McWane, Inc. v. Federal Trade Commission

783 F.3d 814, 2015 U.S. App. LEXIS 6111, 2015 Trade Cas. (CCH) 79,134, 2015 WL 1652200

Opinion

MARCUS, Circuit Judge:

This antitrust case involves allegedly anticompetitive conduct in the ductile iron pipe fittings (“DIPF”) market by McWane, Inc., a family-run company headquartered in Birmingham, Alabama. In 2009, following the passage of federal legislation that provided a large infusion of money for waterworks projects that required domestic pipe fittings, Star Pipe Products entered the domestic fittings market. In response, McWane, the dominant producer of domestic pipe fittings, announced to its distributors that (with limited exceptions) unless they bought all of their domestic fittings from McWane, they would lose their rebates and be- cut off from purchases for 12 weeks. The Federal Trade Commission (“FTC”) investigated and brought an enforcement action under Section 5 of the Federal Trade Commission Act, 15 U.S.C. § 45. The Administrative Law Judge (“ALJ”), after a two-month trial, and then a divided Commission, found that McWane’s actions constituted an illegal exclusive dealing policy used to maintain McWane’s monopoly power in the domestic fittings market. The Commission issued an order directing McWane to stop requiring exclusivity from distributors. McWane appealed, challenging nearly every aspect of the Commission’s ruling.

After thorough review, we affirm the Commission’s order. The Commission’s factual and economic conclusions — identifying the relevant product market for domestic fittings produced for domestic-only projects, finding that McWane had monopoly power in that market, and determining that McWane’s exclusivity program harmed competition — are supported by substantial evidence in the record, as required by our deferential standard of review, and their legal conclusions are supported by the governing law.

I.

A.

. The essential facts developed in this extensive record are these. Pipe fittings join together pipes and help direct the flow of pressurized water in pipeline systems. They are sold primarily to municipal water authorities and their contractors. Although there are several thousand unique configurations of fittings (different shapes, sizes, coatings, etc.), approximately 80% of the demand is for about 100 commonly used fittings.

Fittings are commodity products produced to American Water Works Association (“AWWA”) standards, and any fitting that meets AWWA specifications is interchangeable, regardless of the country of origin. Ductile iron pipe fittings manufacturers rarely sell fittings directly to end users; instead, they sell them to middleman distributors, who in turn sell them to end users. An end user (e.g., a municipal water authority) will issue a “specification” for its project, detailing the pipes, fittings, and other products required. Competing contractors solicit bids for the specified products from distributors, who in turn seek quotes from various manufacturers like McWane.

End users issue either “open specifications,” permitting the use of fittings manufactured anywhere in the world, or “domestic specifications,” requiring the use of fittings made in the United States. An end user might issue a domestic specification either because of its preference or due to' legal procurement requirements: certain municipal, state, and federal laws require waterworks projects to use domestic-only fittings. 1 Domestic fittings sold for use in projects with domestic-only specifications command higher prices than imported fittings or domestic fittings sold for use in projects with open specifications. The majority of specifications are open, and the majority of fittings sold (approximately 80-85%) are imported.

Historicálly, fittings were made by a number of American companies, most of which offered a full line of domestic fittings. However, beginning in the 1980s, importing fitting suppliers — including Star Pipe Products and Sigma Corporation — • began to make significant inroads into the market. By 2005, imported fittings made up the vast majority of ductile iron pipe fittings sales, and the competition from lower-priced and lower-cost imports drove most domestic manufacturers out of . the market.

Today, the overall market for fittings sold in the United States — whether manufactured domestically or abroad, sold into both open-specification and domestic-only projects — is an oligopoly with three major suppliers: McWane, Star, and Sigma. Together they account for approximately 90% of the fittings sold in the United States. There are two national distributors, [¶] Supply and Ferguson, which together account for approximately 60% of the overall waterworks distribution market.

From April 2006 until Star entered the domestic fittings market in late 2009, McWane was the only supplier of domestic fittings. Until 2008, McWane produced fittings at two domestic foundries, one in Anniston, Alabama, (“Union Foundry”) and the other in Tyler, Texas. In 2005, McWane opened a foundry to produce fittings in China, and in 2008 it closed its Texas foundry.

In 2009, looking to take advantage of the increased demand for domestic fittings prompted by ARRA, Star decided to enter the market for domestic DIPFs. In June 2009, Star publicly announced at an industry conference and in a letter to customers that it would offer domestic fittings' starting in September 2009. Star became a “virtual manufacturer” of domestic fittings, contracting with six third-party foundries in the U.S. to produce fittings to Star’s specifications. Star also investigated acquiring its own U.S. foundry, which the Commission found would have been a decidedly less costly and more efficient way to produce domestic fittings.

In response to Star’s forthcoming entry into the domestic DIPF market, McWane implemented its “Full Support Program” in order “[t]o protect [its] domestic brands and market position.” This program was announced in a September 22, 2009 letter to distributors. McWane informed customers that if they did not “fully support McWane branded products for their domestic fitting and accessory requirements;” they “may forgo participation in any unpaid rebates [they had accrued] for domestic fittings and accessories or shipment of their domestic fitting and accessory orders of [McWane] products for up to 12 weeks.” In other words, distributors who bought domestic fittings from other companies (such as Star) might lose their rebates or be cut off from purchasing McWane’s domestic fittings for up to three months. 2 The Full Support Program did contain two exceptions permitting the purchase of another company’s domestic fittings: where McWane products were not readily available, and where the customer bought domestic fittings and accessories along with another manufacturer’s ductile iron pipe.

Internal documents reveal that McWane’s express purpose was to raise Star’s costs and impede it from becoming a viable competitor. McWane executive Richard Tatman wrote, ‘We need to make sure that they [Star] don’t reach any critical market mass that will allow them to continue to invest and receive a profitable return.” In another document, he “observed that ‘any competitor’ seeking to enter the domestic fittings market could face,‘significant blocking issues’ if they are not a ‘full line’ domestic supplier.” McWane I, 155 F.T.C. at 1134. In yet another, McWane employees described the nascent Full Support Program as a strategy to “[floree [distribution to [p]ick their [h]orse,” which would “[floree[ ] Star[ ] to absorb the costs associated with having a more full line before they can secure major distribution.” Mr. Tatman was concerned about the “[e]rosion of domestic pricing if Star emerges as a legitimate competitor,” and another McWane executive wrote that his “chief concern is that the domestic market [might] get[ ] creamed from a pricing standpoint” should Star become a “domestic supplier.”

Initially, the Full Support Program was enforced as threatened. Thus, for example, when the Tulsa,. Oklahoma branch of distributor Hajoca Corporation purchased Star domestic fittings, McWane cut off sales of its domestic fittings to all Hajoca branches and withheld its rebates. 3 Other distributors testified to abiding by the Full Support Program in order to avoid the devastating result of being cut off from all McWane domestic fittings. For example, following the announcement of the Full Support Program, the country’s two largest waterworks distributors, [¶] Supply (with approximately a 28-35% share of the distribution market) and Ferguson (with approximately 25%), prohibited their branches from purchasing domestic fittings from Star unless the purchases fell into one of the Full Support Program exceptions, and even canceled pending orders for domestic fittings that they had placed with Star. Indeed, the Commission found that “Star was rebuffed by some distributors even after offering a more generous rebate than McWane.” However, some distributors also identified other factors that contributed to their decision not to purchase from Star, including “concerns about Star’s inventory, the quality of fittings produced at several different foundries, ... the timeliness of delivery,” and negative past business dealings with Star.

Despite McWane’s Full Support Program, Star entered the domestic fittings market and made sales to various distributors. From 2006 until Star’s entry in 2009, McWane was the only manufacturer of domestic fittings, with 100% of the market for domestic-only projects. By 2010, Star had gained approximately 5% of the domestic fittings market, while McWane captured the remaining 95%. Star grew to just under 10% market share in 2011, leaving the remaining 90% for McWane, and Star was “on pace, at the time of trial, to have its best year ever for [djomestic [flit-tings sales in 2012.” The Commission noted that “many distributors made purchases under the exceptions allowed by the Full Support Program,” but that Star’s sales in total “were small compared to the overall size of the market.” Star estimated that if the Full Support Program had not been in place, its sales would have been greater by a multiple of 2.5 in 2010 and by a multiple of three in 2011.

Star never ended up building or buying a domestic foundry of its own. The Commission found that this was because Star “believed its sales level was insufficient to justify running its own foundry.” Star estimated that the cost of producing fittings at its own domestic foundry would have been significantly lower than the cost of contracting with independent foundries, and that operating its own foundry would have allowed it to appreciably reduce its domestic fittings prices. (This is because the third-party foundries used less specialized and less efficient equipment, had increased logistical costs and higher labor costs, and charged a markup plus a fee for shipping.) The Commission and the ALJ also found that the Full Support Program was a “significant reason” that another distributor, Serampore Industries Private, decided not to enter the domestic fittings market.

During 2009-2010, following Star’s entry into the market and the Full Support Program’s implementation, McWane’s production costs for domestic fittings remained flat,' but it raised its prices for domestic fittings and increased its gross profits. These prices were relatively consistent across all states, regardless of whether Star had entered the domestic fittings market as a rival; Star’s presence in various states did not result in lower prices. McWane “continued to sell its domestic fittings into domestic-only specifications at prices that earnéd significantly higher gross profits than for non-domestic fittings, which faced greater competition.” McWane, Inc. (McWane II), 2014-1 Trade Cas. (CCH) ¶ 78670, 2014 WL 556261, at *17 (F.T.C. Jan. 30, 2014). Star’s average prices, however, were higher than McWane’s in several states.

The duration of the Full Support Program is a matter of some dispute. McWane contends that it ended the Full Support Program in early 2010, eliminating the provision that customers might forego shipments for up to 12 weeks. But the Commission found that McWane had never “publicly withdrawn the policy or notified distributors of any changes,” and that some distributors believed that the policy was “still in effect.” There is also evidence that some distributors started to ignore the Full Support Program in 2010 after they learned of the FTC’s investigation into McWane’s practices.

B.

On January 4, 2012, the FTC issued a seven-count administrative complaint charging McWane, Star, and Sigma 4 with violating Section 5 of the Federal Trade Commission Act. (In February and May of 2012, Star and Sigma entered consent decrees with the FTC without any admission of wrongdoing, leaving McWane as the sole defendant.) The only charge at issue on appeal is found in count six, 5 which alleged that McWane’s exclusivity mandate (the Full Support Program) constituted unlawful maintenance of a monopoly over the domestic fittings market.

The ALJ conducted a two-month trial. On May 8, 2013, he issued a 464-page decision ruling in favor of the complaint counsel on count 6. 6 He specifically found that the sales for projects requiring domestic fittings constituted a separate product market in which McWane had monopoly power. McWane I, 155 F.T.C. at 1239-40, 1375-88. He ruled that McWane’s Full Support Program was an exclusive dealing arrangement that foreclosed Star from a substantial share of the domestic fittings market and, thereby, unlawfully maintained McWane’s monopoly. Both McWane and the complaint counsel appealed the ALJ’s decision to the Commission.

A divided Commission affirmed as to count 6. 7 Like the ALJ, the Commission found that the relevant market was the supply of domestically manufactured fittings for use in domestic-only waterworks projects, because imported fittings are not a substitute for domestic fittings for such projects. McWane II, 2014 WL 556261, at *13. The Commission noted that this conclusion was bolstered by the higher prices charged for domestic fittings used in domestic-only projects. Id. at *14. The Commission also found that McWane had monopoly power in that market, with 90-95% market share from 2010-11 (a much higher share than courts usually require for a prima facie showing of monopoly power) and substantial barriers to entry in the form of major capital outlays required to produce domestic fittings. Id. at *15-18.

The Commission agreed that McWane’s Full Support Program was an unlawful exclusive dealing arrangement that foreclosed Star’s access to distributors for domestic fittings and harmed competition, thereby contributing significantly to the maintenance of McWane’s monopoly power in the market. Id. at *18-28. It noted that [¶] Supply and Ferguson, the country’s two largest waterworks distributors (with a combined 60% market share), prohibited their branches from purchasing domestic fittings from Star after the Full Support Program was announced, except through the program’s limited exceptions. Id. at *23. The practical effect of the program, the Commission found, “was to make it economically infeasible for distributors to drop McWane[ ] ... and switch to Star.” Id. at *24. Unable to attract dis tributors, Star was prevented from generating the revenue needed- to acquire its own foundry, a more efficient means of producing domestic fittings; thus, its growth into a rival that could challenge McWane’s monopoly power was artificially stunted. Id. at *25.

Moreover, the Commission found that there was evidence that McWane’s exclusionary conduct had an impact on price: after the Full Support Program was implemented, McWane raised domestic fittings prices and increased its gross profits despite flat production costs, and it did so across states, regardless of whether Star had entered the market as a competitor. Id. at *27.

Commissioner Wright filed a lengthy dissent. He assumed that McWane was a monopolist in the domestic-only fittings market, agreed that the Full Support Program was an exclusive dealing arrangement, and concluded that there was “ample record evidence” that the program harmed Star. Id. at *46 (Wright, dissenting). However,, he contended that the government “failed to carry its burden to demonstrate that the Full Support Program resulted in cognizable harm to competition.” Id. at *62. He argued that according to modern economic theory, exclusive dealing is harmful to competition (as opposed to merely harmful to a competitor) only if it prevents rivals from attaining a minimum efficient scale needed to constrain a monopolist’s exercise of monopoly power. Id. at *48. Commissioner Wright contended that the government had failed to demonstrate such harm to competition, either through direct or indirect evidence. Specifically, he suggested that the government had failed to show that Star’s inability to afford its own foundry was the equivalent of its being unable to achieve minimum efficient scale, failed to link the market foreclosure to McWane’s alleged maintenance of monopoly power, and miscalculated the relevant foreclosure share. Id. at *58-60. Moreover, he noted that other forms of indirect evidence — including Star’s ability to enter the domestic fittings market and expand despite the existence of the Full Support Program, as well as the short duration and terminability of the exclusive dealing arrangement — cut against a finding that McWane’s conduct was exclusionary. 8 Id. at *61-62.

McWane filed a timely petition in this Court seeking review of the Commissioner’s order on the lone remaining count.

II.

This Court “review[s] the FTC’s findings of fact and economic conclusions under the substantial evidence standard.” Schering-Plough Corp. v. FTC, 402 F.3d 1056, 1062 (11th Cir.2005); see 15 U.S.C. § 45(c) (“The findings of the Commission as to the facts, if supported by evidence, shall be conclusive.”). “Substantial evidence is more than a mere scintilla, and [this Court] require[s] such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Schering-Plough, 402 F.3d at 1062 (quotation omitted). This standard “forbids a court to ‘make its own appraisal of the testimony, picking and choosing for itself among uncertain and conflicting infer ences.’ ” Polypore Int’l, Inc. v. FTC, 686 F.3d 1208, 1213 (11th Cir.2012) (quoting FTC v. Algoma Lumber Co., 291 U.S. 67, 73, 54 S.Ct. 315, 78 L.Ed. 655 (1934)). Indeed, “the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Consolo v. Fed. Mar. Comm’n, 383 U.S. 607, 620, 86 S.Ct. 1018, 16 L.Ed.2d 131 (1966).

We review de novo the Commission’s legal conclusions and the application of the facts to the law. Polypore Int’l, 686 F.3d at 1213. However, “we afford the FTC some deference as to its informed judgment that a particular commercial practice violates the Federal Trade Commission Act.” Schering-Plough, 402 F.3d at 1063; see FTC v. Ind. Fed’n of Dentists, 476 U.S. 447, 454, 106 S.Ct. 2009, 90 L.Ed.2d 445 (1986) (“[T]he identification of governing legal standards and their application to the facts found ... are ... for the courts to resolve, although even in considering such issues the courts are to give some deference to the Commission’s informed judgment that a particular commercial practice is to be condemned as ‘unfair’ [under the Federal Trade Commission Act].”).

McWane challenges three particular determinations by the Commission: its market definition; its finding that McWane monopolized the domestic fittings market; and its finding that the Full Support Program harmed competition. Because the standard of review is essential to our analysis, we explain the applicable standard for each of the Commission’s conclusions. All three determinations are factual or economic conclusions reviewed only for substantial evidence.

First, our caselaw makes clear that “[t]he definition of the relevant market is essentially a factual question.” U.S. Anchor Mfg., Inc. v. Rule Indus., Inc., 7 F.3d 986, 994 (11th Cir.1993). Thus, we review the FTC’s determination of market definition — like all its factual findings — for substantial evidence. See Jim Walter Corp. v. FTC, 625 F.2d 676, 682 (5th Cir.1980) (applying the substantial evidence standard in reviewing the FTC’s finding of market definition). 9

Second, the FTC’s determination that a defendant possesses monopoly power is a factual or economic conclusion that we also review for substantial evidence. No prior case of ours appears to hold this specifically, but this conclusion follows from previous cases that have treated a determination that a defendant possesses market power — a lesser-included element of monopoly power — as a factual finding. See NaBanco, 779 F.2d at 605. Again, other circuits agree. E.g., Realcomp II, Ltd. v. FTC, 635 F.3d 815, 829 (6th Cir.2011) (applying substantial evidence standard to FTC’s finding that defendant pos sessed substantial market power); L.G. Balfour Co. v. FTC, 442 F.2d 1, 13 (7th Cir.1971) (applying substantial evidence standard to FTC’s finding that defendant possessed monopoly power).

Finally, so too with the Commission’s determination that McWane’s conduct harmed competition and lacked offsetting procompetitive benefits. Again, no binding case of ours appears to deal with the particular type of Federal Trade Commission Act violations at issue here, but we have applied the substantial evidence standard to analogous findings under that same act and other antitrust statutes. See Schering-Plough, 402 F.3d at 1068 (examining “whether there is substantial evidence to support the Commission’s conclusion that [defendant’s conduct] restricted] competition” in violation of Section 1 of the Sherman Act and Section 5 of the Federal Trade Commission Act); Foremost Dairies, Inc. v. FTC, 348 F.2d 674, 678-79 (5th Cir.1965) (applying substantial evidence standard to FTC’s finding of injury to competition under the Robinson-Patman Act).

This approach comports with the law in other circuits in a variety of antitrust contexts. The Seventh Circuit put the point most clearly in a Clayton Act case: “[T]he substantial evidence rule (like the clearly erroneous rule) applies to ultimate as well as underlying facts, including economic judgments.... [T]he ultimate question under the Clayton Act — whether the challenged transaction may substantially lessen competition — is governed by the substantial evidence rule.” Hosp. Corp. of Am. v. FTC, 807 F.2d 1381, 1385 (7th Cir.1986) (internal citation omitted). Our sister circuits have applied the substantial evidence standard to analogous economic conclusions in cases brought under the Federal Trade Commission Act, e.g., N.C. State Bd. of Dental Exam’rs v. FTC, 717 F.3d 359, 374 (4th Cir.2013) (applying substantial evidence standard to FTC’s determination that defendant’s behavior “was likely to cause significant anticompetitive harms” in violation of the Federal Trade Commission Act), aff'd, - U.S. -, 135 S.Ct. 1101, 191 L.Ed.2d 35 (2015); Realcomp II, 635 F.3d at 831-34 (applying substantial evidence standard to FTC’s finding that defendant’s policies harmed competition in violation of the Federal Trade Commission Act), and under other antitrust statutes, see, e.g., N. Tex. Specialty Physicians v. FTC, 528 F.3d 346, 370 (5th Cir.2008) (applying substantial evidence .standard to FTC’s determination that defendant’s conduct “amounted to horizontal price-fixing that is unrelated to competitive efficiencies” under Section 1 of the Sherman Act); Gibson v. FTC, 682 F.2d 554, 571 (5th Cir.1982) (applying substantial evidence standard to FTC’s finding of illegal brokerage in violation of Clayton Act § 2(c)); RSR Corp. v. FTC, 602 F.2d 1317, 1320, 1324-25 (9th Cir.1979) (applying substantial evidence standard to FTC’s finding under Section 7 of the Clayton Act that merger was anticompetitive); Fruehauf Corp. v. FTC, 603 F.2d 345, 355 (2d Cir.1979) (same); Yamaha Motor Co. v. FTC, 657 F.2d 971, 977 n. 7 (8th Cir.1981) (same, as to a joint venture).

The ultimate legal conclusion that a defendant’s conduct violates the Federal Trade Commission Act is an “application of the facts to the law,” which we review de novo, Polypore Int’l, 686 F.3d at 1213, except for the limited deference prescribed by Indiana Federation of Dentists, 476 U.S. at 454, 106 S.Ct. 2009. But the Commission’s factual building blocks and economic conclusions — findings of market definition, monopoly power, and harm to competition — are reviewed for substantial evidence.

III.

The Commission found that McWane adopted an exclusionary distribution policy that maintained its monopoly power in the domestic fittings market in violation of Section 5 of the Federal Trade Commission Act, which prohibits “[u]nfair methods of competition in or affecting commerce.” 15 U.S.C. § 45. 10 Although exclusive dealing arrangements are common and can be procompetitive, particularly in competitive markets, see Race Tires Am., Inc. v. Hoosier Racing Tire Corp., 614 F.3d 57, 76 (3d Cir.2010), these arrangements can harm competition in certain circumstances, see Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 45, 104 S.Ct. 1551, 80 L.Ed.2d 2 (1984) (O’Connor, J., concurring) (“Exclusive dealing can have adverse economic consequences by allowing one supplier of goods or services unreasonably to deprive other suppliers of a market for their goods ... ”), abrogated on other grounds by Ill. Tool Works Inc. v. Ind. Ink, Inc., 547 U.S. 28, 126 S.Ct. 1281, 164 L.Ed.2d 26 (2006); Jonathan M. Jacobson, Exclusive Dealing, “Foreclosure, ” and Consumer Harm, 70 Antitrust L.J. 311, 328 (2002) (“The concern [with exclusive dealing arrangements] is ... that- creating or increasing market power through exclusive dealing is the means by which the defendant is likely to increase prices, restrict output, reduce quality, slow innovation, or otherwise harm consumers.”). When a market is competitive, the “competition for the [exclusive] contract is a vital form of rivalry” that can induce the offering firm to provide price reductions or improved services to buyers, to the ultimate benefit of consumers. See Menasha Corp. v. News Am. Mktg. Im-Store, Inc., 354 F.3d 661, 663 (7th Cir.2004). But, notably, in the absence of such competition, a dominant firm can impose exclusive deals on downstream dealers to “strengthen[] or prolong[] [its] market position.” IIIB Philip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 760b7, at 54 (3d ed.2008). Thus, while such arrangements are “not illegal in themselves,” they can run afoul of antitrust laws as “an improper means of maintaining a monopoly.” United States v. Dentsply Int’l, Inc., 399 F.3d 181, 187 (3d Cir.2005).

A violation of Section 5 of the Federal Trade Commission Act premised on monopolization requires proof of “(1) the possession of monopoly power in the relevant market and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident.” Morris Commc’ns Corp. v. PGA Tour, Inc., 364 F.3d 1288, 1293-94 (11th Cir.2004) (quoting United States v. Grinnell Corp., 384 U.S. 563, 570-71, 86 S.Ct. 1698, 16 L.Ed.2d 778 (1966)) (internal quotation mark omitted). Thus, for the Commission’s conclusion that McWane violated the Federal Trade Commission Act to stand, it must have successfully defined the relevant market, demonstrated that McWanehad monopoly power in that market, and showed that McWane’s Full Support Program constituted the illegal maintenance of that monopoly power. McWane challenges all three of the Commission’s determinations, and we address each of them in turn.

A. Monopoly Power in the Relevant Market

1. Market Definition

“Defining the market is a necessary step in any analysis of market power and thus an indispensable element in the consideration of any monopolization ... case arising under section 2.” U.S. Anchor, 7 F.3d at 994. A product market consists of “products that have reasonable interchangeability for the purposes for which they are produced.” United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 404, 76 S.Ct. 994, 100 L.Ed. 1264 (1956). “The reasonable interchangeability of use or the cross-elasticity of demand between a product and its substitutes constitutes the outer boundaries of a product market for antitrust purposes.” U.S. Anchor, 7 F.3d at 995 (footnote omitted). “Cross-elasticity of demand” measures the extent to which modest variations in the price of one good affect customer demand for another good. “[A] high cross-elasticity of demand indicates that the two products in question are reasonably interchangeable substitutes for each other and hence are part of the same market.” Jacobs v. Tempur-Pedic Int’l, Inc., 626 F.3d 1327, 1337 n. 13 (11th Cir.2010).

In defining product markets, this Court has long looked to the factors set forth by the Supreme Court in Brown Shoe Co. v. United States, 370 U.S. 294, 82 S.Ct. 1502, 8 L.Ed.2d 510 (1962), including “industry or public recognition of the submarket as a separate economic entity, the product’s peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.” Polypore Int’l, 686 F.3d at 1217 (quoting U.S. Anchor, 7 F.3d at 995). Again, we are obliged to review the Commission’s market definition for substantial evidence.

A relevant geographic market also must be defined. See, e.g., Am. Key Corp. v. Cole Nat’l Corp., 762 F.2d 1569, 1579 (11th Cir.1985). The Commission (and the ALJ) defined the relevant geographic market as the United States. Neither party contests this determination.

As for the product market, the Commission, agreeing with the ALJ, found that the relevant market was one “for the supply of domestically-manufactured fit tings for use in ... projects with domestic-only specifications.” McWane II, 2014 WL 556261, at *13. It noted that various laws and end-user preferences requiring projects to use domestic fittings precluded imported fittings from being “reasonable substitutes” for those projects, even though the fittings themselves are functionally identical. Id.; see IIB Phillip E. Areeda, Herbert Hovenkamp & John So low, Antitrust Law ¶ 572b, at 430 (3d ed.2007) (“To the extent that regulation limits substitution, it may define the extent of the market.”). The Commission also noted that McWane charged higher prices for (and reaped greater profits from) domestic fittings in domestic-only projects: the ALJ found that McWane charged approximately 20%-95% more for its domestic fittings for domestic-only projects than for open-specification projects. This price differentiation reflected McWane’s ability to target customers with domestic-only project specifications who could not avoid the higher prices by substituting imported fittings. Indeed, Brown Shoe specifically identified “distinct prices” as a factor indicating a separate product market. 370 U.S. at 325, 82 S.Ct. 1502.

McWane contends, however, that domestic and imported fittings are, in fact, interchangeable, because some customers (those whose projects’ specifications are not dictated by law) can “flip” their projects from domestic-only to open, thereby turning imported fittings into a reasonable substitute. However, the Commission found, based on testimony in the record, that “flipping typically only occurs when domestic fittings are unavailable, rather than as a result of competition between domestic and imported fittings.” McWane II, 2014 WL 556261, at *15. This is consonant with the ALJ’s finding that end users with domestic-only preferences “are aware of, but not sensitive to, the price differential between domestic fittings and import fittings.” McWane I, 155 F.T.C. at 999.

McWane also alleges that the Commission’s definition was insufficient as a matter of law because it “was unsupported by an expert economic test,” which McWane claims is a requirement under Eleventh Circuit caselaw. It is true that in some circumstances we have said that a market definition “must be based on expert testimony.” Bailey v. Aligas, Inc., 284 F.3d 1237, 1246 (11th Cir.2002); see Am. Key Corp., 762 F.2d at 1579 (“Construction of a relevant economic market ... cannot ... be based upon lay opinion testimony.”). Such testimony can be insufficient when “conclusory” or “based upon insufficient economic analysis.” Gulfstream Park Racing Ass’n, Inc. v. Tampa Bay Downs, Inc., 479 F.3d 1310, 1313 (11th Cir.2007) (per curiam); see Bailey, 284 F.3d at 1246-47 (finding that plaintiffs expert testimony, which failed to consider alternative products in defining relevant market, was insufficient as a matter of law).

But in this case, the Commission did rely in part on the complaint counsel’s expert witness, Dr. Laurence Schumann, who considered a hypothetical monopolist test and the lack of interchangeability between domestic and imported fittings in domestic-only projects. Nevertheless, McWane claims that the expert’s analysis was insufficient because it did not involve an econometric analysis, such as a cross-elasticity of demand study. However, there appears to be no support in the caselaw for McWane’s claim that such a technical analysis is always required. Indeed, as the Commission correctly noted, “[cjourts routinely rely on qualitative economic evidence to define relevant markets.” McWane II, 2014 WL 556261, at *14. Thus, for example, in Polypore, the Commission’s market definition was affirmed by this Court on the basis of the Brown Shoe factors, apparently without an econometric study. 686 F.3d at 1217-18. Given the identification of persistent price differences between domestic fittings and imported fittings, the distinct customers, and the lack of reasonable substitutes in this case, there was sufficient evidence to support the Commission’s market definition.

2. Monopoly Power

“As a legal matter, Sherman Act § 2 requires that the defendant either have monopoly power or a dangerous probability of achieving it ...” XI Philip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 1800c5, at 22 (3d ed.2011); accord Dentsply, 399 F.3d at 187 (“A prerequisite for [a § 2 violation] is a finding that monopoly power exists.”). Monopoly power is the ability “to control prices or exclude competition.” Grinnell, 384 U.S. at 571, 86 S.Ct. 1698 (quotation omitted). However, “[blecause ... direct proof [of the ability to profitably raise prices substantially above the competitive level] is only rarely available, courts more typically examine market structure in search of circumstantial evidence of monopoly power;” United States v. Microsoft Corp., 253 F.3d 34, 51 (D.C.Cir.2001) (en banc) (per curiam). Courts regularly ask whether the firm has a predominant market share, see Bailey, 284 F.3d at 1246 (“Because demand is difficult to establish with accuracy, evidence of a seller’s market share may provide the most convenient circumstantial measure of monopoly power.”), and look to other circumstantial factors such as “the size and strength of competing firms, freedom of entry, pricing trends and practices in the industry, ability of consumers to substitute comparable goods, and consumer demand,” Dentsply, 399 F.3d at 187.

In determining that McWane had monopoly power, the Commission found that McWane’s market share of the domestic fittings market had been 100% from 2006 until Star’s entry into the market in 2009. McWane’s market share was then approximately 95% in 2010 and approximately 90% in 2011, “far exceeding] the levels that courts typically require to support a prima facie showing of monopoly power.” McWane II, 2014 WL 556261, at *16. It also observed that there were “substantial barriers to entry in the domestic fittings market” both for brand new entrants and for those who already supply imported fittings. Id. Although Star was able to enter the market, the Commission noted that its share remained below 10% in- 2010 and 2011, and, notably, its entry had no effect on McWane’s prices. The Commission reasoned that McWane’s “ability to control prices” in the market “provide[d] direct evidence of [its] monopoly power.” Id. at *18.

The difficulty in this case is that the circumstantial evidence does not all point in the same direction. McWane’s market share during the relevant time period is plainly high enough to be considered predominant. See Eastman Kodak Co. v. Image Technical Servs., Inc., 504 U.S. 451, 481, 112 S.Ct. 2072, 119 L.Ed.2d 265 (1992) (80-95% market share sufficient to establish monopoly power); Grinnell, 384 U.S. at 571, 86 S.Ct. 1698 (87% sufficient); Dentsply, 399 F.3d at 188 (market share between 75-80% is “more than adequate to establish a prima facie case of [monopoly] power”); Colo. Interstate Gas Co. v. Natural Gas Pipeline Co. of Am., 885 F.2d 683, 694 n. 18 (10th Cir.1989) (“[To establish monopoly power,] lower courts generally require a minimum market share of between 70% and 80%.”); Cliff Food Stores, Inc. v. Kroger, Inc., 417 F.2d 203, 207 n. 2 (5th Cir.1969) (“[Something more than 50% of the market is a prerequisite to a finding of monopoly”). Standing alone, this would seem to be sufficient evidence to support the Commission’s conclusion that McWane had monopoly power in the domestic fittings market.

However, there is also evidence that, despite the presence of the Full Support Program, Star was still able to enter the domestic fittings market and expand its market share from 0% in 2009 to approximately 5% in 2010 to approximately 10% in 2011, while McWane’s market share correspondingly declined. McWane contends that this “clear and successful entry” and growth by a competitor precludes a finding of monopoly power by demonstrating a lack of barriers to entry in the market. The Commission disagreed, finding that, despite Star’s entry and growth, substantial barriers to entry existed in both the overall fittings market and the domestic fittings market. The ALJ found (and the Commission agreed) that “a significant capital investment” is required to enter the overall fittings market, McWane I, 155 F.T.C. at 1113, as “new entrants] must overcome existing relationships between existing manufacturers!,] and the [distributors!,] and [e]nd [u]sers,” in addition to “developing] hundreds of patterns and moldings,” id. at 1114. All told, the Commission agreed with the ALJ that a de novo entrant would need approximately three to five years to enter the fittings market. McWane II, 2014 WL 556261, at *16. Star, as an established player in the overall fittings market, did not face all of these obstacles in entering the domestic fittings market. (For example, it had preexisting relationships with some distributors and did not need to alter its sales team.) Nevertheless, the Commission found that significant barriers to entry existed in the domestic market, as Star still needed to purchase its own foundry or contract with third-party domestic foundries. Id.; see Bailey, 284 F.3d at 1256 (“Entry barriers include ... capital outlays required to start a new business----”). Moreover, the Commission found that the Full Support Program itself posed a barrier to entry by shrinking the number of available distributors. In support of this argument, the Commission observed that two other suppliers of imported fittings, Sigma Corporation and Serampore Industries Private, considered entering the domestic fittings market but ultimately concluded that the costs and challenges were too high. McWane II, 2014 WL 556261, at *17.

Some caselaw from other circuits appears to support McWane. See Tops Mkts., Inc. v. Quality Mkts., Inc., 142 F.3d 90, 99 (2d Cir.1998) (“We cannot be blinded by market share figures and ignore marketplace realities, such as the relative ease of competitive entry.... [A competitor’s] successful entry ... refutes any inference of the existence of monopoly power that might be drawn from [the defendant’s] market share.”). 11 But not all courts agree. See Rebel Oil Co. v. Atl. Richfield Co., 51 F.3d 1421, 1440 (9th Cir.1995) (“The fact that entry has occurred does not necessarily preclude the existence of ‘significant’ entry barriers. If the output or capacity of the new entrant is insufficient to take significant business away from the predator, they are unlikely to represent a challenge to the predator’s market power.”); Reazin v. Blue Cross & Blue Shield of Kan., Inc., 899 F.2d 951, 971 (10th Cir.1990) (rejecting defendant’s argument that presence of multiple competitors demonstrated that entry barriers were insubstantial where “no other entrant remotely approached [defendant’s] domination of the market”); Oahu Gas Serv., Inc. v. Pac. Res. Inc., 838 F.2d 360, 366-67 (9th Cir.1988) (“A declining market share may reflect an -absence of market power, but it does not foreclose a finding of such power.” (quotation omitted)). No decision of this Court appears to be directly on point.

In addition to McWane’s overwhelming (albeit declining) market share, the Commission cited the particular importance of Star’s inability to constrain McWane’s pricing for domestic fittings. After Star’s entry, McWane continued to sell domestic fittings for domestic-only products at prices that “earned significantly higher gross profits than for non-domestic fittings, which faced greater competition.” McWane II, 2014 WL 556261, at *17. Indeed, McWane’s prices and profits for domestic fittings rose in 2010, the year after Star’s entry.

On this record, we are unprepared to say that Star’s entry and growth foreclose a finding that McWane possessed monopoly power in the relevant market. Although the limited entry and expansion of a competitor sometimes may cut against such a finding, the evidence of McWane’s overwhelming market share (90%), the large capital outlays required to enter the domestic fittings market, and McWane’s undeniable continued power over domestic fittings prices amount to sufficient evidence that “a reasonable mind might accept as adequate to support” the Commission’s conclusion. Schering-Plough, 402 F.3d at 1062 (quotation omitted).

B. Monopoly Maintenance

Having established that McWane “possess[es] ... monopoly power in the relevant market,” we turn to the question of whether the government proved that McWane engaged in “the willful ... maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident.” Morris Commc’ns, 364 F.3d at 1293-94 (quoting Grinnell, 384 U.S. at 570-71, 86 S.Ct. 1698).

As we’ve observed, exclusive dealing arrangements are not per se unlawful, but they can run afoul of the antitrust laws when used by a dominant firm to maintain its monopoly. Of particular relevance to this case, an exclusive dealing arrangement can be harmful when it allows a monopolist to maintain its monopoly power by raising its rivals’ costs sufficiently to prevent them from growing into effective competitors. See XI Areeda & Hovenkamp, supra, ¶ 1804a, at 116-17 (describing how exclusive contracts can raise rivals’ costs and harm competition); see generally Thomas G. Krattenmaker & Steven C. Salop, Anticompetitive Exclusion: Raising Rivals’ Costs to Achieve Power Over Price, 96 Yale L.J. 209 (1986). The following description seems particularly appropriate here:

[S]uppose an established manufacturer has long held a dominant position but is starting to lose market share to an aggressive young rival. A set of strategically planned exclusive-dealing contracts may slow the rival’s expansion by requiring it to develop' alternative outlets for its product, or rely at least temporarily on inferior or more expensive outlets. Consumer injury results from the delay that the dominant firm imposes on the smaller rival’s growth.

XI Areeda & Hovenkamp, supra, ¶ 1802c, at 76; see ZF Meritor, LLC v. Eaton Corp., 696 F.3d 254, 271 (3d Cir.2012); Dentsply, 399 F.3d at 191.

Tracking this economic argument, the Commission’s theory is that McWane’s Full Support Program was an exclusive dealing policy designed specifically to maintain its monopoly power “by impairing the ability of rivals to grow into effective competitors that might erode the firm’s dominant position.” McWane II, 2014 WL 556261, at *19. To prevail, the FTC must establish that McWane “has engaged in anti-competitive conduct that reasonably appears to be a significant contribution to maintaining monopoly power.” Dentsply, 399 F.3d at 187; accord Microsoft, 253 F.3d at 79 (quoting III Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 650c, at 69 (1996)).

Neither the Supreme Court nor this Cir-. cuit has provided a clear formula with which to evaluate an exclusive dealing monopoly maintenance claim, but the D.C. Circuit has synthesized a structured, “rule of reason”-style approach to monopolization cases that has been cited with approval. See Jacobson, supra, at 364-69; III Areeda & Hovenkamp, supra, ¶ 651, at 97 n. 1. First, the government must show that the monopolist’s conduct had the “anticompetitive effect” of “harm[ing] competition, not just a competitor.” Microsoft, 253 F.3d at 58-59. If the government succeeds in demonstrating this anticompetitive harm, the burden then shifts to the defendant to present procompetitive justifications for the exclusive conduct, which the government can refute. Microsoft, 253 F.3d at 59; Dentsply 399 F.3d at 196; see Eastman Kodak, 504 U.S. at 482-84, 112 S.Ct. 2072 (describing defendant’s proffered “valid business reasons” for its actions and plaintiffs rebuttal). If the court accepts the defendant’s proffered justifications, it must then decide whether the conduct’s procompetitive effects outweigh its anticompetitive effects. Microsoft, 253 F.3d at 59. This approach mirrors rule of reason analysis. See Schering-Plough, 402 F.3d at 1064-65 (outlining a substantially similar burden-shifting approach in “traditional rule of reason analysis”).

The Commission followed this approach. It found that McWane’s Full Support Program was an exclusive dealing policy that harmed competition by foreclosing Star’s access to necessary distributors and contributed significantly to Star’s lost sales and subsequent inability to purchase its own foundry and expand output. It considered McWane’s procompetitive justifications but ultimately found them unpersuasive.

McWane challenges each aspect of the Commission’s ruling: first, it says that its Full Support Program was “presumptively legal” because it was nonbinding and short-term; second, it contends that the government failed to carry its burden of establishing harm to competition; third, it argues that the Commission wrongly rejected its proffered procompetitive justifications. We address each claim in turn.

1. Presumptive Legality

McWane suggests that the Full Support Program lacked the characteristics of anticompetitive exclusive dealing arrangements. Specifically, it urges that the Full Support Program was “presumptively legal” and “[could not] harm competition” because it was short-term and voluntary (rather than a binding contract of a longer term). No binding precedent from the Supreme Court or this Court speaks specifically to this issue, but McWane hangs its hat on caselaw from other circuits. See, e.g., Omega Envtl. v. Gilbarco, Inc., 127 F.3d 1157, 1163 (9th Cir.1997) (“[T]he short duration and easy terminability of these [one-year] agreements negate substantially their potential to foreclose competition.” (footnote omitted)); Roland Mach. Co. v. Dresser Indus., Inc., 749 F.2d 380, 395 (7th Cir.1984) (“Exclusive-dealing contracts terminable in less than a year are presumptively lawful under section 3 [of the Clayton Act].”); Jacobson, supra, at 351-52 & n. 195.

But not all courts agree. The Third Circuit in Dentsply held that where exclusive deals were “technically only a series of independent sales,” they nevertheless constituted antitrust violations because “the economic elements involved — the large share of the market held by [the defendant] and its conduct excluding competing manufacturers — realistically ma[d]e the arrangements ... as effective as those in written contracts.” 399 F.3d at 193. The Dentsply court noted that “in spite of the legal ease with which the relationship can be terminated, the [distributors] have a strong economic incentive to continue [buying defendant’s product].” Id. at 194; see also ZF Meritor, 696 F.3d at 270 (“[.D]e facto exclusive dealing claims are cognizable under the antitrust laws.”); Minn. Mining & Mfg. Co. v. Appleton Papers, Inc., 35 F.Supp.2d 1138, 1144 (D.Minn.1999) (evaluating an exclusive dealing arrangement’s “practical effect” rather than “merely ... its form” in determining whether it was terminable at will (internal quotation marks omitted)). The Third Circuit distinguished opposing cases by noting that those situations primarily involved markets in which firms could viably sell directly to consumers even when foreclosed from distributors, Dentsply, 399 F.3d at 194 n. 2, whereas in Dentsply direct sales were not “a practical alternative 'for most [competing] manufacturers,” id. at 189. Likewise, in the case at hand, both the Commission and the ALJ found that distributors were essential to the domestic fittings market: “No evidence supports the existence of viable alternate distribution channels, including direct sales to end users.” McWane II, 2014 WL 556261, at *23.

This approach is consistent with the Supreme Court’s instruction to look at the “practical effect” of exclusive dealing arrangements. Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 326-28, 81 S.Ct. 623, 5 L.Ed.2d 580 (1961); see also Eastman Kodak, 504 U.S. at 466-67, 112 S.Ct. 2072 (“Legal presumptions that rest on formalistic distinctions rather than actual market realities are generally disfavored in antitrust law. This Court has preferred to resolve antitrust claims on a case-by-case basis, focusing on the ‘particular facts disclosed by the record.’ ” (quoting Maple Flooring Mfrs. Ass’n v. United States, 268 U.S. 563, 579, 45 S.Ct. 578, 69 L.Ed. 1093 (1925))). The Commission adopted this approach, looking to “the reality of [the] marketplace” and finding that “the practical effect of McWane’s program was to make it economically infeasible for distributors to ... switch to Star.” McWane. II, 2014 WL 556261, at *24. Even the dissenting commissioner agreed with this approach. Id. at *55 n. 38 (Wright, dissenting). So do we.

Moreover, the nature of the Full Support Program arguably posed a greater threat to competition than a conventional exclusive dealing contract, as it lacked the traditional procompetitive benefits of such contracts. As we’ve noted, courts often take a permissive view of such contracts on the grounds that firms compete for exclusivity by offering procompetitive inducements (e.g., lower prices, better service). But not here. The Full Support Program was “unilaterally imposed” by fiat upon all distributors, and the ALJ found that it resulted in “no competition to become the exclusive supplier” and no “discount, rebate, or other consideration” offered in exchange for exclusivity. McWane I, 155 F.T.C. at 1114. This is consistent with evidence that McWane’s prices rose, rather than fell, in the wake of the program.

We are disposed to follow the Supreme Court’s instruction that we consider “market realities” rather than “formalistic distinctions” in rejecting McWane’s argument that the specific form of its exclusivity mandate insulated it from antitrust scrutiny.

2. Harm to Competition

We turn then to the first step in the monopolization test: the government must demonstrate that the defendant’s challenged conduct had anticompetitive effects, harming competition.

As with many areas of antitrust law, the federal judiciary’s approach to evaluating exclusive dealing has undergone significant evolution over the past century. Under the approach laid out by the Supreme Court in Standard Oil Co. of California and Standard Stations, Inc. v. United States (Standard Stations), 337 U.S. 293, 69 S.Ct. 1051, 93 L.Ed. 1371 (1949), all that was required for an exclusive deal to violate the Clayton Act was proof of substantial foreclosure — “proof that competition ha[d] been foreclosed in a substantial share of the line of commerce affected.” Id. at 314, 69 S.Ct. 1051. The Supreme Court amended that approach in Tampa Electric, in which it continued to emphasize the importance of substantial foreclosure, but opened the door to a broader analysis. See 365 U.S. at 328-29, 81 S.Ct. 623.

Lower federal courts have burst through that door over the past 50 years, interpreting Tampa Electric as authorizing a rule of reason approach to exclusive dealing cases. See, e.g., ZF Meritor, 696 F.3d at 271 (characterizing Tampa Electric as standing for the proposition that “exclusive dealing agreements ... [are] judged under the rule of reason”); Jacobson, supra, at 322 (noting that “later cases have suggested” that Tampa Electric “authorized full-scale rule of reason analysis”); XI Areeda & Hovenkamp, supra, ¶ 1820b, at 177 (“Most decisions follow the language in the Supreme Court’s Tampa Electric decision indicating that a complete rule of reason analysis is essential, and foreclosure percentages represent only a first step in the inquiry.”). This Court, without specifically citing Tampa Electric, has joined the consensus that exclusive dealing arrangements are “reviewed under the rule of reason.” DeLong Equip. Co. v. Washington Mills Abrasive Co., 887 F.2d 1499, 1508 n. 12 (11th Cir.1989).

The difference between the traditional rule of reason and the rule of reason for exclusive dealing is that in the exclusive dealing context, courts are bound by Tampa Electric’s requirement to consider substantial foreclosure. See Microsoft, 253 F.3d at 69. But foreclosure is usually no longer sufficient by itself; rather, it “serves a useful screening function” as a proxy for anticompetitive harm. Id. Thus, foreclosure is one of several factors we now examine in determining whether the conduct harmed competition. See Jacobson, supra, at 361-64; XI Areeda & Hovenkamp, supra, ¶ 1821d, at 197 (“[Foreclosure percentages] are seldom decisive in and of themselves. Rather, they provide the jumping-off point for further analysis.”). We will also look for direct evidence that the challenged conduct has affected price or output, along with other indirect evidence, such as the degree of rivals’ exclusion, the duration of the exclusive deals, and the existence of alternative channels of distribution. XI Areeda & Hovenkamp, supra, ¶ 1821d, at 197-209. The ultimate question remains whether the defendant’s conduct harmed competition.

To effect anticompetitive harm, a defendant “must harm the competitive process, and thereby harm consumers. In contrast, harm to one or more competitors will not suffice.” Microsoft, 253 F.3d at 58; see also Brooke Grp. Ltd. v. Brown & Williamson Tobacco Co., 509 U.S. 209, 224, 113 S.Ct. 2578, 125 L.Ed.2d 168 (1993). This distinction makes good sense, particularly in a competitive market where injury to a single competitor may not have a significant effect on overall competition due to the persistence of other rivals. However, competitors and competition are linked, particularly in the right market settings: “in a concentrated market with very high barriers to entry, competition will not exist without competitors.” Spirit Airlines, Inc. v. Nw. Airlines, Inc., 431 F.3d 917, 951 (6th Cir.2005). Indeed, this is. one reason' that the behavior of monopolists faces more exacting scrutiny under the antitrust statutes. See Eastman Kodak, 504 U.S. at 488, 112 S.Ct. 2072 (Scalia, J., dissenting) (“Behavior that might otherwise not be of concern to the antitrust laws ... can take on exclusionary connotations when practiced by a monopolist.”); Dentsply, 399 F.3d at 187 (“Behavior that otherwise might comply with antitrust law may be impermissibly exclusionary when practiced by a monopolist.”); IIIB Areeda & Hovenkamp, supra, ¶ 806e, at 423.

Before we proceed, we address a point of disagreement between the Commission, the dissenting commissioner, and the amici: the government’s burden of proof in demonstrating harm to competition. The dissenting commissioner insisted that, given the high likelihood that an exclusive dealing arrangement is actually procompetitive, a plaintiff alleging illegal exclusive dealing must show “clear evidence of anti-competitive effect.” McWane II, 2014 WL 556261, at *51 (Wright, dissenting). Applying that standard, Commissioner Wright concluded that the government had not met its burden for. several reasons, including that it had not sufficiently established that the Full Support Program caused the observed price effects. The Commission countered that Commissioner Wright sought “a new, heightened standard of proof for exclusive dealing cases” that had “no legal support.” Id. at *26 & n. 12 (majority). Although McWane does not articulate its proposed burden of proof using the dissenting commissioner’s language, it agrees in substance that the Commission did not prove harm to competition with sufficient certainty.

We agree with the Commission. Putting aside the possible economic merits of raising the standard of proof for exclusive dealing cases, we can find no foundation for this conclusion in the caselaw. The governing Supreme Court precedent speaks not of “clear evidence” or definitive proof of anticompetitive harm, but of “probable effect.” Tampa Elec., 365 U.S. at 329, 81 S.Ct. 623 (instructing courts to weigh the “probable effect of the [exclusive dealing] contract on the relevant area of effective competition” (emphasis added)); accord ZF Meritor, 696 F.3d at 268 (“Under the rule of reason, an exclusive dealing arrangement will be unlawful only if its ‘probable effect’ is to substantially lessen competition in the relevant market.” (quoting Tampa Elec., 365 U.S. at 327-29, 81 S.Ct. 623)). Indeed, this Court has often articulated the rule of reason — the governing standard for evaluating exclusive dealing claims, DeLong Equip. Co., 887 F.2d at 1508 n. 12—by quoting the Supreme Court’s instruction in Board of Trade of Chicago v. United States, 246 U.S. 231, 238, 38 S.Ct. 242, 62 L.Ed. 683 (1918), to analyze the effects of the challenged conduct, “actual or probable.” E.g., Jacobs v. Tempur-Pedic Int’l, Inc., 626 F.3d 1327, 1334 n. 8 (11th Cir.2010); Schering-Plough, 402 F.3d at 1064 n. 12.

Of course, the FTC’s allegation is not merely that McWane engaged in exclusive dealing, but that it used exclusive dealing to maintain its monopoly power. In the monopolization context, courts have articulated the government’s burden in terms of the causality that must be shown between the defendant’s conduct and the anticompetitive harm. These formulations, too, are framed in terms of probability: “unlawful maintenance of a monopoly is demonstrated by proof that a defendant has engaged in anti-competitive conduct that reasonably appears to be a significant contribution to maintaining monopoly power.” Dentsply, 399 F.3d at 187 (emphasis added); accord Microsoft, 253 F.3d at 79. In Microsoft, the D.C. Circuit found no case supporting the proposition that Sherman Act § 2 liability requires plaintiffs to “present direct proof that a defendant’s continued monopoly power is precisely attributable to its anticompetitive conduct.” Microsoft, 253 F.3d at 79. It noted that “[t]o require that § 2 liability turn on a plaintiff’s ability or inability to reconstruct the hypothetical marketplace absent a defendant’s anticompetitive conduct would only encourage monopolists to take more and earlier anticompetitive action.” Id.; see also III Areeda & Hovenkamp, supra, ¶ 657a2, at 162 (“[T]he government suitor need not show that competition is in fact less than it would be in some alternative universe in which the challenged conduct had not occurred. It is enough to show that anticompetitive consequences are a naturally-to-be-expected outcome of the challenged conduct.”).

We agree with the Commission and our sister circuits that in these circumstances the government must show that the defendant engaged in anticompetitive conduct that reasonably appears to significantly contribute to maintaining monopoly power. As we’ve already discussed, because this determination is an economic conclusion, the Commission’s finding on this count must be supported by substantial evidence.

a) Substantial Foreclosure

“Substantial foreclosure” continues to be a requirement for exclusive dealing to run afoul of the antitrust statutes. Foreclosure occurs when “the opportunities for other traders to enter into or remain in [the] market [are] significantly limited” by the exclusive dealing arrangements. Microsoft, 253 F.3d at 69 (quoting Tampa Elec., 365 U.S. at 328, 81 S.Ct. 623) (internal quotation marks omitted). Traditionally a foreclosure percentage of at least 40% has been a threshold for liability in exclusive dealing cases. Jacobson, supra, at 362. However, some courts have found that a lesser degree of foreclosure is required when the defendant is a monopolist. See Microsoft, 253 F.3d at 70 (“[A] monopolist’s use of exclusive contracts ... may give rise to a § 2 violation even though the contracts foreclose less than the roughly 40% or 50% share usually required in order to establish a § 1 violation.”).

In this case, both the Commission and the ALJ found that the Full Support Program foreclosed Star from a substantial share of the market. Although the Commission did not quantify a percentage, it did note that the two largest distributors, who together controlled approximately 50-60% of distribution, prohibited their branches from purchasing from Star (except through the Full Support Program exceptions) following the announcement of the Full Support Program. Indeed, HD Supply went so far as to cancel pending orders for domestic fittings that it had placed with Star. The Commission also observed that the third-largest distributor was initially interested in purchasing domestic fittings from Star, but followed suit soon after the Full Support Program was announced. Testimony in the record sup ports the Commission’s conclusion that this pattern recurred with other dealers, even when Star promised lower prices than McWane. Thus, for example, U.S. Pipe refused to purchase domestic fittings from Star, despite á promise of lower prices, until September 2010. Likewise with TDG distributors. Executives at Groeniger and Illinois Meter also testified that the Full Support Program deterred them from dealing with Star. Although the Commission did not place an exact number on the percentage foreclosed, it found that the Full Support Program “tie[d] up the key dealers” and that the foreclosure was “substantial and problematic.” McWane II, 2014 WL 556261, at *24 n. 10.

These factual findings are all consistent with the AL J’s determinations, and all pass our deferential review. Nevertheless, McWane challenges the Commission’s conclusion by arguing that Star’s entry and growth in the market demonstrate that, as a matter of law, the Full Support Program did not cause substantial foreclosure. As before, when McWane raised a substantially similar claim to rebut the Commission’s finding of monopoly power, this argument is ultimately unpersuasive. Again, “[t]he test is not total foreclosure, but whether the challenged practices bar a substantial number of rivals or severely restrict the market’s ambit.” Dentsply, 399 F.3d at 191. Our sister circuits have found monopolists liable for anticompetitive conduct where, as here, the targeted rival gained market share — but less than it likely would have absent the conduct.' See Conwood Co. v. U.S. Tobacco Co., 290 F.3d 768, 789-91 (6th Cir.2002). As noted above, exclusive dealing measures that slow a rival’s expansion' can still produce consumer injury. See XI Areeda & Hovenkamp, supra, ¶ 1802c, at 76; accord Dentsply, 399 F.3d at 191; ZF Meritor, 696 F.3d at 271. Given the ample evidence in the record that the Full Support Program significantly contributed to key dealers freezing out Star, the Commission’s foreclosure determination is supported by substantial evidence and sufficient as a matter of law.

b) Evidence of Harm to Competition

Having concluded that the Commission’s finding of substantial foreclosure is supported by substantial evidence, we turn to the remainder of the Commission’s evidence that McWane’s Full Support Program injured competition. The record contains both direct and indirect evidence that the Full Support Program harmed competition. The Commission relied on both, and taken together they are more than sufficient to meet the government’s burden. The Commission found that McWane’s program “deprived its rivals ... of distribution sufficient to achieve efficient scale, thereby raising costs and slowing or preventing effective entry.” McWane II, 2014 WL 556261, at *22. It found that the Full Support Program made it infeasible for distributors to drop the monopolist McWane and switch to Star. This, the Commission found, deprived Star of the revenue needed to purchase its own domestic foundry, forcing it to rely on inefficient outsourcing arrangements and preventing it from providing meaningful price competition with McWane. Id. at *25.

Perhaps the Commission’s most powerful evidence of anticompetitive harm was direct pricing evidence. It noted that McWane’s prices and profit margins for domestic fittings were notably higher than prices for imported fittings, which faced greater competition. Thus, these prices appeared to be supracompetitive. Yet in states where Star entered as a competitor, notably there was no effect on McWane’s prices. Indeed, soon after Star entered the market, McWane raised prices and increased its gross profits — despite its flat production costs and its own internal projections that Star’s unencumbered entry into the market would cause prices to fall. Id. at *27. Since McWane was an incumbent monopolist already charging supra-competitive prices (as demonstrated by the difference in price and profit margin between domestic and imported fittings), evidence that McWane’s prices did not fall is consistent with a reasonable inference that the Full Support Program significantly contributed to maintaining McWane’s monopoly power.

McWane claims, however, that the government did not adequately prove that the Full Support Program was responsible for this price behavior. But as we’ve noted, McWane demands too high a bar for causation. While it is true that there could have been other causes for the price behavior, the government need not demonstrate that the Full Support Program was the sole cause — only that the program “reasonably appear[ed] to be a significant contribution to maintaining [McWane’s] monopoly power.” Dentsply, 399 F.3d at 187. Moreover, under our deferential standard of review, the mere fact that “two inconsistent conclusions” could be drawn from the record “does not prevent [the Commission’s] finding from being supported by substantial evidence.” Consolo, 383 U.S. at 620, 86 S.Ct. 1018.

The Commission also drew on testimony from Star executives that the Full Support Program deprived Star of the sales and revenue needed to invest in a domestic foundry of its own. These estimates were based in part on distributors’ withdrawn requests for quotes or orders in the wake of the Full- Support Program. Indeed, Star had identified a specific foundry to acquire and had entered negotiations to purchase it, but after the announcement of the Full Support Program, decided not to move forward with the purchase. Without a foundry of its own with which to manufacture fittings, Star was forced to contract with six third-party domestic foundries to produce raw casings — a “more costly and less efficient” arrangement on account of higher shipping, labor, and logistical costs; smaller batch sizes; less specialized equipment; and various . other factors. McWane II, 2014 WL 556261, at *25. Star estimated that with its own foundry, it could have reduced costs and substantially lowered its domestic fittings prices.

Moreover, as the ALJ found, some customers, including [¶] Supply and Ferguson, were reluctant to purchase from a supplier that lacked its own foundry, thereby further inhibiting any challenge to McWane’s market dominance. McWane I, 155 F.T.C. at 1157, 1160. Thus, the record evidence suggests that the Full Support Program stunted the growth of Star— McWane’s only rival in the domestic fittings market — and prevented it from emerging as an effective competitor who could challenge McWane’s supracompetitive prices.

We also consider it significant that alternative channels of distribution were unavailable to Star. In cases where exclusive dealing arrangements tie up distributors in a market, courts will often consider whether alternative channels of distribution exist. See Dentsply, 399 F.3d at 193; Omega Envtl., 127 F.3d at 1162-63; XI Areeda & Hovenkamp, supra, ¶ 1821d4, at 203-09. If firms can use other means of distribution, or sell directly to consumers, then it is less likely that their foreclosure from distributors will harm competition. In Denstply, the Third Circuit found exclusive deals with distributors to be anticompetitive where direct sales of the market’s products (artificial teeth) to consumers was not “practical or feasible in the market as it exists and functions.” 399 F.3d at 193. The Commission found the same in the domestic fittings market, and the dissent agreed. Thus, Star’s foreclosure from the major distributors was particularly likely to harm competition in this market.

Finally, the clear anticompetitive intent behind the Full Support Program also supports the inference that it harmed competition. Anticompetitive intent alone, no matter how virulent, is insufficient to give rise to an antitrust violation. See Microsoft, 253 F.3d at 60. But, as this Court has said, “[e]vidence of intent is highly probative ‘not because a good intern tion will save an otherwise objectionable regulation or the reverse; but because knowledge of intent may help the court to interpret facts and to predict consequences.’ ” Graphic Prods. Distribs., Inc. v. ITEK Corp., 717 F.2d 1560, 1573 (11th Cir.1983) (quoting Bd. of Trade of Chi., 246 U.S. at 238, 38 S.Ct. 242). For a monopolization charge, intent is “relevant to the question whether the challenged conduct is fairly characterized as ‘exclusionary’ or ‘anticompetitive’.... [T]here is agreement on the proposition that ‘no monopolist monopolizes unconscious of what he is doing.’ ” Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585, 602, 105 S.Ct. 2847, 86 L.Ed.2d 467 (1985) (quoting United States v. Aluminum Co. of Am., 148 F.2d 416, 432 (2d Cir.1945)); see also Microsoft, 253 F.3d at 59 (“Evidence of the intent behind the conduct of a monopolist is relevant only to the extent it helps us understand the likely effect of the monopolist’s conduct.”).

In this case, the evidence of anticompetitive intent is particularly powerful. Testimony from McWane executives leaves little doubt that the Full Support Program was a deliberate plan to prevent Star from “reaching] any critical market mass that will allow them to continue to invest and receive a profitable return” by “[f]orc[ing] Star[ ] to absorb the costs associated with having a more full line before they can secure major distribution.” Indeed, the plan was implemented as a reaction to concerns about the “[e]rosion of domestic pricing if Star emerges as a legitimate competitor.” Although such intent alone is not illegal, it could reasonably help the Commission draw the inference that the witnessed price behavior was the (intended) result of the Full Support Program.

Not all of the evidence adduced in this case uniformly points against McWane. For example, as we’ve previously noted, Star was not completely excluded from the domestic fittings market; it was able to enter, and grow despite the presence of the Full Support Program. However, it is still perfectly plausible to conclude on this record that Star’s growth was meaningfully (and deliberately) slowed and its development into a rival that could constrain McWane’s monopoly power was stunted. Cf. Microsoft, 253 F.3d at 71 (stating that defendant’s exclusionary conduct kept the rival’s product “below the critical level necessary for [the targeted rival] or any other rival to pose a real threat to [the defendant’s] monopoly”). Also, the Full Support Program was not a binding contract of a lengthy duration. As noted above, these characteristics do not render the program presumptively lawful, but they also do not point in the FTC’s favor as an indirect indicator of anticompetitive harm. Nevertheless, the direct and indirect evidence of anticompetitive harm is more than sufficient to pass our deferential review. Again, the Commission’s conclusion that the Full Support Program harmed competition is supported by substantial evidence and sound as a matter of law.

3. Procompetitive Justifications

Having established that the defendant’s conduct harmed competition, the burden shifts to the defendant to offer procompetitive justifications for its conduct. As the Commission explained, “[cjognizable justifications are typically those that reduce cost, increase output or improve product quality, service, or innovation.” McWane II, 2014 WL 556261, at *30 (collecting cases); see also XI Areeda & Hovenkamp, supra, ¶ 1822a, at 213 (“A justification is reasonable’ if it reduces the defendant’s costs, minimizes risk, or lessens the danger of free riding....”). Such justifications, however, cannot be “merely pretextual.” Morris Commc’ns, 364 F.3d at 1296; see Eastman Kodak, 504 U.S. at 483-84, 112 S.Ct. 2072.

McWane offers two; neither is persuasive. First, McWane says that the Full Support Program was necessary to retain enough sales to keep its domestic foundry afloat. The Commission rightly rejected this argument; as other courts have recognized, such a goal is “not an unlawful end, but neither is it a procompetitive justification.” Microsoft, 253 F.3d at 71. And as the Commission noted, the steps McWane took to preserve its sales volume “were not the type of steps, such as a price reduction, that typically promote consumer welfare by increasing overall market output.” McWane II, 2014 WL 556261, at *30. McWane’s sales “did not result from lower prices, improved service or quality, or other consumer benefits,” but rather from reducing the output of its only rival. Id.

Second, McWane offers the more sophisticated argument that the Full Support Program was needed to keep Star from “ ‘cherrypick[ing]’ the core of [the] domestic fittings business by making only the top few dozen fittings that account for roughly 80% of all fittings sold,” while leaving McWane alone to sell the remaining 20%. But even if McWane had good business reasons to adopt such a strategy, and such conduct could result in increased efficiency in the right market conditions, McWane offers no reasons to think that such conditions exist in this case. As the Commission noted, a full-line supplier like McWane could instead compete “by lowering its price for [the more common] products and increasing its price for the less common products.” Id. at *31. Again, McWane has not explained why such a strategy would not work, how the collapse of the full line of products would harm consumers, or why full-line forcing was instead necessary. Thus, this argument is also unpersuasive.

Moreover, McWane’s internal documents belie the notion that .the Full Support Program was designed for any pro-competitive benefit. As the Commission noted, McWane executives discussed the Full Support Program in terms of maintaining domestic prices and profitability by preventing Star from becoming an' effective competitor. For example, McWane executive Richard Tatman said that his “chief concern” with Star becoming a domestic fittings supplier was that “the domestic market [might] get[ ] creamed from a pricing standpoint,” and identified the biggest risk factor of Star’s entry as the “[e]rosion of domestic pricing if Star emerged as a legitimate competitor.” In a document encouraging the adoption of an exclusive dealing arrangement, Tatman opined that not doing so would allow Star to “drive profitability out of our business.” And in an e-mail, he stated, with regard to Star, “we need to make sure that they, don’t reach any critical mass that will allow them to continue to invest and receive a profitable return.” The Supreme Court has looked to evidence that proffered justifications for conduct “are merely ... an excuse to cover up different and anticompetitive reasons.” Jacobson, supra, at 367-68 (citing Eastman Kodak, 504 U.S. at 483, 112 S.Ct. 2072). McWane’s damning internal documents seem to be powerful evidence that its procompetitive justifications are “merely pretextual.”

IV.

All told, the Commission’s factual and economic conclusions are supported by substantial evidence and its legal conclusions comport with the governing law. The Commission’s determination of the relevant market and its findings of monopoly power and anticompetitive harm pass our deferential review, and we agree that the conduct amounts to a violation of Section 5 of the Federal Trade Commission Act.

Accordingly, we AFFIRM.

1 . In particular, the American Recovery and Reinvestment Act of 2009 ("ARRA”), Pub.L. No. 111-5, 123 Stat. 115, provided more than $6 billion to fund water infrastructure projects, all with domestic-only specifications. Pennsylvania and New Jersey state laws also require domestic materials in public projects, as do Air Force bases, certain federal programs, and various municipalities. See, e.g., 73 P.S. § 1884, 1886; N.J. Stat. Ann. § 52:33-3; McWane, Inc. (McWane I), 155 F.T.C. 903, 994-95 (2013).

2 . McWane emphasizes that the policy deliberately used the words “may” and “or” to convey "a weak stance.” However, McWane’s Vice President and General Manager Richard Tatman recognized that “[a]lthough the words ‘may’ and ‘or’ were specifically used, the market has interpreted the communication in the more hard line ‘will’ sense.”

3 . McWane maintains that this was the only example of the Full Support Program’s enforcement: “McWane. never enforced the rebate program against any other distributor.” Of course, the goal of the program was not necessarily to enforce the punishments but to dissuade customers from leaving McWane in the first place.

4 . In a series of events irrelevant to the resolution of this appeal, Sigma entered the domestic fittings market as an authorized distributor of McWane’s domestic fittings. See McWane II, 2014 WL 556261, at *10-11.

5 . Counts 1, 2, and 3 alleged an earlier conspiracy among McWane, Sigma, and Star to stabilize prices in the non-domestic fittings market. Counts 4 and 5 alleged that McWane's distribution agreement with Sigma violated the Federal Trade Commission Act. Count 7 alleged that the same conduct targeted in Count 6 amounted to attempted monopolization.

6 . The ALJ dismissed counts 1-3 but ruled in favor of the complaint counsel on counts 4-7.

7 . The Commission dismissed the other six counts. As to Count 7, attempted monopolization, the Commission deemed it "unnecessary to ask whether McWane attempted to monopolize the market” since it had found that McWane had actually done so. McWane II, 2014 WL 556261, at *31 n. 16.

8 . Former FTC Commissioner Rosch — whom Commissioner Wright replaced on the Commission in January 2013 — had issued similar criticisms in his dissents at both the pleading and summary judgment stages of the case.

9 . In Bonner v. City of Prichard, 661 F.2d 1206, 1209 (11th Cir.1981) (en banc), this Court adopted as binding precedent all decisions of the old Fifth Circuit handed down prior to October 1, 1981.

A recent opinion of this Court stated that we review the FTC’s finding of market definition for "clear error.” Polypore Int’l, 686 F.3d at 1217. Clear error is the traditional standard used to review a district court's factual findings, and we employ it in reviewing a finding of market definition by a district court judge. See, e.g., United States v. Engelhard Corp., 126 F.3d 1302, 1305 (11th Cir.1997); Cable Holdings of Ga., Inc. v. Home Video, Inc., 825 F.2d 1559, 1563 (11th Cir.1987); Nat’l Bancard Corp. (NaBanco) v. VISA U.S.A., Inc., 779 F.2d 592, 604 (11th Cir.1986). Poly-pore drew its "clear error” language from just such a case. 688 F.3d at 1217 (citing Engelhard, 126 F.3d at 1305). But substantial evidence, not clear error, is the "traditional ... standard used by courts to review agency decisions.” Am. Tower LP v. City of Huntsville, 295 F.3d 1203, 1207 (11th Cir.2002). Indeed, Polypore itself noted the correct standard of review for the FTC's factual findings earlier in the opinion. See 686 F.3d at 1213.

Other circuits follow this distinction, reviewing the FTC's market definition finding for substantial evidence while reviewing a district court’s market definition finding for clear error. Compare, e.g., Olin Corp. v. FTC, 986 F.2d 1295, 1297-98 (9th Cir.1993) (reviewing FTC’s market definition for substantial evidence), and ProMedica Health Sys., Inc. v. FTC, 749 F.3d 559, 566 (6th Cir.2014) (same), petition for cert. filed, No. 14-762 (Dec. 30, 2014), with, e.g., JBL Enters., Inc. v. Jhirmack Enters., Inc., 698 F.2d 1011, 1016 (9th Cir.1983) (reviewing district court’s market definition for clear error), and United States v. Cent. State Bank, 817 F.2d 22, 24 (6th Cir.1987) (per curiam) (same).

Moreover, Polypore’s language cannot be squared with the old Fifth Circuit's approach in Jim Walter. In that case, the Court asked "whether there is substantial evidence to support the FTC's finding of a national market for tar and asphalt roofing products.” 625 F.2d at 683. After determining that the FTC's market definition was founded "primarily on the casual observations of industry representatives and an economist,” the Court held that the FTC’s proposed market was "not supported by substantial evidence” and remanded "for reconsideration of the appropriate ... market.” Id. Jim Walter plainly held that the FTC’s market definition is reviewed for substantial evidence. Although Polypore may be read to say otherwise, in the case of an intracircuit conflict, the earlier case is binding. See Monison v. Amway Corp., 323 F.3d 920, 929 (11th Cir.2003).

10 . The Commission acknowledged that violations of Section 2 of the Sherman Act (monopolization) also constitute "unfair methods of competition” under Section 5 of the Federal Trade Commission Act, and therefore relied on Section 2 caselaw in its analysis. See McWane II, 2014 WL 556261, at *11 n. 7 (citing Cal. Dental Ass’n v. FTC, 526 U.S. 756, 762 & n. 3, 119 S.Ct. 1604, 143 L.Ed.2d 935 (1999); FTC v. Motion Picture Adver. Serv. Co., 344 U.S. 392, 394-95, 73 S.Ct. 361, 97 L.Ed. 426 (1953)); see also William Holmes & Melissa Mangiaracina, Antitrust Law Handbook § 7:2 (2014) (“For the most part ... the [Federal Trade Commission Act] has been held coterminous with the Sherman and Clayton Acts.”). Both parties (and the dissenting Commissioner) agree that this is the correct analytical approach.

11 . It is worth noting, however, that the defendant in Tops Markets had a lower market share than McWane — 74% as opposed to over 90% — and the plaintiffs "failed to produce any ... evidence to rebut [the defendant's] assertion” that the market contained no barriers to entry. 142 F.3d at 99. In this case, as we noted, the complaint alleged and the Commission found significant entry barriers.