Last week, China rejected on antitrust ground Coca-Cola’s acquisition of the China Huiyuan Juice Company.  Today, the AP (via the NY Times) reports:

In its first detailed explanation of the March 18 ruling, the commerce ministry said it looked at China’s beverage market and concluded that Coca-Cola’s dominance in carbonated drinks could be used to promote sales of Huiyuan Juice Group, stifling competition and leading to higher prices.

…

“Whether Huiyuan is a national brand is not a factor that needs to be considered in an antimonopoly investigation and has nothing to do with the commerce ministry’s rejection of this acquisition,” a ministry spokesman, Yao Jian, said in statement on its Web site.

…

Coca-Cola’s “dominant position in the carbonated beverage market could be transferred to the fruit juice market,” Mr. Yao said. “This would seriously cripple or deprive other fruit juice producers of the ability to compete. It would harm competition in the fruit juice market, forcing customers to accept higher prices and fewer products.”

He said Coca-Cola might have used its carbonated beverage brands to promote Huiyuan or bundle products together for sale.

…

The company [Coca-Cola] has a 16 percent share of China’s soft drink market — which also includes fruit and vegetable juices, bottled teas and sports drinks — and adding Huiyuan would have raised that to 18 percent, according to the consulting firm Euromonitor.

Yesterday, the U.S. Court of Appeals for the Fourth Circuit issued an opinion in a RPM case.  According to the Court: 

In this Sherman Act suit, plaintiffs, who provide pest control services to individual customers, allege that defendants, who manufacture pesticides, illegally conspired with their distributors to set minimum resale prices of certain termiticide products. Specifically, plaintiffs claim that defendant manufacturers Bayer CropScience LP and Bayer Corp. (hereinafter collectively referred to as “Bayer”) and BASF Corp. each engaged in the practice known as “resale price maintenance” or “vertical price fixing” — Bayer with its product Premise and BASF with its product Termidor. Defendants counter that United States v. General Electric Co., 272 U.S. 476 (1926), held that a manufacturer may lawfully set minimum prices for its products when there is a genuine principal-agent relationship between the manufacturer and its distributors, and that such relationships existed here. Plaintiffs rejoin that Leegin Creative Leather Products, Inc. v. PSKS, Inc. , 127 S. Ct. 2705 (2007), implicitly overruled General Electric, and in the alternative argue that the agency relationships between defendants and their distributors were a sham. Because Leegin did not eliminate the agency defense to a claim of resale price maintenance and the agency relationships between defendants and their distributors were genuine, we find no basis for antitrust liability and thus affirm the district court’s grant of summary judgment to defendants.

… Plaintiffs insist in their briefs and at oral argument that afterLeegin the agency defense under General Electric to a claim of resale price maintenance is no longer viable. This argument fails because the two cases dealt with two separate elements of antitrust liability: General Electric addressed what types of relationships constitute agreements to set prices for purposes of the Sherman Act, while Leegin concerned whether such agreements, once proven, should be considered per se unlawful.

You can read the whole opinion here.

Due to indexing issues in the database behind The Antitrust Review, most links on this blog are currently broken. We are working to fix the problem. Please pardon us in the meantime.

Fixed. Please let us know if you still see broken links.

One of the effects of the Supreme Court’s recent LinkLine decision was to extend the reach of Trinko. Trinko dealt with the relation of regulation (in the telecomm industry) and antitrust. The decision proceeded in three steps:

1. The 1996 Telecommunications Act declares the antitrust laws to be applicable.
2. The claim in Trinko, which concerned non-price terms in the resale of telecomm services under a duty to deal flowing from the 1996 Act, was a novel claim that would have extended the reach of the antitrust laws (specifically §2 Sherman Act).
3. Because of the dangers of false positives inherent in antitrust enforcement, and because of the presence of an regulatory scheme designed to prevent and sanction antitrust violations, §2 should not be extended to reach the claims in Trinko, absent an antitrust duty to deal.

The Ninth Circuit, in its Linkline decision, which was overruled by the Supreme Court, used this structure of the Trinko decision to argue that Trinko should not preclude Linkline’s price-squeeze claim, since the price squeeze was not an extension of the antitrust laws (having been established since Alcoa), and because the regulatory scheme in Linkline did not (effectively) address antitrust violations.

But in the opinion of the court in Linkline, Chief Justice Roberts writes that Trinko

… makes clear that if a firm has no antitrust duty to deal with its competitors at wholesale, it certainly has no duty to deal under terms and conditions that the rivals find commercially advantageous. … The nub of the complaint in both Trinko and [Linkline] are identical—the plaintiffs alleged that the defendants (upstream monopolists) abused their power in the wholesale market to prevent rival firms from competing effectively in the retail market. Trinko holds that such claims are not cognizable under the Sherman Act in the absence of an antitrust duty to deal.

This re-interpretation extends Trinko in significant ways. First, the specific application to the telecommunications industry and its regulatory scheme is gone; Trinko á la Linkline is broadly applicable to any vertically-integrated upstream monopolist. Second, the holding is no longer confined to an extension of the antitrust laws, but reaches all of §2 (hence the possibility of a slippery slope, as Hanno pointed out). Finally, the prensence and effectiveness of a regulatory scheme, which provides a regulatory duty to deal and which may or may not address abuses of market power, has been taken out of the equation: the only thing that matters now is the absence of an antitrust duty to deal. And we know from Trinko that antitrust duties to deal probably don’t exist except perhaps under Aspen Skiing, that is in the case of prior business relationships discontinued for no good reason other than exclusion of a rival.

Earlier today, the Chinese government rejected on antitrust ground Coca-Cola’s acquisition of the China Huiyuan Juice Company.  The New York Times reports:

The ministry of commerce said Coke’s bid to acquire the China Huiyuan Juice Company was rejected on antitrust reasons. The government said the deal would allow Coke to dominate a huge segment of the beverage market. In a statement released Wednesday, the commerce ministry said it was worried that Coke would “set up some exclusive terms to restrict competition in the juice market,” drive up the consumer prices and squeeze out smaller beverage makers.

The statement is available, in Chinese, here (if/when we can find the statement in English, we will post a link to it too).

The FTC and Whole Foods have reached a settlement argeement.  In short, “Whole Foods agreed to divest itself of 31 Wild Oats stores in 12 states, including 19 stores that already have been closed, and one Whole Foods store. Whole Foods also agreed to relinquish the rights to the Wild Oats brand, which could be sold to a potential competitor.”  More at the WSJ Law Blog.

The new edition of The Antitrust Source is out.

In a new amendment to Korea’s Monopoly Regulation and Fair Trade Law, Korea will abolish the 30-day waiting period that previously applied to pre-closing merger notification. While the parties may therefore file as soon or late after the triggering event (such as the signing of a definitive agreement) as they like, the reportable transaction may not close before it is cleared by the the KFTC. The amendments will go in effect in about 3 months.

HT to Kim & Chang.

As the NY Times reported last week, the Treasury Department has agreed to take a 30 to 40 percent stake in Citigroup; the acquired securities will be voting common stock. Following up on my recent posts about HSR reportability, Jon Baker  raised this interesting question:

If the US Government owns a substantial equity stake in one bank and then acquires a substantial equity stake in a competing bank, would it have to file HSR notification (assuming that a private firm in the same situtaiton would be required to file)?

Let me start by saying that the fact of a substantive overlap (a stake in another bank is already held by the acquiror), does not affect HSR reportability, which turns solely on whether there has been an acquisition (of sufficient size) between persons (of sufficient size), to which no exemptions apply. HSR reportability uses formal criteria that are linked only indirectly to the question under Section 7: will the acquisition substantially lessen competition?

So, what are the exemptions that apply to acquisitions of voting securities by governments? There are three:

The first, §7A(c)(4), exempts “transfers to or from a Federal agency or a state or political subdivision thereof.” If the Treasury Department, which is officially an “executive agency” of the U.S., will acquire the voting securities directly, the acquisition is exempt from HSR reportability. That is the easy part of the answer about the federal bank bailout.

The second exemption is buried in the Rules’ definition of “entity.” Rule 801.1(a)(2) defines “entity” as just about everything,

Provided, however, That the term “entity” shall not include any foreign state, foreign government, or agency thereof (other than a corporation engaged in commerce), nor the United States, any of the States thereof, or any political subdivision or agency of either (other than a corporation engaged in commerce).

It’s a backward way of exempting acquisitions by governments, states, or the US: If the entity whose assets or voting securities are being acquired, or who is acquiring assets or voting securities, isn’t an “entity,” then whoever ultimately controls the non-entity can’t be a person. (Person means the ultimate parent entity and all entities which it controls directly or indirectly, 801.1(a)(1); more on this in my HSR Primer posts.) No entity, no acquiring or acquired person, no acquisition, no filing obligation.

The definition in 801.1(a)(2) raises a few questions, some of which have come up in the FTC’s informal interpretations. What exactly are “political subdivisions or agencies of either” the U.S. or the States? Are agencies of subdivisions of States, such as the agency of a municipality, not “entities”? What about multistate agencies, such as the Port Authority of New York and New Jersey? The correct answer is that the exemption should apply, as Axinn, Fogg, Stoll, Prager, Nisa point out in §6.09[4] of their treatise.

The question that is harder to answer is what “engaged in commerce” means. (Note the drafting oversight that restricts the exception from the exemption to corporations - a State-owned LLC or other unincorporated entity engaged in commerce is not an “entity,” a State-owned corporation engaged in commerce is.) Engaged in “commerce” must mean something other than interstate commerce, which is defined in 801.1(l), with reference to the statutory jurisdictional grants under the Commerce Clause. The definition of “entity” in the Rules, however, distinguishes between governmental and commercial activities. The first example that springs to mind here at The Antitrust Review is the Hofbräuhaus in Munich, which is owned by the State of Bavaria (did you know that brewing wheat beer or “Weissbier” was a ducal privilege and that the Hofbräuhaus held the monopoly on brewing it for over 200 years? Antitrust is everywhere). The subtext behind the definition of “commercial” for purposes of exempting government corporations from HSR reportability should be the state-action doctrine, as formulated in Parker v. Brown and Midcal. That line of cases shows that we are reluctant to exempt state and city governments from antitrust scurtiny.

There is a third exemption for acquisitions by governments, 802.52. It applies only to foreign governments, and I won’t go into any detail here. But note that foreign states, governments, and their agencies (other than a corporation engaged in commerce) also are not “entities” under the Rules, so that 802.52 only becomes relevant for corporations engaged in commerce whose ultimate parent entity is a foreign state, government, or agency.

The answer to the original question about the bank bailout is that HSR will most likely not apply. But if the banks are corporations engaged in commercial activity, they will be “entities” under the Rules and HSR potentially applies as long as the acquisition isn’t a “transfer to or from a Federal agency or a State or a political subdivision thereof” (in other words, unless the statutory exemption of §7A(c)(2) applies, which the Rules cannot abrogate). HSR reportability of course does not affect the bailout’s legality under §7 Clayton Act one way or another.

Here is a summary for the graphically inclined:

I will be at the ABA White Collar Crime this Wednesday through Friday in San Francisco.  If you are there, be sure to say hello.

In the better-late-than never category, President Obama has nominaed Jon Leibowitz to be the Chairman of the Federal Trade Commission.  Although predictions can be dangerous, a safe one is that he will be easily confirmed by the Senate.

In the last post of this series, I posed a challenge: to find the UPE, or ultimate parent entity, and acquired person in a hypothetical.  To find the answer, three steps are necessary:

1. Identify the acquired entity, that is (1) the enitity whose assets are being acquired, or (2) issuer whose voting securities are being acquired, or (3) the unincorporated entity whose membership interests are being acquired.
2. Trace the chain of control up, using the definitions for control for corporations, unincorporated entities, and trusts, until the entity is reached which itself is not controlled by another entity. That is the UPE. (Individuals are never controlled, and therefore always UPEs.)
3. Trace the control down from the UPE (or UPEs) until all controlled entities are identified. They are the entities “included” in the person. The person is the UPE plus all entities controlled by it, directly or indirectly, 801.1(a)(1). Without finding all entities included in the acquired or acquiring person, the size-of-person test cannot be applied.

Remember 801.1(c)(8): A person holds all assets and voting securities held by the entities included within it; in addition to its own holding, an entity holds all assets and voting securities held by the entities which it controls directly or indirectly.

In the hypothetical the acquired entity is H Inc. It is the corporation whose voting securities are being acquired. Tracing the chain of control up, you’ll find three UPEs (marked with UPE crowns): Mrs. Z, Mr Z, and A Inc.

click to enlarge

Some explanations:

• Mrs. Z and Mr Z are both UPEs, even though neither of them individually meets the definition of control for corporations (holing 50%+ voting securities of H Inc, or having the right to appoint 50%+ of directors). The reason is that the holdings of spouses (and their minor children) are aggregated under Rule 801.1(c)(2).  Mrs. Z and Mr Z are therefore treated as if they were each holding 55% of H Inc’s voting security, which is enough to give them control. Since individuals can’t be controlled by another entity, they are each a UPE and will have to file as acquired persons. Only one notification form is required for spouses, however, 803.2(a).
• F LLC controls H Inc because it has the right to appoint 50% directors (even though it holds less than 50% voting securities).
• Mr Y does not control F LLC: While he has the right to appoint the entire management board of F LLC, for HSR purposes unincorporated entities are “controlled” only by entities with a right to 50% of profits or assets upon dissolution. This distinguishes the treatment of unincorporated entites from that of corporations.
• Instead F LLC is controlled only by E Inc., which has a right to 99% of F LLC’s profits or assets upon dissolution.
• But who controls E Inc? There is no single entity that either holds 50%+ of E Inc’s voting securities or has a right to appoint 50%+ of directors. The four shareholders of E Inc are Mr Y (1% voting securities or vs), B Inc (29% vs), C Inc (30% vs), and D Inc (40% vs). B and C are controlled by A Inc, their sole shareholder, and D Inc is controlled by both C Inc and L Inc (each 50% vs). This means that A Inc. indirectly controls E Inc., since A Inc, through B, C, and D controls a total of 99% of the voting securities of E Inc. Even though E Inc at first blush looks like it is controlled by no other single entity, it is indirectly controlled by A Inc. (L Inc does not indirectly control E Inc, even though it controls D Inc, since D Inc has only a 40% stake in E Inc.)
• Finally, at the top, A Inc. is a UPE because it is not controlled by anyone else: Mrs. X 25% stake is insufficient, as will be the stakes of all the public shareholders included in the 75% public float.

Once the UPEs have been identified the next step follows: tracing control down until all controlled entities have been identified. The following graph shows the analysis for A Inc only. (The other two UPE’s, Mrs. Z and Mr Z, each control H Inc, I Inc, and J Inc.)

click to enlarge

The only additional explanation required here concerns G GP, the general partner entity of G LP (G GP is some unincorporated entity, like an LLC). G GP is controlled by E Inc, which has the right to 51% of G GP’s profits or assets upon dissolution. But G GP does not control G LP for HSR purposes, even though in practical terms G GP will make all business decisions of G LP, as its general partner. The Rules’ definitions of control are rigid and formalistic:  G GP determines the competitive behavior of G LP, but only profits and assets upon dissolution are considered HSR control. Since 99% of those rights rest with the dispersed limited partners, no one controls G LP. (This is a common situation for investment funds organized as limited partnerships, or other unincorporated entities where governance rights and economic rights are divorced from each other.)

In closing it’s important to emphasize that control is not exclusive under HSR. An entity may be controlled by more than one other entity. For example, a corporation may have a 50%+ shareholder but another person who may appoint the majority of directors.  In fact, a corporation may be controlled by a maximum of 4 entities (ignoring spouses as shareholders):

• The first holding 50% voting securities,
• The second holding the other 50% voting securities,
• The third having the contractual right to appoint 50% of the board of directors, and
• The fourth having the contractual right to appoint the other 50% of the board.

An unincorporated entity can also be controlled by a maximum of 4 entities:

• The first having the right to 50% profits,
• The second the right to the other 50% profits,
• The third having the right to 50% assets upon dissolution, and
• The fourth the right to the other 50% assets upon dissolution.

(The right to assets upon dissolution refers to the residual distribution after all claims and preferences have been resolved; if the rights are not clear, determine control of an unincorporated entity as if it were being dissolved now.)

The next post in this series will deal with unincorporated entities in more detail, answering the question whether they are “assets” or “voting securities” under §7A.

Previous posts: §7 and HSR. The Basic Test. Parsing Acquisitions. Getting to the Person.

The LinkLine holding is not particularly surprising. Trinko applies at the wholesale level and Brooke Group at the retail level.

1. Wholesale — If there is no antitrust duty to deal in the first place, then the upstream firm may charge whatever price it wants. In fact, Roberts is pretty clear about the fact that the upstream firm can put the downstream competitor out of business in a number of ways: stop selling, higher prices, crummy service, etc. It all depends on whether there is an antitrust duty to deal. Here, there isn’t. (See FN.2)
2. Retail — Downstream pricing can only be an antitrust violation if it satisfies the (i) below cost and (ii) recoupment prongs of Brooke Group, which LinkLine doesn’t.

What’s interesting about LinkLine is not so much what it says (aside from some very quotable language), but what it doesn’t address. Unlike Scalia in Trinko who went out of his way to address and limit the application of essential facilities and duties to deal in general, LinkLine is silent on what exactly constitutes the all important duty to deal. Roberts cites Aspen as one source of a duty to deal, and then quotes Areeda’s “Epithet” article but only for the limited and rather uncontroversial proposition that “court’s shouldn’t be rate regulators, etc.”, which he then proceeds to discuss for about a page and a half.

So LinkLine (unlike Trinko) is a rather disciplined and limited decision, only applying (a moderate interpretation of) Trinko and Brooke Group to a set of facts that involves both elements. The only “novelty” is the disaggregation of a price squeeze claim in an upstream duty to deal and a downstream predatory pricing claim. The key question, however, has been left open: What constitutes a duty to deal?

A somewhat different question is whether Roberts’ consistent application of the “lesser included” argument, i.e., if you don’t have to sell at all, then whatever else you do to your buyer (price, quality, delay, etc.) can’t be an antitrust violation, may turn out to be a slippery slope. This line of reasoning is, of course, prominent in the IP context. As long as a firm stays within the scope and duration of the patent grant, it is all but immune from antitrust challenge (Fed. Cir, DOJ). I wonder what this will do, over time, to tying. In a tying case, one almost never has a duty to deal in the first place. The offense is conditioning the sale on the purchase of other stuff. In fact, the “lesser included” reasoning may well be extended to vertical restraints in general, all of which have some form of conditioning attached to them (I’ll sell you my stuff, but only if (i) you don’t resell it in California; (ii) you don’t sell anyone else’s stuff, etc.). Extrapolating from LinkLine, one might argue that as long as one can refuse to sell in the first place, there is little that one can do to downstream buyers that could run afoul of the antitrust laws.

The decision is available here.  The decision begins:

The plaintiffs in this case, respondents here, allege that a competitor subjected them to a “price squeeze” in violation of §2 of the Sherman Act. They assert that such a claim can arise when a vertically integrated firm sells inputs at wholesale and also sells finished goods or services at retail. If that firm has power in the wholesale market, it can simultaneously raise the wholesale price of inputs and cut the retail price of the finished good. This will have the effect of “squeezing” the profit margins of any competitors in the retail market. Those firms will have to pay more for the inputs they need; at the same time, they will have to cut their retail prices to match the other firm’s prices. The question before us is whether such a price-squeeze claim may be brought under §2 of the Sherman Act when the defendant is under no antitrust obligation to sell the inputs to the plaintiff in the first place. We hold that no such claim may be brought.

Previous AR coverage is here and here.

Here are the print and the screen versions of my essential facilities presentations at the ABA IP and Antitrust Conference in Berkeley earlier this February.