Amicus Curiae Brief — Ohio v. Am. Express Co., 138 S. Ct. 355 (2017) (No. 16-1454)

Supreme Court brief2017

Ask Donna

What actually matters in this document.

Text

No. 16-1454

In the Supreme Court of the Hnited States

STATE OF OHIO, ET AL.,

Petitioners,

AMERICAN EXPRESS COMPANY, ET Al.,

Respondents.

ON WRIT OF CERTIORARI TO THE UNITED STATES COURT

OF APPEALS FOR THE SECOND CIRCUIT

BRIEF FOR DISCOVER FINANCIAL SERVICES

AS AMICUS CURIAE IN SUPPORT OF

PETITIONERS

ELIZABETH P. PAPEZ*

ANDREW C. NICHOLS

Winston & Strawn LLP

1700 K Street, N.W.

Washington, DC 20006

(202) 282-5000

epapez@winston.com

*Counsel of Record

Counsel for Amicus Curiae

QUESTION PRESENTED

This case asks how Section I of the Sherman Act,

which bans unreasonable restraints of trade, applies

to “two-sided” platforms that unite distinct customer

groups. Such platforms are ubiquitous, ranging from

eBay (serving buyers and sellers), to newspapers

(serving readers and advertisers). Here, credit-card

networks bring cardholder customers together with

merchant customers for ordinary transactions. When

doing so, Respondents American Express Company

and American Express Travel Related Services Com-

pany (“Amex”) contractually bar merchant customers

from steering cardholder customers to credit cards

that charge merchants lower prices. Applying the

“rule of reason,” the district court held that: (1) the

Government proved that Amex’s anti-steering provi-

sions were anticompetitive because they stifled com-

petition among credit-card companies for the prices

charged to merchants, and (2) Amex failed to estab-

lish any procompetitive benefits. The Second Circuit

reversed. It held that, to prove that the anti-steering

provisions were anticompetitive (and so to transfer

the burden of establishing procompetitive benefits to

Amex), the Government bore the burden to show not

just that the provisions had anticompetitive pricing

effects on the merchant side, but also that those anti-

competitive effects outweighed any benefits on the

cardholder side. The question presented is:

Under the “rule of reason,” did the Government's

showing that Amex’s anti-steering provisions stifled

price competition on the merchant side of the credit-

card platform suffice to prove anticompetitive effects

and thereby shift to Amex the burden of establishing

any procompetitive benefits from the provisions?

TABLE OF CONTENTS

Page

e RR CEN enna i

INTEREST OF AMICUS CURIAE. 1

r 13

A. Discover Offers Breakthrough Value

Propositions to Cardholders and

IIT —— p — ů ů ů ĩ 4

B. Discover Attempts to Compete By

Offering Merchants Lower Prices 5

C. The District Court Finds the NDPs

Unreasonably Restrain Trade..................... 9

D. The Court of Appeals Reinstates the

— —A—ͤ—1k ———— 13

SUMMARY OF ARGUMENT. . 15

rr 20

I. The Opinion Below Unjustifiably Departs From

This Court's Approach to Market Definition 20

Il. The Court of Appeals’ Market Definition

Distorts the Rule-of-Reason Inquiry and

Insulates an Admitted Restraint on Price

Competition from Antitrust Scrutiny.................. 25

IIL. The Decision Below Justifies Its Departure

from Settled Antitrust Principles on Grounds

That Violate the Sherman Act’s Central Tenet .. 28

— ——— —

ill

TABLE OF AUTHORITIES

Page(s)

Cases

All. Richfield Co. v. USA Petroleum Co,,

, .. 21, 29

Brown Shoe Co. v. United States,

. SSE 10, 21, 22, 23

Cal. Dental Ass n v. FTC,

/ / 9, 18

Cargill, Inc. v. Monfort of Colo., Inc.,

e aeictiesaipiasiunbiibaiinted 30

Catalano, Inc. v. Target Sales, Inc.,

e . 26

Eastman Kodak Co. v. Image Tech. Sucs., Inc.,

e ... 21. 23

F. J. C. v. Super. Ct. trial Lawyers Asen

e ......

N. Pac. Ry. Co. v. United States,

e .. 27

Vat ! Collegiate Athletic Ass n v. Bd. of

Regents of Univ. of Okla.,

| Ee ce 20, 25, 26

Nat'l Soc’y of Prof'l Eng’rs v. United States,

— „ — 20, 26, 29

State Oil Co. v. Khan,

522 U.S. 3 (1997) . . ..... .. - 3, 5, 19, 29

Sullivan v. NFL,

34 F.3d 1091 (1* Cir. 1994), cert.

denied, 513 U.S. 1190 (1995) ...r. 27

Times-Picayune Publ g Co. v. United States,

BAB U.S. 594 (1953) .. . . 22, 23

United States v. E. I. du Pont de Nemours & Co.

351 U.S. 377 (1956) . . . . e

United States v. Gen. Motors Corp.,

384 U.S. 127 (1966) .... . . . . . . 26

United States v. Phila. Nat! Bank,

374 U.S. 321 (1963) . . . . . . .. . . . . 27

United States v. Socony- Vacuum Oil Co.,

310 U.S. 150 (1940) . . . . . . 25

United States v. Topco Assocs., Inc.,

405 U.S. 596 (1972). . . . . 29

Statutes

16 U.. C. § 1. 1, 9, 13, 17, 18-21, 25-26, 29, 31

Other Authorities

Philip E. Areeda & Herbert Hovenkamp,

An Analysis of Antitrust Principles

and Their Application “|562(e),

(Supp. 2017). . . . .. ... 16, 31

INTEREST OF AMICUS CURIAE*

Discover Financial Services operates the Discover

payment network and, along with certain affiliates

and third parties, issues Discover-branded payment

cards to consumers. As detailed in the district court's

opinion, e.g., Pet. App. 70a-86a, Discover competes

directly with Respondents American Express Compa-

ny and American Express Travel Related Services

(collectively, “Amex”) on both sides of the “two-sided”

(merchant-cardholder) payment platform described in

the decisions below. Pet. App. 39a-40a. Discover

competes with Amex, Visa, and MasterCard in selling

network services to merchants and acquiring banks.

Pet. App. 70a, 117a. And Discover competes with

Amex and numerous Visa- and MasterCard-affiliated

banks in issuing payment cards to cardholders. Ibid.

Discover has a direct interest in this action be-

cause the opinion below reinstates Amex network

rules—known as “nondiscriminatory provisions“

(“NDPs”) or “anti-steering rules“ — that bar mer-

chants from steering transactions to Discover or other

payment card networks that offer merchants lower

transaction fees. EK. g., Pet. App. 100a-10 la, 203a-

207a. Discover's unrebutted trial testimony that the

NDPs thwarted Discover’s attempt to compete for

merchant business on the basis of price was central to

the district court’s conclusion that the NDPs violated

Section 1 of the Sherman Act. Pet. App. 197a, 203a-

205a. As Petitioners and the United States explain

* Pursuant to Rule 37.3(a), Discover has received consent for its

filing from all parties. In accordance with Rule 37.6, no counsel

for any party has authored this brief in whole or in part, and no

person or entity other than Discover has made a monetary con-

tribution to the preparation or submission of this brief.

in their briefs, Discover's experience “vividly illus-

trates” how the NDPs “block[] price competition” and

stifle innovatlion]' in payment systems. Brief for

the United States as Respondent Supporting Peti-

tioners at 32 (“U.S. Br.“) (citing Pet. App. 203a));

Brief for Petitioners (Pet. Br.“) at 46.

The opinion below does not contest these findings.

It reinstates the NDPs on the basis that the “District

Court’s erroneous market definition caused its anti-

competitive effects finding to come up short.” Pet.

App. 49a. The opinion below begins by emphasizing

that “cardholders and merchants * * * comprise dis-

tinct yet equally important and interdependent sets

of consumers sitting on either side of the payment-

card platform.” Pet. App. 50a. It then holds that the

district court “erred” in “declin{ing] to * * collapsſe]

the [cardholder] issuance and [merchant] network

services markets into a single platform-wide market

for transactions” on payment card networks. Pet.

App. 32a (internal quotation marks omitted). Adopt-

ing this new market definition, the court of appeals

held that the NDPs’ undisputed interference with

network price competition for merchant business did

not satisfy Plaintiffs’ “initial burden” of showing “an

actual adverse effect on competition as a whole in the

relevant market,” Pet. App. 49a-50a (internal quota-

tion marks omitted), because it did not “show that the

NDPs made all Amex consumers on both sides of the

platform—i.e., both merchants and cardholders—

worse off overall.” Pet. App. 51a.

Discover has an obvious interest in this Court's

review of the legal analysis the court of appeals em-

ployed to disregard the NDPs’ admitted interference

with Discover's attempt at price competition. See

Pet. App. 197a, 206a; U.S. Br. 32-24. Discover also

has a direct interest in the economically sound and

predictable application of antitrust law to the distinct

but related markets and competitive processes in-

volved in the successful operation of payment net-

works. As the court of appeals observed, Discover op-

erates the same type of “two-sided” credit and charge-

card platform that Amex does, Pet. App. 23a, and

provides network services to “separate, yet deeply in-

terrelated, markets” on either side of this platform, ibid

(internal quotation marks omitted). Accordingly, Dis-

cover is acutely aware of the need to “balance the two

sides of its platform” by, among other things, pricing

its services to “reflect the unique demands of the con-

sumers on each side.” Pet. App. 9a.

This Court's approach to the market definition

and burden-shifting issues in this case could signifi-

cantly affect this competitive balancing process in

many different contexts. Accordingly, Discover re-

spectfully submits this brief in the hope that it will

aid the Court in resolving the question presented in

accordance with business and market realities rele-

vant to advancing the “primary” antitrust aim of

“protect[ing! interbrand competition” within and

across payment networks. State Oil Co. v. Khan, 522

U.S. 3, 15 (1997).

STATEMENT

The history of the NDPs (or “anti-steering” rules)

at issue in this case is addressed in the government

briefs. See Pet. Br. 6-9; U.S. Br. 5-7. In general,

these rules “bar[{] merchants from (1) offering cus-

tomers any discounts or nonmonetary incentives to

use credit cards less costly for merchants to accept,

(2) expressing preferences for any card, or (3) disclos-

ing information about the costs of different cards to

merchants who accept them.” Pet. App. 4a. This en-

forcement action was originally brought against Visa

and MasterCard anti-steering rules as well as the

Amex NDPs addressed below. Pet. App. 66a. But Vi-

sa and MasterCard resolved the claims against them

in a 2012 consent decree, so the case proceeded to tri-

al only on the Amex rules. Pet. App. 66a-67a.

Discover—the smallest major payment netwerk by

charge volume and only successful new network en-

trant in decades, Pet. App. 15 la, 154a—testified as a

government witness about the NDPs’ competitive ef-

fects, Pet. App. 154a, 203a-207a, 212a-214a, 219a-

220a, notably in thwarting Discover's strategy to gain

market share “by pricing its network services ‘very

aggressively for merchants,” Pet. App. 203a (quoting

Tr. 821:8—16').

A. Discover Offers Breakthrough Value

Propositions to Cardholders and Mer-

chants

“Discover was initially owned and operated by one

of the nation’s Jargest retailers, Sears, which market-

ed Discover's cards to its already significant popula-

tion of private label cardholders.” Pet. App. 154a

n.24. Leveraging this initial platform, Discover was

able to gain a foothold in the broader payment card

industry by offering “breakthrough value proposi-

tions“ to both cardholders and merchants. Id. at

203a. The value proposition on the cardholder side

was that, “in 1986, most credit cards had annual

fees,” but “Discover had no annual fees,” offered “24

by 7 customer experience,” and “was the first card to

' Unless otherwise indicated, all transcript references in this

brief are to the trial testimony of Discover's President, Roger

Hochschild.

have any form of rewards with providing cash back

on every transaction.” Tr. 821:9-13. Discover com-

plemented these cardholder incentives with a value

proposition to merchants. Discover “pric[ed] its net-

work services ‘very aggressively for merchants|,] set-

ting all-in discount rates significantly below those of

its competitors.” Pet. App. 203a-204a (quoting id. at

821:14). Discover was able to do this through a “very

focused effort on keeping [its] expenses as low as pos-

sible.” Tr. 821:24-25. “[B]y operating very efficiently”

in a market in which other networks were “charging

a lot to merchants and charging a lot to cardholders,”

id. at 821:22-25, Discover could offer “a good value

proposition to [its] merchant customers on one side

and a good proposition to [its] cardholders on the oth-

er side,” id. at 822:1-3.

B. Discover Attempts to Compete By Offer-

ing Merchants Lower Prices

Although Discover is accepted at over nine million

merchant locations (roughly the same number as Vi-

sa and MasterCard and more than Amex), see Pet.

App. 184a-185a, Discover’s network has by far the

smallest share of U.S. credit and charge card volume.

See Pet. App. 15la (estimating 2013 shares of total

charge volume at Visa 45%, Amex 26.4%, MasterCard

23.3% and Discover 5.3%). As Discover explained at

trial, it also trails rivals in certain categories of card

issuance, as well as in volume of loan amounts out-

standing. See Tr. 816:24-817:1.

The evidence in the trial record identifies the bat-

tle for share of charge volume in the market for net-

work services to merchants, as well as the battle for

share of consumer spending or loans in the market

for branded card-issuance to consumers, as the focal

points of horizontal competition among payment card

networks and the banks that issue network-branded

credit and charge cards. See Tr. 814:14-16 (Discover

“competes against Visa, MasterCard and American

Express” in the provision of network services to mer-

chants and banks, and “measures its market share as

a credit card network in terms of percent of total

sales volume for general purpose credit cards”); Tr.

816:5-18 (Discover also competes with Amex and

“well over a hundred banks” in issuing credit cards to

consumers and measures its share of that market

based on “two metrics * * * share of credit card loans

outstanding * * * and share of credit card sales”); JA.

129-30 (Visa document stating that “every share

point that is shifted in the market to the bankcards’

advantage -- to the advantage of Visa and Master-

Card -- shifts almost $80 million in pre-tax profit

away from American Express”); id. at 190 (Amex

presentation stating that “the only real way to prove

that our customers prefer us is to measure market

share” in each of the various “business[s]” Amex

serves).

Networks can “increase consumer preference for

using” their brand and affiliated services, J.A. at 129,

by competing on either or both “sides” (merchant or

cardholder) of the “two-sided” platform described in

the opinions below. Merchants and cardholders are

“distinct yet interrelated” groups of consumers who

purchase “distinct yet interrelated” products from

payment card networks. Pet. App. 117a-119a; Tr.

814-816. This “interrelatedness”—which gives rise to

the so-called “chicken and egg problem” of merchant

acceptance turning on cardholder acceptance and

vice-versa, Tr. 821:4—means that a network can im-

prove its competitive position across the entire pay-

ment platform by competing for market share in the

product markets on either side of it. Tr. 828-831.

“Sensing an increase in merchant dissatisfaction

in the late 1990s amidst a series of price increases by

its competitors, Discover saw an opportunity to lever-

age its position as the lowest-priced network to gain

share.” Pet. App. 204a (citations omitted). Accord-

ingly, in “1999, the network launched a ‘major cam-

paign’ aimed at highlighting the pricing disparity be-

tween it and its competitors in order to persuade

merchants to ‘shift their business to [Discover’s] low-

er- priced network.” Jbid. (quoting Tr. 833:4—11). As

Discover's then-President explained: Discover in-

tend[s] to partner with merchants in helping them

control payment costs and propose[s] that they steer

customers to the lower-cost Discover cards.” Pet.

App. 204a (citing Tr. 834:13—20; PX1277 at 090,

‘094-95 (noting that Discover wanted “to help [mer-

chants] save money by encouraging their customers

to pay with Discover Card”).)

To kickstart the campaign, Discover “sent a letter

to every merchant on its network, alerting them to

[its] competitors’ recent price increases and inviting

the merchant to save money by shifting volume to

Discover.” Pet. App. 204a (citing Tr. 836:6—837:18).

Discover also “met with a number of larger mer-

chants to offer discounts from the network’s already

lower prices if they would steer customers to Discov-

er.” Ibid. In these meetings Discover “suggested a

number of means by which merchants could achieve

this share shift, including point-of-sale signage,” Pet.

App. 204a-205a (citing Tr. 839:22—842:3; PX1292 at

991-94), and the use of pass-through pricing that

would convert the network fee reductions into lower

retail prices that generate customer loyalty to mer-

chants, Pet. App. 205a (citing Tr. 847:8-848:14).)

Discover believed that over the long term, this strate-

gy and “offering further discounts to large merchants

would be profitable for the network” because it would

generate “greater transaction volume, and resulting

increases in discount and interest revenue.” Pet.

App. 205a (citing Tr. 837:19-25).

The district court found that Discover’s merchant

discount campaign was a paradigmatic example of

horizontal price competition that would have influ-

enced consumer behavior in an unrestrained market.

Discover testified that it was willing to “sacrifice

some revenue per transaction” in order to grow the

number of transactions and get[] other revenues on

those transactions.” Tr. 837:21-24. And several

“merchants testified that they would, in fact, steer if

given the opportunity.” Pet. App. 222a (citations

omitted). But Discover's efforts at price competition

“failed to produce ‘any significant movement in share’

due to the anti-steering rules maintained at the time

by Visa, MasterCard, and American Express.” Pet.

App. 205a (quoting Tr. 848:15-849:15).

Several merchants advised Discover that they

could not “express a preference for Discover” or oth-

erwise “steer share to Discover's lower-priced net-

work” without violating the anti-steering rules “im-

posed by the other payment networks.” (Pet. App.

205a (citing Tr. 848:15-849: 15, 852:24—-853:15).2. As

For example, the record contains unrefuted evidence that

the NDPs preclude merchants from:

* “fo}ffering a 10% discount off the posted purchase price,

free shipping, free checked bags, gift cards, or any other

monetary incentive for using their Discover card

* “[p)osting a sign saying ‘We Prefer Discover at the point

the district court observed, merchant “steering” is in

other industries “both procompetitive and ubiqui-

tous indeed as “routine[]” as “placing a particular

brand of cereal at eye level.” Pet. App. 67a. But it is

“absent in the credit card industry because the NDPs

eliminate merchants’ ability to “attempt to influence

customers’ purchasing decisions” in favor of a particu-

lar network’s lower-priced services. Jbid.; see also id.

at 204a-205a.

In the face of this competitive restraint, Discover

“abandoned” its low-price campaign because giving

merchants a “discount without getting anything in

return didn’t make business sense” for Discover. Pet.

App. 206a (quoting Tr. 854:7-15) (“To the extent that

offering a lower price was not going to give [Discover]

any business benefits, it was leaving money on the

table” that it could channel into other competitive

endeavors.). Accordingly, Discover refocused its

strategy on closing the ſelompetitive [g]ap” with ri-

vals by “raising [its] discount rates” to “more closely

align” with Visa and MasterCard. Ibid.

C. The District Court Finds the NDPs Un-

reasonably Restrain Trade

After a seven-week bench trial, the district court

issued a 150-page opinion holding that Amex’s NDPs

unreasonably restrained trade in violation of the

Sherman Act. Pet. App. 69a-7la, 112a, 114-122a,

259a. Applying this Court’s precedents on market

of sale”; and

* “[a)nswering the phone by saying ‘Thank you for calling

us, we proudly accept the Discover card’ or posting a

sign that says Thank You For Us/ag Discover.”

Pet. App. 100a-101a.

definition and the “three-step burden shifting frame-

work” that governs “rule of reason” review, see Pet.

App. 108a, llla-112 (citing United States v. El. du

Pont de Nemours & Co., 351 U.S. 377, 404 (1956);

Eastman Kodak Co. v. Image Tech. Sucs., Inc., 504

U.S. 451, 482 (1992); Brown Shoe Co. v. United

States, 370 U.S. 294, 336 (1962)), the district court

weighed all the trial evidence and found that the

NDPs unreasonably restrained trade because they

“block[ed]” inter- network price competition, “stifle[d]”

innovation, and resulted in higher costs for both

“merchants and cardholders.” Pet. App. 193a, 203a-

222a (finding that “inflated merchant discount rates

are passed on to all customers—Amex cardholders

and non-cardholders alike—in the form of higher re-

tail prices”).

1. In defining the product market relevant to its

analysis, the district court recognized that a “pay-

ment-card network sits at the center of a two-sided

platform that ‘comprises at least two separate, yet

deeply interrelated, markets: a market for card issu-

ance, in which Amex and Discover compete with

thousands of Visa- and MasterCard-issuing banks;

and a network services market, in which Visa, Mas-

terCard, Amex, and Discover compete to sell ac-

The first step of this framework required the government

plaintiffs to establish that the NDPs were “prima facie anticom-

petitive.” Cal. Dental Ass'n v. F.T.C., 526 U.S. 756, 771 (1999).

Because the district court found that the government made such

a showing, the burden shifted to Amex to identify any “procom-

petitive justificationſs for the challenged rules. Ibid see Pet.

App. 110a. And because Amex did so, the third step of the in-

quiry shifted the burden back to the government to show that

the competitive benefits “could have been achieved through less

restrictive means.” Pet. App. 110a (citation omitted).

ceptance services” to merchants and their affiliates.

Pet. App. 23a (quoting Pet. App. 70a). The NDPs,

however, reside only in contracts governing the provi-

sion of network services to merchants. Pet. App. 19a,

95a. And the record contains no evidence that such

services are interchangeable with the services that

networks provide to issuers and their cardholders.

Accordingly, and “[djespite the two-sided nature of

the platform” that comprises Amex’s overall payment

network, Pet. App. 23a, the district court concluded

that the “relevant product market for purposes of its

analysis of Amex’s NDPs is the market for general

purpose credit and charge card network services” to

merchants. Pet. App. 121la.

In reaching this conclusion the district Wourt ex-

pressly recognized the relationship between the mar-

ket for network services to merchants and the market

for network services to issuing banks and their card-

holders. See Pet. App. 12la. The district court's

analysis simply tracks record evidence of the services

the NDPs contractually restrain—network services

governed by “merchant agreements,” Pet. App. 22a,

95a—in identifying the primary product market rele-

vant to its analysis of the NDPs’ competitive effects.

It then defines the boundaries of that market in keep-

ing with this Court's direc.ion that antitrust product

markets should generally encompass only “products

that have reasonable interchangeability” and “cross-

elasticity of demand between the product itself and

substitutes for it.” EJ. du Pont, 351 U.S. at 404; see

also Brown Shoe, 370 U.S. at 325; see Pet. App. 11 1a-

112a.

This analysis did not foreclose consideration of

Amex’s arguments about the NDPs’ competitive im-

pact on the market for cardholder services. Pet. App.

7la, 12la-122a. It merely recognized the “reality,”

Pet. App. 118a, that the services Amex provides to

merchants and cardholders are not interchangeable

within the meaning of this Court’s market-definition

precedents, and thus deferred to the latter stages of

its rule-of-reason analysis full consideration of the

NDPs’ competitive effects on markets beyond the one

for the merchant services the NDPs contractually re-

strain. See Pet. App. 22a, 7 la, 239a-240a.

2. Applying the first step of the rule-of-reason

analysis, the district court concluded that the gov-

ernment had met its initial burden of showing that

the NDP’s were prima facie anticompetitive because,

among other things, unrefuted trial evidence estab-

lished that the rules actually and adversely affected

competition by “effectively deny[ing}” other networks

the ability to “offer{] merchants a low price in return

for greater volume.” Pet. App. 203a.

Like its analysis of market definition, the district

court’s analysis of the government’s prima facie case

did not foreclose consideration of the NDPs’ proffered

procompetitive effects in the market for network ser-

vices to issuers and their cardholders, or of the NDPs’

role in mediating network competition across the

“two-sided” payment platform Amex operates. It

simply shifted the burden to Amex to address pro-

competitive justifications for the NDPs, including and

specifically the justification that the “NDPs enhance

overall competition in the credit and charge card in-

dustry * * * by inhibiting competition in the network

services market for merchants—thereby ensuring

that Amex’s spend-centric model continues to be

fueled by high merchant discount fees—in favor of

greater competition in the interrelated but distinct

issuing market.” Pet. App. 238a-239a.

Weighing all the evidence, the district court found

that the NDPs’ admitted restraint on network price

competition for merchant business was not justified

by the foregoing or other purported benefits to issu-

ers, their cardholders, or overall platform competi-

tion. Pet. App. 7la, 229a, 239a-240a; see also id. at

228a-258a. Notably, the district court found that

Amex’s NDPs actually precluded other networks (and

specifically Discover) from reaping a “competitive re-

ward for offering merchants lower swipe fees” and

“thereby suppressſed / an important avenue of horizon-

tal interbrand competition.” E.g., Pet. App. 197a (in-

ternal quotation marks and citations omitted; em-

phasis added). Based on this and other evidence, the

district court found that “the failure of Discover’s low-

price value proposition is emblematic of the harm

done to the competitive process by Amex’s rules

against merchant steering,” Pet. App. 206a (emphasis

added), and that this harm resulted in “higher prices

for merchants and their customers,” Pet. App. 228a.

Accordingly, the district court concluded that the

NDPs unreasonably restrained trade in violation of

the Sherman Act and enjoined their enforcement.

Pet. App. 71a, 2 59a.

D. The Court of Appeals Reinstates the NDPs

The court of appeals reversed and reinstated the

NDPs on the grounds that the district court's errone-

ous “definition of the relevant market in this case is

fatal to its conclusion that Amex violated § 1.” Pet.

App. 3la. According to the panel, the district court

“erred in excluding the market for cardholders from

its relevant market definition,” and specifically in

“declin{ing] to * * collapsſel the issuance and net-

work services markets into a single platform-wide

market for [payment card] transactions.” Pet. App.

32a.

Applying this new market definition, id. at 32a-

40a, the court of appeals held that even direct evi-

dence that the NDPs “blocked” network price compe-

tition for merchant business was insufficient to meet

the government’s initial burden of showing that the

NDPs are prima facie anticompetitive. See Pet. App.

Zla, 39a-40a, 49a-53a. The reason, the panel stated,

is that even this undisputed evidence did not prove

“net” anticompetitive effects across both “sides” (mer-

chant and cardholder) of Amex’s “two-sided platform.”

Pet. App. 49a, 51-53a.

In reaching this conclusion, the court of appeals

did not contest Discover’s trial testimony or purport

to disturb the district court’s finding that the NDPs

“block Amex-accepting merchants from encouraging

their customers to use any * * * card other than an

American Express card, even where that card is less

expensive for the merchant.” Pet. App. 100a-101la. It

simply redefined the “relevant market” to include

network services to cardholders because such services

are “necessarily affected” by (though not inter-

changeable with) network services to merchants, and

then disregarded the NDPs’ restraint on competition

for the latter as inadequate to show that the “NDPs

made all Amex consumers on both sides of the plat-

form—i.e., both merchants and cardholders—worse

off overall.” Pet. App. 51a; see also id. at 39a-54a.

The court of appeals’ market definition and onsu-

ing assessment of the government’s prima facie case

depart from this Court’s precedents because the rec-

ord contains no evidence that network services to

merchants are interchangeable with, or in any sense

substitutes for, network services to cardholders. The

opinion below grounds this departure in competition

policy concerns purportedly unique to “two-sided”

platforms and the particulars of Amex’s business

model. See Pet. App. 49a-53a. Specifically, the court

of appeals reasoned that because the “NDPs simulta-

neously affect competition for merchants and card-

holders by protecting the critically important revenue

that Amex receives from its relatively high merchant

fees,” a “reduction in revenue that Amex earns from

merchant fees may decrease the optimal level of

cardholder benefits, which in turn may reduce the in-

tensity of competition among payment-card networks

on the cardholder side of the market.” Pet. App. 50a.

The court of appeals then held that the government’s

failure to rebut this potential competitive benefit in

its prima facie case—and specifically its failure to

identify a “net price affecting consumers on both sides

of the platform“ required reversal of the district

court's judgment and reinstatement of the NDPs.

Pet. App. 53a-54a.

SUMMARY OF ARGUMENT

“(T]he primary purpose of the antitrust laws is to

protect interbrand competition.” State Oil, 522 U.S.

at 15 (citation omitted). To compete for market share

against other brands, a payment card network must

“balance the two sides of its platform” by, among oth-

er things, pricing its services to “reflect the unique

demands of the consumers on each side.” Pet. App.

9a. Such balancing must be responsive to market

forces on all sides of the platform, because competi-

tion at the platform level depends on the competi-

tiveness of each network’s offerings in each product

market the platform serves. This phenomenon is not

unique to payment-card platforms. It applies equally

to any platform (a newspaper serving advertisers and

readers, or a travel booking business serving hotels

and guests) in which platform-level competition turns

on rivals’ ability to offer a dynamic range of “different

price pairs” to the markets the platform serves in the

hope of finding one that “best satisflies] consumer

preferences.” Economists’ Cert. Amicus Br. 11.

Amex successfully competes with other networks

by offering the price pair described in the opinions

below. E.g., Pet. App. 50a. The problem with the

NDPs is that they protect Amex’s competitive model

by restraining other networks from using their own

price pairs to compete freely with Amex.

The antitrust laws dictate that “competition

should choose the optimal mix of revenue between

the two sides” of the platform. Philip E. Areeda &

Herbert Hovenkamp, Antitrust Law: An Analysis of

Antitrust Principles and Their Application “562(e), at

101 (Supp. 2017) (Areeda & Hovenkamp). According-

ly, in this case as in others, the central aim of the an-

titrust laws is best served by adhering to this Court's

precedents respecting the economic boundaries of

product markets, and allowing sufficient evidence of

anticompetitive effects in any such market—

including those that are part of a broader network or

platform—to trigger antitrust scrutiny under the

burden-shifting framework this Court has long ap-

plied to rule-of-reason analysis. Any other ap-

proach—and particularly one that excuses direct and

undisputed evidence of anticompetitive restraints in

one or more product markets comprising a network

platform—will inhibit competition both within and

across the platform precisely because the product

markets it serves are related.

The opinion below disregards these principles in

order to protect Amex’s chosen method of competing

with MasterCard and Visa over competition more

broadly. The record here provides no reason to en-

dorse this approach, and many reasons to reject it.

1. The court of appeals’ departure from this

Court's longstanding definition of an antitrust prod-

uct market is not supported by the law or record, and

is not nevessary to protect competition among opera-

tors of “two-sided” payment card networks. Although

the markets for network services to merchants and

cardholders are related, the record shows—and Amex

concedes, Br. in Opp. 16—that a network cannot

“substitute” its competitive offerings to one group

with its competitive offerings to the other. In such

circumstances, this Court’s precedents counsel that

the threshold question whether a challenged prac-

tice—here, a contractual restraint that networks im-

pose on merchants—has “anticompetitive effects” is

properly assessed in the market for the restrained

service and its substitutes.

This Court’s precedents further counsel that

where, as here, there is direct evidence that a chal-

lenged practice restrains horizontal price competition

in the market for a particular service and its reason-

able substitutes, a district court may rely on it to find

anticompetitive effects necessary to satisfy the first

step of the rule-of-reason analysis and shift the bur-

den to the defendant to address any procompetitive

justifications for the restraint. That is because on

such a record, a district court may fairly conclude

that questions about how the challenged practice “af-

fect” competition in “related” but distinct product

markets are properly addressed to the reasonable-

ness—not the existence—of the practice’s anticompet-

itive effects.

There was no need to alter or dispense with the

foregoing approach in this case merely because the

NDPs restrain competition in a service market that

sits on one side of a “two-sided” platform. Indeed,

market definition is particularly important where a

challenged practice applies only to customers in one

of several distinct product markets served by a

broader platform (here, NDPs in merchant but not

cardholder contracts), because an overbroad market

definition will short-circuit the antitrust inquiry and

excuse at the outset practices that undeniably re-

strain competition in product markets that must be

competitive in their own right to foster free and fair

competition across the platform.

Discover's trial testimony again illustrates the

point. In an effort to compete with Visa, MasterCard

and Amex both within and across the network plat-

form at issue here, Discover attempted to engage in

price competition for merchant business that the

NDPs admittedly “block[ed].” Pet. App. 100a. But

the court of appeals’ unprecedented approach to mar-

ket definition caused the court to disregard this com-

petitive restraint as legally irrelevant. Pet. App. 49a-

53a. In so doing, the court of appeals immunized

from antitrust scrutiny a conceded restraint not just

on network competition for merchant business, but

also—because that market is part of a broader plat-

form—on the competitive pairings networks can offer

across the platform.

2. The danger inherent in the court of appeals’

market definition is evident in the court’s articulation

of the proof required to establish that the NDPs are

“prima facie anticompetitive.” Cal. Dental Ass’n v.

FTC, 526 U.S. 756, 771 (1999); Pet. App. 32a, 39a-

40a, 49a-53a. The court of appeals held that undis-

puted evidence of price restraints this Court has long

recognized as sufficient to satisfy step one of the rule-

of-reason analysis did not suffice here, Pet. App.

205a-207a, 227a-228a, because the “two-sided” nature

of Amex’s network “platform” required prima facie

evidence that the “NDPs made all Amex consumers

on both sides of the platform—i.e., both merchants

and cardholders—worse off overall.” Ibid.

This Court should reject that approach and ad-

vance the “primary” antitrust aim of protectſingl in-

terbrand competition,” State Oil, 522 U.S. at 15, by

reaffirming the application of settled antitrust prin-

ciples to the district court's undisturbed factual find-

ings. Applying those principles, the district court

correctly began its analysis of the challenged practice

here (Amex’s NDPs) by identifying the service market

they contractually restrain (network services to mer-

chants), and then defining the boundaries of that

market based on the economic substitution test in

this Court’s precedents. The district court then cor-

rectly focused on proof of anticompetitive effects with-

in that market to determine whether the NDPs were

prima facie anticompetitive for purposes of step one

of the rule-of-reason analysis.

The record evidence that the NDPs restrain price

competition in a core product market served by

Amex’s platform was under this Court’s precedents

sufficient to support the district court’s finding that

the NDPs were prima facie anticompetitive, and to

shift the burden to Amex to identify any procompeti-

tive justifications for them.

3. The district court’s adherence to this settled

burden shifting approach on the record here did not

condemn the Amex business model the court of ap-

peals sought to protect in the opinion below. It simp-

ly refused to protect it at the expense of restraining

competition by Discover or other networks willing to

offer different price or service combinations consum-

ers might find more attractive. In short, the district

court’s analysis would allow “competition,” not courts

or particular competitors, to drive network price and

service offerings both within and across the payment

card platform at issue here. That is the right result,

and the Court should reject the court of appeals’ con-

trary analysis, which uses the “rule of reason” to do

exactly what this Court has held it should not: name-

ly, treat “competition itself [ajs unreasonable.” Nat’

Soc y of Profl Eng’rs v. United States, 435 U.S. 679,

696 (1978).

ARGUMENT

I. The Opinion Below Unjustifiably Departs

From This Court’s Approach to Market Defi-

nition

Although the text of Section 1 of the Sherman Act

prohibits “[e]very contract * * * in restraint of ſinter-

state] trade or commerce,” 15 U.S.C. § 1, this Court

has long held that the statute “prohibit[{s] only unrea-

sonable restraints of trade.” E.g., Nat’ Collegiate

Athletic Ass n v. Bd. of Regents of Univ. of Okla., 468

U.S. 85, 98 (1984) (NCAA) (emphasis added). Market

definition is important in assessing whether a chal-

lenged practice “unreasonabl{y]” restrains competi-

tion, because the competitive impact of a challenged

practice depends, among other things, on the availa-

bility of substitutes for the product or service it alleg-

edly restrains.

Accordingly, this Court has long held that the

“relevant market for antitrust purposes is determined

by the choices available to” consumers of the re-

strained product or service, e.g., Kodak, 504 U.S. at

481482, and is therefore defined by “products that

have reasonable interchangeability” with it. Jd. at

482 (quoting du Pont, 351 U.S. at 404). Interchange-

ability requires similar products or services as well as

similar customer needs and preferences that result in

“cross-elasticity of demand between the product itself

and substitutes for it.” J. du Pont, 351 U.S. at 404;

see also Pet. App. 11lla-112a; Brown Shoe, 370 U.S.

at 325.

The court of appeals purported to acknowledge

this definition of a product market. Pet. App. 32a.

But it abandoned it in defining the market in this

case to include any product, service, or customer for

which the “NDPs affect competition.” Pet. App. 51a.

The legal precedents the court cites, Pet. App. 32a,

focus on product substitutes, not affectſed] markets.

See id. And the record contains no evidence that the

services Amex provides to cardholders are in any way

“substitutes” for the services Amex provides to mer-

chants. E.g., Kodak, 504 U.S. at 482. Indeed, the

district court found exactly the opposite. See Pet.

App. 114a-122a, 127a.

Lacking any relevant precedent for defining a

product market to include services or products that

are not interchangeable, the court of appeals simply

asserted that its novel market definition was neces-

sary to protect “critically important revenue” Amex

needed to compete against Visa and MasterCard.

Pet. App. 50a. But that is not enough under the anti-

trust laws, which protect “competition, not competi-

tors.” Atl. Richfield Co. v. USA Petroleum Co., 495

U.S. 328, 338 (1990). As Discover's experience illus-

trates, the court of appeals’ market definition under-

mines both product-market and platform-level com-

petition by excusing a clear impediment to price ri-

valry in the product market the NDPs directly re-

strain, and that must be competitive in order to foster

competitive offerings across the platform.

1. As the district court correctly observed, the

services that Amex and other networks provide to

merchants and cardholders are part of a broader plat-

form, but are nonetheless “distinct [and] involveſe]

different sets of rivals and the sale of separate,

though interrelated, products and services to sepa-

rate groups of consumers.” Pet. App. 119a.4 The dis-

trict court was thus fully justified in relying on record

evidence that the NDPs appear only in contracts for

network services to merchants to define the services

they contractually restrain, and then defining the

market for those services in keeping with the product

substitution analysis in this Court’s precedents in or-

der to assess whether the NDPs had anticompetitive

effects sufficient to satisfy the first step of the rule-of-

reason inquiry. Pet. App. 119a-120a.

This Court’s analysis in Times—Picayune illus-

trates the point. The plaintiff in that case challenged

(as an anticompetitive tying arrangement) a newspa-

‘ Similarly in Brown Shoe Co. v. United States, 370 U.S 294

(1962), this Court held that although all the defendants pro-

duced some form of shoes, “the record supports the district

court's finding that the relevant lines of commerce are mens,

women’s, and children’s shoes. These product lines are recog-

nized by the public; each line is manufactured in separate

plants; each has characteristics peculiar to itself rendering it

generally noncompetitive with the others; and each is, of course,

directed toward a distinct class of customers.” Id. at 326.

per requirement that all advertisers who wished to

run content in the paper’s morning publication would

also have to run content in the paper's (distinct but

related) evening publication. Times-Picayune Publ’g

Co. v. United States, 345 U.S. 594, 596-597 (1953). In

assessing the product market relevant to its antitrust

analysis of the challenged restraint, this Court recog-

nized that “every newspaper is a dual trader in sepa-

rate though interdependent markets” serving both

advertisers and readers. Id. at 610. But because the

restraint in issue concernſed] solely one of these

markets” (because it applied only to advertisers, not

readers), the Court defined the relevant market sole-

ly in terms of competition for advertisers. /bid.

The record here is similar. See Pet. App. 119a-

120a. Although “merchants and their customers

jointly make the decision of which method of payment

is used for any given transaction,” Pet. App. 127a, the

NDPs appear only in network contracts with mer-

chants. Accordingly, the “relevant consumer’ for

purposes of assessing price sensitivity” impacted by

the NDPs “is the merchant,” ibid., and the network

services market is the proper market for assessing

whether the NDPs have anticompetitive effects suffi-

cient to establish a prima facie case under step one of

the rule of reason.

2. The court of appeals did not apply Times

Picayune (or du Pont, Kodak, or Brown Shoe), in de-

fining the relevant market in this case. Pet. App.

118a. Nor did it explain how a network—much less a

plaintiff—would go about calculating the “net price

affecting consumers on both sides of the platform” it

describes. Pet. App. 53a. That is not surprising, be-

cause as noted, its analysis sweeps in distinct and

complex markets, customers, and services that are

not interchangeable, and are the subject of proposed

innovations (like Discover’s proposal to “offer[{] mer-

chants equity ownership in the network,” Tr. 839:3-

4), that would complicate any meaningful or reliable

attempt to measure what the court of appeals termed

a “net price.” Pet. App. 53a. Accordingly, the court of

appeals was left to defend its analysis as necessary to

protect Amex’s ability to compete against Visa and

MasterCard. Pet. App. 48a-49a & n.51.

Again that is not a reason for insulating the NDPs

from antitrust scrutiny, particularly by collapsing

two distinct service markets with very different cus-

tomers and competitive conditions, see Pet. App.

117a-119a; Tr. 814-16, into a sweeping “platform”

market that will unnecessarily and intractably com-

plicate the question whether a challenged practice is

prima facie anticompetitive, and if so in what regard.

The court of appeals identified no reason to introduce

such uncertainty into the analysis on the record here,

which concerns a contractual restraint that appears

only in network agreements with merchants, and

that supported the district court’s initial assessment

of the NDPs’ competitive effects in the market for

those services.

This traditional approach to market definition did

not foreclose consideration of Amex’s justifications for

the rule. See Pet. App. 7 Ia, 11la-122a. It simply left

them to a later stage of rule-of-reason analysis. See

id. at 71a, 239a-240a. In contrast, the court of ap-

peals’ unprecedented approach to market definition

short-circuited antitrust review of the NDPs notwith-

standing their admitted restraint on price competi-

tion in a product market the Sherman Act dictates

must be open to competition in its own right.

II. The Court of Appeals’ Market Definition Dis-

torts the Rule-of-Reason Inquiry and Insu-

lates an Admitted Restraint on Price Compe-

tition from Antitrust Scrutiny

The significance of the court of appeals’ novel

market definition is apparent from the court's de-

scription of the evidence required to establish a pri-

ma facie case that the NDPs had an anticompetitive

effect in that market “as a whole.” Pet. App. 49a-50a.

The district court found on the record here that the

NDPs’ admitted restraint on horizontal price compe-

tition in the market for network services resulted in

increased prices and reduced choice for both mer-

chants and cardholders. E.g., Pet. App. 39a-40a, 67a,

100a-101la, 203a-207a, 219a-220a. Yet the court of

appeals held that even this extraordinary evidence

was insufficient to establish a prima facie case that

the NDPs had an anticompetitive in the “relevant

market” because the evidence did not “show that the

NDPs made all Amex consumers on both sides of the

platform—i.e., both merchants and cardholders—

worse off overall.” Pet. App. 51a.

This analysis is difficult to square with this

Court’s precedents condemning competitive re-

straints that “impede[] the ordinary give and take of

the marketplace and substantially deprivell custom-

er|s]” of the chance “to utilize and compare prices.”

Profl Eng’rs , 435 U.S. at 692-93; Catalano, Inc. v.

Target Sales, Inc., 446 U.S. 643, 648-649 (1980) (hold-

ing that because credit terms “must be characterized

as an inseparable part of price,” restraining credit

terms constituted an unlawful restraint on price

competition under the Sherman Act). Because “price

is the ‘central nervous system’ of the economy,” Unit-

ed States v. Socony-Vacuum Oil Co., 310 U.S. 150,

224-26 & n.59 (1940), price competition is “an object

of special solicitude under the antitrust laws.” Unit-

ed States v. Gen. Motors Corp., 384 U.S. 127, 148

(1966); NCAA, 468 U.S. at 106 & n.30 (“[a] restraint

that has the effect of reducing the importance of con-

sumer preference in setting price” is inconsistent

“with thſel fundamental goal of antitrust law”). Ac-

cordingly, this Court’s precedents support the district

court’s conclusion that the record evidence here was

sufficient to establish a prima facie case that the

NDPs unreasonably restrain trade, and thus to shift

the burden to Amex to identify procompetitive justifi-

cations for the rules.

The court of appeals’ contrary conclusion does not

serve the purpose of the antitrust laws it purports to

apply, because it preemptively terminates the anti-

trust analysis out of concern for procompetitive justi-

fications that could be (and in the district court were)

addressed in the second and third stages of the bur-

den shifting analysis. See Pet. App. 239a-240a (find-

ing that “even if * * * cross-market balancing is ap-

propriate under the rule of reason in a two-sided con-

text, here Defendants have failed to establish that

the NDPs are reasonably necessary to robust compe-

tition on the cardholder side of the GPCC platform, or

that any such gains offset the harm done in the net-

work services market”).

Given the district court’s undisturbed findings on

“cross-market balancing,” see Pet. App. 239a-240a,

this case does not require the Court to resolve pre-

cisely when or to what extent the “two-sided” nature

of a payment network permits a defendant to justify

an anticompetitive practice in one market by refer-

ence to its procompetitive effect in a related market.“

Reversal is warranted under the settled principle

that “unrestrained interaction of competitive forces

will yield the best allocation of our economic re-

sources, the lowest prices, the highest quality and the

greatest material progress.” N. Pac. Ry. Co. v. United

States, 356 U.S. 1, 4 (1958).

Scrutinizing competitive restraints on either side

of the “two-sided” platform the court of appeals de-

scribed is critical to ensuring competition both within

and across the platform. Yet that is precisely what

the court of appeals’ analysis precludes. There is no

basis in the law or record for affirming this approach.

This Court has in some contexts discouraged such cross-

market defenses on the ground that courts have an “inability to

weigh, in any meaningful sense, destruction of competition in

one sector of the economy against promotion of competition in

another sector.” United States v. Topco Association, Inc., 405

U.S. 596, 609-10 (1972); cf. United States v. Phila. Nat. Bank,

374 U.S. 321, 370 (1963) (“If anticompetitive effects in one mar-

ket could be justified by procompetitive consequences in anoth-

er, the !ogical upshot would be that every firm in an industry

could, without violating § 7, embark on a series of mergers that

would make it in the end as large as the industry leader.”). That

said, this Court has entertained such arguments in cases where

the challenged practice impacts multiple related markets, see,

e.g., NCAA, 468 U.S. at 117-19, and some lower courts have

likewise recognized the need in some circumstances to “balance

the anticompetitive effects on competition in one market with

certain procompetitive benefits in other markets.” Sullivan v.

NFL, 34 F.3d 1091, 1112 52 (Ist Cir. 1994), cert. denied, 513

U.S. 1190 (1995).

III. The Decision Below Justifies Its Depar-

ture from Settled Antitrust Principles on

Grounds That Violate the Sherman Act’s

Central Tenet

The court of appeals’ departure from the law and

record is particularly troubling because it is grounded

in precisely the type of policy judgment the Sherman

Act forbids. The court of appeals’ decision turns on

the conclusion that protecting Amex’s “unique” meth-

od of competing with Visa and MasterCard, Pet. App.

87a, is more important than allowing Discover's low-

price version or, for that matter, any inter-network

price competition on merchant network fees at all.

Id. at 39a-40a, 48a-53a. As noted, “the primary pur-

pose of the antitrust laws is to protect interbrand

competition.” State Oil, 522 U.S. at 15. In upholding

a “restraint that effectively blocks interbrand ccmpe-

tition on price” because Amex could compete better

without such price pressure, Pet. App. 235a, the pan-

el opinion endorses exactly the “frontal assault on the

basic policy of the Sherman Act” this Court has ad-

monished against. Profl Eng’rs, 435 U.S. at 695; Pet.

App. 235a, 240a-24 la.

The antitrust laws do not authorize courts “to

draw lines between ‘good’ competition and ‘bad’ com-

petition,” Pet. App. 20 la, but rather “reflect a stead-

fast ‘legislative judgment that ultimately [all forms

of] competition will produce not only lower prices, but

also better goods and services.” Ibid. (quoting F. T. C.

v. Super. Ct. Trial Lawyers Ass’n, 493 U.S. 411, 423

(1990)). Any line-drawing among types of competi-

tion is a task reserved exclusively for Congress. See

United States v. Topco Assocs., Inc., 405 U.S. 596, 611

(1972) (“If a decision is to be made to sacrifice compe-

tition in one portion of the economy for greater com-

petition in another portion this too is a decision that

must be made by Congress and not by private forces

or by the courts.”). And to date Congress’s judgment

has been to protect “competition, not competitors.”

Atl. Richfield, 495 U.S. at 338 (quoting Brown Shoe,

370 U.S. at 320 and explaining that tjo hold that

the antitrust laws protect competitors from the loss of

profits due to [nonpredatory] price competition would,

in effect, render illegal any decision by a firm to cut

prices in order to increase market share”) (alteration

in original) (citation omitted)).

The panel opinion disregards these fundamental

principles in holding that the NDPs and Amex busi-

ness model are more important than price competi-

tion, including the interbrand price competition the

NDPs admittedly precluded Discover from successful-

ly pursuing. Pet. App. 203a-207a, 219a. See Cargill,

Inc. v. Monfort of Colo., Inc., 479 U.S. 104, 116-117

(1986) (describing the “perverse result” that would

accompany a holdlingl that the antitrust laws pro-

tect competitors from the loss of profits due to * * *

price competition”). As the district court’s opinion -

lustrates, such a stark departure from this Court’s

precedents was not necessary to ensure proper con-

sideration of the NDPs’ impact on competition within

and across the “two-sided” platform at issue here.

Based on the seven-week trial record, the district

court found the failure of Discover’s low-price cam-

paign “emblematic of the harm done to the competi-

tive process by Amex’s rules against merchant steer-

ing.” Pet. App. 206a. And it found that enjoining the

NDPs would redress this harm by allowing Discover

and other networks “aggressively [to] pursue a strat-

egy of lowering [their merchant] prices” in exchange

for volume while still robustly competing for card-

holders through rewards and steering benefits. Pet.

App. 219a-220a.

On this record, the district court did not “err{]” in

“declin[ing] ‘to define the relevant product market to

encompass the entire multi-sided platform.” Pet.

App. 39a (quoting Pet. App. 119a). It simply (and

correctly) recognized that the merchant “side” of the

platform constitutes a distinct product market under

this Court’s precedents, Pet. App. 114a-122a, and en-

gaged in a rule-of-reason analysis that credited price

restraints in this market as sufficient to establish a

prima facie case before shifting the burden to Amex

to identify procompetitive justifications that ulti-

mately failed to carry the day. After weighing all the

evidence, the court concluded that competition, ra-

ther than Amex’s rules, should decide the “optimal

mix of revenue as between the two sides.“ Areeda &

Hovenkamp %562(e), at 101 (Supp. 2017). That is

what the Sherman Act contemplates, and what the

panel opinion disregards in conflict with the statute

and this Court’s decisions.

CONCLUSION

This Court should reverse the judgment below.

Respectfully submitted.