Federal Trade Commission v. Whole Foods Market, Inc.

KAVANAUGH, Circuit Judge,

dissenting.1

The Federal Trade Commission has sought a preliminary injunction to block the Whole Foods-Wild Oats merger as anticompetitive under § 7 of the Clayton Act. As in many antitrust cases, the analysis comes down to one issue: market definition. Is the relevant product market here all supermarkets? Or is the relevant product market here only so-called “organic supermarkets”? If the former, as Whole Foods argues, the Whole Foods-Wild Oats merger would be lawful because it would not lessen competition in the broad market of all supermarkets: Whole Foods and Wild Oats together operate about 300 of the approximately 34,000 supermarkets in the United States. If the latter, as the FTC contends, the merger may be unlawful: Whole Foods and Wild Oats are the only significant competitors in the alleged organic-store market and their merger would substantially lessen competition in such a narrowly defined market.

More than a year ago, after a lengthy evidentiary hearing and in an exhaustive and careful opinion, the District Court found that the record evidence overwhelmingly supports the following conclusions: WTiole Foods competes against all supermarkets and not just so-called organic stores; the relevant market for evaluating this merger for antitrust purposes is all supermarkets; and the merger of Whole Foods and Wild Oats would not substantially lessen competition in a market that includes all supermarkets. The court therefore denied the FTC’s motion for a preliminary injunction.

Also more than a year ago, a three-judge panel of this Court unanimously denied the FTC’s request for an injunction pending appeal, thereby allowing the WTiole Foods-Wild Oats deal to close. Since then, the merged entity has shut down, sold, or converted numerous Wild Oats stores and otherwise effectuated the merger through many changes in supplier contracts, leases, distribution, and the like.

The Court’s splintered decision in this case seeks to unring the bell. In my judgment, this Court got it right a year ago in refusing to enjoin the merger, and there is no basis for a changed result now. Both a year ago and now, the same central question has been before the Court in determining whether to approve an injunction: whether the FTC demonstrated the necessary “likelihood of success” on its § 7 case. A year ago, the Court said no. Now, the Court says yes. The now-merged entity, the industry, and eonsum-*1052ers no doubt will be confused by this apparent judicial about-face.

The law does not allow the FTC to just snap its fingers and temporarily block a merger. Even at the preliminary injunction stage, the relevant statutory text and precedents expressly require that the FTC show a “likelihood of success on the merits.” FTC v. H.J. Heinz Co., 246 F.3d 708, 714 (D.C.Cir.2001); see also 15 U.S.C. § 53(b) (“likelihood of ultimate success”); cf. Munaf v. Geren, — U.S. -, 128 S.Ct. 2207, 2218-19, 171 L.Ed.2d 1 (2008). Because “[mjerger enforcement, like other areas of antitrust, is directed at market power,” Heinz, 246 F.3d at 713, the FTC therefore needs to make a sufficient showing that the merged company could exercise market power and profitably impose a “small but significant and nontransitory increase in price,” typically meaning a five percent or greater price increase. Horizontal Merger Guidelines § 1.11 (internal quotation marks omitted); see 15 U.S.C. § 18. As the District Court concluded, the FTC did not come close to presenting that kind of evidence in this case; the FTC completely failed to make the economic showing that is Antitrust 101.

By seeking to block a merger without a sufficient showing that so-called organic stores constitute a separate product market and that the merged entity could impose a significant and nontransitory price increase, the FTC’s position — which Judge Brown and Judge Tatel largely accept— calls to mind the bad old days when mergers were viewed with suspicion regardless of their economic benefits. See generally RobeRt H. BoRK, The Antitrust Paradox (1978). I would not turn back the clock. I agree with and would affirm the District Court’s excellent decision denying the FTC’s motion to enjoin the merger of Whole Foods and Wild Oats. See FTC v. Whole Foods Mkt., Inc., 502 F.Supp.2d 1 (D.D.C.2007).

I

A

Section 7 of the Clayton Act prohibits mergers “where in any line of commerce or in any activity affecting commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly.” 15 U.S.C. § 18. The Horizontal Merger Guidelines jointly promulgated by two Executive Branch agencies (the Department of Justice and the FTC) implement that statutory directive and recognize that the key initial step in the analysis is proper product-market definition. See Horizontal Merger Guidelines § 1.11; see also 2B Phillip E. Areeda & Herbert Hoveniíamp, Antitrust Law ¶ 536, at 284-85 (3d ed.2007). Proper produet-market analysis focuses on products’ interchangeability of use or cross-elasticity of demand. A product “market can be seen as the array of producers of substitute products that could control price if united in a hypothetical cartel or as a hypothetical monopoly.” Id. ¶ 530a, at 226. In the merger context, the inquiry therefore comes down to whether the merged entity could profitably impose a “small but significant and nontransitory increase in price” typically defined as five percent or more. See Horizontal Merger Guidelines § 1.11 (internal quotation marks omitted). If the merged entity could profitably impose at least a five percent price increase (because the price increase would not cause a sufficient number of consumers to switch to substitutes outside of the alleged product market), then there is a distinct product market and the proposed merger likely would substantially lessen competition in that market, in violation of § 7 of the Clayton Act.

In considering whether the merged entity could increase prices, courts of course recognize that “future behavior must be *1053inferred from historical observations.” 2B AREEDA & HoVENKAMP, ANTITRUST LAW ¶ 530a, at 226. Therefore, the courts scrutinize existing markets to assess the probable effects of a merger.

This approach was applied sensibly by Judge Hogan in his thorough and leading opinion in FTC v. Staples, 970 F.Supp. 1066 (D.D.C.1997). There, Judge Hogan found that office products sold by an office superstore were functionally interchangeable with office products sold at other types of stores, but he nonetheless found that office-supply superstores constituted a distinct product market. One key fact led Judge Hogan to that conclusion: In areas where Staples was the only office superstore, it was able to set prices significantly higher than in areas where it competed with other office superstores (Office Depot and OfficeMax). See id. at 1075-76. For example, the FTC presented “compelling evidence” that Staples’s prices were 13 percent higher in areas where no office-superstore competitors were present. Id. Judge Hogan ultimately concluded that “[t]his evidence all suggests that office superstore pnces are affected primarily by other office superstores and not by non-superstore competitors.” Id. at 1077 (emphasis added). For that reason, the Court enjoined the merger of Staples and Office Depot.

B

Consistent with the statute, the Executive Branch’s Merger Guidelines, and Judge Hogan’s convincing opinion in Staples, the District Court here carefully analyzed the economics of supermarkets, including so-called organic supermarkets. The court considered whether Whole Foods charged higher prices in areas without Wild Oats than in areas with Wild Oats. After an evidentiary hearing and based on a painstaking review of the evidence in the record, the court concluded that “Whole Foods prices are essentially the same at all of its stores in a region, regardless of whether there is a Wild Oats store nearby.” FTC v. Whole Foods Mkt., Inc., 502 F.Supp.2d 1, 22 (D.D.C.2007). That factual conclusion was supported by substantial evidence offered by Dr. Scheff-man, Whole Foods’s expert, and by the lack of any credible evidence to the contrary.

Dr. Scheffman analyzed Wfiiole Foods’s actual prices across stores and concluded that “there is no evidence that [Whole Foods] and [Wild Oats] price higher” where they face no competition from so-called organic supermarkets compared with where they do face such competition. Scheffman Expert Report ¶ 292, at 113. At a regional level, his studies revealed that only a “very small percentage” of products vary in price within a region, indicating that “prices are set across broad geographic areas.” Id. ¶ 300, at 116. He also analyzed prices at the individual store level, examining how many products sold at a specific store have prices that differ from the most common price in the region. He found that “differences in prices across stores are generally very small (less than one half of one percent) and there is no systematic pattern as to the presence or absence of [organic-supermarket] competition.” Id. ¶ 305, at 116.

Moreover, the record evidence in this case does not show that Whole Foods changed its prices in any significant way in response to exit from an area by Wild Oats. In the four cases where Wild Oats exited and a Whole Foods store remained, there is no evidence in the record that Whole Foods then raised prices. Nor was there any evidence of price increases after Whole Foods took over two Wild Oats stores.

The facts here contrast starkly with Staples, where Staples charged significantly different prices based on the presence or *1054absence of office-superstore competitors in a particular area. The evidence there showed that Staples charged prices 13 percent higher in markets without office-superstore competitors than in markets with such competitors. There is nothing remotely like that in this case.

In the absence of any evidence in the record that Whole Foods was able to (or did) set higher prices when Wild Oats exited or was absent, the District Court correctly concluded that Whole Foods competes in a market composed of all supermarkets, meaning that “all supermarkets” is the relevant product market and that the Whole Foods-Wild Oats merger will not substantially lessen competition in that product market.

In addition to the all-but-dispositive price evidence,2 the District Court identified other factors further demonstrating that the relevant market consists of all supermarkets.

The record shows that Whole Foods makes site selection decisions based on all supermarkets and checks prices against all supermarkets, not only so-called organic supermarkets. As Dr. Scheffman concluded, Whole Foods “price checks a broad set of competitors ... nationally, regionally and locally.” Id. ¶ 224, at 86. This “demonstrates that [Whole Foods] views itself as competing with a broad range of supermarkets and that these supermarkets, in fact, constrain the prices charged by [Whole Foods].” Id. Those other supermarkets include conventional supermarkets such as Safeway, Albertson’s, Weg-man’s, HEB, and Harris Teeter, as well as so-called organic supermarkets like Wild Oats. Id. ¶¶ 225-26, at 86-87. As Professors Areeda and Hovenkamp have explained, a “broad-market finding gains some support from long-standing documents indicating that A or B producers regard the other product as a close competitor.” 2B Areeda & HovenKAMp, Antitrust Law ¶ 562a, at 372. The point here is simple: Whole Foods would not examine the locations of and price cheek conventional grocery stores if it were not a competitor of those stores. Whole Foods does not price check Sports Authority; Whole Foods does price check Safeway.

The record also demonstrates that conventional supermarkets and so-called organic supermarkets are aggressively competing to attract customers from one another. After reviewing a wide variety of industry information and trade journals, Dr. Scheffman concluded that “ ‘[o]ther’ supermarkets are competing vigorously for the purchases made by shoppers at [Whole Foods] and [Wild Oats].” Scheffman Expert Report ¶212, at 77. Whole Foods “recognizes the fact that it has to appeal to a significantly broader group of consumers than organic and natural focused consumers.” Id. ¶ 279, at 108. The record shows that Whole Foods has made progress: Most products that Whole Foods sells are not organic. Conversely, conventional supermarkets have shifted towards “emphasizing fresh, ‘natural’ and organic” products. Id. ¶215, at 80. “[M]ost of the major chains and others are expanding into private label organic and natural products.” Id. ¶ 220, at 85; see also id. ¶ 219, at 83-85 (listing changes in other supermarkets).

So the dividing line between “organic” and conventional supermarkets has blurred. As the District Court aptly put it, the “train has already left the station.” *1055Whole Foods, 502 F.Supp.2d at 48. The convergence undermines the threshold premise of the FTC’s case. This is an industry in transition, and Whole Foods has pioneered a product differentiation that in turn has caused other supermarket chains to update their offerings. These are not separate product markets; this is a market where all supermarkets including so-called organic supermarkets are clawing tooth and nail to differentiate themselves, beat the competition, and make money.

The District Court’s summary of the evidence warrants extensive quotation:

In sum, while all supermarket retailers, including Whole Foods, attempt to differentiate themselves in some way in order to attract customers, they nevertheless compete, and compete vigorously, with each other. The evidence before the Court demonstrates that conventional or more traditional supermarkets today compete for the customers who shop at Wfliole Foods and Wild Oats, particularly the large number of cross-shopping customers— or customers at the margin — with a growing interest in natural and organic foods. Post-merger, all of these competing alternatives will remain. Based upon the evidence presented, the Court concludes that many customers could and would readily shift more of their purchases to any of the increasingly available substitute sources of natural and organic foods. The Court therefore concludes that the FTC has not met its burden to prove that “premium natural and organic supermarkets” is the relevant product market in this case for antitrust purposes.

Id. at 36.3

II

In an attempt to save its merger case despite its inability to meet the test reflected in the Merger Guidelines and applied in Staples, the FTC cites marginally relevant evidence and advances a scatter-shot of flawed arguments.

First, the FTC says that so-called organic supermarkets like Wfliole Foods and Wild Oats constitute their own product market because they are characterized by factors that differentiate them from conventional supermarkets. Those factors include intangible qualities such as customer service and tangible factors such as a focus on perishables.

This argument reflects the key error that permeates the FTC’s approach to this case. Those factors demonstrate only product differentiation, and product differentiation does not mean different product markets. “For antitrust purposes, we apply the differentiated label to products that are distinguishable in the minds of buyers but not so different as to belong in separate markets.” 2B Phillip E. Areeda & HERBERT HoVENKAMP, ANTITRUST LAW ¶ 563a, at 385 (3d ed.2007). As the District Court noted, supermarkets including so-called organic supermarkets differentiate themselves by emphasizing specific benefits or characteristics to attract customers to their stores. See FTC v. Whole Foods Mkt., Inc., 502 F.Supp.2d 1, 24-26 (D.D.C.2007). They may differentiate *1056themselves along dimensions such as “low price, ethnic appeal, prepared foods, health and nutrition, variety within a product category, customer service, or perishables such as meats or produce.” Stanton Expert Report ¶ 23, at 6.

The key to distinguishing product differentiation from separate product markets lies in price information. As Professors Areeda and Hovenkamp have stated, differentiated sellers “generally compete with one another sufficiently” that the prices of one are “greatly constrained” by the prices of others. 2B Aueeda & Hovenkamp, AntitRust Law ¶ 563a, at 384. To distinguish differentiation from separate product markets, courts thus must “ask whether one seller could maximize profit” by charging “more than the competitive price” without “losing too much patronage to other sellers.” Id. ¶ 563a, at 385. Here, in other words, could so-called organic supermarkets maximize profit by charging more than a competitive price without losing too much patronage to conventional supermarkets? Based on the evidence regarding Whole Foods’s pricing practices, the District Court correctly found that the answer to that question is no. So-called organic supermarkets are engaged in product differentiation; they do not constitute a product market separate from all supermarkets.

Second, the FTC points to internal Whole Foods studies and other evidence showing that if a Wild Oats near a Whole Foods were to close, most of the Wild Oats customers would shift to Whole Foods. But that says nothing about whether Whole Foods could impose a five percent or more price increase and still retain those customers (and its other customers), which is the relevant antitrust question. In other words, the fact that many Wild Oats customers would shift to Whole Foods does not mean that those customers would stay with Whole Foods, as opposed to shifting to conventional supermarkets, if Whole Foods significantly raised its prices. And even if one could infer that all of those former Wild Oats customers would so prefer Whole Foods that they would shop there even in the face of significant price increases, that would not show whether Whole Foods could raise prices without driving out a sufficient number of other customers as to make the price increases unprofitable. In sum, this argument is a diversion from the economic analysis that must be conducted in antitrust cases like this. The District Court properly found that the expert evidence in the record leads to the conclusion that Whole Foods could not profitably impose such a significant price increase.4

Third, the FTC cites comments by Whole Foods CEO John Mackey as en-*1057dence that Whole Foods perceived Wild Oats to be a unique competitor. Even if Mackey’s comments were directed only to Wild Oats, that would not be evidence that Whole Foods and Wild Oats are in their own product market separate from all other supermarkets. It just as readily suggests that Whole Foods and Wild Oats are two supermarkets that have similarly differentiated themselves from the rest of the market, such that Mackey would be especially pleased to see that competitor vanish. Beating the competition from similarly differentiated competitors in a product market is ordinarily an entirely permissible competitive goal. Saying as much, as Mackey did here, does not mean that the similarly differentiated competitor is the only relevant competition in the marketplace. Moreover, Mackey nowhere says that the merger would allow Whole Foods to significantly raise prices, which of course is the issue here. In any event, intent is not an element of a § 7 claim, and a CEO’s, bravado with regard to one rival cannot alter the laws of economics: Mere boasts cannot vanquish real-world competition — here, from Safeway, Albertson’s, and the like. As Judge Easterbrook has explained, “Firms need not like their competitors; they need not cheer them on to success; a desire to extinguish one’s rivals is entirely consistent with, often is the motive behind, competition.” A.A. Poultry Farms, Inc. v. Rose Acre Farms, Inc., 881 F.2d 1396, 1402 (7th Cir.1989). And “[i]f courts use the vigorous, nasty pursuit of sales as evidence of a forbidden ‘intent’, they run the risk of penalizing the motive forces of competition.” Id. “Intent does not help to separate competition from attempted monopolization....” Id.

Fourth, the FTC says that a study by its expert, Dr. Murphy, demonstrates that Whole Foods’s profit margins decreased in geographic areas where it competed against Wild Oats. But the relevant inquiry under the Merger Guidelines is prices. And Dr. Murphy did not determine whether Whole Foods prices ever differed as a result of competition from Wild Oats.

Moreover, there was only a slight difference between Whole Foods margins when Wild Oats was in the same area and when it was not. The overall difference was 0.7 percent, which Dr. Murphy himself recognized was not statistically significant. The FTC’s evidence on margins is wafer-thin and does not suffice to show that organic stores constitute their own product market.

Fifth, the FTC points to evidence that Whole Foods’s entry into a particular area, unlike the entry of conventional supermarkets, caused Wild Oats to lower its prices. Dr. Murphy’s reliance on Wild Oats’s reaction to Whole Foods’s entry is questionable. Dr. Murphy based his entire analysis on a meager two events, hardly a large sample size. In addition, Dr. Murphy’s analysis did not control for the reaction of conventional supermarkets to Whole Foods’s entry. In other words, he assumed that the relevant product market was so-called organic supermarkets (the point he was trying to prove) and therefore assumed that all changes in Wild Oats’s prices were directly caused by Whole Foods’s entry. But if conventional supermarkets also lowered prices to compete with Whole Foods when Whole Foods entered, Wild Oats’s price decreases may well have been due to the overall reduction in prices by all supermarkets in the area. If that were true, the relevant product market would obviously be all supermarkets, not just so-called organic supermarkets. Dr. Murphy’s analysis never confronted that possibility or the complexity of how competition works in this market; his analysis appears to have assumed the conclusion and reasoned backwards from there.

*1058Moreover, the fact that Whole Foods and Wild Oats went toe-to-toe on occasion does not mean that they did not also go toe-to-toe with conventional supermarkets, which is the key question. And it is revealing that despite having access to the necessary data for six such events, Dr. Murphy did not analyze the effect of a Wild Oats exit on Whole Foods’s prices. As Dr. Scheffman wrote: “A number of [Wild Oats] stores have closed.... [Dr. Murphy] has done no analysis to assess the effects of those store exits in the local shopping areas.... This is a curious omission, since such evidence, if reliable and reliably analyzed, would be relevant to the issue of what happens in local market areas in which a [Wild Oats] store closes.” Scheffman Rebuttal ¶ 63, at 21.

The bottom line is that, as the District Court found, there is no evidence in the record suggesting that Whole Foods priced differently based on the presence or absence of a Wild Oats store in the area. That is a conspicuous' — and all but disposi-tive — omission in Dr. Murphy’s analysis and in the FTC’s case.

Sixth, the FTC cites the openings of three Earth Fare stores near Whole Foods stores in North Carolina, which caused decreases in Whole Foods’s prices in those areas. But soon after those entries, Whole Foods’s prices returned to normal levels. So the record hardly shows the sort of “nontransitory” price changes that are the touchstone of product-market definition. See Merger Guidelines § 1.11. A price increase ordinarily must last “for the foreseeable future,” id., considered by some to be more than a year, to qualify as “non-transitory” See 2B Areeda & Hovenkamp, AntitRust Law ¶ 537a, at 290. Moreover, the entry of a Safeway store in Boulder, Colorado, had a similar short-term impact on Whole Foods, indicating that whatever inference should be drawn from the Earth Fare entries cannot be limited to so-called organic supermarkets but rather applies to conventional supermarkets.

The FTC’s reference to Earth Fare mistakenly focuses on a few isolated trees instead of the very large forest indicating a competitive market consisting of all supermarkets. In short, I fail to see how Whole Foods’s temporary price changes to compete against three Earth Fare stores in North Carolina could possibly be a hook to block this nationwide merger of Whole Foods and Wild Oats.

Ill

A

The opinions of Judge Brown and Judge Tatel rest on two legal points with which I respectfully but strongly disagree.

First, the Court’s decision resuscitates the loose antitrust standards of Brown Shoe Co. v. United States, 370 U.S. 294, 82 S.Ct. 1502, 8 L.Ed.2d 510 (1962), the 1960s-era relic. See, e.g., Brown Op. at 1039 (“We look to the Brown Shoe indi-cia... .”); Tatel Op. at 1046 (“Brown Shoe lists ‘distinct prices’ as only one of a non-exhaustive list of seven ‘practical indicia’ that may be examined to determine whether a separate market exists.”) (citation omitted). This is a problem because Brown Shoe’s brand of free-wheeling antitrust analysis has not stood the test of time. See, e.g., EineR Elhauge & DaMIEn GeRADIn, Global Antitrust Law and ECONOMICS 874 (2007) (“Modern practice takes a much more rigorous approach to market definition [than Brown Shoe ]”); 4 Phillip E. Areeda & Herbert Hoveneamp, Antitrust Law ¶ 913a, at 62 (2d ed. 2006) (“One alternative that we do not recommend is a return to Brown Shoe’s language of ‘sub-markets’ ”).

As demonstrated in this Court’s most recent merger case, the practical indicia test of Brown Shoe no longer guides courts’ merger analyses because it does *1059not sufficiently account for the basic economic principles that, according to the Supreme Court, must be considered under modern antitrust doctrine. See FTC v. H.J. Heinz Co., 246 F.3d 708, 715-16 (D.C.Cir.2001) (not applying Brown Shoe practical indicia test; instead using the economically grounded Herfindahl-Hirsch-man Index test for market definition employed in FTC v. Staples, Inc., 970 F.Supp. 1066 (D.D.C.1997)); cf. Leegin Creative Leather Prods, v. PSKS, Inc., — U.S. -, 127 S.Ct. 2705, 2718, 168 L.Ed.2d 628 (2007) (“the antitrust laws are designed primarily to protect interbrand competition”); State Oil Co. v. Khan, 522 U.S. 3, 14, 118 S.Ct. 275, 139 L.Ed.2d 199 (1997) (“Our analysis is also guided by our general view that the primary purpose of the antitrust laws is to protect interbrand competition.”); Hosp. Corp. of Am. v. FTC, 807 F.2d 1381, 1386 (7th Cir.1986) (Posner, J.) (noting the “most important developments that cast doubt on the continued vitality of such cases as Brown Shoe”). Judge Bork forcefully catalogued the flaws in the Brown Shoe approach 30 years ago in his landmark antitrust book; indeed, his cogent critique helped usher Brown Shoe and several other cases to the jurisprudential sidelines. See RobeRT H. BoRK, The Antitrust Paradox 210, 216 (1978) (“It would be overhasty to say that the Brown Shoe opinion is the worst antitrust essay ever written.... Still, all things considered, Brown Shoe has considerable claim to the title.... Brown Shoe was a disaster for rational, consumer-oriented merger policy.”); George L. Priest, The Abiding Influence of The Antitrust Paradox, 31 Harv. J.L. & Pub. Pol’y 455, 459 (2008) (praising Judge Bork’s criticism of the “now notorious, though then mainstream” Brown Shoe opinion).

The Court’s revival of the loose Brown Shoe standard threatens to reverse this trend and to upend modern merger practice.5

Second, the opinions of Judge Brown and Judge Tatel both dilute the standard for preliminary injunction relief in antitrust merger cases, such that the FTC apparently need not establish a “likelihood of success on the merits.” Heinz, 246 F.3d at 714. In particular, Judge Brown and Judge Tatel rely heavily on their belief that: “In this circuit, the standard for likelihood of success on the merits is met if the FTC has raised questions going to the merits so serious, substantial, difficult and doubtful as to make them fair ground for thorough investigation, study, deliberation and determination by the FTC in the first instance and ultimately by the Court of Appeals.” Tatel Op. at 1042 (internal quotations and citations omitted); see also id. at 1042; Brown Op. at 1035 (indicating that “the FTC will usually be able to obtain a preliminary injunction blocking a merger” by satisfying the same test).

In applying this watered-down test for issuing a preliminary injunction in FTC merger cases, Judge Brown and Judge Tatel rely on language contained in our opinion in Heinz. However, Heinz only assumed this particular gloss on the “likelihood of success on the merits” requirement for preliminary injunctions based on *1060a concession in the case. See Heinz, 246 F.3d at 715 (D.C.Cir.2001) (“This specific standard was articulated by the court below, and it is a standard to which the appellees have not objected.”) (citation omitted). Heinz did not hold that this gloss was the proper meaning of 15 U.S.C. § 53(b) in FTC preliminary injunction merger cases.6

This “serious questions” standard is inconsistent with the relevant statutory text. The statute unambiguously requires that courts consider “the Commission’s likelihood of ultimate success” when the FTC seeks to preliminarily enjoin a merger. 15 U.S.C. § 53(b).7

There is a significant difference, moreover, between the relaxed “serious questions” standard applied by Judge Brown and Judge Tatel and the traditional likelihood of success standard — as the Supreme Court explained just a few months ago in Munaf v. Geren, — U.S.-, 128 S.Ct. 2207, 171 L.Ed.2d 1 (2008), rev’g sub nom. Omar v. Harvey, 479 F.3d 1 (D.C.Cir.2007). To be sure, that case did not involve a merger; but the Supreme Court there did address the general likelihood-of-success preliminary injunction standard— the same standard that is expressly articulated in 15 U.S.C. § 53(b). The District Court in the Omar litigation — like Judge *1061Brown and Judge Tatel here — had concluded that a preliminary injunction was justified because the ease presented questions “so serious, substantial, difficult and doubtful, as to make them fair ground for litigation and thus for more deliberative investigation.” Omar v. Harvey, 416 F.Supp.2d 19, 23-24 (D.D.C.2006) (citation omitted). This Court then affirmed the District Court’s preliminary injunction. See Omar v. Harvey, 479 F.3d 1, 11 (D.C.Cir.2007) (concluding that the Court “need not address” the merits of petitioner’s claims).

But the Supreme Court unanimously rejected that lesser “serious questions” standard as too weak and not equivalent to the “likelihood of success” necessary for a preliminary injunction to issue. See Munaf, 128 S.Ct. at 2219 (“We begin with the basics.... [A] party seeking a preliminary injunction must demonstrate, among other things, ‘a likelihood of success on the merits.’ ”) (citations omitted); see also Winter v. NRDC, — U.S. -, 129 S.Ct. 365, 374, 172 L.Ed.2d 249 (2008) (“A plaintiff seeking a preliminary injunction must establish that he is likely to succeed on the merits”) (citing Munaf, 128 S.Ct. at 2218-19). And the Supreme Court directly criticized the approach of the District Court and this Court in the Omar litigation: “one searches the opinions below in vain for any mention of a likelihood of success as to the merits.” Munaf 128 S.Ct. at 2219.

The Court in this case repeats the same mistake made in Omar of watering down the preliminary injunction standard. Both Judge Brown and Judge Tatel approve the FTC’s request for preliminary injunction without making the essential “likelihood of success” finding that is required by the statutory text and Supreme Court precedent. See Brown Op. at 1035, 1041; Tatel Op. at 1041-42, 1042-43. To the extent the “serious questions” standard they apply was ever appropriate for preliminary injunction merger cases, the combination of the clear statutory text in 15 U.S.C. § 53(b) and the Supreme Court decision in Munaf convincingly demonstrates that it is not the proper standard now.

In short, the approach of Judge Brown and Judge Tatel revives the moribund Brawn Shoe practical indicia test and applies an overly lax preliminary injunction standard for merger cases. I respectfully disagree on both counts. In my judgment, the FTC may obtain a preliminary injunction only by establishing a likelihood of success — namely, a likelihood that, among other things, the merged entity would possess market power and could profitably impose a significant and nontransitory price increase.8

*1062B

In reaching her conclusion, Judge Brown also relies on a distinction between marginal consumers and core consumers. See, e.g., Brown Op. at 1041 (“In sum, the district court believed the antitrust laws are addressed only to marginal consumers. This was an error of law, because in some situations core consumers, demanding exclusively a particular product or package of products, distinguish a submarket.”). But the FTC never once referred to, much less relied on, the distinction between marginal and core consumers in 86 pages of briefing or at oral argument. The terms “marginal consumer” and “core consumer” are nowhere to be found in its briefs.

In any event, I respectfully disagree with Judge Brown’s emphasis on core customers. For a business to exert market power as a result of a merger, it must be able to increase prices (usually by five percent or more) while retaining enough customers to make that price increase profitable. See 2B Phillip E. Areeda & HeRbert Hovenxamp, Antitrust Law ¶ 501, at 109 (3d ed. 2007) (“A defendant firm has market power if it can raise price without a total loss of sales.”). If too many “marginal” customers are turned off by a price hike, then the hike will be unprofitable even if a large group of die-hard “core” customers remain active clients. Therefore, a focus on core customers alone cannot resolve a merger case. The question here is whether Whole Foods could increase prices by five percent or more without losing so many marginal customers as to make the price increase unprofitable. See id. ¶ 536, at 284. As discussed above, the FTC has not come close to making that showing. Moreover, there is no support in the law for that singular focus on the core customer. Indeed, if that approach took root, it would have serious repercussions because virtually every merger involves some core customers who would stick with the company regardless of a significant price increase. So under this “core customer” approach, many heretofore permissible mergers presumably could be blocked as anticompetitive. That cannot be the law, and it is not the law.

In a related vein, Judge Brown repeatedly suggests that Whole Foods and Wild Oats engage in “price discrimination” — more specifically, Judge Brown asserts that organic supermarkets “discriminate on price between their core and marginal customers, thus treating the former as a distinct market.” Brown Op. at 1041. But this assertion has no factual support in the record. For antitrust purposes, price discrimination normally involves one seller charging different prices to different customers for the same prod*1063uct. See 2b Phillip E. AREeda & HERBERT Hovenkamp, Antitrust Law ¶ 517a (noting as an indication of market power “systematic price discrimination, as when a seller can identify two (or more) groups of customers with different demands and charge each group different prices even though its cost of serving each group is the same”). If there is price discrimination in an industry, then under certain circumstances a relevant market may be defined to include only those customers who pay the higher price. See Horizontal Merger Guidelines § 1.12. In this case, however, neither Judge Brown nor the FTC has pointed to any evidence suggesting either that price discrimination occurred before this merger or that the merged entity will be able to price-discriminate. In other words, there is no reason to think that “core” as opposed to non-core customers ever pay higher prices for the same products in organic supermarkets.

IV

In the end, the FTC’s case is weak and seems a relic of a bygone era when antitrust law was divorced from basic economic principles. The record does not show that Whole Foods priced differently based on the presence or absence of Wild Oats in the same area. The reason for that and the conclusion that follows from that are the same: Whole Foods competes in an extraordinarily competitive market that includes all supermarkets, not just so-called organic supermarkets. The merged entity thus could not exercise market power such that it could profitably impose a significant and nontransitory price increase. Therefore, there is no sound legal basis to block this merger.

The issues presented in this case are important to antitrust regulators and practitioners, to potentially merging companies, and ultimately to the overall economy. The splintered panel opinions will create enormous uncertainty, debate, and litigation over the meaning and effect of this decision. And to the extent common principles and holdings are derived from the opinions of Judge Brown and Judge Tatel, those principles will authorize the FTC to obtain preliminary injunctions and block mergers based on a watered-down preliminary injunction standard and without sufficient regard for the economic principles that have undergirded modern antitrust law. That will give the FTC far greater power to block mergers than the statutory text or Supreme Court precedents permit.

I respectfully dissent.

. In light of changes made by Judge Brown and Judge Tatel to their opinions in response to the petition for rehearing — most notably, the fact that Judge Tatel no longer joins Judge Brown's opinion, meaning there is no majority opinion for the Court — this dissent contains changes throughout, including a new Part III, from the dissenting opinion released on July 29, 2008.

. Judge Tatel's opinion disparages the evidence about Whole Foods's prices, calling it "all-but-meaningless” and implicitly suggesting that Whole Foods manipulated its prices just for the expert study. Tatel Op. at 1047. But Judge Tatel offers no evidence for that suggestion.

. A showing that the merged entity would possess market concentration in a defined product market is necessary but not sufficient to establish an antitrust violation. See United States v. Baker Hughes Inc., 908 F.2d 981, 985 (D.C.Cir. 1990) (listing factors that might militate against finding an antitrust violation, even assuming market concentration exists). I need not address the other necessary components of the FTC’s case, however, because the FTC has not satisfied the threshold requirement of showing that the merged entity would have such market concentration.

. According to Judge Tatel’s opinion, the FTC’s expert purported to say that Whole Foods could impose a five percent or greater price increase because of the number of Wild Oats customers who would switch to Whole Foods rather than conventional supermarkets. Tatel Op. at 1044 (citing Rebuttal Expert Report of Kevin M. Murphy ¶ 32 (July 13, 2007)). But that ambiguous statement constituted a single, unexplained sentence in the middle of a lengthy report. Moreover, the expert apparently based his conclusion entirely on the so-called “Project Goldmine” analysis of diversion ratios associated with store closures — that is, of the number of Wild Oats customers who would switch to Whole Foods in the event that a Wild Oats store closes and Whole Foods prices remain constant. As the expert himself appeared to acknowledge, see Murphy Report ¶ 32 (noting that "marginal and average diversion ratios could be different”), the data do not necessarily shed any light on how many customers would continue to shop at a merged Wild Oats-and-Whole Foods entity in the event that the entity uniformly increased prices. All of this no doubt explains why the FTC never even mentioned this aspect of its expert’s report in the argument section of its opening brief.

. As two antitrust commentators perceptively stated: “The basic problem with the FTC's position in Whole Foods was that it lacked the pricing evidence it had in Staples, which showed that customers did not go elsewhere if the office superstores increased their prices. Whole Foods is an attempt by the FTC to persuade a court that if you take a CEO’s statements about a merger and stir it in with evidence showing the existence of several 'practical indicia’ from Brown Shoe, the resulting mixture should trump objective evidence about how customers would react in the event of a price increase.” Carlton Var-ner & Fleather Cooper, Product Markets in Merger Cases: The Whole Foods Decision (Oct.2007), www.antitrustsource.com.

. The gloss on § 53(b) appears to have arisen originally in other circuits around the middle of the 20th century in connection with a more general view that a lighter "likelihood of success” standard is appropriate whenever the balance of equities weighs strongly in favor of issuing an injunction. Compare FTC v. Beatrice Foods Co., 587 F.2d 1225, 1229 (D.C.Cir. 1978) (Appendix to Statement of MacKinnon & Robb, JJ.) (citing Hamilton Watch Co. v. Benrus Watch Co., 206 F.2d 738, 740 (2d Cir.1953) (which noted in the FTC merger context that "if the other elements are present (i.e., the balance of hardships tips decidedly toward plaintiff), it will ordinarily be enough that the plaintiff has raised questions going to the merits so serious .... ”)), with Omar v. Harvey, 416 F.Supp.2d 19, 28 (D.D.C.2006) (citing Washington Metro. Area Transit Comm'n v. Holiday Tours, 559 F.2d 841, 842-44 & n. 1 (D.C.Cir.1977) (which noted outside the FTC merger context that courts may generally apply the relatively lax "serious questions” approach only "when confronted with a case in which the other three [preliminary injunction] factors strongly favor interim relief')). But as explained below in footnote 7, Congress in 1973 codified a preliminary injunction standard for FTC merger cases that specifically directs courts to consider the Commission's "likelihood of ultimate success.” 15 U.S.C. § 53(b). And as explained in the text, the Supreme Court recently repudiated the "serious questions” approach to preliminary injunctions in general by requiring a likelihood of success showing in all cases, regardless of whether the balance of equities weighs in favor of the injunction. See Munaf v. Geren,-U.S.-, 128 S.Ct. 2207, 2219, 171 L.Ed.2d 1 (2008).

. In justifying his adoption of the “serious questions" test for likelihood of success, Judge Tatel highlights the "unique ‘public interest’ standard in 15 U.S.C. § 53(b).” Tatel Op. at 1043 (citing FTC v. Exxon Corp., 636 F.2d 1336, 1343 (D.C.Cir.1980)); see also id. at 1043. But the statute explicitly preserves the traditional likelihood of success requirement. See § 53(b) ("Commission's likelihood of ultimate success”). What makes § 53’s standard for preliminary injunctions “unique,” as we have explained, is that the FTC need not show irreparable harm and, secondarily, that private equities are subordinated to public equities. See FTC v. Weyerhaeuser Co., 665 F.2d 1072, 1081-83 (D.C.Cir.1981) ("The case law Congress codified removes irreparable damage as an essential element of the preliminary injunction proponent's case and permits the judge to presume from a likelihood of success showing that the public interest will be served by interim relief.”); see also Heinz, 246 F.3d at 727 n. 25; Exxon Corp., 636 F.2d at 1343. Far from reading the "likelihood of ultimate success” language out of the statute, we have recognized that the statutory phrase "weighing the equities and considering the likelihood of ultimate success” was specifically added by the Conference Committee and that this "deliberate addition” should not "be brushed aside as essentially repetitive or meaningless.” Weyerhaeuser, 665 F.2d at 1081.

. The precedential effect of today’s splintered decision is muddied somewhat by the fact that Judge Brown and Judge Tatel have issued individual opinions concurring in the judgment. That said, it is of course well-settled that the mere fact that there is no majority opinion does not mean that the decision constitutes no precedent for future cases. This happens quite frequently with splintered Supreme Court decisions where there is no majority opinion. As the Supreme Court has repeatedly explained, in the vast majority of cases without a majority opinion there is still a binding holding of the Court — even if it can occasionally be difficult to determine. This is known as the Marks principle. See Marks v. United States, 430 U.S. 188, 193, 97 S.Ct. 990, 51 L.Ed.2d 260 (1977); King v. Palmer, 950 F.2d 771, 783 (D.C.Cir.1991) (en banc) ("implicit agreement” between judges can produce a "controlling” principle of law); see generally Planned Parenthood of Southeastern Pennsylvania v. Casey, 947 F.2d 682, 691-97 (3d Cir.1991). Like the Supreme Court, this Court has routinely recognized that a decision without a majority opinion usually still constitutes a binding precedent. See, e.g., In re Navy Chaplaincy, 534 F.3d 756, 759 n. 2 (D.C.Cir.2008) (construing Hein v. Freedom From Religion Foundation,-U.S.-, 127 S.Ct. 2553, 168 L.Ed.2d 424 (2007)); Shur-*1062berg Broadcasting of Hartford, Inc. v. FCC, 876 F.2d 902, 910 (D.C.Cir.1989) ("a lower federal court must do its level best to extract the holding that commanded a majority in each case to arrive at the governing principles and limitations”), rev’d on other grounds sub nom. Metro Broad., Inc. v. FCC, 497 U.S. 547, 110 S.Ct. 2997, 111 L.Ed.2d 445 (1990); Martin v. Malhoyt, 830 F.2d 237, 247 n. 28 (D.C.Cir.1987) (citing Marks and noting that Justice Black's concurrence in Barr v. Matteo, 360 U.S. 564, 79 S.Ct. 1335, 3 L.Ed.2d 1434 (1959), "provides the 'narrowest grounds' for the Court’s disposition of the case and thus constitutes the Court's holding”). Only in very rare cases do the opinions making up a majority of a court contain no common principles or common ground on which to derive any precedential holding of the court. See Nichols v. United States, 511 U.S. 738, 743-46, 114 S.Ct. 1921, 128 L.Ed.2d 745 (1994) (construing Baldosar v. Illinois, 446 U.S. 222, 100 S.Ct. 1585, 64 L.Ed.2d 169 (1980)); King, 950 F.2d at 782-85.

It is unclear whether district courts and future courts of appeals will construe this case as one of those rare situations that falls entirely outside the Maries rule. At a minimum, this confused decision will invite years of uncertainty and litigation over what the holding of this case is — a separate but important problem with the Court’s approach.