an obvious effect of excluding or restricting competition. This, on the one hand, showed that the appellees’ “product incompatibility” conduct did not constitute abuse of market dominance prohibited by the Anti-Monopoly Law, and, on the other hand, corroborated the conclusion that the appellees did not hold market dominance.

Second, whether the appellees constituted tying prohibited by the Anti-Monopoly Law. Under Article 17 of the Anti-Monopoly Law, an undertaking with market dominance that, without legitimate reasons, ties goods or attaches other unreasonable transaction conditions to a transaction, constitutes abuse of market dominance. The appellant claimed that the appellees bundled QQ Software Manager with the instant messaging software and installed QQ Doctor in the name of upgrading QQ Software Manager, which did not conform to transaction practice, consumer habits, or the functions of the goods, restricted consumers’ right of choice, and lacked legitimate reasons; and that the allocation of the burden of proof in the first-instance judgment regarding the exclusionary or restrictive effect on competition of the alleged tying conduct was erroneous. In this regard, the court in its effective judgment held that the appellant’s appeal ground that the appellees had committed abuse of market dominance could not be established.

(Judges of the effective judgment: Wang Chuang, Wang Yanfang, Zhu Li)

Relevant Legal Provisions

Articles 17, 18, and 19 of the Anti-Monopoly Law of the People’s Republic of China

Free PDF download of the complete article.