Keywords: Anti-Monopoly Law, merger control, MOFCOM, Glencore/Xstrata, Marubeni/Gavilon, Baxter/Gambro, MediaTek/MStar,

There were some interesting twists in the public and private enforcement of China's Anti-Monopoly Law (AML) during 2013. In the area of merger control, the Chinese Ministry of Commerce (MOFCOM) published four conditional clearance decisions: Glencore/Xstrata, Marubeni/Gavilon, Baxter/Gambro and MediaTek/MStar. Each decision turns on its own facts, but there are recurring themes:

  • MOFCOM is prepared to find market power notwithstanding relatively low market share levels.
  • There is a continued preference for the imposition of elaborate and onerous hold-separate arrangements as a condition for clearance.
  • MOFCOM has sought commitments to supply certain key products to the Chinese market on favorable terms as a pre-condition to clearance.
  • MOFCOM will shy not away from imposing extraterritorial remedies even where the competition economics basis for seeking the commitment might not be that clear cut.
  • Coordinated effects theories of harm arise with some regularity in MOFCOM's published decisions and this year's cases offer further examples.

Some of these themes appear driven by industrial policy considerations and indicate that any transaction that involves key industries—food and agriculture in Marubeni/Gavilon, minerals and ores in Glencore/Xstrata, important inputs for Chinese manufacturers in MediaTek/MStar—will be scrutinized closely and regulated with an eye toward broader strategic interests. By contrast, Baxter/Gambro is something of an anomaly in so far as political considerations were not in issue.

We explore below these four cases in detail and consider the implications for MOFCOM's future practice.


Following a lengthy review lasting the best part of a year, on April 16, 2013, MOFCOM gave a conditional green light to the acquisition of Xstrata by Glencore. The long review period is not the only way in which the decision stands out. Particularly striking is MOFCOM's decision to seek an extraterritorial divestiture in circumstances where the competition arguments that might justify this decision appear to be borderline.

Also striking is the fact that the regulator, for the first time, published a detailed scheme of the remedial commitments it accepted from the parties as a pre-condition to clearance. This document offers a valuable insight into MOFCOM's practice and merits careful consideration. It is also a further indication of MOFCOM's ongoing and developing commitment to a more transparent process.

Although neither Glencore nor Xstrata own or operate productive assets in the relevant markets in China, MOFCOM took great interest in the transaction, focusing on the importance of China as a major market for the parties and China's reliance on imports of raw materials of central importance to the wider Chinese economy. Explaining that import volumes of copper, zinc and lead concentrate accounted for 68.5, 28.7 and 27.3 percent, respectively, of total supplies on the Chinese market in 2011, and that the parties had relatively high market shares in the production and supply of these products globally and in China, MOFCOM focused its review on these markets and ultimately concluded that the acquisition may have the effect of eliminating or restricting competition in them.

That said, the analysis of the parties' market power in the production and supply of copper concentrates suggests that MOFCOM is willing to find market power at what might otherwise be considered moderate market share levels. The regulator explained that Glencore and Xstrata are among the world's leading producers and suppliers of copper concentrate and that the global market shares of Glencore and Xstrata for the production of copper concentrate were 1.5 and 6.1 percent, respectively, in 2011, with a combined share of 7.6 percent, collectively ranking third in the world. Further, the global market shares of the parties in 2011 with respect to the supply of copper concentrate were 5.3 and 4 percent, respectively, with a combined share of 9.3 percent, ranking first in the world. As regards the market shares of Glencore and Xstrata for the supply of copper in China itself, these were 9 and 3.1 percent for 2011, giving a combined share of 12.1 percent—a leading position on the market.

While the published decision does elaborate other reasons for concluding the parties would have market power post-merger in the copper concentrate market (vertical integration and foreclosure concerns, barriers to entry), it is worth noting that the market share levels discussed above are all well below the 25 percent figure used by, for example, the European Commission to establish a threshold below which an absence of restrictive effects can be presumed. Whether MOFCOM will always be willing to find market power at such levels is not clear. Arguably, the central issue for the regulator in the case was an understandable concern that China was uniquely dependent on overseas supplies of a critical commodity.

The analysis of competition concerns in the markets for zinc and lead concentrate is, in many respects, comparable to the analysis for copper, albeit that MOFCOM noted that China is less dependent on imports as indicated above. MOFCOM therefore took the view that softer behavioral commitments would suffice in these markets.

The most notable of the restrictive conditions imposed—and likely the cause of the protracted review as the regulator and the parties sought to agree an acceptable compromise—is the required divestiture by September 2014 of Xstrata's Las Bambas copper mine in Peru at a price not lower than the higher of the fair market value of the mine or the sum of all costs incurred in developing it. The purchaser must be approved by MOFCOM, though Glencore has undertaken to use its best efforts to submit the details of all potential buyers of Las Bambas to MOFCOM by August 31, 2014. Whether this means MOFCOM would then select a buyer is not entirely clear.

If Glencore fails to comply with the structural remedy and sell Las Bambas within the required time, the remedy scheme provides that Glencore must submit a proposal to MOFCOM for the appointment of a divestiture trustee empowered to sell, without a reserve price, Glencore's interest in one of a number of alternative copper mining projects in Latin America or Southeast Asia as might be specified by MOFCOM.

In addition to the structural remedy, MOFCOM sought conduct remedies: behavioral conditions were imposed relating to the pricing and volumes of copper, zinc and lead concentrates supplied to Chinese customers. For a period of eight years, Glencore agreed to supply the Chinese market with 900,000 tons of copper concentrate annually at a regulated price—although the minimum volume is subject to adjustment in line with Glencore's actual levels of production. In the case of supplies of zinc and lead concentrates, Glencore agreed that during the eight-year supply commitment period, its offer conditions would be "fair, reasonable, and consistent with the then prevailing terms used in the international market."


MOFCOM published its conditional approval of Marubeni's acquisition of Gavilon Holdings hot on the heels of the decision in Glencore/Xstrata. The US$5.6 billion grain deal between one of Japan's largest trading companies and the third-largest North American grain company took just under one year for MOFCOM to clear.

First notified by the parties in June 2012, the notification was withdrawn and resubmitted in January 2013 at the end of a Phase III review and after an initial remedies proposal had been rejected. The transaction was ultimately cleared several days after the parties agreed to operate Marubeni's and Gavilon's China soybean export businesses through separate and independent legal entities backed up by firewall mechanisms to safeguard against the exchange of competitively sensitive information. The final remedies scheme includes the following elements:

  • Marubeni and Gavilon will set up two independent legal entities for the purpose of exporting and selling soybeans on the China market.
  • Marubeni's soybean subsidiary and Gavilon's soybean subsidiary will maintain structural and operational independence with respect to personnel, sourcing, marketing, sales and pricing functions. Post-completion, Marubeni's soybean subsidiary will not source soybeans from Gavilon's US assets except on an arm's-length basis.
  • Extensive hold-separate remedies of this kind are not unusual in the China context, and MOFCOM imposed similar arrangements in Western Digital/Hitachi in 2012. What is telling, is not so much the remedies as such but more the basis for seeking them in the first place. MOFCOM's decision rehearses the key considerations:
  • China is the world's largest importer of soybeans. In 2012, China's imported volume of soybeans accounted for 60 percent by volume of total worldwide soybean trade, and 80 percent of China's domestic supply.
  • China imported 58.38 million tons of soybeans in 2012, implying a total domestic market of 72.975 million tons (MOFCOM makes no reference however to total market size as it defines the relevant market as a market for imports).1
  • Marubeni shipped 10.5 million tons of soybeans to China in 2012, implying an approximate market share of 14 percent—or 18 percent if the relevant market is defined by reference to imports alone as in the MOFCOM decision.2 Marubeni ranked first among suppliers of imported soybeans in China.
  • Gavilon's global soybean sales in 2012 amounted to 5.1 million tons. This seems to translate into a global market share of approximately 5 percent.3 MOFCOM does not provide any market share for Gavilon in China but the figures provided in the decision allow one to determine that its share would be less that 1 percent.4
  • While noting that the supply of soybeans in China was highly dependent on imports (80 percent of all supplies were imported in 2012 as mentioned above), MOFCOM explained that the downstream domestic China market for soybean crushing was highly fragmented and characterized by small scale production with weak countervailing bargaining power.

Overall, the key consideration appears to have been that the deal would significantly boost Marubeni's access to global soybean resources through the acquisition of Gavilon's capacity for soybean origination, storage and logistics in North America, thus enhancing Marubeni's ability to import soybeans into China. This would result in what MOFCOM terms a "materially strengthening" of Marubeni's "control" over the import market for soybeans.

Nonetheless, a more orthodox assessment of the facts might lead one to question whether the parties have any particular level of market power on the relevant market for soybeans. In this respect, it is notable that MOFCOM appears not to have considered in any detail the degree of competitive constraint provided by Marubeni's rivals or the ability of competitors to expand in response to attempts by the merged firm to increase prices and/or lower output. And, as indicated above, it is striking that MOFCOM chose to define the relevant market as a market for imports into China thus, by implication, taking the view that domestic supplies were somehow not relevant to the assessment. Whatever the rationale for such an approach—the decision is silent on the point—the effect would be to overstate the parties' market position, albeit that on the facts of the case, not by very much.


Baxter/Gambro was conditionally approved on August 8, 2013, after MOFCOM opened a so-called "Phase III" review—an agreed extension to Phase II.

MOFCOM focused its competition assessment on the relevant global and domestic China markets for continuous renal replacement therapy (CRRT) equipment and related consumables (collectively, the markets for CRRT equipment) and haemodialysis (HD) equipment. MOFCOM concluded that the transaction would likely eliminate or restrict competition in these markets and specifically, for MOFCOM's purposes, the China market for the relevant products.

The pie charts below show the parties' respective market shares in the CRRT China markets as disclosed by MOFCOM in its decision. Interestingly, MOFCOM also offers a detailed assessment of concentration levels pre- and post-transaction using the Herfindahl-Hirschmann Index (HHI).5 MOFCOM's findings in this respect are shown here in chart form.

In relation to CRRT equipment, MOFCOM found that the transaction would significantly increase concentration levels in the relevant markets (the HHI "delta" or difference between the HHI levels pre- and post-acquisition serves as an indicator or proxy for the change in concentration levels brought about by the merger) and eliminate Baxter's closest competitor, resulting in high combined market shares held by the merged firm post acquisition and affording it what MOFCOM regarded as a "dominant position" in the China markets for CRRT equipment.

With respect to HD equipment, MOFCOM concluded that the transaction would likely result in "coordinated effects" in the relevant markets in China, with the two main competitors—the merged entity and a third party, Nipro Medical Corporation—holding a combined 48 percent market share. Although the "increment" in the merged firm's market share was a modest 3 percent (according to MOFCOM Gambro's market share was 19 percent, while Baxter's was 3 percent), the key consideration would appear to have been the existence of an agreement between Baxter and Nipro for the production of HD products. MOFCOM was concerned that this agreement created a risk that competitively sensitive information would be exchanged on such matters as production costs and quantities. The agreement would "facilitate the mutual coordination" of the two key players on the China HD market, according to MOFCOM. On this point, it might be noted that MOFCOM appears generally more open to deploying a coordinated effects theory of harm as compared with authorities elsewhere. That said, the existence of structural links between players in a market is generally recognized as an important consideration when assessing the likelihood of coordinated effects.

In light of its concerns, MOFCOM approved the transaction subject to the following:

  • The divestiture of Baxter's global CRRT business; and
  • The discontinuation of the Baxter-Nipro agreement for the production of hemodialyzers in so far as it related to China.

Other authorities, notably the EU Commission and the Australian Competition and Consumer Commission, accepted a similar remedy in relation to CRRT equipment. By contrast, the EU Commission noted in its press release announcing it had cleared the transaction that it "found that Baxter and Gambro are not particularly close competitors in HD and will continue to face, after the merger, significant competition from a range of dynamic market participants." Different regulators reaching different conclusions is, of course, an inevitable feature of global merger control.


MediaTek/MStar was first filed with MOFCOM on July 6, 2012. Following a number of rounds of unsuccessful negotiations between the parties and MOFCOM over the scope of a possible remedies package, the filing was eventually cleared on August 26, 2013.

MOFCOM noted that both merging parties are mainly engaged in the design and manufacture of integrated circuit chips for multimedia display and wireless communications devices. After concluding that the transaction would not likely have any anticompetitive effects on the market for mobile phone baseband chips, MOFCOM focused on the market for LCD TV control chips. The published decision is equivocal as to whether the geographic scope of this market is global or national, but MOFCOM's main concern was clearly the China market which, in any event, had its own special features in MOFCOM's view.

The pie chart below shows the parties' respective market shares in the relevant China market as set out in MOFCOM's decision. As in Baxter/Gambro, MOFCOM offered a detailed assessment of concentration levels using the HHI system of indicators. MOFCOM's findings in this respect are again shown here in chart form.

MOFCOM considered the relevant Chinese market for LCD TV control chips to be highly concentrated before the merger and noted that the transaction would "obviously" change the structure of the market. In MOFCOM's view, the acquisition would eliminate MediaTek's closest competitor and remaining suppliers, and the threat of new entry, would not constitute an effective competitive restraint post merger. MOFCOM concluded that the merged entity would become the "dominant" player in China with a market share of 80 percent.

In view of these concerns, MOFCOM imposed a rather striking set of behavioral remedies:

  • MStar's LCD TV control chip business (Morningstar) must be maintained as an "independent competitor" on the market;
  • MediaTek's exercise of its shareholder rights in Morningstar are to be strictly limited and subject to prior approval by MOFCOM with the exception of rights to receive dividends and information necessary for producing consolidated financial statements;
  • Directors of Morningstar may only be appointed/removed with the approval of MOFCOM;
  • MediaTek and Morningstar must maintain their R&D investments at no less than pre-acquisition levels;
  • MediaTek and Morningstar are prohibited from exchanging competitively sensitive information and using customers as conduits for the exchange of such information. Board members and senior executives who breach this obligation may be dismissed;
  • Cooperation between MediaTek and Morningstar is to be subject to prior approval;
  • Certain customary pre-acquisition practices regarding the supply of LCD TV control chips and after-sales service levels must be maintained post merger;
  • Should MediaTek and/or Morningstar merge in the future with another party active in the LCD TV control chip market, they must seek prior approval from MOFCOM. This applies regardless of the turnover of the undertaking concerned; and
  • MediaTek and Morningstar must comply with certain arrangements intended to control the prices of LCD TV control chips and related products sold on the China market. In particular, prices in China must not be higher than the prices of similar products sold by MediaTek and Morningstar outside China.

Looking at these remedies, the obvious question might be why MOFCOM would choose such a structure over a divestiture, which would seem equally capable of addressing the competition concerns identified while placing less of a burden on MOFCOM in terms of monitoring. On the other hand, a straightforward divestiture would not have afforded MOFCOM the leverage it acquired over LCD TV control chip pricing in China in view of the pricing commitment described above.


While a review of these recent cases clearly highlights the importance of industrial policy in a MOFCOM review, conditionally cleared cases are the exception rather than the rule. Such decisions make up a very small percentage of the cases filed with MOFCOM: approximately 97 percent of filings are cleared unconditionally. Standard competition considerations are therefore the prime concern although industrial policy could well trump this in a given case.

This article is extracted from an article initially published in The Asia-Pacific Antitrust Review 2014:

Originally published Spring 2014


1 MOFCOM does not in fact give any market share percentages for soybeans in its decision. It explains however that China imported 58.38 million tons of soybeans in 2012 and that this accounted for 80 percent of China's domestic supply. Accordingly China's total market size would be in the region of 72.975 million tons. Assuming Marubeni shipped 10.5 million tons of soybeans to China in 2012 (MOFCOM's figure), this implies a market share of 14 percent of all domestic supply or 18 percent if the relevant market is defined by reference to imports alone as MOFCOM does.

2 See footnote 1 above.

3 MOFCOM advises that in 2012, China's imported volume of soybeans accounted for 60 percent by volume of total worldwide soybean trade. As China imported 58.38 million tons of soybeans in 2012, this would suggest the global market amounted to 97.3 million tons. If Gavilon's global soybean sales in 2012 amounted to 5.1 million tons (MOFCOM's figure), this translates into a global market share of approximately 5 percent.

4 MOFCOM advises that in China Gavilon is active in the trading of "bulk agricultural produce such as yellow corn, soybeans, soy meal, and feed and food ingredients". MOFCOM further notes that in 2012, Gavilon exported to China a total quantity of about 400,000 tons of bulk agricultural produce. Even assuming this entire volume was constituted by soybeans (which it would not be), this translates into a market share of less than 1 percent even where the relevant market is defined solely by reference to imports.

5 The Herfindahl-Hirschmann Index measures concentration levels in a given market by summing the squares of the individual market shares of all the firms in the market.

Learn more about our Antitrust & Competition practice.

Visit us at

Mayer Brown is a global legal services organization comprising legal practices that are separate entities (the Mayer Brown Practices). The Mayer Brown Practices are: Mayer Brown LLP, a limited liability partnership established in the United States; Mayer Brown International LLP, a limited liability partnership incorporated in England and Wales; Mayer Brown JSM, a Hong Kong partnership, and its associated entities in Asia; and Tauil & Chequer Advogados, a Brazilian law partnership with which Mayer Brown is associated. "Mayer Brown" and the Mayer Brown logo are the trademarks of the Mayer Brown Practices in their respective jurisdictions.

© Copyright 2014. The Mayer Brown Practices. All rights reserved.

This article provides information and comments on legal issues and developments of interest. The foregoing is not a comprehensive treatment of the subject matter covered and is not intended to provide legal advice. Readers should seek specific legal advice before taking any action with respect to the matters discussed herein. Please also read the JSM legal publications Disclaimer.