Anti-competitive effects of abuse of collective dominance
Competition focus on abuse of collective dominance
Recent NDRC cases and relevant issues
Suggestions for application and implementation of collective abuse rules


The concept of an abuse of collective market dominance stems from Article 102 of the Treaty on the Functioning of the European Union (TFEU) and was first acknowledged by the EU General Court in 1992.(1) According to Article 102 of the TFEU, 'collective abuse' refers to when two or more undertakings which are connected in some way abuse their concentrated market dominance.

The National Development and Reform Commission (NDRC) recently undertook two investigations in which it applied the rules of collective abuse for the first time.

Anti-competitive effects of abuse of collective dominance

Undertakings can usually follow one of two anti-competitive paths to achieve, maintain and exercise their market power:

  • collusion – which could be described as 'marrying' a competitor and accounts for the anti-competitive effects of horizontal monopolistic conduct or cartels; and
  • exclusion – which could be described as 'killing' or 'maiming' a competitor and accounts for the anti-competitive effects of most vertical monopolistic conduct, especially the abuse of market dominance.(2)

In theory, the determination of collective dominance requires some kind of connection between or coordination among multiple undertakings. However, in terms of actual anti-competitive effects, collective abuse is founded on the anti-competitive path of exclusion, as is invariably the case where multiple undertakings conduct abusive behaviour to exclude their common competitors by means of foreclosure.

Consequently, it is rational to prohibit such types of behaviour through the application of antitrust laws. While the market shares of respective undertakings may not reach the statutory threshold for market dominance when considered individually, in practice, a strong link between such undertakings may result in an integration of market power that enables them to act effectively in the same manner as a single undertaking with market dominance.

Competition focus on abuse of collective dominance

Internal relationship between multiple undertakings
One of the significant preconditions for a finding of collective abuse is the existence of a strong link between separate undertakings. In order to establish the existence of such a collective entity on the market, factors that could have given rise to a connection between the undertakings concerned must be examined.(3) Such factors may flow from the nature and terms of an agreement between the undertakings or from the way in which it was implemented,(4) provided that the agreement has led the undertakings to present themselves or act as a collective entity. This may be the case if the undertakings have concluded cooperation agreements that led them to coordinate their conduct on the market or if ownership interests or other legal connections have led them to coordinate.(5)

However, the existence of such an agreement or legal connection is not indispensable to a finding of collective market dominance. Such a finding may also be based on:

  • other connecting factors;
  • an economic assessment; or
  • an assessment of the structure of the market in question.(6)

It follows that the structure of the market and the way in which the undertakings interact on it may also result in a finding of collective market dominance.(7)

External relationship beyond multiple undertakings
Based on the so-called 'rule of reason', in order to evaluate the extent of market foreclosure that such abusive behaviour may produce, several market economic factors – which would similarly be applied when considering a traditional abuse of dominance by a single undertaking – must be analysed.

As regards market structure, it will be easier to satisfy a finding of collective abuse if multiple undertakings compete with one another in an oligopolistic market whose structure still affords the undertakings sufficiently competitive conditions. For instance, to demonstrate collective abuse, each undertaking must be able to monitor whether the other undertakings are adhering to a common policy. Therefore, there must be sufficient market transparency – which is frequently the case in an oligopolistic market – for all of the undertakings concerned to learn, with sufficient precision and speed, of the other undertakings' market conduct.(8) In addition, market signals in the oligopolistic market must allow multiple undertakings to carry out abusive behaviour solely through relying on tacit cooperation or coordination, rather than by way of a real agreement.

Further, the difficulty level for new parties to enter the market should be considered, as it would when assessing the strength of potential rivals in the case of single market dominance. In doing so, the relationship between the collective entities and other potential competitors should be analysed, as an easy new entry could destroy the collective entities' so-called 'Nash equilibrium'(9) and render such abusive behaviour ineffective for the purpose of distorting market competition.

In addition, the duration of the abusive behaviour should be considered, as – to be effective – implementation of a common policy among undertakings must be sustainable over time. This presupposes the existence of sufficient deterrent mechanisms, which are sufficiently severe to convince all undertakings concerned that it is in their best interest to adhere to the common policy.(10) If the duration of the abusive behaviour is unsustainable over a relatively long period, the undertakings will have insufficient incentive to follow the common policy, as breaking rank may lead to greater profitability.

Finally, certain other factors may be critical for evaluating the competitive effects of abusive behaviour on a case-by-case basis, such as:

  • the ability of the undertakings to control the trade terms in the market;
  • the financial and technical status of the undertakings; and
  • the extent of undertakings' reliance on other undertakings for transactions, by which the foreclosure brought by abusive behaviour would lead other competitors in the relevant market to lose essential trading opportunities and subsequently be excluded from the market.

Recent NDRC cases and relevant issues

The NDRC recently initiated two investigations on this matter and on July 28 2017 it issued Decisions 1 and 2 2017 on the Administrative Penalty, in which it penalised two Chinese pharmaceutical undertakings for alleged collective abuse in the isoniazid active pharmaceutical ingredient market. This was the first time that the Chinese enforcement authorities applied the rules of the Anti-monopoly Law to a case concerning collective abuse.

As stipulated in Article 19 of the law:

"Undertakings are assumed to have a dominant market position if any one of the following conditions is fulfilled: (i) The market share of one undertaking accounts for 1/2 in the relevant market;(ii) The joint market share of two undertakings amounts for 2/3 in the relevant market; or (iii) The joint market share of three undertakings amounts for 3/4 in the relevant market. In case that the circumstances of the undertakings fall under the conditions (ii) or (iii) and any of the undertakings has a market share of less than 10%, that undertaking shall not be assumed to have a dominant market position. Undertakings that are assumed to have a dominant market position shall not be considered to have a dominant market position provided that there is opposite evidence."

Accordingly, the NDRC applied this article in its assessment of the behaviour of the pharmaceutical undertakings in question. However, although it used the rule of reason to evaluate the competitive effects of the abusive behaviour in accordance with an assessment of the economic factors listed in Article 18 of the law, an analysis of whether a connective link existed, and the relationship between each undertaking in question, were missing in its decision – both of which are considered significant conditions for the application of this rule (as discussed above).

Suggestions for application and implementation of collective abuse rules

Establishing and improving corresponding guidelines
As mentioned above, from a legislative perspective, the Anti-monopoly Law provisions covering collective abuse are rationally based on competition theory. However, a practical standard and other means for applying this concept must still be established. Based on the Constitution in Law and the Economics for Antitrust Law, detailed applicable standards and means for a competitive evaluation in enforcement regulations or antitrust guidelines should arguably be issued by governments rather than promulgated in antitrust law. More specifically, from a horizontal antitrust agreement perspective, lessons may also be drawn from the relevant approaches to assessment and the standard of proof applied for such conduct, as the relationship between multiple undertakings in a collective abuse case is, to some extent, similar to that between multiple parties in the case of concerted conduct (ie, cartels).(11)

Competitive legal risks for enterprises
Although the above NDRC cases were the first – and, to date, the only – instance of a Chinese enforcement authority applying the rules of collective abuse since the Anti-monopoly Law was implemented in 2008, enterprises in China should still be conscious of the need to avoid relevant legal risks.

Due to the Anti-monopoly Law, the cases investigated by the NDRC and the potential for the reapplication of these rules in future, full consideration and monitoring of the relevant legal risks should be a core part of enterprises' compliance functions. More specifically, enterprises should act as soon as possible to avoid the corresponding behaviours listed in Article 17 of the law (eg, refusal to deal and exclusive dealing) if:

  • they are involved in an oligopolistic market;
  • their joint market share exceeds the proportions stipulated in Article 19 of law for the relevant market; and
  • they each have a market share of more than 10%.

In practice, if an enterprise is involved in an oligopolistic market where essential trading information can easily be transferred through a relatively transparent market mechanism, it is important that it monitors its market share in relation to that of its competitors, even if there are no cooperative agreements between the enterprise and its competitors. In addition, such enterprises should:

  • remain alert;
  • establish appropriate firewalls to isolate sensitive information from the competition; and
  • refrain from sharing the same distribution network with their competitors.

For further information on this topic please contact Hao Zhan, Ying Song or Chen Tian at AnJie Law Firm by telephone (+86 10 8567 5988) or email ([email protected], [email protected] or [email protected]). The AnJie Law Firm website can be accessed at‚Äč


(1) The EU General Court's judgment in Italian Flat Glass (joined Cases T-68, 77 and 78/89).

(2) See Thomas G Krattenmaker and Steven C Salop, "Anti-competitive Exclusion: Raising Rivals' Costs to Achieve Power over Price", Vol 96, The Yale Law Journal, 209, 238-242 (1986).

(3) Joined Cases C-68/94 and C-30/95, paragraph 221.

(4) Case C-393/92 (Almelo), paragraphs 41 to 43.

(5) See the Directorate-General for Competition's discussion paper on the application of Article 82 of the TFEU to exclusionary abuses (2005), paragraph 45.

(6) Supra, note 3.

(7) See the European Commission's Guidelines on the Assessment of Horizontal Mergers, issued under the EU Merger Regulation (139/2004/EC), paragraphs 39 to 57.

(8) See Case T-342/99, Airtours plc v Commission [2002] ECR II-2585, paragraph 62; and Case T- 193/02, Piau, paragraph 111.

(9) In game theory, the 'Nash equilibrium' is a concept involving a non-cooperative game involving two or more players, under which each player is assumed to know the equilibrium strategies of the other player(s) and no player has anything to gain by changing only his or her own strategy. See Martin J Osborne and Ariel Rubinstein, A Course in Game Theory, Cambridge, MA: MIT, 1994.

(10) Supra, note 11.

(11) See Ning Du, Demonstration of the Standard of Proof for Concerted Conduct, Price: Theory & Practice, Vol 1, 2016.