Abuse of Market Dominance Under China’s Anti-Monopoly Law: The Case of Tetra Pak

The announcement (in Chinese), issued by the Anti-Monopoly and Anti-Unfair Competition Enforcement Bureau of the SAIC, is available at http://​www.​saic.​gov.​cn/​xw/​zyxw/​201704/​t20170424_​261958.​html.

The Chinese version of the SAIC’s full decision is available at http://​www.​saic.​gov.​cn/​fldyfbzdjz/​jzzfgg/​201703/​P020170309853689​741679.​pdf.

The National Development and Reform Commission (“NDRC”) has fined U.S. chipmaker Qualcomm $975 million USD in 2015. This fine was the highest announced to date by Chinese competition authorities.

For an introduction of the decision and a preliminary discussion of its implications, see Zhang et al. (2016). In addition, Han et al. (2017) focuses on the SAIC’s analysis of Tetra Pak’s loyalty rebate schemes and compares it to the approach taken by the European Commission in the Intel case.

For an assessment of the enforcement of the AML’s abuse of dominance provisions, see Liu and Qiao (2012).

Specifically, Article 18 identifies the factors to be evaluated when determining the market dominance position, such as market share, the capacity to control the market, barriers to entry, and so on. Article 19 presents a quantitative approach to the identification of market dominance; for example, paragraph 1 states that a company can be presumed to have a dominant market position if its market share accounts for 50% or more in the relevant market. But this is not to say that a firm would immediately be absolved if its market share falls below 50%. The 50% threshold merely creates a presumption of dominance, and nowhere in the AML is it stated that firms with less than 50% market share cannot be dominant. .

For a discussion of private antitrust litigation under the AML in China, see Lu and Tan (2013).

In China, it is the responsibility of the Anti-Monopoly Commission (AMC) under the State Council to promulgate antitrust guidelines. In practice, the AMC usually assigns the drafting work to certain competition agencies and will review and finalize the draft. For Guidelines on the Definition of Relevant Market, effective as of May 24, 2009, see http://​www.​gov.​cn/​zwhd/​2009-07/​07/​content_​1355288.​htm. For Regulations on the Prohibition of Abuse of Dominant Market Position, effective as of February 1, 2011, see http://​www.​saic.​gov.​cn/​fldyfbzdjz/​zcfg/​xzgz/​201101/​t20110107_​233538.​html. Unofficial translation of the two aforementioned guidelines is available at http://​www.​lawinfochina.​com/​display.​aspx?​lib=​law&​id=​7575 and http://​www.​lawinfochina.​com/​display.​aspx?​lib=​law&​id=​8540, respectively.

Hongchan Yang, Director General of the Anti-Monopoly and Anti-Unfair Competition Enforcement Bureau of the SAIC, Speech at the 6th Annual China Competition Policy Forum (August 30, 2017), http://​www.​legaldaily.​com.​cn/​Finance_​and_​Economics/​content/​2017-09/​01/​content_​7303776.​htm?​node=​76109.

The SAIC’s antitrust enforcement decisions are regularly published at http://​www.​saic.​gov.​cn/​fldyfbzdjz/​jzzfgg/​. As of August, 2017, the final decisions of 52 cases have been released, including 25 cases of monopoly agreement and 27 cases of abuse of dominance.

Greatview, Global Offering (November 26, 2010), http://​www.​greatviewpack.​com/​UploadFile/​File/​Global%20​Offering.​pdf. Statistics and forecasts in the report are based on a filled volume conversion methodology, in order to account for packs of different sizes. As used in this article, aseptic packaging or aseptic cartons refer to multi-layered aseptic cartons that are made from paperboard, polyethylene, and aluminum. “Aseptic” means free from bacterial contamination. The downstream application of aseptic packaging ranges from milk to juice, yogurt, ready-to-drink tea, coffee, soups, creams, and wine. Since the SAIC identified Tetra Pak’s market dominance and abusive practices from 2009 through 2013, we will pay particular attention to industry statistics and company facts during the relevant time period.

In China, aseptic packaging was mainly applied in the dairy sector. According to an industry report that was submitted by Tetra Pak, as of 2012, around 80% of white milk and over half of other liquid dairy products in China were packaged in aseptic cartons. Therefore, the rapid growth of the dairy sector had greatly contributed to increasing the demand for aseptic cartons.

USDA Foreign Agricultural Service, 2011. See https://​www.​fas.​usda.​gov/​data for more detail.

“The dairy consumption per capita was 32.4 kg, which was less than 1/3 of the global average.” Hongbin Gao, Vice-Minister of the Ministry of Agriculture, speech at the 3rd Dairy Conference of China (June 19, 2012), http://​roll.​sohu.​com/​20120708/​n347568318.​shtml.

For more recent facts about Tetra Pak, see http://​www.​tetrapak.​com/​about/​facts-figures.

See Tao (2005, p. 3). Moreover, when Tetra Pak entered the Chinese market, it was requested by the Chinese government to play a leading role in the improvement of the country’s food processing and packaging sector.

See Tao and Wei (2012, p.1).

For more detail, a company report of Greatview (July 30, 2014) presented by the Hong Kong and Shanghai Banking Corporation (“HSBC”), focusing on Greatview’s performance and growth potential, is available at http://​doc.​xueqiu.​com/​147850fce82703fe​70e1bfd5.​pdf.

Here, paper-based aseptic cartons refer to multi-layered aseptic cartons that are mainly made from paperboard. Specifically, the SAIC summarized that paper-based aseptic packaging is better suited for liquid dairy food than other packaging forms and for the following reasons: First, aseptic packaging materials have much better antibacterial qualities than non-aseptic packaging materials, and, as a result, dairy food packaged in aseptic cartons can be stored at room temperature; due to the lack of cold-chain logistics in many areas of China, aseptic packaging could not be substituted by any non-aseptic packaging forms. Second, paper-based aseptic cartons are lighter, thus making transportation easier, and have lower costs, as compared to non-paper-based aseptic cartons.

Regarding the materials market, its upstream industry is mainly the supply of raw paper. Raw paper for packaging materials includes kraft-base paper (“brown paper”) and white-base paper (“white paper”). Brown paper has advantages over white paper in terms of costs and quality, so it is typically preferred for producing high-quality paperboard. As of 2013, only two companies could supply brown paper in China: Foshan Huaxin Packaging, and its subsidiary Zhuhai Hongta Renheng Paper. Following the pattern in the decision, we use “Hongta” to denote both companies.

“Contestable” volume is that for which the buyer is willing and able to substitute an alternative product, while “non-contestable” volume is that for which the buyer is unable or unwilling to accept a substitute. Note that the assessment of contestable and non-contestable portion of demand has been made in many loyalty discount cases.

For a general discussion of alternative viewpoints surrounding analytical framework of loyalty discounts, see EC (2005), Spector (2005), Gates (2013), Moore and Wright (2015), Salinger (2017), Salop (2006, 2017), and Steuer (2017).

http://​ec.​europa.​eu/​competition/​sectors/​ICT/​intel_​provisional_​decision.​pdf;

FTC Docket No. 9341 (Aug. 4, 2010),

www.​ftc.​gov/​sites/​default/​files/​documents/​cases/​2010/​08/​100804inteldo_​0.​pdf;

Settlement Agreement between AMD and Intel, 05-441 (D. Del. Nov. 11, 2009),

www.​sec.​gov/​Archives/​edgar/​data/​50863/​0000050863090002​13/​exh101.​htm;

(2013).

Case COMP/E-1/38.113—Tomra, European Comm’n Decision, 2006; Case T-155/06, Tomra Systems and Others v. European Comm’n (GC Sep 9, 2010); Case C-549/10 P, Tomra Systems and Others v. European Comm’n (ECJ April 19, 2012).

American Bar Association Section of Antitrust Law, Presidential Transition Report: The State of Antitrust Enforcement, Antitrust Source 36 (Jan. 2017), www.​americanbar.​org/​groups/​antitrust_​law/​publications/​the_​antitrust_​source_​201701.​html.

See, for example, Greenlee et al. (2008), Klein and Lerner (2016), Murphy et al. (2014), Fumagalli and Motta (2017) and Durkin (2017). Through predatory pricing, a firm lowers its price below some measure of cost, drives equally efficient rivals out of the market, and then raises its price to recoup the profit lost while it was pricing below cost.

See, for example, Salop (2017). The article summarizes those two paradigms and suggests that conditional pricing practices are generally better characterized as belonging to the RRC foreclosure paradigm than to the predatory one. See also Joshua D. Wright, Commissioner of the FTC, “Simple but Wrong or Complex but More Accurate? The Case for an Exclusive Dealing-Based Approach to Evaluating Loyalty Discounts”, remarks at the Bates White 10th Annual Antitrust Conference (June 3, 2013), www.​ftc.​gov/​sites/​default/​files/​documents/​public_​statements/​simple-wrong-or-complex-more-accurate-case-exclusive-dealing-based-approach-evaluating-loyalty/​130603bateswhite​.​pdf-

To support his view, Salop (2017) claims that the price–cost test is fundamentally flawed. “Passing the test does not rule out anticompetitive exclusion and failing the test does not prove anticompetitive exclusion.” In addition, Salop believes that the test fails to provide reliable evidence of consumer harm, and it is premised on the erroneous idea that only equally efficient competitors are worth protecting.

The majority of judges on the Meritor panel explained that “contracts in which discounts are linked to purchase (volume or market share) targets are frequently challenged as de facto exclusive dealing arrangements on the grounds that the discounts induce customers to deal exclusively with the firm offering the rebate.” Moreover, it is important to mention that the judges also stressed that “a plaintiff’s characterization of its claim as an exclusive dealing claim does not take the price–cost test off the table” and “when price is the clearly predominant mechanism of exclusion, the price–cost test tells us that, so long as the price is above-cost, the procompetitive justifications for, and the benefits of, lowering prices far outweigh any potential anticompetitive effects.” Therefore, the judges declined to employ a price–cost test because it was found that price was not the primary method of exclusion. For more detail, see supra note 32.

In the initial consideration of the Intel case, the European Commission staff in fact performed a price–cost test as part of the rule-of-reason analysis of Intel’s loyalty rebates. On May 13, 2009, the Commission imposed a fine of €1.06 billion on Intel. Intel then brought an action against the Commission’s decision before the EU General Court. By judgment of June 12, 2014, the General Court ruled that an as-efficient-competitor test is not necessary for a finding of abuse of dominance, arguing that the use of market-share discounts by a dominant firm is per se illegal. Intel brought an appeal against the judgment of the General Court before the Court of Justice. By judgment of September 9, 2017, the European Court of Justice reversed the General Court’s decision, endorsing a rule of reason approach, and ordering the General Court to carry out an assessment of “the actual or potential effect of Intel’s conduct on competition.” For more detail, see supra note 30.

As an illustration, the SAIC’s decision included a chart that showed how a customer’s total payment changed as the purchase volume of a particular type of packaging materials increased in 2013. The chart shows that after reaching a certain volume threshold, the total payment dropped—which implies a negative price at the margin.

Note that our example is for illustrative purposes only. In practice, Tetra Pak’s retroactive rebates program contained dozens of volume grades and served hundreds of customers, which was much more complex than the hypothetical AUDs scheme that we consider here.

Chao and Tan (2017) consider a very similar setting in which the downstream customer demands at most 10 units but exhibits stepwise willingness-to-pay (i.e., the customer is willing to pay at most 10 for the first unit, 9 for the second unit, and so on). In that case, the customer receives a lower surplus with AUDs.

Although in our simple example the rival is treated as a follower in the price-setting process and only linear pricing is feasible for the rival, the insights above still hold in an extended game in which the pricing options of the two firms are endogenized. More specifically, Chao, Tan, and Wong (2018) also consider a situation in which an unconstrained firm and a capacity-constrained firm are treated symmetrically for making pricing decisions: Both are free to choose whether to commit themselves to liner pricing or AUDs schemes. It turns out that only the unconstrained firm has an incentive to adopt AUDs, which suggests that the asymmetry in capacity between the firms allows the unconstrained firm to take advantage of the quantity commitment through AUDs, while the smaller rival is induced to offer linear pricing and is partially foreclosed.

Article 53, paragraph 2 of the AML states that “where any party concerned is dissatisfied with any decision made by the anti-monopoly authority, other than the decisions prescribed in the preceding paragraph, it may lodge an application for administrative reconsideration or initiate an administrative lawsuit in accordance with law.”

Recently, Deng and Katsoulacos (2017) suggest that current fining policies in the EU and throughout the world typically base fines on revenue in the relevant market. A typical reason is that revenue-based penalties have the lowest implementability costs and tend to have less legal uncertainty than do penalties that are based on estimates of illegal gains. They also cite Lin (2016), who suggests that illegal gains confiscations were imposed in a very small portion of the cases investigated by the SAIC or the NDRC.

On June 17, 2016, the NDRC requested comments with regard to the Draft Guidelines on the Determination of Illegal Gains and Fines in Relation to Undertakings’ Monopoly Conduct, available at http://​www.​ndrc.​gov.​cn/​yjzq/​201606/​t20160617_​807549.​html. Illegal gains are defined by the NDRC as the additional profits earned from monopoly agreements during the period of the illegal conduct, which is equal to “actual income” minus “hypothetical income.” Actual income refers to the income earned during the period that the illegal conduct lasts, and hypothetical income refers to the income that the company could have obtained in the relevant market if there was no illegal conduct during the period concerned.

For a more detailed discussion on abuse of dominance under the 1986 Canadian Competition Act, see Church and Ware (1998).