Pulse Network v. Visa

Case: 18-20669 Document: 00516267971 Page: 1 Date Filed: 04/05/2022
United States Court of Appeals
for the Fifth Circuit United States Court of Appeals
Fifth Circuit
FILED
April 5, 2022
No. 18-20669
Lyle W. Cayce
Clerk
Pulse Network, L.L.C.,
Plaintiff—Appellant,
versus
Visa, Incorporated,
Defendant—Appellee.
Appeal from the United States District Court
for the Southern District of Texas
USDC No. 4:14-CV-3391
Before Smith, Willett, and Duncan, Circuit Judges.
Stuart Kyle Duncan, Circuit Judge:
This appeal concerns an antitrust dispute between Pulse and Visa,
competitors in the multi-billion-dollar debit network market. After litigation
had been dawdling for years, the district court dismissed Pulse’s Sherman
Act claims against Visa for lack of antitrust standing. We reverse in part,
remand for further proceedings, and direct reassignment to a different judge.
I. Background
First, a brief sketch of the debit network market (infra I.A), Visa’s
challenged policies (infra I.B), and the district court proceedings (infra I.C).
Case: 18-20669 Document: 00516267971 Page: 2 Date Filed: 04/05/2022
No. 18-20669
A. The Debit Network Market
1. The Market Structure
To pay for breakfast at the local coffee shop, you swipe (or tap) your
debit card. So begins an invisible process that transfers your money to the
shop. The electronic architecture that makes this possible is a “debit
network.” This diagram shows roughly how it works:
Located at the central hub of the diagram, the debit network links the
merchant’s bank (or “acquirer”) with the cardholder’s bank (or “issuer”).
Data races back and forth between acquirer and issuer. If the issuer approves
the transaction, the price of breakfast zips from your account to the coffee
shop’s.
There are two kinds of debit networks. A “PIN network” is used
when you complete a sale by punching in your personal identification
number. A “signature network” is used when you sign your name. Nearly all
debit cards enable one signature network and at least one PIN network. 1
Notably, though, the line between the two kinds of networks has blurred:
1
The network logos appear on the back of your card.
2
Case: 18-20669 Document: 00516267971 Page: 3 Date Filed: 04/05/2022
No. 18-20669
companies have developed “PINless” technology that lets PIN networks
process sales that would otherwise route through signature networks.
Debit networks are not free. Two kinds of fees are collected on every
transaction. First, debit network companies collect “network fees,” which
are their primary revenue. These are paid by both merchants and issuers.
They are typically low—averaging a few cents per transaction—and slightly
higher for signature than for PIN networks. Second, issuers collect
“interchange fees” from merchants’ banks. These make up the largest
portion of the prices merchants pay for debit transactions. 2 Both kinds of fees
are big business. In 2019, issuers and merchants paid $2.94 billion and $5.32
billion, respectively, in network fees, and issuers received $24.31 billion in
interchange fees. 3
The debit network market is “two-sided,” meaning debit network
companies compete for business from both merchants and issuers. Issuers
choose which PIN and signature networks to enable on cards; merchants
choose which of those networks to route sales over. Thus, debit network
companies compete by (1) convincing issuers to include their networks on
cards and (2) convincing merchants to route sales over their networks.
Success means pleasing both sides, because effects on one side ripple over to
the other. If a network’s fees go up, issuers may not choose it, lowering that
network’s value to merchants. If in turn merchants opt not to use that
network, it has even less value to issuers, triggering “a feedback loop of
declining demand.” Ohio v. Am. Express Co., 138 S. Ct. 2274, 2281 (2018).
2
See Notice of Proposed Rulemaking, Debit Card Interchange Fees and Routing,
75 Fed. Reg. 81,722, 81,723–24 (Dec. 28, 2010); Final Rule, Debit Card and Interchange
Fees and Routing, 76 Fed. Reg. 43,394, 43,396 (July 20, 2011).
3
See Board of Governors of the Federal Reserve System, 2019 Interchange Fee
Revenue, Covered Issuer Costs, and Covered Issuer and Merchant Fraud Losses Related
to Debit Card Transactions (May 2021) at 12.
3
Case: 18-20669 Document: 00516267971 Page: 4 Date Filed: 04/05/2022
No. 18-20669
2. The Market Players
Having sketched the market, we bring in the relevant players: Visa and
Pulse. Both operate debit networks. Pulse has a PIN network; Visa has a
signature network (“Visa Debit”) and a PIN network (“Interlink”). Both
companies have also developed PINless options: Pulse’s “Pulse Pay
Express” and Visa’s “PAVD” (short for “PIN-authenticated Visa Debit”).
The signature debit network market is dominated by Visa and
Mastercard, which are the signature network on 99% of debit cards. Of the
two, Visa is the bigger dog, currently with a 70–75% share of all signature
network transactions. The PIN debit network market is more crowded. It
includes not only Interlink and Pulse but also Maestro, STAR, NYCE,
ACCEL, and Shazam, among others.
Federal law affects the debit network market. The “Durbin
Amendment” to the 2010 Dodd-Frank Act regulates the market in two ways
relevant here. 4 First, the Amendment forces issuers to enable at least two
unaffiliated debit networks on all debit cards. See 15 U.S.C. § 1693o-
2(b)(1)(A); 12 C.F.R. § 235.7(a)(1). For Visa-branded debit cards (on which
Visa’s signature network is enabled), this means issuers must enable at least
one non-Visa PIN network on each card. Second, the Amendment gives
merchants total autonomy to choose which debit network to route
transactions over. See 15 U.S.C. § 1693o-2(b)(1)(B); 12 C.F.R. § 235.7(b). 5
4
See Dodd-Frank Wall Street Reform and Consumer
Protection Act, Pub. L. No. 111–203, 124 Stat. 1376, § 1075 (2010); 15 U.S.C. §§ 1693et seq. See also generally NACS v. Bd. of Gov. of Fed. Reserve Sys., 746 F.3d 474, 479–81 (D.C.
Cir. 2014); TCF Nat’l Bank v. Bernanke, 643 F.3d 1158, 1164–65 (8th Cir. 2011) (discussing
Durbin Amendment).
5
Before the Durbin Amendment, Visa had “all-Visa” exclusive arrangements with
issuers where the only networks enabled on a Visa debit card were Visa’s signature network
and Interlink. By 2010, Visa processed around 45% of all PIN debit network transactions in
the United States.
4
Case: 18-20669 Document: 00516267971 Page: 5 Date Filed: 04/05/2022
No. 18-20669
B. Visa’s Alleged Anti-Competitive Actions
In response to the Durbin Amendment, Visa made certain changes to
its policies relevant here: PAVD, FANF, and volume-based agreements.
First, Visa instituted its PAVD program. This requires issuers to
enable Visa’s PAVD technology (i.e., Visa’s PINless system) on all Visa debit
cards they issue. This guarantees that Visa can compete for PIN transactions
on every Visa-branded card, even if the issuer has not enabled Interlink
(Visa’s PIN network) on that card.
Second, Visa instituted the “Fixed Acquirer Network Fee”
(“FANF”). Instead of charging merchants only a per-transaction fee, Visa
began charging them 6 a fixed monthly fee for using its debit networks.
Merchants must pay this up-front fee so long as they accept payment from
any Visa product during the month. Visa continued to charge per-transaction
fees, but they were substantially reduced from previous levels. Given the
incentives created by this new pricing structure and Visa’s market
dominance, Pulse claims the FANF has these effects: (1) merchants can’t
refuse to pay the fixed monthly fee because, realistically, they can’t stop
accepting Visa cards, and (2) to recoup the fixed fee, merchants must route
debit transactions through Visa’s networks, which charge lower per-
transaction fees than do Visa’s rivals.
Third, Visa entered various volume-based agreements with issuers
and merchants. These agreements offer incentives to merchants to route a
certain number of transactions each month over Visa’s networks. Similarly,
Visa offers incentives to issuers—“rebates, discounts, and other
incentives”—if certain numbers of transactions occur on Visa networks each
month.
6
The up-front fee is actually charged to acquirers (merchants’ banks). But they
pass the cost along to merchants.
5
Case: 18-20669 Document: 00516267971 Page: 6 Date Filed: 04/05/2022
No. 18-20669
C. Pulse’s Lawsuit
Pulse sued Visa in 2014, alleging the three policies just described
violate federal and state antitrust statutes. 7 The case was assigned to Judge
Lynn Hughes of the Southern District of Texas. There, the case languished
for four years. In 2017, despite the fact that little discovery had been allowed,
Visa moved for summary judgment on both the merits and antitrust standing.
About a year later, the district court held Pulse lacked antitrust standing and
dismissed the case. The court’s terse decision appeared to rest on three
holdings.
First, the court concluded Pulse had suffered no injury-in-fact. It
reasoned that “[e]ven if Visa stopped using [the challenged strategies], Pulse
would not necessarily win more business.” It noted that “Mastercard, a
major market participant second only to Visa, has adopted a pricing structure
like Visa’s,” and that “[t]he rise of fixed fees would not stop if Visa were
barred from having them.” Second, the court held Pulse did not suffer an
antitrust injury. It reasoned that any injury inflicted by Visa’s policies was felt
by merchants and issuers, not Pulse, and that Visa’s policies increased
competition rather than harmed it. Third, the court appeared to hold that
Pulse was too remote a plaintiff. In its view, because merchants, issuers, and
acquirers were the parties potentially harmed by Visa’s conduct, “[t]hey are
better and more directly positioned to challenge Visa if they think that this
conduct violates the antitrust laws.”
Pulse timely appealed. Oral argument was first heard by a panel on
October 9, 2019. Following argument, one judge recused. The case was
7
Specifically, Pulse brought claims of monopolization and attempted
monopolization under § 2 of the Sherman Act, 15 U.S.C. § 2; claims of restraint of trade,
exclusive dealing, and illegal tying under § 1 of the Sherman Act; claims of tortious
interference with prospective business relationships under Texas law; and claims under the
Texas Free Enterprise and Antitrust Act.
6
Case: 18-20669 Document: 00516267971 Page: 7 Date Filed: 04/05/2022
No. 18-20669
reassigned to a different panel, which—following delays caused by the
pandemic and a hurricane—heard argument on January 5, 2022.
II. Antitrust Standing
On appeal, Pulse argues the district court erred in granting summary
judgment based on Pulse’s lack of antitrust standing. We review summary
judgments de novo. In re La. Crawfish Producers, 852 F.3d 456, 462 (5th Cir.
2017); see Fed. R. Civ. P. 56(a).
The Clayton Act provides that “any person who shall be injured in his
business or property by reason of anything forbidden in the antitrust laws may
sue therefor in any district court of the United States.” 15 U.S.C. § 15(a).
The Supreme Court has read this language to impose on antitrust plaintiffs
threshold requirements that go beyond Article III standing. See Atl. Richfield
Co. v. USA Petroleum Co. (ARCO), 495 U.S. 328, 334 (1990) (discussing
precedents); see also 2A Philip E. Areeda et al., Antitrust Law
§ 335 (4th ed. 2014) (“Antitrust standing . . . requires more than the
constitutional minimum for the ‘case or controversy’ that brings jurisdiction
to Article III courts.”). Our precedents distill those requirements to three
elements: “1) injury-in-fact, an injury to the plaintiff proximately caused by
the defendants’ conduct; 2) antitrust injury; and 3) proper plaintiff status,
which assures that other parties are not better situated to bring suit.” Doctor’s
Hosp. of Jefferson, Inc. v. Se. Med. All., Inc., 123 F.3d 301, 305 (5th Cir. 1997)
(citing McCormack v. NCAA, 845 F.3d 1338, 1341 (5th Cir. 1988)); see also
Areeda § 335c, at 77.
The parties primarily debate the second element, antitrust injury,
which describes
injury of the type the antitrust laws were intended to prevent
and that flows from that which makes defendants’ acts
unlawful. The injury should reflect the anticompetitive effect
either of the violation or of anticompetitive acts made possible
7
Case: 18-20669 Document: 00516267971 Page: 8 Date Filed: 04/05/2022
No. 18-20669
by the violation. It should, in short, be “the type of loss that the
claimed violations . . . would be likely to cause.”
Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977) (ellipsis
in original) (quoting Zenith Radio Corp. v. Hazeltine Rsch., Inc., 395 U.S. 100,
125 (1969)). Antitrust injury fleshes out the basic idea that “[t]he antitrust
laws were enacted for the protection of competition, not competitors.” ARCO,
495 U.S. at 338 (quotation marks omitted).
“At its most fundamental level, the antitrust injury requirement
precludes recovery for losses resulting from competition, even though such
competition was actually caused by conduct violating the antitrust laws.”
Areeda § 337a, at 102. The premise is that marketplace conduct can
simultaneously impair and enhance competition. See ibid. “Conduct in
violation of the antitrust laws may have three effects, often interwoven: In
some respects the conduct may reduce competition, in other respects it may
increase competition, and in still other respects effects may be neutral as to
competition.” ARCO, 495 U.S. at 343–44 (cleaned up). So, we must isolate
which aspect of the defendant’s allegedly illegal conduct adversely affected
the plaintiff. See id. at 342–44; Brunswick, 429 U.S. at 488. Even if a
defendant’s conduct violates the antitrust laws—and hence carries certain
anticompetitive effects—a given plaintiff lacks antitrust standing unless its
asserted injury reflects “an anticompetitive aspect of the defendant’s
conduct.” ARCO, 495 U.S. at 339 (emphasis omitted).
The district court found Pulse failed to show antitrust standing as to
each of the challenged Visa policies—PAVD, FANF, and volume-based
agreements. We therefore address antitrust standing separately as to each
policy. In doing so, we assume arguendo that each policy violates the antitrust
laws. See Sanger Ins. Agency v. HUB Int’l, Ltd., 802 F.3d 732, 738 (5th Cir.
2015) (“In analyzing this [antitrust] standing issue, we assume that
[plaintiff’s] allegations . . . amount to an antitrust violation.”) (citing
Doctor’s Hosp., 123 F.3d at 306; 2A Areeda § 335f, at 91).
8
Case: 18-20669 Document: 00516267971 Page: 9 Date Filed: 04/05/2022
No. 18-20669
A. PAVD
Pulse contends it has antitrust standing to contest Visa’s PAVD
program. Pulse alleges the program is an illegal tying arrangement that
requires issuers to enable PAVD (and thereby Visa PIN transactions) on any
Visa signature debit card. 8 As a result, Pulse can no longer be the exclusive
PIN network on Visa cards, and, as merchants choose to route PIN
transactions via Visa, Pulse loses transaction volume and revenue. We
disagree with Pulse.
Before PAVD, Visa debit cards usually included one signature
network and one PIN network. Visa reserved the signature slot for itself and,
in compliance with the Durbin Amendment, reserved the PIN slot for a
nonaffiliate. By obtaining exclusive placement on Visa debit cards as the sole
PIN network, Pulse benefited from that effective exclusion of Visa from the
PIN network market. But the PAVD program gives merchants a competing
option. Whereas Pulse previously was the only PIN network on Visa
signature debit cards, PAVD now guarantees merchants the choice of routing
PIN transactions via Pulse’s or Visa’s network. As merchants choose Visa’s
over Pulse’s, Pulse loses PIN debit volume and revenue.
This brings us to the core of Pulse’s alleged injury: merchants, when
given the option of Visa (through PAVD) or Pulse, are choosing Visa. Pulse,
understandably, would prefer that merchants be denied that choice. Antitrust
law does not assist Pulse in achieving that goal.
8
An illegal tying arrangement is one where the seller “exploit[s] . . . its control over
the tying product to force the buyer into the purchase of a tied product that the buyer either
did not want at all, or might have preferred to purchase elsewhere on different terms.” Ill.
Tool Works, Inc. v. Indep. Ink, Inc., 547 U.S. 28, 34–35 (2006) (internal citation omitted); see
also Areeda § 340c2, at 170 (“A dominant seller can exploit its market power directly by
charging a price higher than the competitive price would be, or indirectly by forcing the
buyer to buy a second product. The seller may have reasons to prefer the second route, just
as society may choose to condemn it as an unlawful tie, because it ‘introduces an alien
factor’ into competition among rival producers of that second product.”).
9
Case: 18-20669 Document: 00516267971 Page: 10 Date Filed: 04/05/2022
No. 18-20669
Loss from competition itself—that is, loss in the form of customers’
choosing the competitor’s goods and services over the plaintiff’s—does not
constitute antitrust injury, even if the defendant is violating antitrust laws in
order to offer customers that choice. See Brunswick, 429 U.S. at 487–88. A
plaintiff that sues a rival, complaining that the rival’s mere presence in the
market causes it injury, seeks to gain not the opportunity to compete in the
marketplace but only “the benefits of increased concentration.” Id. at 488.
Such a plaintiff seeks not “to share shelf space with its competitor” but to
have “that shelf space all to itself.” NicSand, Inc. v. 3M Co., 507 F.3d 442,
454 (6th Cir. 2007) (en banc). To be sure, the defendant might have violated
the antitrust laws to place itself on the shelf next to the plaintiff, but it would
be “inimical to the purposes of [the antitrust] laws” to recognize the plaintiff
as being injured by the defendant’s presence on that shelf. Brunswick, 429
U.S. at 488. Pulse has therefore not shown antitrust injury here. 9
Pulse counters that its loss of exclusive-dealing arrangements can
constitute antitrust injury because exclusive dealing may be the only way for
non-dominant firms, such as Pulse, to compete. We disagree. Pulse cites
multiple cases to support its “loss-of-exclusivity” theory of injury. 10 But
those cases teach only the well-established proposition that exclusive-dealing
arrangements are not per se antitrust violations. 11 Whether exclusive-dealing
9
To be sure, Pulse has shown injury-in-fact. It claims Visa’s conduct caused it to
lose PIN debit volume and revenue, and that Visa impeded its efforts to compete with its
PINless products. These allegations of economic injury establish injury in fact. See, e.g.,
Energy Mgmt. Corp. v. City of Shreveport, 397 F.3d 297, 302 (5th Cir. 2005). But Pulse still
lacks antitrust standing because of the lack of antitrust injury. ARCO, 495 U.S. at 343–44.
10
E.g., Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 334 (1961); FTC v.
Motion Picture Advert. Serv. Co., 344 U.S. 392, 396 (1953); Hornsby Oil Co. v. Champion
Spark Plug Co., 714 F.2d 1384, 1392 n.6 (5th Cir. 1983).
11
See Tampa Elec., 365 U.S. at 327 (“In practical application, even though a
contract is found to be an exclusive-dealing arrangement, it does not violate [antitrust law]
unless the court believes it probable that performance of the contract will foreclose
competition in a substantial share of the line of commerce affected.”); Motion Picture
Advert., 344 U.S. at 395-96 (recognizing that exclusive-dealing agreements are not per se
10
Case: 18-20669 Document: 00516267971 Page: 11 Date Filed: 04/05/2022
No. 18-20669
arrangements are legal is a question separate from whether conduct that
limits exclusivity, like Visa’s here, causes antitrust injury. In this case, the
answer is no.
Neither does the calculus change if we construe Pulse’s injury as the
loss of the ability to negotiate for exclusivity instead of the loss of exclusivity
itself. True, Pulse is not exactly suing to deny Visa participation in the market
for PIN transactions—even if Pulse’s suit were successful, Visa could still
offer issuers incentives to enable PAVD on Visa debit cards, and Pulse
presumably would offer competing incentives. And it might be, as Pulse
claims, that “many issuers would prefer not to enable PAVD” because the
associated transaction fees, though lower for merchants, are higher for
issuers. Nevertheless, the injury to Pulse—as distinguished from any possible
injury to issuers—is, ultimately, a loss of transaction volume for having to
compete with Visa for merchant transactions. 12 That kind of loss is not for
antitrust laws to remedy. Brunswick, 429 U.S. at 489.
Perhaps exclusive dealing is the only way Pulse can facilitate its
expansion as a non-dominant firm. But antitrust law wasn’t made to help a
smaller firm expand where competition limits its ability to do so on its own. 13
Congress may enact legislation—such as the Durbin Amendment—
specifically to assist smaller firms, but it is not for the courts to retrofit
antitrust violations); Hornsby, 714 F.2d at 1392 n.6 (“Exclusive dealing arrangements have
not received the more stringent per se treatment.”).
12
This also suggests that parties other than Pulse are better situated to bring suit
and that Pulse therefore lacks “proper plaintiff status.” See ARCO, 495 U.S. at 345–46
(noting that “a competitor will be injured and hence motivated to sue only when a vertical,
maximum-price-fixing arrangement has a procompetitive impact on the market,” so the
competitor’s suit “would not protect the rights of dealers and consumers under the
antitrust laws”).
13
See Cargill, Inc. v. Monfort of Colo., Inc., 479 U.S. 104, 116 (1986) (“[I]t is in the
interest of competition to permit dominant firms to engage in vigorous competition . . . .”).
11
Case: 18-20669 Document: 00516267971 Page: 12 Date Filed: 04/05/2022
No. 18-20669
antitrust law to further such goals. 14 Even assuming Visa’s PAVD program is
an illegal tie, Pulse’s injury—decreased PIN debit volume and revenue as
merchants choose Visa over Pulse—results from increased competition and
is therefore not antitrust injury.
B. FANF
Pulse also argues it has antitrust standing to contest FANF. It alleges
the FANF pricing structure has caused merchants to use its debit network
less, decreasing Pulse’s revenue. Visa orchestrates this injury, Pulse claims,
in two integrated steps. First, Visa uses its market dominance to foist on
merchants a high fixed fee they wouldn’t ordinarily accept. Second, Visa then
uses the revenues from that unavoidable upfront fee to artificially lower its
per-transaction fees, which effectively forecloses rivals like Pulse from
competing. Visa responds that Pulse is really harmed only by the increased
competition created by FANF (i.e., cheaper per-transaction fees), rather than
some anticompetitive aspect of the pricing structure. And injury from
increased competition, Visa reminds us, is no concern of the antitrust laws.
We agree with Pulse.
Visa might have a point if Pulse were complaining only that Visa had
slashed its per-transaction prices. See, e.g., Felder’s Collision Parts, Inc. v. All
Star Advert. Agency, Inc., 777 F.3d 756, 760–61 (5th Cir. 2015) (“Low prices
benefit consumers and are usually the product of the competitive
marketplace that the antitrust laws are aimed at promoting.” (citing Brooke
Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 223 (1993))).
Pulse claims more than price competition is afoot, though. After the Durbin
Amendment loosened Visa’s grip on the debit network market, Visa began
shedding merchants to Pulse and other networks because its pricing wasn’t
14
See Brunswick, 429 U.S. at 488 (“Congress is free, if it desires, to mandate
damages awards for all dislocations caused by unlawful mergers despite the peculiar
consequences of so doing.”).
12
Case: 18-20669 Document: 00516267971 Page: 13 Date Filed: 04/05/2022
No. 18-20669
competitive on a per-transaction basis. Instead of improving its product or
competing on price, however, VISA began charging the FANF to
merchants—and then using some of those revenues to reduce per-
transaction fees. This integrated fee structure, argues Pulse, forces
merchants to pay a higher total cost (fixed plus per-transaction fees) than
before, and yet Visa’s market share and profits have recovered.
This alleged scheme inflicts antitrust injury on Pulse. Under Pulse’s
theory, it doesn’t lose customers to Visa in a fair fight over per-transaction
fees. Rather, Pulse loses customers because Visa abuses its dominance in the
debit card market. Merchants have no choice but to pay Visa’s high fixed
monthly fee. They recoup that expense by routing more transactions through
Visa’s network, which charges lower per-transaction fees than competitors.
But Visa can achieve that only by leveraging the upfront fees to artificially
deflate its per-transaction fees. We must assume this pricing structure
violates the antitrust laws. See Sanger Ins. Agency, 802 F.3d at 738; Doctor’s
Hosp., 123 F.3d at 306. When we do, the link between Pulse’s injury and
Visa’s alleged anticompetitive conduct becomes plain. Pulse is squeezed out
of the market because Visa exploits its dominance to impose supra-
competitive prices on merchants and simultaneously undercut competitors’
per-transaction fees. That is textbook antitrust injury. See Andrx Pharms., Inc.
v. Biovail Corp. Int’l, 256 F.3d 799, 816–17 (D.C. Cir. 2001) (“Irrespective of
consumer injury, an excluded competitor . . . suffers a distinct injury if it is
prevented from selling its product.”). 15
Visa’s counterarguments do not persuade us.
First, Visa argues that Pulse can’t show antitrust injury because
“Pulse does not contend that Visa’s lowered per-transaction fees are
15
Pulse obviously suffers injury-in-fact from the FANF, as it contributed to Pulse’s
losing volume and market share. These injuries are real harms that Visa allegedly intended
to inflict. Allegations of economic harm are enough to establish injury in fact. See supra note
9. The district court plainly erred to the extent it concluded otherwise.
13
Case: 18-20669 Document: 00516267971 Page: 14 Date Filed: 04/05/2022
No. 18-20669
predatory” and “injuries that flow from non-predatory price cuts are not
antitrust injuries.” For this argument, Visa relies heavily on the Supreme
Court’s ARCO decision. See 495 U.S. at 340. It quotes the Court’s
statements that “[l]ow prices benefit consumers regardless of how those
prices are set” and that “[w]hen prices are not predatory, any losses flowing
from them cannot be said to stem from an anticompetitive aspect of the
defendant’s conduct.” Id. at 340–41.
Visa’s argument misperceives Pulse’s antitrust claim. Pulse isn’t
complaining about low prices but about high prices—i.e., the supra-
competitive overall prices Visa can charge merchants by exploiting its market
dominance. To be sure, part of Visa’s scheme is to use the upfront fixed fee
to artificially deflate its per-transaction charges as to which it faces direct
competition. But, as Pulse points out, “that is just a manifestation of an
integrated strategy of using market and monopoly power to charge supra-
competitive prices.”
ARCO is inapposite. There, an oil company allegedly conspired with
its dealers to set maximum resale prices for gas. A competitor of those dealers
sued on the theory that this was a “vertical, maximum-price-fixing
agreement,” at the time a per se Sherman Act violation. 495 U.S. at 331–33. 16
The Supreme Court found the competitor lacked antitrust injury. Id. at 336–
38. The anticompetitive effects of the vertical agreement—while harmful to
the dealers bound by it and their consumers—were actually beneficial to the
competitor, which could undercut those dealers on prices or services. 17 The
Court also rejected the competitor’s alternative argument that the agreement
16
That is no longer the case. See State Oil Co. v. Khan, 522 U.S. 3, 7 (1997)
(overruling Albrecht v. Herald Co., 390 U.S. 145 (1968), and holding that an alleged vertical
maximum price-fixing agreement is subject to the rule of reason).
17
See id. at 336–37 (“Respondent was benefited rather than harmed if petitioner’s
pricing policies restricted ARCO sales to a few large dealers or prevented petitioner’s
dealers from offering services desired by consumers such as credit card sales.”).
14
Case: 18-20669 Document: 00516267971 Page: 15 Date Filed: 04/05/2022
No. 18-20669
injured it by setting prices too low. As the Court explained, antitrust injury
cannot be founded on a claim that firms have “lower[ed] prices but
maintain[ed] them above predatory levels.” Id. at 337. In other words, harm
from “nonpredatory price competition” does not arise from “an
anticompetitive aspect of the defendant’s conduct.” Id. at 338–39 (citing
Brunswick, 429 U.S. at 487); see also Felder’s, 777 F.3d at 760–62.
This context shows why Visa’s reliance on ARCO is unavailing. In
that case, antitrust injury was absent because the plaintiff competitor was not
harmed (and instead was benefited) by the anticompetitive aspects of the
alleged antitrust violation. Here, by contrast, Pulse is injured precisely by the
anticompetitive aspects of Visa’s conduct, i.e., the integrated FANF
structure that excludes Pulse from the market. Moreover, ARCO discussed
predatory pricing in the context of antitrust claims targeting the low prices
set by a price-fixing agreement. 495 U.S. at 338–41. Pulse, by contrast, isn’t
challenging FANF because it imposes low or below-cost pricing. Rather, it
argues that FANF abuses Visa’s market power, specifically by imposing
supra-competitive prices on merchants while manipulating prices in a way
that excludes competitors from the market.
Second, Visa argues we should disregard FANF’s integrated pricing
structure and instead treat the fixed fees and the per-transaction fees
separately. Visa relies on the statement in ARCO that antitrust injury must
be “attributable to an anti-competitive aspect of the practice under scrutiny.”
ARCO, 495 U.S. at 334 (emphasis added) (citing Cargill, Inc. v. Monfort of
Colo., Inc., 479 U.S. 104, 109–10 (1986)). On this view, Pulse’s injury can be
attributed only to the low per-transaction fees—not to the fixed fees—and
hence only to the effects of price competition. We disagree.
The Supreme Court has time and again reminded us that analysis
“rest[ing] on formalistic distinctions rather than actual market realities are
generally disfavored in antitrust law.” Am. Express, 138 S. Ct. at 2285
15
Case: 18-20669 Document: 00516267971 Page: 16 Date Filed: 04/05/2022
No. 18-20669
(citation omitted). 18 So we cannot blind ourselves to the ample record
evidence that Visa created the FANF to function as an integrated program.
As Pulse puts it, “Visa’s fixed fees and per-transaction fees are two
components of a single integrated price structure that raises overall prices for
merchants while artificially deflating Visa’s per-transaction charges, where
Visa faces direct competition from Pulse and others.” Pulse’s claimed injury
stems directly from the combined effect of those two components—the fixed
fee allowing Visa to subsidize its per-transaction fee, imposing supra-
competitive overall costs on merchants while excluding competitors from the
market. To separate those components when assessing antitrust injury, as
Visa wants us to do, would falsify the “actual market realities” at play here.
Eastman Kodak Co. v. Image Tech. Servs., Inc., 504 U.S. 451, 466 (1992). We
won’t do that.
Third, Visa claims Pulse is not a proper plaintiff to challenge FANF
because merchants and issuers pay the FANF, not Pulse. We again disagree.
Antitrust standing requires “proper plaintiff status, which assures
that other parties are not better situated to bring suit.” Doctor’s Hosp., 123
F.3d at 305. This inquiry focuses on proximate causation. 19 Our circuit
considers factors such as (1) “whether the plaintiff’s injuries or their causal
link to the defendant are speculative”; (2) “whether other parties have been
more directly harmed”; and (3) “whether allowing this plaintiff to sue would
risk multiple lawsuits, duplicative recoveries, or complex damage
18
See also NCAA v. Alston, 141 S. Ct. 2141, 2158 (2021) (“Whether an antitrust
violation exists necessarily depends on a careful analysis of market realities.” (citations
omitted)); Doctor’s Hosp., 123 F.3d at 305 (explaining “antitrust injury for standing
purposes should be viewed from the perspective of the plaintiff’s position in the
marketplace”).
19
See, e.g., Lexmark Int’l, Inc. v. Static Control Components, Inc., 572 U.S. 118, 126
(2014) (citing Associated Gen. Contractors of Cal., Inc. v. Carpenters, 459 U.S. 519, 532–33
(1983)).
16
Case: 18-20669 Document: 00516267971 Page: 17 Date Filed: 04/05/2022
No. 18-20669
apportionment.” McCormack, 845 F.2d at 1341; see also Norris v. Hearst Trust,
500 F.3d 454, 465 (5th Cir. 2007).
Pulse is a proper plaintiff to challenge FANF. Pulse claims FANF
squeezes it out of the debit network market, reducing Pulse’s transaction
volume and market share. Based on the record, a reasonable jury could find a
non-speculative causal link between these claimed injuries and FANF. See
McCormack, 845 F.2d at 1341. 20 Moreover, Pulse’s claimed harm—being
driven from the market by FANF’s abusive structure—is distinct from any
increased costs FANF may visit on merchants or issuers. Those harms are
no more direct than the ones Pulse claims as an excluded competitor. See
ibid. 21 Finally, no merchant or issuer could recover for Pulse’s competitive
injuries, so there is no chance of duplicative recoveries. Ibid. 22
20
Conceding Pulse has lost volume and market share, Visa attributes those losses
to business failures unrelated to Visa’s conduct. Maybe, maybe not. But it is Visa which
moved for summary judgment, and so Pulse gets the benefit of all reasonable inferences
from the record. La. Crawfish Producers, 852 F.3d at 462. A reasonable jury could conclude
from the record that Visa’s policies deprived Pulse of the opportunity to compete for
business from at least one major merchant.
21
See also Norris, 500 F.3d at 467 (holding plaintiffs lacked standing because they
were “neither consumers nor competitors in the market attempted to be constrained”);
TCA Bldg. Co. v. Nw. Res. Co., 861 F. Supp. 1366, 1380 (S.D. Tex. 1994) (“As a competitor
for sales . . . in a market which the Defendants have allegedly monopolized, which has
allegedly lost sales due to the Defendants’ allegedly unlawful agreement to exclude
competitors, no party is in a better position to vindicate the purposes of the antitrust laws
than [the plaintiff].”).
22
That is, no merchant or issuer could recover from Visa for Pulse’s lost profits
and market share. See, e.g., Den Norske Stats Oljeselskap As v. Heeremac V.O.F., 241 F.3d
420, 438 (5th Cir. 2001) (Jones, J., dissenting) (stating that had the majority reached the
issue of antitrust standing, the plaintiff was a proper plaintiff because “[t]here is no
suggestion that any unnamed party can seek to recover for the same damages [the plaintiff]
suffered” (emphasis added)); see also Andrx Pharm., 256 F.3d at 817 (finding a competitor
had antitrust standing because his “alleged injury [was] not measured by or derived from”
the injury suffered by “consumer plaintiffs”).
17
Case: 18-20669 Document: 00516267971 Page: 18 Date Filed: 04/05/2022
No. 18-20669
C. Volume-Based Agreements
Finally, Pulse argues it suffers antitrust injury from Visa’s volume-
based routing agreements with merchants and issuers. These agreements,
Pulse alleges, are “designed to lock up the market and thereby protect Visa’s
lucrative signature debit business from competition from Pulse Pay Express
and other debit networks’ PINless products.” Pulse claims the agreements
thus constitute “exclusive-dealing or quasi-exclusive-dealing agreements,”
which Visa has employed to suppress competition and reduce Pulse’s market
share in PINless transactions. See, e.g., ZF Meritor, LLC v. Eaton Corp., 696
F.3d 254, 270 (3d Cir. 2012). 23 Pulse argues the district court erred in ruling
it lacks antitrust standing to challenge the volume-based agreements. We
agree.
As it did with respect to FANF, Visa argues that the agreements
merely amount to “non-predatory price competition.” See ARCO, 495 U.S.
at 340–41. That’s a merits question, however. At this stage we must assume
that Pulse will prove the agreements violate the antitrust laws as anti-
competitive exclusive-dealing arrangements. See Doctor’s Hosp., 123 F.3d at
306. Based on that assumption, Pulse has shown antitrust injury. Similar to
its claims against FANF, Pulse isn’t claiming that it’s losing a fair price war
against Visa. Instead, it’s claiming that Visa has used its market dominance
to strong-arm merchants into avoiding Pulse Pay Express.
Visa also makes the factual argument that its agreements with
merchants and issuers are “short term, freely terminable, contain[ ] no
penalties for non-performance, and impose[] no obligations or commitments
23
As the Third Circuit has explained, “[a]n exclusive dealing arrangement is an
agreement in which a buyer agrees to purchase certain goods or services only from a
particular seller for a certain period of time.” ZF Meritor, 696 F.3d at 270. Such
arrangements, while not always anti-competitive, “may be used by a monopolist to
strengthen its position, which may ultimately harm competition.” Ibid. (citing, inter alia,
Tampa Elec., 365 U.S. at 327–29; 11 Herbert Hovenkamp, Antitrust Law
¶ 1800a, at 3 (3d ed. 2011)).
18
Case: 18-20669 Document: 00516267971 Page: 19 Date Filed: 04/05/2022
No. 18-20669
on . . . merchants.” But the record reveals fact disputes on that point. For
instance, Pulse deposed an officer from Kroger, a major merchant, who
stated that Visa fined Kroger repeatedly for using competing PIN debit
networks instead of Visa’s signature debit network and threatened to revoke
Kroger’s ability to accept any Visa debit card. So, what to make of Visa’s
agreements with merchants and issuers is a fact question for a jury, not a
summary judgment issue for a court. And a reasonable jury could find that
some of Visa’s volume-based agreements amount to exclusive-dealing
contracts designed to squeeze Pulse out of the PINless transaction market. 24
III. Reassignment on Remand
Pulse asks us to reassign the case to a different judge on remand. Our
supervisory authority permits us to reassign cases, see 28 U.S.C. § 2106;
Liteky v. United States, 510 U.S. 540, 554 (1994), but this should be done
“infrequently and with great reluctance,” Miller v. Sam Houston State Univ.,
986 F.3d 880, 892 (5th Cir. 2021) (quoting United States v. Winters, 174 F.3d
478, 487 (5th Cir. 1999)); see also Johnson v. Sawyer, 120 F.3d 1307, 1333 (5th
Cir. 1997) (reassignment is “extraordinary” and “rarely invoked”) (citation
omitted). To assess whether to reassign, we consider three factors:
(1) whether the original judge would reasonably be expected
upon remand to have substantial difficulty in putting out of his
mind or her mind previously-expressed views or findings
determined to be erroneous or based on evidence that must be
rejected, (2) whether reassignment is advisable to preserve the
appearance of justice, and (3) whether reassignment would
24
That also shows why Pulse is injured-in-fact by the agreements and why it’s a
proper plaintiff to challenge them. Though the agreements are with merchants and issuers
(and so may harm them in some way), Pulse suffers distinct and direct harm because the
agreements are allegedly designed to hurt Pulse’s market share. See also 9 AREEDA § 1800,
at 10 (“If the exclusive arrangement is anticompetitive at all, it is because the arrangement
forecloses rivals from adequate sales outlets. Thus, the condition that makes tying or
exclusive dealing anticompetitive in the first place is that customers lack sufficient options
to purchase . . . elsewhere.”).
19
Case: 18-20669 Document: 00516267971 Page: 20 Date Filed: 04/05/2022
No. 18-20669
entail waste and duplication out of proportion to any gain in
preserving the appearance of fairness.
Miller, 986 F.3d at 892–93 (citation omitted). 25 Applying these factors, we
conclude reassignment is warranted.
Pulse’s overarching contention is that the district judge had pre-
judged the case against Pulse from the outset. This is a serious accusation,
but unfortunately there is record support for it. For example, at an initial
conference in 2015, the judge repeatedly insisted that the challenged Visa
policies did not harm competition and that merchants “were not forced to
pay” the FANF. These are some of the key disputed issues underlying
Pulse’s claims. Pulse also points out that the district judge candidly revealed
his disdain for antitrust law and antitrust plaintiffs. For instance, the judge
remarked that “there are more bad antitrust cases than any other single
category,” theorized that “[t]he only real monopolies are ones supported by
the government,” and suggested that the Standard Oil Company wasn’t a
real monopoly. Viewed in isolation, any one of these admittedly gratuitous
comments might be harmless. Taken together, however, they raise concerns
that the judge harbored ingrained skepticism about Pulse’s claims from the
jump.
What happened over the ensuing four years of proceedings only
sharpens those concerns. Most significantly, the district judge repeatedly
stymied Pulse’s legitimate requests to engage in critical discovery. As Pulse
points out, “four years in[to the litigation], Pulse ha[d] not been allowed to
25
Our cases also articulate a second, simpler test: a case should be reassigned
“when the facts might reasonably cause an objective observer to question the judge’s
impartiality.” In re DaimlerChrysler Corp., 294 F.3d 697, 701 (5th Cir. 2002) (cleaned up);
see also Miller, 986 F.3d at 893 (characterizing second test as “more lenient”). But the two
tests are “redundan[t]” because “the second factor of the first test is virtually identical to
the single question the simpler test asks.” Willey v. Harris Cty. Dist. Att’y, 27 F.4th 1125,
1137 (5th Cir. 2022) (quoting United States v. Khan, 997 F.3d 242, 249 n.4 (5th Cir. 2021)).
So, we needn’t apply the second test.
20
Case: 18-20669 Document: 00516267971 Page: 21 Date Filed: 04/05/2022
No. 18-20669
take any party discovery from Visa—no document requests, no
interrogatories, no depositions, nothing.” At least eight of Pulse’s requests
for party and non-party discovery were denied—including discovery directed
to the core issue of whether Visa was using FANF to subsidize its per-
transaction fees. 26 The judge also denied Pulse’s request to participate in
discovery in a related MDL involving Visa, even after Visa sought substantial
third-party discovery from Pulse in that MDL. Indeed, instead of allowing
Pulse to engage in discovery, the judge required Pulse to provide information
to Visa. 27 The sum total of this approach left Pulse, despite years of litigation,
without any discovery on aspects of Visa’s policies central to its case.
Finally, the district court’s substantive rulings lend further support to
Pulse’s arguments for reassignment. For instance, after Visa moved to
dismiss Pulse’s case in 2015, the district court took nine months to issue a
one-sentence order denying the motion. The order stated in full: “While the
complaint is not compellingly lucid, Pulse Network, LLC, has alleged
sufficient facts that probably adequately state a claim for relief.” Two years
later—despite the lack of meaningful discovery—Visa was allowed to move
for summary judgment on both the merits and antitrust standing. The court
then waited another ten months to resolve the motion. Its order consisted
in—to borrow from a previous case involving the same judge that was also
reassigned on remand—“a [seven]-page opinion with few citations to either
26
In one illustrative exchange, Pulse explained it needed discovery because “we
need to know exactly how Visa is implementing [the strategies] we’re complaining about.”
The court denied the request—first telling Pulse that it “has no evidence” that Visa uses
the FANF to finance discounts on per-transaction fees, and then refusing to allow Pulse
the chance to obtain that evidence: “No. You’re not going to get discovery to find out that
– You only suspect that it’s below cost.”
27
The district court suggested that Pulse should be required to produce unilateral
discovery because it is the plaintiff: “since [Pulse] brought the lawsuit, it’s going to have to
show Visa what it’s done before I make them reveal their records. They started it. They’re
going to have to take the lead in furnishing the data they have that reflects the injury they
say they inflicted.”
21
Case: 18-20669 Document: 00516267971 Page: 22 Date Filed: 04/05/2022
No. 18-20669
record evidence or relevant legal authority . . . consist[ing] almost entirely of
conclusory statements.” United States ex rel. Little v. Shell Expl. & Prod. Co.,
602 F. App’x 959, 975 (5th Cir. 2015) (unpublished).
In light of all this, our three-factor test counsels reassignment. First,
we conclude that “the [district] judge would likely have substantial difficulty
putting out of his mind his previously expressed views” concerning antitrust
law in general and Pulse’s claims in particular. Khan, 997 F.3d at 249.
Second, we find that “the appearance of justice has been compromised” by
the judge’s remarks and by the course of proceedings discussed above. Ibid.;
see also United States v. Varner, 948 F.3d 250, 256 (5th Cir. 2020) (“Federal
judges should always seek to promote confidence that they will dispense
evenhanded justice.” (citing Canon 2(A), Code of Conduct for United States
Judges)). While the third factor cautions against reassignment for fear of
“waste and duplication out of proportion to any gain in preserving the
appearance of fairness,” Johnson, 120 F.3d at 1333 (citation omitted), that
concern lacks traction here. As discussed, little discovery was allowed over
four years of litigation and the case has only now proceeded past standing.
Reassignment won’t make the new judge start over because even after so
much time the case has barely started. 28
28
We have reassigned this district judge’s cases before. See, e.g., Khan, 997 F.3d at
249 (reassignment where same judge sentenced defendant, was reversed, and on remand
“declined to reconsider the sentence in any respect, showing that he [wa]s adamant against
further consideration of the substance of the record”); Miller, 986 F.3d at 892–93
(prejudicial comments and peremptory rulings justified reassignment); United States v.
Swenson, 894 F.3d 677, 683, 685 (5th Cir. 2018) (reassignment warranted where judge did
not explain discovery rulings and attributed counsel’s mistakes to her sex); Shell, 602 F.
App’x at 975 (reassignment where judge ignored our instructions after an appeal, resulting
in a second appeal); Latiolais v. Cravins, 574 F. App’x 429, 437 (5th Cir. 2014) (concluding
judge’s comments demonstrated it would be exceedingly difficult for him to put aside the
views he expressed about the evidence).
22
Case: 18-20669 Document: 00516267971 Page: 23 Date Filed: 04/05/2022
No. 18-20669
IV. Conclusion
We REVERSE the summary judgment in part, REMAND the case
for further proceedings consistent with this opinion, and DIRECT the Chief
Judge of the Southern District of Texas to assign the case to a different
district judge.
23