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

OHIO, ET AL., PETITIONERS

v.

AMERICAN EXPRESS COMPANY, ET AL.

ON PETITION FOR A WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

BRIEF FOR THE UNITED STATES IN OPPOSITION

J&FFREY B. WALL

Acting Solicitor General

Counsel of Recor

PATRICIA A. BRINK

Director of Civil

Enforcement

KnisTen C. LIMARZI

ROBERT B. NICHOLSON

JAMES J. FREDRICKS

N yn G, LEVIN

Department of Justice

Washington, D.C. 20530-0001

SupremeCtB ricfequsdaj.gov

(202) 514-2217

QUESTION PRESENTED

The private respondents operate the American Ex-

press credit-card network. In this antitrust enforce-

ment action, the United States and a group of States

challenged “anti-steering” rules that respondents im-

pose on merchants that accept American Express cards.

Those rules prohibit merchants from encouraging their

customers to use other credit cards, which often charge

the merchants lower fees. Applying the rule of reason,

the district court held that the anti-steering rules vio-

late Section 1 of the Sherman Act, 15 U.S.C. 1. The

court found that the rules stifle price competition

among credit-card networks, allowing a// networks to

raise their merchant fees and forcing merchants to

charge higher prices to retail consumers as a result.

The court of appeals reversed. It did not question

the clistrict court’s finding that the anti-steering rules

thwart price competition, raise merchant fees, and in-

flate retail prices. But the court emphasized that the

credit-card industry is “two-sided,” in that networks

compete to attract both merchants and cardholders.

And the court held that the United States and the States

had not made a prima facie showing that the anti-steering

rules unreasonably restrain trade because they had not

sufficiently demonstrated that the rules harm caidhold-

ers as Well as merchants. The question presented is as

follows:

Whether proof that respondents’ anti-steering rules

thwart interbrand price competition, raise merchant

fees, and inflate retail prices was sufficient to make out

a prima facie case under the rule of reason.

(1)

TABLE OF CONTENTS

Page

SEIN III TIP sis oncicnenctsvenisieatnehinesnnmigsemenubbencsentuinislebianamnnneeiniiiadaina 1

ESRI Rp ap ASNT REN ONC pS OS NN 1

ETE Se cee MC ae SA AN GaP aan Sen ae tooo eRAE STAT 2

BE ainictisstsnevndiceasinensninapniciiitenseritaianetiinssisieiininalamsoainaibetintnbinendinanniani 9

UIT wxstcnissecssesibsantncetacneonncnianeenciieabdtinatadiatieanitddabiiniiammidinsninidiid 21

TABLE OF AUTHORITIES

Cases:

Associated Gen. Contractors of Cal., Inc. v.

California State Council of Carpenters,

ieee i iiitatatatnineiinicetmnenneneiamnancesion 18

Board of Trade v. United States, 246 U.S. 231 (1918).......... 4

Broadcast Music, Inc. v. CBS, Inc.,

IIE icciiesinnsnnpiicennicintpinraieeieinaiinniteptiaiinianbatenenaiins 17

California Dental Ass'n v. FTC, 526 U.S. 756 (1999)....4, 15

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

ee ee ee i iniisicsnanectniesinenirhanidiciisennniecomninnnesceneese 11, 12

FTC v. Actavis, Inc., 183 S. Ct. 2223 (2018)............ccccceeecees 16

Leegin Creative Leather Prods., Inc. v. PSKS, Inc.,

ee i aeslicosnicniiineeeicvainsienecnameniicingeniétnincmaesccns 4,14

NCAA v. Board of Regents of the Univ. of Okla.,

a I irrtiietisicisnieninciicinnriciniatninanisigiasidiatisisieminntnpascians 18

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

SI iss tiniterecitnscisiniesillictinrtanebininiinnitniabeiiniie 8

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

ee I iaalindininreihinitildcienntintindniateicthiniptanatennimcncoses 17

Rothery Storage & Van Co. v. Atlas Van Lines, Inc.,

792 F.2d 210 (D.C. Cir. 1986), cert. denied,

Fe arent tcntpeeeainiabiaaiienentinensennnaiensenene 11

State Oil Co. v. Khan, 522 U.S. 3 (1997)......cccccccesssesseerenerses 14

(IIT)

IV

Cases—Continued: Page

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

TL ee 11, 13

US Airways, Inc. v. Sabre Holding Corp.,

No. 11-ev-2725, 2017 WL 1064709

(S.D.N.Y. Mar. 21, 2017), appeal pending,

No. 17-960 (2d Cir. filed Apr. 5, 2017).......ccccccccccccccscsoseceee 20

United States v. Continental Can Co.,

ES RAS ea ll

United States v. E.1. du Pont de Nemours & Co.:

a Oe 4, 11

ES TE eae ar a RE 11

United States v. Socony-Vacuum Oil Co.,

I 17

United States v. Visa U.S.A., Inc., 344 F.3d 229

(2d Cir. 2003), cert. denied, 543 U.S. 811 (2004).............. 20

Statute and rule:

Sherman Act, 15 U.S.C. 1 et seg.:

RASS FS CRE NEN CD ME Sle RE 3

ERA a ans ne es aE a Eee 10, 19

Miscellaneous:

2B Phillip E. Areeda et al., Antitrust Law:

An Analysis of Antitrust Principles and Their

I I ectetneciienicitintastmninintinstrnessnzcrnssen 11

Phillip E. Areeda & Herbert Hovenkamp,

Antitrust Law: An Analysis of Antitrust

Principles and Their Application

Ne aa it eictcieinilactuenstabesieintendiniomniinioscshnescinsbenedincensies 12, 16, 19

Jn the Supreme Court of the Anited States

No. 16-1454

OHIO, ET AL., PETITIONERS

Vv.

AMERICAN EXPRESS COMPANY, ET AL.

ON PETITION FOR A WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

BRIEF FOR THE UNITED STATES IN OPPOSITION

OPINIONS BELOW

The opinion of the court of appeals (Pet. App. 1a-58a)

is reported at 838 F.3d 179. The opinion of the district

court (Pet. App. 63a-259a) is reported at 88 F. Supp. 3d

143.

JURISDICTION

The judgment of the court of appeals was entered on

September 26, 2016. A petition for rehearing was de-

nied on January 5, 2017 (Pet. App. 324a-326a). On

March 24, 2017, Justice Ginsburg extended the time

within which to file a petition for a writ of certiorari to

and including May 5, 2017. On April 24, 2017, Justice

Ginsburg further extended the time within which to file

a petition to and including June 2, 2017, and the petition

was filed on that date. The jurisdiction of this Court is

invoked under 28 U.S.C. 1254(1).

(1)

2

STATEMENT

1. The private respondents operate one of the coun-

try’s four major credit-card networks, American Ex-

press (Amex). In 2013, the Amex network captured 26.4

percent of all credit-card transaction volume. Pet. App.

13a. The rest was divided among Visa (45 percent),

MasterCard (23.3 percent), and Discover (5.3 percent).

Ibid. Together, the four networks processed nearly

$2.4 trillion in credit-card transactions. /d. at 74a.

Amex cards are accepted by merchants that account

for more than 90 percent of credit-card transactions by

dollar value, including virtually all of the country’s larg-

est merchants. Pet. App. 188a, 224a. To accept Amex

cards, a merchant must enter into a contract with re-

spondents and agree to pay a “merchant discount fee”

equal to a percentage of each transaction. /d. at 15a.

For example, a merchant that accepts an Amex card for

a $100 purchase might receive only $97—the purchase

price less a three-percent merchant discount fee. Visa

and MasterCard also charge merchant discount fees,

though their fees are structured somewhat differently

because Visa and MasterCard use a different business

model. Jd. at 13a-14a.’

Respondents have traditionally charged merchants

higher fees than other networks, which respondents use

in part to fund rewards and other benefits for Amex

cardholders. Pet. App. 68a, 175a-180a. Beginning

' Visa and MasterCard operate “open-loop” systems in which “ac-

quiring” banks act as intermediaries between the credit-card net-

works and merchants, and “issuing” banks act as intermediaries be-

tween the networks and individual cardholders. Pet. App. 13a-14a,

8la-83a (citation omitted). Respondents, in contrast, operate a

“closed-loop” system in which they typically maintain direct rela-

tionships with cardholders and merchants. /d. at 14a-15a, 88a-84a.

3

around 1990, Visa and MasterCard highlighted their

lower fees and urged merchants to encourage their cus-

tomers to use Visa and MasterCard instead of Amex—

for example, by displaying signs stating “We Prefer

Visa.” Jd. at 91a-92a. Combined with other marketing

strategies, those competitive efforts “were remarkably

effective” and caused a substantial drop in respondents’

market share. /d. at 19a; see id. at 92a.

Respondents reacted by tightening the “anti-steering”

rules in their merchant contracts to “to stifle any fur-

ther steering or preference campaigns.” Pet. App. 200a,

202a; see id. at 19a, 92a-93a. The anti-steering rules bar

merchants that accept Amex cards from offering cus-

tomers discounts or incentives to use other cards, ex-

pressing a preference for other cards, or even disclosing

their relative costs of accepting different cards. /d. at

19a-20a, 100a-101a. The rules prohibit those actions

even when a customer does not have an Amex card or

presents a different card. /bid. Respondents “actively

monitor|[|” compliance with the rules and have “vigor-

ously enforce[d]” them to stamp out a variety of at-

tempts by merchants to encourage their consumers to

use less-expensive rival cards. Jd. at 102a-103a; see id.

at 103a-104a.

2. In 2010, the United States and a group of States

sued respondents, along with Visa and MasterCard, al-

leging that the networks’ anti-steering rules unreason-

ably restrained trade in violation of Section 1 of the

Sherman Act, 15 U.S.C. 1. Pet. App. 21a-22a. Visa and

MasterCard entered into consent judgments and re-

scinded their anti-steering rules. /d. at 22a. Respond-

ents proceeded to trial. After a seven-week bench trial,

the district court held that the anti-steering rules vio-

late Section 1 under the rule of reason. /d. at 63a-259a.

4

a. “The rule of reason is the accepted standard for

testing whether a practice restrains trade in violation of

[Section] 1.” Leegin Creative Leather Prods., Inc. v.

PSKS, Inc., 551 U.S. 877, 885 (2007). [t looks to “the

restraint’s history, nature, and effect,” ibid. (citation

omitted), and asks whether a restraint “merely regu-

lates and perhaps thereby promotes competition or

whether it is such as may suppress or even destroy com-

petition,” Board of Trade v. United States, 246 U.S. 231,

238 (1918). The plaintiff in a rule-of-reason case bears

the initial burden to show that the challenged restraint

is “prima facie anticompetitive.” California Dental

Ass'n v. FTC, 526 U.S. 756, 771 (1999). If the plaintiff

makes that showing, the burden shifts to the defendant

to establish any “procompetitive justification.” Jbid.;

see Pet. App. 108a-109a.

b. The district court began its rule-of-reason analy-

sis by defining the relevant antitrust market, which es-

tablishes the “context for the remainder of [the] analy-

sis.” Pet. App. 111la; see id. at 111la-148a. An antitrust

market is defined as those products “that have reason-

able interchangeability for the purposes for which they

are produced,” such that customers would switch from

one product to another if faced with a price increase.

United States v. EI. du Pont de Nemours & Co., 351 U.S.

377, 404 (1956). Here, the district court concluded that

the market affected by the anti-steering rules is the

market for “general purpose credit and charge card net-

work services” of the type that respondents provide to

merchants. Pet. App. 112a-113a.

The district court rejected respondents’ contention

that this market definition should be expanded to in-

clude services to cardholders as well as merchants. Pet.

App. 114a-122a. The court agreed with respondents

5

that the credit-card business is “two-sided,” in that

credit-card networks compete to attract both mer-

chants and cardholders. /d. at 121a-122a; see id. at 77a-

78a. The court emphasized, however, that the services

respondents provide to merchants are not interchange-

able with the services they provide to cardholders. /d.

at 118a-119a.

Although the district court rejected respondents’

proposed market definition, it recognized that the “two-

sided” nature of respondents’ platform is relevant to the

antitrust inquiry because “the antitrust significance of

a restraint that nominally affects conduct on only one

side of [respondents’] platform cannot be assessed with-

out considering its impact on the other side of the plat-

form.” Pet. App. 121a-122a. The court thus considered

the interdependence between the merchant and card-

holder sides of respondents’ platform throughout its

rule-of-reason analysis. /bid.

ec. Under the rule of reason, the plaintiff has two

ways of establishing a prima facie case that the chal-

lenged restraint adversely affects competition. The in-

direct method requires a showing that the defendant

has market power and that the restraint is potentially

anticompetitive. Pet. App. 108a-109a. The direct

method requires a showing of “an actual adverse effect

on competition.” /d. at 108a (citation omitted). Here,

the district court held that the United States and the

States had carried their burden under both methods.

Id. at 148a-228a.

i. The district court first held that respondents have

market power. Pet. App. 148a-19la. The court ex-

plained that respondents capture 26.4 percent of the

market for network services, a concentrated market

with significant barriers to entry. Jd. at 150a-156a. The

6

court noted that respondents’ market power is magni-

fied by “cardholder insistence”—the fact that many

merchants cannot practically refuse to accept Amex

cards because many of their customers would shop else-

where if they did so. /d. at 156a-165a. The court also

relied on respondents’ pricing, including their proven

ability to significantly increase fees without causing

merchants to stop accepting Amex cards. /d. at 165a-

180a.

ii. The district court further held, in the alternative,

that respondents’ anti-steering rules had caused “ac-

tual, sustained adverse effects on competition.” Pet.

App. 198a (citation omitted); see id. at 19la-228a. The

court found that “[p)rice competition is a critical avenue

of horizontal interbrand competition, and yet it is frus-

trated to the point of near irrelevance in the network

services market as a result of [respondents’ anti-steering

rules}.” Jd. at 195a. The court explained that “[s}teering

is a lynchpin to inter-network competition on the basis

of price” because, once a merchant decides to accept

cards from a particular network, the choice of the card

used for a particular transaction lies with the card-

holder. /d. at 196a. By barring merchants from encour-

aging cardholders to use less-expensive cards, the anti-

steering rules ensure that “there is virtually no check

on the networks’ incentive or ability to charge higher

prices to merchants, so long as the network’s pricing is

below the level at which a rational merchant would drop

acceptance entirely.” /d. at 197a.

The district court also found that the anti-steering

rules “render it nearly impossible” for a new network to

enter the market “by offering merchants a low-cost-

alternative to the existing networks.” Pet. App. 208a.

The court explained that Discover had tried to pursue

7

such a low-fee strategy in the 1990s and had been

thwarted by rules that “denied merchants the ability to

*** steer share to Discover’s lower-priced network.”

Id. at 205a. Discover therefore had “abandoned its low-

price business model” and had raised its merchant fees

to align with those charged by Visa and MasterCard.

Id. at 206a.

The district court further found that, by stifling price

competition, the anti-steering rules “allowed all four

networks to raise their [merchant] fees more easily and

more profitably.” Pet. App. 207a; see, e.g., id. at 166a-

172a (describing respondents’ “Value Recapture” initi-

ative, which significantly raised prices over a five-year

span without causing merchants to stop accepting Amex

ecards). The court emphasized that those higher mer-

chant fees had “also resulted in increased prices for con-

sumers” because merchants “pass most, if not all, of

their additional costs along to consumers in the form of

higher retail prices.” Jd. at 210a-21 1a.

d. The district court’s findings placed on respond-

ents the burden of proving that the anti-steering rules

had procompetitive benefits, and the court held that re-

spondents had not carried that burden. Pet. App. 228a-

258a. Inter alia, the court rejected respondents’ con-

tention that the rules are justified to protect Amex’s

“differentiated business model,” which relies on charg-

ing higher merchant fees to offer more generous card-

holder rewards. /d. at 229a-236a. The court held that,

to find the anti-steering rules reasonable “because they

shield [respondents’] preferred business strategy from

a legitimate form of interbrand competition, especially

competition on the basis of price, would amount to ‘noth-

ing less than a frontal assault on the basic policy of the

8

Sherman Act.’” /d. at 235a (quoting National Soc’y of

Prof'l Eng’rs v. United States, 435 U.S. 679, 695 (1978)).

3. The court of appeals reversed and directed the

entry of judgment for respondents. Pet. App. la-58a.

a. The court of appeals first held that the district

court had “erred in excluding the market for cardhold-

ers from its relevant market definition.” Pet. App. 32a;

see id. at 3la-40a. The court emphasized the “interde-

pendence” of credit-card networks’ competition for

merchants and their competition for cardholders, and it

stated that separating those two avenues of competition

into different antitrust markets could allow “legitimate

competitive activities in the market for [cardholders] to

be penalized no matter how output-expanding such ac-

tivities would be.” /d. at 35a.

b. The court of appeals next held that the district

court had erred in holding that respondents have mar-

ket power. Pet. App. 40a-48a. It concluded that the dis-

trict court should not have focused on increases in re-

spondents’ merchant fees because respondents use a

portion of those fees to provide greater cardholder re-

wards (which provide the functional equivalent of re-

duced prices for cardholders). /d. at 43a-44a. The court

stated that the district court should have calculated

Amex’s “two-sided price”—that is, the aggregate amount

charged to both merchants and cardholders. /d. at 44a

(citation omitted). The court also held that the district

court had erred in relying on “cardholder insistence” as

evidence of market power. /d. at 45a-48a. The court

reasoned that cardholders insist on using Amex cards

only because they like the rewards and other services

that Amex provides, and it believed that “so long as

Amex’s market share is derived from cardholder satis-

faction, there is no reason to intervene.” Jd. at 48a.

9

c. Finally, the court of appeals held that the district

court had erred in holding that the United States and

the States had made a prima facie case that the anti-

steering rules have an actual adverse effect on competi-

tion. Pet. App. 49a-53a. The court did not question the

district court’s finding that the anti-steering rules stifle

price competition among the networks and thereby

cause merchants (and their customers) to pay more.

But the court held that such harms were insufficient to

establish a prima facie case because the district court

had “failed to consider the two-sided net price account-

ing for the effects of the [anti-steering rules] on both

merchants and cardholders.” /d. at 49a. To prove anti-

competitive effects in the form of higher prices, the

court stated, the United States and the States were re-

quired to provide at minimum a “reliable measure of

American Express’s two-sided price that appropriately

account[ed] for the value or cost of the rewards paid to

cardholders.” /d. at 53a (citation omitted). The court

also stated that the United States and the States bore

the “initial burden” to show that the anti-steering rules

“made all Amex consumers on both sides of the platform

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

overall.” Jd. at 5la.

4. The court of appeals denied the United States’ pe-

tition for rehearing and rehearing en banc. Pet. App.

324a-326a.

ARGUMENT

Petitioners contend (Pet. 18-25, 30-35) that the dis-

trict court’s findings established a prima facie case that

the anti-steering rules unreasonably restrain trade, and

that the court of appeals erred in holding otherwise.

The United States agrees. The court of appeals seri-

ously departed from sound antitrust principles, and its

10

decision leaves in place restraints that thwart price

competition in an important sector of the economy and

inflate the retail prices paid by all consumers.

Nonetheless, the United States has not sought this

Court’s review because the case does not satisfy the

Court’s traditional certiorari standards. See Sup. Ct. R.

10. The court of appeals rested its decision almost en-

tirely on the “two-sided” nature of the credit-card in-

dustry, and neither this Court nor any other circuit has

squarely considered the application of the antitrust

laws to two-sided platforms as such. Consistent with its

usual practice of awaiting further percolation in the

lower courts before taking up such novel legal issues,

the Court should deny review here.

1. The district court found that respondents’ anti-

steering rules stifle competition among credit-card net-

works, thereby increasing merchant fees, blocking low-

cost competitors, and inflating the retail prices paid by

all consumers. Although the court of appeals did not

overturn any of those findings, it concluded that they do

not establish even a prima facie case of an antitrust vi-

olation because (a) the relevant market must be defined

to include cardholders as well as merchants, and (b)

proof that a restraint thwarts competition and inflates

prices on one side of a two-sided market is insufficient

to establish a prima facie case that it unreasonably re-

strains trade. As petitioners explain (Pet. 18-25), both

of those holdings were wrong.”

* The court of appeals also held that the United States and the

States had failed to establish a prima facie case under the indirect

method because respondents do not have market power. Pet. App.

40a-48a. Petitioners do not seek further review of that holding. See

Pet. i, 18-25.

1]

a. A rule-of-reason analysis focuses on the relevant

antitrust market, which is “the area of effective compe-

tition” directly and immediately affected by the chal-

lenged restraint. United States v. E.J. du Pont de

Nemours & Co., 353 U.S. 586, 593 (1957). An antitrust

market is “composed of products that have reasonable

interchangeability for the purposes for which they are

produced.” United States v. E.J. du Pont de Nemours

& Co., 351 U.S. 377, 404 (1956) (citation omitted); see,

e.g., Hastman Kodak Co. v. Image Tech. Servs., Inc.,

504 U.S. 451, 482 (1992); United States v. Continental

Can Co., 378 U.S. 441, 449 (1964). “Because the ability

of consumers to turn to other suppliers restrains a firm

from raising prices above the competitive level, the def-

inition of the ‘relevant market’ rests on a determination

of available substitutes.” Rothery Storage & Van Co. v.

Atlas Van Lines, Inc., 792 F.2d 210, 218 (D.C. Cir. 1986)

(Bork, J.), cert. denied, 479 U.S. 1033 (1987). The mar-

ket thus “must be drawn narrowly” to include only prod-

ucts that are reasonable substitutes. Times-Picayune

Publ’g Co. v. United States, 345 U.S. 594, 612 n.31

(1953); see 2B Phillip E. Areeda et al., Antitrust Law:

An Analysis of Antitrust Principles and Their Appli-

cation 1 565a, at 430 (4th ed. 2014) (“[A] relevant mar-

ket consists only of goods that are reasonably close swb-

stitutes for one another.”).

The court of appeals articulated the correct legal

standard, explaining that the relevant antitrust market

should be defined to include “products ‘reasonably in-

terchangeable by consumers for the same purposes.”

Pet. App. 32a (citation omitted). But the court never

explained how the services that respondents provide to

merchants are “reasonably interchangeable” with the

services that respondents provide to cardholders. They

12

are not. A retailer facing an increase in respondents’

merchant fees could not, for example, avoid those fees

by becoming an Amex cardholder instead of a merchant

that accepts Amex cards.

Rather than faithfully applying the “reasonably in-

terchangeable” standard, the court of appeals empha-

sized that respondents’ competition for merchants and

their competition for cardholders are interdependent.

Thus, the court of appeals observed that “the price

charged to merchants necessarily affects cardholder

demand, which in turn has a feedback effect on mer-

chant demand.” Pet. App. 39a. But as petitioners ex-

plain (Pet. 20-21), it is common for prices in one market

to affect prices in another. That sort of indirect effect

does not mean that the relevant products are reasona-

bly interchangeable or that they should be collapsed

into a single market for purposes of antitrust analysis.

See, e.g., Eastman Kodak, 504 U.S. at 463, 481-482 (dis-

tinguishing the markets for photocopier replacement

parts and services from the market for photocopiers).

“(T]he fact that a firm obtains its profits from two dif-

ferent, non-substitutable groups does not serve to place

the two groups into the same market.” Phillip E.

Areeda & Herbert Hovenkamp, Antitrust Law: An

Analysis of Antitrust Principles and Their Applica-

tion 1 565, at 104 (Supp. 2017) (Areeda & Hovenkamp).

The court of appeals also placed great weight on the

fact that the credit-card industry is two-sided, so that

networks must attract both merchants and cardholders

in order to succeed. Pet. App. 39a-40a. But where, as

here, the different sides of a firm’s two-sided platform

involve distinct competition and products that are not

substitutes, they are properly analyzed as separate anti-

trust markets. That point is well illustrated by this

13

Court’s decision in Times-Picayune Publishing, which

involved a newspaper publisher’s requirement that ad-

vertisements appear in both its morning and evening

papers. The Court explained that “every newspaper is

a dual trader in separate though interdependent mar-

kets” serving advertisers and readers—that is, that the

publisher in effect had a two-sided platform. 345 U.S.

at 610. But because the challenged restraint “con-

cern|ed| solely one of these markets,” the Court limited

the relevant market to reasonable substitutes for news-

paper advertising; it did not treat the two sides of the

platform taken together as a single market. J/bid.; see

id. at 612 & n.31.

Like the markets for newspaper advertisers and

readers, the markets for merchants and cardholders are

distinct spheres of competition, “involving different sets

of rivals and the sale of separate, though interrelated,

products and services to separate groups of consumers.”

Pet. App. 119a. By collapsing the two into a single anti-

trust market, the court of appeals severed market

definition from its purpose and “preventjed] the relevant-

market inquiry from accurately answering the questions

for which it is asked.” Law Professor Amicus Br. 5.

b. Even under the court of appeals’ flawed definition

of the relevant antitrust market, the district court’s

findings were sufficient to establish a prima facie case

that the anti-steering rules unreasonably restrain com-

petition.

i. Respondents’ anti-steering rules are vertical re-

straints subject to the rule of reason, rather than to the

per se rule that is often applied to horizontal restraints.

Pet. App. 29a-30a. But unlike resale-price-maintenance

agreements and many other vertical restraints, the anti-

steering rules do not “reduc[e] intrabrand competition

14

—the competition among retailers selling the same

brand”—-in order to “stimulate interbrand competi-

tion.” Leegin Creative Leather Prods., Inc. v. PSKS,

Inc., 551 U.S. 877, 890 (2007) (emphases added); see,

e.g., State Oil Co. v. Khan, 522 U.S. 3, 14-15 (1997). In-

stead, the purpose and effect of the anti-steering rules

is to suppress interbrand competition between respond-

ents and their rival credit-card networks by making it

impossible for merchants to encourage consumers to

use cards that cost the merchants less. To that end, the

anti-steering rules restrict not simply the merchants’

interactions with Amex customers, but their pricing and

other behavior towards non-Amex customers as well.

The district court’s extensive factual findings de-

scribe the unsurprising results of suppressing price

competition among competing credit-card networks.

The anti-steering rules “create a competitive environ-

ment in which there is virtually no check on the net-

works’ incentive or ability to charge higher prices to

merchants,” so long as a network’s prices remain below

the level that would cause the merchant to stop accept-

ing its cards altogether. Pet. App. 197a. The record

demonstrates that the rules blocked Discover’s low-cost

strategy and prevented merchants from steering their

customers to use lower-cost cards. /d. at 201la-202a,

208a-209a. As a result, the anti-steering rules “have al-

lowed all four networks to raise their swipe fees.” Jd. at

207a; see, e.g., id. at 166a-172a. Those higher fees, in

turn, have resulted in “higher retail prices” for consum-

ers—including both consumers who use Amex cards

and those who do not. /d. at 210a-21la. The financial

impact on Amex cardholders is offset, at least in part,

by respondents’ cardholder rewards. /bid. But con-

sumers who use cash, debit cards, and lower-reward

15

credit cards—who tend to be less affluent—also face

higher retail prices yet “do not receive any of the pre-

mium rewards or other benefits conferred by American

Express on the cardholder side of its platform.” Jd. at

211a.

ii. The court of appeals deemed those findings le-

gally insufficient to establish even a prima facie case

that the anti-steering rules unreasonably restrain trade

because—in the court’s view—they did not adequately

account for the anti-steering rules’ purported benefits

to Amex cardholders. At times, the court appeared to

fault the United States and the States for failing to ac-

count for all possible procompetitive benefits of the

anti-steering rules at the first step of the burden-

shifting framework. For example, the court stated that

the United States and the States bore the “initial bur-

den” of “show[ing] that the [anti-steering rules} made

all Amex consumers on both sides of the platform—.e.,

both merchants and cardholders—worse off overall.”

Pet. App. 51a; see id. at 49a n.52 (faulting the district

court for “failling] to take into account offsetting bene-

fits to cardholders”); id. at 54a (requiring proof of “net

harm” to cardholders and merchants).

To the extent that the court of appeals’ opinion can be

read to require antitrust plaintiffs to account for a re-

straint’s potential procompetitive benefits in establish-

ing a prima facie case, it seriously distorted the burden-

shifting framework that governs rule-of-reason litiga-

tion. Under that framework, a plaintiff carries its initial

burden by showing that the challenged restraint is

“nrima facie anticompetitive.” California Dental Ass’n

v. FTC, 526 U.S. 756, 771 (1999). That showing “place[s]

the burden of procompetitive justification on [the de-

fendant],” ibid., which is responsible for establishing

16

any “legitimate justifications,” FTC v. Actavis, Inc., 133

S. Ct. 2223, 2236 (2013); see Areeda & Hovenkamp 1

1505, at 171 (“The defendant, being the author of the

restraints, is in a better position to explain why they are

profitable and in consumers’ best interests.”). Under

established law, it is thus the defendant’s burden to es-

tablish a challenged restraint’s procompetitive benefits,

not the plaintiff’s initial burden to anticipate and refute

them.

iii. The court of appeals’ opinion is also susceptible

to a narrower reading, under which the court deemed

the district court’s factual findings insufficient to make

out a prima facie case under the particular theory that

the court of appeals understood the United States and

the States to have advanced here. The court of appeals

stated that the United States and the States could have

carried their burden of proving actual anticompetitive

effects by showing “that cardholders engaged in fewer

credit-card transactions (7.e., reduced output), that card

services were worse than they might otherwise have

been (7.¢e., decreased quality), or that Amex’s pricing

was set above competitive levels within the credit-card

industry (i.e., supracompetitive pricing).” Pet. App.

52a. The court understood the United States and the

States to have relied on supracompetitive pricing, but it

* In one sentence of its opinion, the court of appeals acknowledged

that “[wJhether the [anti-steering rules] had pro-competitive effects

on cardholders—let alone whether any alleged procompetitive effects

on cardholders outweigh ‘anticompetitive’ effects on merchants—has

no bearing on whether [the United States and the States) carried

their initial burden.” Pet. App. 5la. That statement, however, came

only two sentences after the court’s assertion that the United States

and the States bore the “initial burden” of “show[ing] that the [anti-

steering rules} made ali Amex consumers on both sides of the plat-

form * * * worse off overall.” /bdid.

17

deemed proof of inflated prices charged to merchants

insufficient because—in the court’s view—the relevant

market includes cardholders as well.

The court of appeals stated, for example, that the

district court had “failed to consider the two-sided net

price accounting for the effects of the [anti-steering

rules] on both merchants and cardholders.” Pet. App.

49a. The court emphasized that “the revenue earned

from merchant fees funds cardholder benefits,” which

effectively reduce prices on the cardholder side of the

market. Jd. at 50a. The court concluded that the United

States and the States could not carry their burden of

showing actual anticompetitive effects using evidence of

increased prices absent “a reliable measure of [respond-

ents’}] two-sided price that appropriately accounts for

the value or cost of the rewards paid to cardholders.”

Id. at 58a (citation omitted).

The court of appeals erred in holding that proof that

the anti-steering rules block competition and inflate

fees charged to merchants was insufficient to establish

a prima facie case absent evidence of an increase in

respondents’ “two-sided” price. Federal antitrust law

“rests on the premise that the unrestrained interaction

of competitive forces will yield the best allocation of our

economic resources, the lowest prices, the highest qual-

ity and the greatest material progress.” Northern Pac.

Ry. Co. v. United States, 356 U.S. 1, 4 (1958). Preserv-

ing interbrand price competition is especially important

because price is the “central nervous system of the

economy,” United States v. Socony-Vacuum Oil Co.,

310 U.S. 150, 226 n.59 (1940), and “competitive pricing

fis] the free market’s means of allocating resources.”

Broadcast Music, Inc. v. CBS, Inc., 441 U.S. 1, 23 (1979).

18

Indeed, the “Sherman Act was enacted to assure cus-

tomers the benefits of price competition.” Associated

Gen. Contractors of Cal., Inc. v. California State Coun-

cil of Carpenters, 459 U.S. 519, 538 (1983).

Given the central importance of price competition to

the antitrust laws, the district court’s finding that the

anti-steering rules blocked such competition and in-

flated prices charged to merchants was sufficient to es-

tablish a prima facie case that the rules unreasonably

restrain trade. “Under the Sherman Act the criterion

to be used in judging the validity of a restraint on trade

is its impact on competition.” NCAA v. Board of Re-

gents of the Univ. of Okla., 468 U.S. 85, 104 (1984). The

district court’s factual findings establish that, at a min-

imum, respondents’ anti-steering rules had an actual

adverse effect on competition on the merchant side of

the two-sided market identified by the court of appeals.

Although we are not aware of any prior decision

squarely addressing the issue, such a showing of ad-

verse effects on competition in one side of a two-sided

market should suffice to establish a prima facie case

even absent proof that the challenged restraint in-

creased the defendant’s aggregate or “two-sided” price.‘

‘ In fact, the United States and the States showed—and the dis-

trict court found—that the higher merchant fees made possible by

respondents’ anti-steering rules “were not wholly offset by addi-

tional rewards expenditures or otherwise passed through to card-

holders.” Pet. App. 209a. The anti-steering rules thus did “result{]

in a higher net price” on both sides of respondents’ platform. /bid.;

see id. at 166a-167a (“Because [respondents’] Value Recapture ini-

tiatives were not paired with offsetting adjustments on the card-

holder side of the platform, the resulting increases in merchant pric-

ing are properly viewed as changes to the net price charged across

Amex’s integrated platform.”).

19

Respondents are free to charge higher merchant fees

in order to fund more generous cardholder rewards.

Under the antitrust laws, however, that strategy must

succeed or fail under conditions of market competition.

As the district court explained, the Sherman Act “does

not permit [respondents] to decide on behalf of the en-

tire market which legitimate forms of interbrand com-

petition should be available and which should not.” Pet.

App. 240a; accord Areeda & Hovenkamp 1 1505, at 170.

Yet that is exactly what the court of appeals allowed:

Respondents’ anti-steering rules have essentially forced

the entire credit-card industry to channel competition

away from merchant fees and into cardholder rewards.

Even if price competition on the cardholder side of the

market entirely offset the higher merchant fees—which

it did not, see note 4, supra—restraints that create such

a substantial distortion in market competition are

properly viewed as at least prima facie anticompetitive.

The court of appeals erred in holding otherwise.

2. Although the court of appeals’ decision was erro-

neous, this case does not satisfy this Court’s traditional

standards for certiorari. See Sup. Ct. R. 10. Most im-

portantly, the court of appeals’ decision does not di-

rectly conflict with any decision of this Court or another

court of appeals. This Court has decided antitrust cases

involving two-sided platforms, including Times-Picayune

Publishing. But the Court has not squarely considered

questions of market-definition or proof of anticompeti-

tive effects in cases involving two-sided platforms as

such. And so far as we are aware, no other court of ap-

peals has specifically considered the application of the

Sherman Act to two-sided platforms either.

In addition, the scope of the court of appeals’ deci-

sion is unclear. /nter alia, the court emphasized that

20

“(mJarket definition is a deeply fact-intensive inquiry,”

and it did not suggest that its holding in this case would

extend to all cases involving industries that might be

described as two-sided. Pet. App. 32a (citation omit-

ted). To the contrary, the court distinguished an earlier

decision in which it had defined separate markets for

cardholders and merchants in the very industry that is

at issue here. /d. at 33a-36a (citing United States v.

Visa U.S.A., Inc., 344 F.3d 229 (2d Cir. 2003), cert. de-

nied, 543 U.S. 811 (2004)); see id. at 35a (“The relevant

market in this case is not the same as the relevant mar-

ket in Visa.”).°

Further percolation in the lower courts may be espe-

cially useful because of the idiosyncratic character of

the agreements at issue here. Because Amex and the

merchants that accept Amex cards stand in a vertical

rather than a horizontal relationship, the parties and

the courts below treated those contracts as vertical

agreements subject to rule-of-reason analysis. Those

agreements also have a meaningful horizontal aspect,

however, since they restrict the terms on which mer-

chants may deal with non-Amex as well as Amex cus-

tomers and effectively prevent price competition among

competing credit-card networks. See pp. 13-15, supra.

* The only district court that has applied the court of appeals’

decision to a purportedly two-sided market concluded that “(t]he

relevant market for purposes of antitrust analysis may not be two-

sided even though the defendant operates a two-sided platform.”

US Airways, Inc. v. Sabre Holding Corp., No. 11-cv-2725, 2017 WL

1064709, at *8 (S.D.N.Y. Mar. 21, 2017), appeal pending, No. 17-960

(2d Cir. docketed Apr. 5, 2017). That court upheld a jury’s finding

that the market at issue was one-sided, concluding that the court of

appeals’ decision in this case “does not invalidate the jury's finding

of a one-sided market *** in a different industry and with very

different facts.” Jd. at *10.

21

The agreements therefore might roughly be analogized

to contracts in which a manufacturer forbids retailers

to charge more for the manufacturer’s goods than for

specified goods obtained from the manufacturer’s com-

petitors. To date, however, there is no meaningful body

of precedent addressing the antitrust status of agree-

ments of that character.

If the court of appeals adheres to the approach it fol-

lowed here—and particularly if it reads the decision be-

low broadly—this Court’s intervention may be war-

ranted in an appropriate future case. But the Court or-

dinarily awaits the development of a conflict among the

lower courts before exercising its certiorari jurisdic-

tion. At minimum, that course will allow the Second Cir-

cuit to clarify the scope and content of its holding. And

if this Court’s review is ultimately warranted, additional

percolation in the lower courts may assist the Court in

its application of general antitrust principles to two-

sided platforms and to agreements of the sort at issue

here.

CONCLUSION

The petition for a writ of certiorari should be denied.

Respectfully submitted.

JEFFREY B. WALL

Acting Solicitor General

PATRICIA A. BRINK

Director of Civil

Enforcement

KRISTEN C. LIMARZI

ROBERT B. NICHOLSON

JAMES J. FREDRICKS

NICKOLAI G. LEVIN

Attorneys

AUGUST 2017