The Chinese competition authority went through a major restructuring in November 2021. The Anti-Monopoly Bureau under the State Administration for Market Regulation (“SAMR”) was expanded to three separate bureaus to cover respectively conduct investigation, merger review and competition policy, which together constitute the new independent State Anti-Monopoly Bureau (“SAMB”) within SAMR. Looking back on the merger review cases during the past few years (2018-2022), there are some notable new trends: (1) SAMR appears to be more likely to find competition concerns in vertical and conglomerate mergers than competition authorities in other major jurisdictions; (2) FRAND supply commitments to Chinese customers have become common in most conduct remedy cases; (3) SAMR has embraced a new behavioral remedy of helping a third party to enter the market; and (4) SAMR is less willing to allow automatic sunset clauses after an initial remedy term and instead often requires parties to apply at some certain points in the future for and justify termination of the remedy conditions based on changed market conditions.
Yizhe Zhang & Peter Wang[1]
I. INTRODUCTION
This article summarizes several new trends in merger review by the Chinese competition authority — SAMR — that have emerged over the past 4 years (2018-2022).[2] These include the increased concern regarding conglomerate effects, the expansive use of FRAND commitments, the introduction of a new type of behavioral remedy, and the frequent use of open remedy terms without automatic sunset clauses. These developments reflect SAMR’s increased focus on domestic industries and have profound implications for future transactions.
II. AGGRESSIVE ENFORCEMENT AGAINST CONGLOMERATE TRANSACTIONS
Compared with competition authorities in other jurisdictions, SAMR is more receptive to theories of harm based on conglomerate effects. During the past few years, SAMR imposed conditional approvals in 6 conglomerate transactions, 5 of which had been unconditionally cleared by the European Commission (“EC”) and the U.S. competition authorities.
A closer examination of these cases shows that high market shares (typically more than 50 percent) in non-overlapping products sold to the same group of customers can easily lead to conglomerate concerns in China. SAMR’s conglomerate concerns or theories of harm can include tying, imposing potentially less favorable trade terms post-transaction, degrading interoperability and refusals to deal. Chart 1 below reflects categories of SAMR’s concerns in recent conglomerate cases.
Tying is the leading concern expressed by SAMR in conglomerate transactions. For example, in Infineon/Cypress (2020), SAMR found that two groups of products, i.e. automotive-grade IGBT/automotive-grade MCU, and automotive-grade NOR flash memory/automotive-grade MCU, have the same customer group. The post-transaction entity could tie automotive-grade MCU with automotive-grade IGBT or automotive-grade NOR flash memory to coerce downstream customers to procure Infineon’s MCU.[3]
Degrading interoperability is another typical concern for SAMR in technology deals. For example, in AMD/Xilinx (2022), SAMR was concerned that the post-transaction entity would be likely to degrade interoperability between its field-programmable gate arrays (“FPGAs”) and CPUs or GPUs produced by third parties, leveraging its strong market power (over 50 percent market share) in FPGAs, to eliminate or restrict competition.[4]
Chart 1. Types of Concerns from SAMR in Conglomerate Cases (2018-2022)
No. |
Case |
Tying and/or bundling |
Product integration |
Degrading interoperability |
Less favorable trade terms post- transaction |
1 |
AMD/Xilinx (2022) |
ü |
n/a |
ü |
ü |
2 |
Nvidia/Mellanox (2020) |
ü |
n/a |
ü |
ü |
3 |
Infineon/Cypress (2020) |
ü |
ü |
ü |
ü |
4 |
KLA-Tencor/Orbotech (2019) |
ü |
n/a |
n/a |
n/a |
5 |
UTC/Rockwell Collins (2018) |
ü |
n/a |
ü |
ü |
6 |
Essilor/Luxottica (2018) |
ü |
n/a |
n/a |
n/a |
In contrast, EC and U.S. antitrust authorities cleared 5 of these cases unconditionally, only imposing divesture remedies in UTC/Rockwell Collins (2018) to address horizontal, rather than conglomerate, concerns.
In UTC/Rockwell Collins (2018), SAMR found that Rockwell Collins had a 40-45 percent share in China’s avionics market, and UTC had high shares in certain markets, including engine nacelles (40-45 percent), auxiliary flight control actuators (30-35 percent), ice-detector systems (90-95 percent), power systems (75-80 percent), and fire control systems (50-55 percent). SAMR therefore was concerned that the post-transaction entity would adopt a tying or bundling strategy to leverage its market power in avionics into neighboring products, or vice versa. Similarly, SAMR found that UTC had a dominant position in air data sensors (55-60 percent) and integrated air data systems (95-100 percent), leading it to conclude that the post-transaction entity would have the ability and incentive to leverage that market strength into air data computers.[5]
The EC also examined conglomerate links between the parties’ activities in (a) aircraft engines and avionics, (b) environmental control systems and galley cooling, and (c) pilot controls, flight control and actuation. However, the EC found that the parties had no such ability and incentive to foreclose competitors through the use of tying or bundling strategies.[6]
The U.S. Department of Justice (“DOJ”), in its Competitive Impact Statement, did not discuss conglomerate concerns, but rather addressed only horizontal concerns in two relevant markets, namely pneumatic ice protection systems for aircraft, and trimmable horizontal stabilizer actuators for large aircraft. Given that both parties were close competitors in such concentrated markets, the DOJ proposed a divesture as a remedy.[7]
III. THE EXPANSIVE USE OF FRAND COMMITMENTS
A fair, reasonable and non-discriminatory (“FRAND”) commitment is typically made by patentees of standard essential patents (“SEPs”) to standard setting organizations (“SSOs”). The FRAND doctrine aims to avoid the acquisition or abuse of dominant market position by SEP holders as a result of the standard setting process. Nevertheless, a FRAND commitment in and of itself is ambiguous as to what level of royalty constitutes a FRAND royalty in the context of SEP licensing.
FRAND commitments were first introduced into Chinese merger remedies in MOFCOM’s conditional clearance of Google/Motorola Mobility (2012). In this case, Google was required to continue to obey FRAND obligations related to the patents of Motorola Mobility.[8] MOFCOM imposed more detailed FRAND obligations on SEP holders in Microsoft/Nokia (2014).[9] In these two merger cases, the FRAND obligations were strictly limited to SEP licensing.
Recently, SAMR has further expanded the FRAND doctrine to non-SEP cases to ensure fair post-transaction supply to Chinese customers. Since 2018, SAMR has imposed FRAND commitments on behalf of Chinese customers in 11 out of 16 conditional approval cases involving behavioral remedies.
SAMR has imposed FRAND conditions even in cases where the parties’ shares were relatively moderate. For example, in Cisco/Acacia (2021), the post-transaction entity’s total shares in the coherent digital signal processor market would have been 45-50 percent worldwide and 40-45 percent in China respectively. In Globalwafers/Siltronic (2022), the FRAND supply obligation was expanded beyond the relevant product of concern (8-inch wafer) to the supply of 6-inch and 12-inch wafers, which were defined as separate relevant markets by SAMR. In AMD/Xilinx (2022), the FRAND obligation was expanded beyond the relevant product of concern (FGPA) to AMD’s CPU and GPU processors, even though SAMR found AMD’s 2020 CPU and GPU worldwide market shares to be below 5 percent and 10 percent respectively.[10]
The specific meaning of FRAND is subject to SAMR’s interpretation. However, conditions imposed in the above cases seem to suggest that FRAND as envisioned by SAMR in merger remedies typically implies: (1) that no discriminatory treatment should be imposed against customers facing similar trading conditions; (2) that refusals, restrictions or delays to supply products should be prohibited; and (3) suppliers should not provide Chinese customers with terms less favorable nor levels of service inferior to those given pre-transaction. Chart 2 below summarizes specific obligations relating to FRAND in 11 recent cases.[11]
Chart 2. Obligations relating to FRAND Commitments in SAMR Merger Cases (2018 – Jan. 2022)
IV. THE INTRODUCTION OF A NEW BEHAVIORAL REMEDY
In SK Hynix/Intel (2021),[12] the Chinese competition authority embraced a new type of behavioral remedy: requiring the post-transaction entity to assist third-party competitors to enter the relevant markets (“Market Entry Assistance”). In that case, SAMR expressed competition concerns over unilateral and coordinated effects in the (global and Chinese) markets for PCIe enterprise-class SSDs and SATA enterprise-class SSDs.[13]
The SAMR noted high market entry barriers in both relevant markets, given that enterprise-class SSDs are mainly used in data center servers; customers have extremely high requirements for product quality and stability; and new entrants usually face financial and customer-recognition obstacles.
It is not clear from the decision what the appropriate circumstances are for requiring such a market entry assistance remedy. If SAMR is concerned about eliminating a competitor, divestiture (which maintains an existing competitor in the market) appears to be a more logical remedy, but the criteria are not yet well-defined.
The remedy of helping new entry also may face some practical issues in implementation.
First, the scope of the mandated “assistance” is unclear. Assistance is a broad term that could cover not only financing or loan efforts, but potentially also the transferring or licensing of intellectual property rights (“IPRs”) such as patents, copyrights, trademarks, and even know-how. Financing or loans readily can be obtained from foreign and China financial markets rather than from merging parties. However, the transferring or licensing of IPRs is always a complex battlefield, particularly between incumbent companies and potential competitors. The post-transaction entity and the third-party competitor would need to negotiate, for example, the number and types of IPRs required to enter into the relevant markets, whether the IPRs include SEPs and non-SEPs, and applicable royalty rates. The third-party competitor might also ask for the post-transaction entity’s assistance in product quality control and servicing.
Second, the meaning of “entry” can be interpreted differently. Market entry can be understood as either a one-time effort, such as licensing a patent portfolio, or as a continuous effort over a certain period of time, such as assisting in building up a new production line. Given the complexity of certain business environments, a third-party competitor may not be able to grow to become a stable supplier. It is not clear whether the merging parties have an obligation to ensure successful third-party entry, and at what point their assistance obligation would end.
Third, there are coordination risks relating to the potential exchange of competitively sensitive information. Absent a clean team and firewall during the assistance period, the risk of exchanging competitively sensitive information would exist between the post-transaction entity and the third-party competitor or new entrant. However, at present it appears that in the merger review context SAMR is more concerned about reducing alleged vertical and conglomerate effects (particularly on Chinese customers) rather than any risks of facilitating horizontal coordination.
V. SUNSET AND OPEN-ENDED TERMINATION CLAUSES
The Interim Provisions on the Review of Concentration of Undertakings (2020) (“SAMR Interim Provisions”) provide two methods of termination: (1) automatic termination of conditions upon expiry of the remedy term if there was no breach of the conditions, and (2) termination of conditions upon SAMR’s review and approval.[14]
Among the 16 cases involving conduct remedies from 2018 to January 2022 (see Chart 3 below), SAMR adopted automatic sunset clauses in only 6 cases and required open-ended termination clauses in 10 cases.
Termination of conditions only based on SAMR’s approval, even after expiry of the original remedy term, requires a substantive evaluation of market conditions years after, places the burden of proof on the merging parties, and thus essentially constitutes a second merger review potentially extending the remedy term indefinitely. Such practically unlimited remedy terms in China have become a source of frequent frustration for merging parties.
Chart 3. The Number of Cases with Sunset and Open-Ended Termination Clauses (2018 – Jan 2022)
Sunset Clauses in SAMR Cases
ZF/Wabtec (2020)
Case |
Sunset Clause |
Cisco/Acacia (2021) |
Lift automatically 6 years after the decision. |
Lift automatically 6 years after the decision. |
|
Infineon/Cypress (2020) |
Lift automatically 5 years after the decision. |
Novelis/Aleris (2020) |
Lift automatically 10 years after the decision. |
Zhejiang Garden High Tech/DSM (2019) |
Lift automatically 5 years after the decision. |
KLA-Tencor/Orbotech (2019) |
Lift automatically 5 years after the decision. |
Open-Ended Termination Clauses in SAMR Cases
Case |
Open-Ended Termination Clause |
AMD/Xilinx (2022) |
Lift by application and upon SAMR approval after 6 years’ initial term. |
Globalwafers/Siltronic (2022) |
Lift by application and upon SAMR approval after 5 years’ initial term. |
SK Hynix/Intel (2021) |
Lift by application and upon SAMR approval after 5 years’ initial term. |
ITW/MTS (2021) |
Lift by application and upon SAMR approval after 5 years’ initial term. |
Nvidia/Mellanox (2020) |
Lift by application and upon SAMR approval after 6 years’ initial term. |
II-VI/Finisar (2019) |
Lift by application and upon SAMR approval after 3 years’ initial term. |
Cargotec/TTS (2019) |
Lift by application and upon SAMR approval after 5 years’ initial term. |
Linde/Praxair (2018) |
Lift by application and upon SAMR approval after 5 years’ initial term. |
UTC/Rockwell Collins (2018) |
Lift by application and upon SAMR approval after 5 years’ initial term. |
Essilor/Luxottica (2018) |
Lift by application and upon SAMR approval after 5 years’ initial term. |
VI. CONCLUSION
SAMR has and exercises significant discretion in its merger review cases and is not shy of adopting unique theories of harm and imposing China-specific conditions. For complex global transactions that potentially require remedies, parties and their counsel must thoughtfully assess the antitrust risks associated with SAMR’s review, and plan their China filings accordingly.
1 Yizhe Zhang & Peter Wang are partners of Jones Day. Yichen Wu and David Wu, associates in Jones Day’s Beijing Office, assisted in preparation of the article. The views expressed in this article are those of the authors and do not necessarily reflect the views of Jones Day.
2 SAMR’s Anti-Monopoly Bureau was reorganized and changed its name to the State Anti-Monopoly Bureau (“SAMB”) in November 2021. SAMB currently oversees three bureaus within SAMR, i.e., the Anti-Monopoly Enforcement Bureau I, the Anti-Monopoly Enforcement Bureau II and the Competition Policy Coordination Bureau. The functions of each bureau are available at https://www.samr.gov.cn/jg/.
3 SAMR Conditional Approval of Infinon’s Acquisition of Cypress (2020) available at https://www.samr.gov.cn/fldj/tzgg/ftjpz/202004/t20200408_313950.html.
4 SAMR Conditional Approval of AMD’s Acquisition of Xilinx (2020) available at https://www.samr.gov.cn/fldj/tzgg/ftjpz/202201/t20220127_339441.html.
5 SAMR Conditional Approval of UTC’s Acquisition of Rockwell Collins (2018) available at https://www.samr.gov.cn/fldj/tzgg/ftjpz/201811/t20181123_332679.html.
6 EC Case M.8658 – UTC/Rockwell Collins Decision on May 4, 2018 available at https://ec.europa.eu/competition/mergers/cases/decisions/m8658_2749_3.pdf.
7 U.S. v. UTC and Rockwell Collins, Inc. (Case Number: 1:18-cv-02279-RC) (2019) available at https://www.justice.gov/atr/case/us-v-united-technologies-corp-and-rockwell-collins-inc.
8 MOFCOM Conditional Approval of Google’s Acquisition of Motorola Mobility (2012) available at http://fldj.mofcom.gov.cn/article/ztxx/201205/20120508134324.shtml.
9 MOFCOM Conditional Approval of Microsoft’s Acquisition of Nokia’s Device and Service Businesses (2014) available at http://fldj.mofcom.gov.cn/article/ztxx/201404/20140400542415.shtml.
10 EC Decision of AMD/Xilinx Case (2021), Para 108 and 126, available at https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32021M10097&from=EN.
11 The cases are Essilor/Luxottica (2018), UTC/Rockwell Collins (2018), KLA-Tencor/Orbotech (2019), DSM/Zhejiang Garden (2019), Infineon/Cypress (2020), Nvidia/Mellanox (2020), ZF/Wabco (2020), Cisco/Acacia (2021), SK Hynix/Intel (2021), GlobalWafers/Siltronic (2022), and AMD/Xilinx (2022).
12 SAMR Conditional Approval of SK Hynix’s Acquisition of Intel’s SSD and NAND Businesses (2021) available at https://www.samr.gov.cn/fldj/tzgg/ftjpz/202112/t20211222_338317.html.
13 In contrast, the EC concluded, among other things, that the post-transaction entity will still have to live up to the pace of a very dynamic market, and Samsung will still hold the largest market share and continue to be the market leader, making unilateral effect less likely to be a serious concern. See EC Case M.10059 – SK Hynix / Intel’s NAND and SSD Business, Article 6(1)(b) Non-Opposition, May 20, 2021, Paragraphs 136-38, available at https://ec.europa.eu/competition/elojade/isef/case_details.cfm?proc_code=2_M_10059.
14 Article 46 of the SAMR Interim Provisions on the Review of Concentration of Undertakings (2020) available at https://gkml.samr.gov.cn/nsjg/fgs/202010/t20201027_322664.html.