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

No. 16-1454

In the Supreme Court of the United States

STATES OF OHIO, CONNECTICUT, IDAHO, ILLINOIS, IOWA,

MARYLAND, MICHIGAN, MONTANA, RHODE ISLAND, UTAH,

AND VERMONT,

Petitioners,

Vv.

AMERICAN EXPRESS COMPANY, AND AMERICAN EXPRESS

TRAVEL RELATED SERVICES COMPANY, INC.,

Respondents.

ON WRIT OF CERTIORARI TO THE UNITED STATES

COURT OF APPEALS FOR THE SECOND CIRCUIT

BRIEF FOR AMICI CURIAE JOHN M. CONNOR,

MARTIN GAYNOR, DANIEL MCFADDEN, ROGER

NOLL, JEFFREY M. PERLOFF, JOSEPH A. STIGLITZ,

LAWRENCE J. WHITE, AND RALPH A. WINTER IN

SUPPORT OF PETITIONERS

ANTHONY J. BOLOGNESE

Counsel of Record

BOLOGNESE & ASSOCIATES, LLC

1500 JFK Boulevard, Suite 320

Philadelphia, PA 19102

(215) 814-6750

A Bolognese@bolog

.*

Counsel for Amici Curiae

December 14, 2017

i

TABLE OF CONTENTS

INTEREST OF AMICI CURIAE...........0.2:02sc00eeeseee000 1

SUMMARY OF ARGUMENT ...............::::ccceeeeeeeeeeeees 1

ET eiihtiiinctecstetinniiinernimncmrenmeimmmnenateremnes 9

DOI NIIIIE sisncccescsncoseriencesrnnsetsceninsennctios 9

Il. PLATFORM COMPETITION IN TWO-SIDED

FE cekecesnieitiiicinastvarisnpnimetiniseuserswsusantsin 13

iil. THE AMEX RESTRAINTS HARM

HORIZONTAL COMPETITION AMONG

CREDIT CARD PLATFORMS ..............0::c:c0000 16

IV. CROSS-MARKET EXTERNALITIBG.............. 24

V. MARKET POWER, MARKET DEFINITION

RTEP PEED chtcncscnctvscciccnsccsscceseesveenvesence 29

VI. INCREASED AMEX OR CREDIT CARD

OUTPUT DOES NOT DEMONSTRATE THE

ABSENCE OF ANTICOMPETITIVE IMPACT34

I oeccnennincensesnninsenevsiesensmnensscetnveenionototeinne 36

il

TABLE OF AUTHORITIES

Cases

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

I i lis 16

Times Picayune Pub. Co. v. United States, 345 U.S.

a iasiemibaaibioion 5, 30

L_S. Airways, Inc. v. Sabre Holdings Corp., 2017 U.S.

Dist. LEXIS 40932 (S.D.N.Y. Mar. 21, 2017)....... 10

United States v. Microsoft Corp., 253 F.3d 34 (D.C.

EEE RRS Ti ee EA Sat As cv NE Oe OE 5

Other Authorities

A. Boik & K. Corts, The Effects of Platform Most-

Favored Nation Clauses on Competition and Entry,

Ee eiteticceccescnsevndnnccastiiasenceves 12

A. Dixit and J. Stiglitz, Monopolistic Competition

and Optimum Product Diversity, Am. Ec. Rev.

tiie hades ees Spaiilehdenniplensdiedcboniiener 25

D. Carlton & R. Winter, Vertical MF'N’s and the

Credit Card No-surcharge Rule ..................... 11, 12

G. Stigler, Price and Non-Price Competition, 76 J.

Pe I Bi GI icatiecnceseccnccecnceessccersasenisneiceses 24

J. Johnson, The Agency Model and MFN Clauses

ty GREER TREE aie, reset ae ee 12

J. Rochet and J. Tirole, Two-Sided Markets: A

Progress Report, Rand J. of Econ. 37(3) (2006)...... 9

Lear, Can “Fair” Prices Be Unfair? A Review of Price

Relationship Agreements, UK Office of Fair

Trading, Paper #1438 (2012)..............:ccccccceeeeeeeees 12

iii

M. Rysman, 7he Economics of Two-Sided Markets, ¢/.

of Econ. Perspectives 23(3) (2009)

S. Salop & S. Morton, Developing an Administrable

MFN Policy, Antitrust (2013) .............0cccccceecceeeeees 12

S. Schuh et al., An Economic Analysis of the 2011

Settlement Between the Department of Justice and

Credit Card Networks, J. of Competition Law &

Econ. (2012)

l

INTEREST OF AMICI CURIAE

The amici are eight economists — scholars and

experts in competition, industrial organization, and

the economic analysis of antitrust issues.' Amici

respectfully submit that sound antitrust economics

supports judgment in favor of the Petitioners. The

Second Circuit’s ruling sets new and flawed antitrust

standards for analyses of two-sided markets,

standards that are based on a_ fundamental

misunderstanding of the economics of such markets.

Unless corrected by this Court, the decision will

result in anticompetitive distortions not only in the

credit card industry, but also in the many extant and

emerging industries that involve or are alleged to

involve two-sided platforms.

SUMMARY OF ARGUMENT

The question presented to this Court is

whether, under the "Rule of Reason," a showing that

i The attached Appendix identifies the amici The amici

include Joseph Stiglitz, who is currently consulting and is an

expert witness for a number of national supermarket and

drugstore chains that have challenged Amex’s, Visa's, and

MasterCard’s restraints in related lawsuits pending in the

No counsel for a party authored this brief in whole or in

part, and no party or their counsel made a monetary

contribution intended to fund the preparation or submission of

this brief. No person other than amici or their counsel made a

monetary contribution to its preparation or submission.

Counsel for Petitioners and Respondents have consented

to the filing of this brief through blanket consents to the filing

of Amicus Curiae briefs.

2

the Amex Merchant Restraints? suppressed price

competition on the merchant side of the credit card

platform suffices to prove anticompetitive effects and

thereby shift to Amex the burden of establishing any

procompetitive benefits from the Restraints? The

Second Circuit held that for the Government to prove

that the Amex Restraints were anticompetitive, the

Government bore the burden to show aot just that

the restraints had anticompetitive pricing effects on

the merchant side, but also that those

anticompetitive effects were not outweighed by any

benefits on the cardholder side.

The amici submit that the proper analysis of

the Amex Restraints is straightforward — the

Restraints impede competition among credit cards in

the pricing of their services to merchants.

Demonstration of this paradigmatic anticompetitive

effect is, amici submit, sufficient to shift to Amex the

burden of demonstrating procompetitive effects from

its Restraints.

In this brief, amici show that the Second

Circuit was incorrect in its analysis of two-sided

platforms and markets. But, more importantly,

amici submit that it is unnecessary for a court to

delve into the difficult economics of two-sided

2 The Court of Appeals summarized the Amex Merchant

Restraints (or “Amex Restraints” or “Restraints”) as “barring

merchants from (1) offering customers any discounts or

nonmonetary incentives to use credit cards less costly for

merchants to accept, (2) expressing preferences for any card, (3)

disclosing information about the costs of different cards to

merchants who accept them.” Pet. App. 4a.

3

markets in the first stage of a Rule of Reason

analysis of the Amex Restraints. Contrary to the

Second Circuit's ruling, the proper analysis in the

first stage of the Rule of Reason analysis is not

whether the two-sided aspects of Amex’s business

lead to impacts from its Restraints on both sides.

Rather, the critical determination is whether a

restraint that impedes price competition among

credit card companies regarding merchant fees

demonstrates an anticompetitive impact, thus

satisfying plaintiffs’ initial burden. Amici submit

that the answer is clearly that it does. If the two-

sided nature of Amex’s business somehow justifies

such suppression of horizontal price competition at

the merchant level, this should be for Amex to

demonstrate in the second stage of the analysis.

A tworsided platform is characterized by a

seller that brings together two different sets of users

(the two sides), and where increased usage on each

side benefits the other side (referred to in economics

as two-sided externalities). The Court of Appeals

assumed that because Amex must attract both

cardholders and merchants, it must be a two-sided

platform with the requisite significant feedback

effects from Amex’s merchant pricing on cardholders

and vice versa. But the fact that a firm such as

Amex must attract two sets of users does not mean

there are such feedback effects. The Court of Appeals

3 The Court of Appeals characterized this as “price

changes on one side can result in demand changes on the other

side.” U.S. v. Am. Express Co., 2016 WL 5349734, at *3 (2d Cir.

Sep. 26, 2016). This is not strictly correct as it is increased

usage on one side that increases value on the other side. Of

course, typically changes in prices change usage.

4

allows for a very broad conception of two-sided

platforms, one under which almost every firm—

ranging from grocery stores (who need food buyers

and food producers) to delivery companies (who need

senders and receivers)—could qualify as a two-sided

platform.‘

Regardless, assuming Amex is a_ two-sided

platform, then its pricing on one side of its platform

would impact demand on the other side. This

complexity is perhaps what led the Court of Appeals

to rule that the assessment of the anticompetitive

impact of restraints supporting high prices to one

side must consider the impact on the other side. But

that is certainly not a sufficient basis to change

established antitrust analysis. For example, for any

firm selling complements (complements are products

used together like tennis rackets and tennis balls),

changes in the price of one good will change the

demand for the other good.5 The fact of such pricing

interrelationships does not imply that the proper

analysis of an anticompetitive act regarding one

product must take account of the impact on the other

products. Nor do such pricing interrelationships

imply that the relevant economic market for

4 Before reconfiguring the established and _ well-

understood Rule of Reason analysis for a mature network like

Amex, careful analysis should be conducted to ascertain the

significance of any feed-back effects from pricing to each side

from the other side (two-sided externalities), and whether there

are ways of appropriately incorporating the analysis of

allegedly two-sided markets within the standard Rule of Reason

framework.

a This positive cross-elasticity of demand is what

economically defines complements.

5

assessment of competitive impact should include all

goods impacted by a price change of one product.

While the Court of Appeals ruling purports to

be based on the economics of “two-sided markets,” it

departs sharply from prior antitrust analyses and

rulings involving two-sided markets. Formal

economic analysis of two-sided platforms is relatively

new, but antitrust analysis of industries involving

two-sided platforms is not. For over fifty years,

courts have analyzed the competitive impact of

restraints on one side of a two-sided platform by

focusing on how competition among competing

suppliers on that side of the market is affected.

In this case, the Court of Appeals departs from

this standard analysis by requiring a plaintiff to

show that a competitive harm on one side of a

platform (here the merchant side) is not offset by

purported benefits on the other side of that same

6 See, e.g., Times Picayune Pub. Co. v. United States, 345

U.S. 594, 610 (1953) (confining competitive analysis to

“advertising market, not in readership” while noting that “every

newspaper is a dual trader in separate though interdependent

markets.”); U/nited States v. Microsoft Corp., 253 F.3d 34 (D.C.

Cir. 2001) (en banc) (analysis of restraints on entry of

competing browsers).

Under the Court of Appeals ruling, in Times Picayune at

the first stage of the Rule of Reason analysis, the government

would have to show that any benefits to readers from high

advertising prices do not outweigh the anticompetitive effect of

those high prices. In Microsoft, the government would have to

show that any benefits to, for example, application developers

from Microsoft's monopoly over browsers, do not outweigh the

adverse effects of that monopoly.

6

platform (the cardholder side). To the extent that a

Court departs from established Rule of Reason

analysis and allows harms and benefits to two

distinct sets of consumers to be balanced, an

approach amici do not endorse, then Amex, as the

proponent and enforcer of the Restraints, is in the

best position to understand and quantify any

cognizable benefits to cardholders resulting from any

relevant two-sidedness of its business. Requiring the

government or a private plaintiff victim of a restraint

of trade to prove that its harm is not offset by

benefits to third parties would create an

unwarranted, inefficient and unreasonable burden.

If not stymied by the Amex Restraints, credit

card platforms (Amex, Visa, MasterCard, and

Discover) should compete against each other on

price. The price competition would be on each side of

the platform. Each credit card platform offers a pair

of prices consisting of the price charged to merchants

for the use of the platform’s card acceptance services,

and a price charged to cardholders (including

benefits in the form of “rewards” and other charges).

One platform may choose to compete by offering a

high merchant price and high rewards to

cardholders. Another platform may choose to offer

lower merchant fees and lower cardholder rewards,

expecting that merchants will steer its customers to

that platform’s cards by offering discounts or other

incentives at the point of sale due to the

comparatively low merchant fees.

With competition in the market, each network

chooses its preferred price pair. Market forces,

which can include differential pricing or other

benefits such as a preferred checkout line at the

7

point of sale, would sort out how much of that

platform’s services will be demanded — that is, its

success in the market. The Amex Restraints directly

interfere with this competition among the credit card

platforms. With the Amex Restraints in place, a

rival credit card platform or a new entrant that

attempts to compete against Amex with lower

merchant fees will likely be unsuccessful. The Amex

Restraints prohibit merchants from passing that low

fee on to the cardholders — the people who chose

which card to use. Thus, a credit card platform with

lower merchant prices than Amex can garner little

additional sales from its low merchant prices; it will

be driven by the Amex Restraints towards Amex

pricing.

Competition requires that (1) credit card

platforms can freely choose their prices, (2)

consumers have choices of card network rewards

versus the merchant-offered discounts and

incentives, and (3) competitive market forces

determine how much of alternative platforms’

services will be demanded. The Amex Restraints

prevent the price system and competition from

working properly. Prices should guide decision

making. But the Restraints prevent the merchant

from reflecting the high Amex price it pays in the

prices it charges customers for the use of this high-

priced payments service.

Amici respectfully submit that the proper

analysis of restraints imposed on one side of a two-

sided platform is the established Rule of Reason

analysis previously recognized by the Second Circuit

and adopted by many other Circuits. Under this

standard analysis, the first step is to determine if a

8

restraint, whether on one or both sides of a platform,

injures competition between and among platforms on

the “side” where the restraint is imposed.’ If a

plaintiff satisfies this showing, the defendant can

then show procompetitive benefits that may or may

not offset the anticompetitive impacts.®

The Second Circuit erred by disregarding the

critical antitrust issue — the impacts of the Amex

Merchant Restraints on competition among credit

card platforms. The Second Circuit erroneously (1)

assumed that the characterization of the Amex

service as a two-sided platform should fundamentally

change the traditional antitrust principles that

govern the Amex Restraints, (2) misunderstood the

nature of competition in two-sided markets, (3)

incorrectly defined the relevant market in which to

assess anticompetitive impact as including both

merchants and cardholders, and (4) presumed that

reductions in Amex cardholder prices are a

procompetitive benefit, and then placed the burden

on plaintiffs to prove that the harm from supra-

competitive merchant fees are not outweighed by

benefits to third-party cardholders.

7 In the credit card market, all firms operate platforms

dealing with both sides of the market. Amici hence refer to

competition among platforms. In other settings, some marset

participants may operate on only one side. Generally, the

proper inquiry at the first stage would be on the s ‘e where the

restraint is imposed.

‘ As amici explain below, the impact on the prices

charged to the other side of the platform is not a procompetitive

benefit.

9

If the ruling of the Court of Appeals stands,

the adverse impact will be substantial. Credit card

platforms process trillions of dollars of transactions

in the United States annually. More importantly,

firms operating in two-sided markets using the

internet, such as Amazon, Uber, Facebook, Google,

and Airbnb, are multiplying in number and size. The

Court of Appeals decision gives firms in these rapidly

developing markets latitude to act anticompetitively

through an action on one side of their platform as

long as they can point to some indirect or secondary

benefit on the other side of the platform populated by

a different set of consumers.° The Court of Appeals

ruling will make analyses of such conduct needlessly

complex and, perhaps, effectively place such

restraints beyond reach of the antitrust laws.

ARGUMENT

I. “TWO-SIDEDNESS”

As noted, a two-sided platform brings together

two sets of users, in a situation where the usage by

each set (side) significantly affects the other set (two-

sided externalities).'!° Two-sided industries have

included newspapers, television and radio, computer

operating systems, dating services, and travel

9 All firms that intermediate transactions between

suppliers and consumers can raise aspects of two-sidedness.

10 See Jean-Charles Rochet and Jean Tirole, 7wo-Sided

Markets: A Progress Report, Rand J. of Econ. 37(3) (2006), pp.

645-67; M. Rysman, The Economics of Two-Sided Markets, J. of

Econ. Perspectives 23(3) (2009), pp. 125-143. The literature

mainly focuses on the optimal pricing of a platform, and not the

more difficult issue at hand — the effect of restraints on

competition among platforms.

10

reservation services.'! In such two-sided industries

where price on one side affects usage on the other,

the profitable pricing of the two-sided platform is a

simultaneous two-sided price-pair decision.

The Court of Appeals emphasized that Amex

must attract both cardholders and merchants to its

network.!? But this does not distinguish Amex from

other firms that need to attract two different sets of

users. Nearly any firm intermediating between two

groups could offer the Court of Appeals’ flawed

analysis that a restraint that raises the firm’s price

to the one side might pass antitrust muster if the

restraint provides the firm with revenue that it

spends on enhancing the quality of the products it

offers to the “other side.” A new antitrust doctrine

should not follow from a mere labeling of a

conventional impact just because it rises in a two-

sided platform. '8

i We write “have included” because as_ tworsided

platforms mature, the externalities from each side to the other

can become unimportant and insignificant, rendering the two-

sidedness of no relevance. For example, a mature flight

reservation system may not attract another airline if it adds

more users, and vice versa. If so, there are no remaining

significant two-sided externalities. See U.S. Airways, Inc. v.

Sabre Holdings Corp., 2017 U.S. Dist. LEXIS 40932 at 32

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

12 See Pet. App. 49a-50a.

18 In a recent paper co-authored by one of the amuci, the

authors conclude that “[t]he two-sidedness of credit card

markets does not require a new set of economic principles for

assessing competition policy because the difference between the

credit card setting and a conventional one-sided market is

essentially a matter of labeling... Creating different legal rules

for the same economic conduct depending on whether the

ll

The economic literature analyzing two-sided

platforms is new, complex, and evolving." If the

Court of Appeals were correct that different and

novel economic analysis is required in two-sided

markets — a proposition with which amici disagree —

such analysis should be applied only after a rigorous

and careful demonstration that two-sided market

characteristics exist and are important to the

competitive impact of a restraint.

This is especially important in the case of

Amex. Only recently has the economic literature

considered the impact of restraints on competition

that allow a firm like Amex to charge higher prices to

one side (here, merchants), which results in

increased prices charged to a// the merchants’

customers, and which, through competition on the

other side (cardholders), may result in lower prices to

market can be described as one-sided or two-sided is a mistake

that could lead to widespread confusion in the evaluation of

vertical restrictions.” D. Carlton & R. Winter, Vertical MFN's

and the Credit Card No-surcharge Rule, at 40 (working paper

available at https://goo.gl/kKd2Ck).

14 As an example of its incomplete understanding of aspects

of two-sided platforms, the Court of Appeals cites a 2013

working paper by Filistrucchi et al. (Since published in J.

Comp. Law & Econ., 2014.) See Pet. App. 7a n.3 (citing Lapo

Filistrucchi et al., Market Definition in Two-Sided Markets:

Theory and Practice 5). The analysis therein emphasizes a

possible important distinction for competitive analysis between

two-sided transactions markets (where, as with payment cards,

the two sides directly interact with one another), and two-sided

non-transactions markets (where, like newspapers, the two

sides do not interact with one another), While noting this

paper, the Court of Appeals court simply disregards the

distinction.

12

that side.'° The Court of Appeals analysis and ruling

fails to account for this recent learning. '®

Instead, the Court of Appeals took a novel and

unprecedented approach to analyzing the competitive

impact of the Amex Restraints. Rather than

determining whether the _ Restraint injured

competition among platforms — that is, among

competing credit card firms — on the one side, the

court ruled that the proper antitrust analysis must

“consider the two-sided net price accounting for the

effects of the [restraints] on both merchants and

16 The price to the cardholder side may include annual

fees, transactions fees net of discounts, and rewards. There can

be no presumption that restraints leading to higher prices

charged to merchants result in lower cardholder prices, much

less an overall lower “total price.” Nonetheless, as Amici

emphasize below, even if there is a reduction in the total price,

that does not imply increased efficiency.

16 The recent economic literature finds that restraints

such as the Amex Restraints “typically raise platform fees and

retail prices, and also curtail entry or skew positioning

decisions by potential entrants pursuing low-end business

models.” A. Boik & K. Corts, The Effects of Platform Most-

Favored Nation Clauses on Competition and Entry, J. of Law &

Econ., Abstract (2016). See, also, S. Schuh et al., An Economic

Analysis of the 2011 Settlement Between the Department of

Justice and Credit Card Networks, J. of Competition Law &

Econ. (2012); S. Salop & F. Scott Morton, Developing an

Administrable MFN Policy, Antitrust (2013); Lear, Can “Fair”

Prices Be Unfair? A Review of Price Relationship Agreements,

UK Office of Fair Trading, Paper #1438 (2012); J. Johnson, The

Agency Model and MFN Clauses (Jan. 25, 2017) (available at

https://goo.g/Vbi3tV); D. Carlton & R. Winter, Vertical MF'N’s

and the Credit Card No-surcharge Rule (available at

https://goo.gVkKd2Ck).

13

cardholders.”'? Furthermore, the Court held that it

was the plaintiffs’ burden to show that the Amex

Restraints had an adverse net effect, defined by the

Court as the sum of the prices to merchants and

cardholders. Only after the plaintiff has passed this

hurdle would Amex be obligated to come forward

with any evidence of a procompetitive justification.

Il. PLATFORM COMPETITION IN TWO-SIDED

MARKETS

The Court of Appeals’ emphasis on the net

two-sided price does not assess the competitive

impact of the Amex Restraints. If a market were

demonstrated to consist of two-sided platforms, and

if benefits to consumers using one side of a platform

(here cardholders) result from restraints that harm

the other side (merchants), then a simple summing of

these benefits and harms is not informative as to the

restraint’s impact on competition.

The correct approach is rather to determine

whether a restraint on one side of a_ tworsided

market interferes with competition among platforms

in the market. The competitive impact of restraints

such as those imposed by Amex — restraints that

directly alter and impede horizontal competition

among platforms — is properly demonstrated by

showing the interference with competitive setting of

the platforms’ pricing, whether on one side or both.

The Court of Appeals disregarded this most critical,

and amici submit dispositive, economic issue — how

the Amex restraints affect competition among Amex,

Visa, MasterCard, Discover, and potential new

17 Pet, App. 49a.

14

entrants by distorting the competitive process

leading to the platforms’ prices. '8

To better understand competition in two-sided

markets, consider the example of platforms

intermediating between hotels and travelers. These

platforms can be two- sided if the demand for the

services by travelers increases when more and better

hotels use the service, and vice versa. If a platform

lowers its price to travelers and this increases the

number of travelers using the platform, then the

platform can be more valuable to hotels. Similarly,

lowering the platform fee charged to hotels can

increase the number of hotels using the platform,

making the platform more valuable to travelers. In

this situation, the hotel booking service providers are

competing in a two-sided market.

Two-sided platforms compete, in part, via the

prices offered by each platfor: ‘0 each side they

serve. For example, one hotel booking service may

charge a high price to hotels and a relatively low

price to travelers, while other platforms may expect

more equal prices to be more profitable, resulting in

a better mix of hotels and travelers. Competition is

likely to result in competing platforms offering

different price pairs, and those offering the price-

pairs that best satisfy consumer preferences will

thrive.

18 Visa and MasterCard operate as what is called four:

party systems (cardholders, merchants, issuers, and acquirers,

see Figure 2 at Pet. App. 55a°56a) in which the Visa and

MasterCard platforms deal with acquiring banks that compete

for merchants and issuing banks that compete for cardholders.

This difference from Amex and Discover does not impact our

analysis.

15

The important economic point is that in two-

sided markets, the relevant competition occurs at the

platform level (ie., competition among the credit

card companies). A competitive two-sided market,

through consumers’ choices, will effectively decide

the preferred and competitive prices to each side,

which, of course, determines the overall “price levels”

(the sum of the prices) for the two sides. It is this

platform competition that is directly interfered with

by the Amex restraints on the merchant side of the

Amex platform.

Rather than asking whether the Amex

Restraints prevented competitive market forces from

determining the price pairs offered by the competing

credit card platforms, the Court of Appeals focused

only on the impact of the Amex Restraints on the

sum of the Amex prices to both sides of its platform.

This is a fundamental economic error. Whether the

sum of the prices remains the same or even falls does

not indicate whether restraints are or are not

anticompetitive. The total price, or change in total

price, simply does not indicate whether the restraints

are distorting and interfering with competition

among platforms. When restraints hamper the

process of platform competition, anticompetitive

harm follows because the restraints alter the price

pairs themselves, regardless of whether the sum of

the prices increases, decreases, or remains

unchanged. !9

19 See P. Areeda & H. Hovenkamp, Antitrust Law § 562e,

p. 101 (Supp. 2017) (stating that the Court of Appeals erred

because “competition should choose the optimal mix of revenue

between the two sides”).

16

Ill. THE AMEX RESTRAINTS HARM

HORIZONTAL COMPETITION AMONG

CREDIT CARD PLATFORMS

The Amex Restraints bar merchants

purchasing Amex services from differentially pricing

the use of Amex cards versus other credit cards. The

Restraints even bar merchants from providing their

customers with accurate information about the prices

charged to merchants by Amex and alternative

cards. This interferes with transparency in prices, a

hallmark of competitive and efficient markets. The

Amex Restraints are vertical restraints, imposed by

a supplier on its customers. However, the Amex

vertical restraints have direct horizontal effects

because they interfere with horizontal competitors’

pricing.

The required competitive analysis of a vertical

price restraint is set forth in Leegin Creative Leather

Prods., Inc. v. PSKS, Inc., 551 U.S. 877 (2007), and

summarized in the Court of Appeals’ decision.”

Leegin concerned resale price maintenance (RPM).

This well-established analysis of RPM finds the first

stage of the Rule of Reason satisfied by showing an

increase in the retail price. The burden then shifts

to the supplier imposing the restraint to show

offsetting procompetitive benefits. The Court of

Appeal failed to follow this approach, and its failure

to do so is not justified by sound ecouomic principles.

With RPM, the suppression of competition at

the retail/merchant level, is a cost to the supplier

imposing the restraint, because the direct effect is

20 Pet. App. 30a-31a.

17

reduced demand for the supplier’s product from the

higher retail price. Therefore, to be of benefit to the

supplier, the decision to impose RPM can be

presumed to have some non-price, demand-

enhancing effects. In contrast, Amex’s suppression of

price competition at the retail/merchant level

provides first-order benefits to Amex — the Restraints

effectively impede its credit card competitors from

competing against Amex’s price to merchants.

Because of this direct and intended reduction in

horizontal price competition among Amex and its

competitors, the Amex Restraints, unlike RPM,

cannot be presumed to be motivated by non-price,

overall demand-enhancing effects.

Additionally, the direct impact of RPM on

competing suppliers is to increase the demand for the

competitors’ products. In contrast, the Amex

restraints provide no direct benefits to Amex’s

platform competitors; rather, the restraints directly

interfere with the other credit card platforms’ ability

to compete with Amex in pricing to merchants. Yet

for analysis of RPM, simply showing an increase in

the retail prices to the buyers is sufficient to satisfy

the first prong of the Rule of Reason analysis — an

anticompetitive impact. It is not sound economics to

reject the rebuttable presumption that the Amex

Restraints are anticompetitive simply because the

services may be two-sided.

The practical effect of the Amex Restraints is

to force merchants that purchase Amex services (and

pay the merchant fee) to set no (or equal) prices for

the use of all cards, regardless of their relative cost

to the merchant. The result is that merchants’

customers paying with credit cards perceive no

18

difference in selecting one credit card rather than

another, and the customers will be motivated to

choose the card considered to offer the highest

cardholder benefits. Consequently, as the district

court below correctly found, if a payment card

platform seeks to compete for transactions with

Amex by offering an identical net price, but with

lower prices on the merchant side, then its effort will

be impeded, not because there is no demand for the

platform’s services, but because the Amex Restraints

effectively suppress the use of prices to drive demand

on the merchant side.2! Therefore, platforms (such

as Discover) that attempt to compete with Amex by

charging lower merchant fees and equal or possibly

lower rewards will realize little benefit from the low

merchant fees.22 As a consequence, the Amex

Restraints suppress horizontal competition among

credit card platforms to increase transactions by

charging lower merchant prices. Similarly, the Amex

Restraints impede competition on the cardholder side

by preventing platforms from offering lower prices to

merchants in exchange for merchants offering more

immediate and more valuable benefits or discounts

to cardholders at the point of sale.

As a result, with the Amex Restraints in place,

competing platforms will be motivated to raise their

merchant price — that is, they will be driven towards

the Amex business model. In so doing, the platforms

will have to abandon other competitive business

21 See Pet. App. 194a-203a.

22 See Dist. Ct. Op., Pet. App. 203a-207a (holding that

Amex’s merchant restraints effectively deny other networks the

opportunity to pursue a business model that differentiates itself

by offering merchants a low price for greater volume).

19

models that they, the retail consumers, the

cardholders, and the merchants might prefer. The

Amex Restraints directly interfere with competitors’

ability to compete with alternative platform models

offering different and potentially efficient price pairs.

This is regardless of whether such competitive price

pairs have equal, lower, or even higher total two-

sided net prices.

The competitive impact of merchant pricing

restraints like Amex’s on new entry is also far

reaching. The Amex Restraints result in all credit

card networks competing for transactions mainly

through the issuing side of the platform where card

issuers (Amex, Discover, and the thousands of Visa

and MasterCard issuing banks) provide cardholder

benefits at some future point in time. Cardholder

benefits can be a way to reduce the net prices paid by

the cardholders. But at best, the benefits are a

discount on the credit card bill from the issuer, and

the discount accomplishes a price reduction only with

a lag in time. The Amex rules prohibit a merchant

from accepting payment cards from a competing

network that offered rewards in the form, for

example, of a point-of-sale discount or other benefits

received at the time of the sale such as a preferred

checkout line. Under the Amex Restraints, any such

point-of-sale benefits would be considered differential

pricing, which is not allowed. Thus, the Amex

Restraints directly interfere with possible entry of

innovative and potentially efficient alternative

23 See Dist. Ct. Op., Pet. App. 216a-217a (finding that

without the Amex Restraints, all four card networks’ merchant

prices would decrease).

20

platform pricing systems, and this adverse entry

effect of the restraints occur in an industry in which

the Court of Appeals notes “continues to be

characterized by formidable barriers to entry.”*4 This

important anticompetitive impact was not given

proper weight by the Court of Appeals. 25

With Restraints in place that impede

horizontal competition regarding pricing’ to

merchants, competition on the cardholder side may

or may not increase, and such competition might or

might not result in a change to the net prices

summed over the two sides. However, such a

reduction in the cardholder price, if it occurs, is not a

“pro-competitive” or an efficiency-enhancing benefit

that offsets the interference with competition on the

merchant side. It is, rather, a further economic

distortion and inefficiency directly due to the Amex

Restraints, enhancing Amex’s market power and

ability to charge supra-competitive prices and impose

anticompetitive restraints.

The Court of Appeals recognized that in two-

sided markets, a platform must “find an effective

method for balancing the prices on the two sides of

24 Pet. App. 17a.

2 The Court of Appeals described Amex’s entry strategy

in the 1950s — “Amex ... attracted merchants by setting its

merchant fee slightly lower than the contemporary Diner's Club

merchant fee.” Pet. App. lla. The Amex Restraints protect

Amex from other competitors following this same entry

strategy.

21

the market.”26 This is the essence of competition in

two-sided markets — identifying and offering a price

pair that attracts both sides to use the platform.

Selecting and offering a preferred price pair,

however, is quite different from the situation in

which restraints on one side of the market allow for

price increases that, through competition on the

other side, may lead to price reductions to the second

side. Even if the total two-sided price were

unaffected, which is unlbs»ly, the resulting price pair

is distorted, other platforms’ pricing is distorted, and,

as discussed further below, the higher merchant fees

distort merchants’ prices to all their customers.

The potential adverse consequences of the

Court of Appeals’ approach — in which indirect effects

on the second side of a two-sided platform must be

taken into account in the initial assessment of

anticompetitive effects — can be readily understood

by viewing the court’s analysis through the lens of

traditional anticompetitive conduct. Consider a case

in which Amex, Visa, MasterCard, and Discover

agree to fix prices by charging higher and perhaps

equal merchant fees. The obvious anticompetitive

harm is the direct interference in platform

competition regarding the prices paid by merchants.

And with merchant restraints in place like those of

Amex, which effectively restrain other credit card

platforms incentives to offer lower merchant prices,

competition through entry cannot solve the problem.

However, if this hypothetical cartel of credit card

platform suppliers does not control competition on

26 Pet. App. 8, n.4 (citing J. Rochet & J. Tirole, An

Economic Analysis of the Determination of Interchange Fees in

Payment Card Systems, 2 Rev. Network Econ. 69, 71 (2003)).

22

the cardholder side, then the result is likely to be

increased cardholder benefits with reduced

cardholder prices, and in the long run, perhaps (but

unlikely given the small number of credit card

competitors) full dissipation of all the profits earned

from the high merchant fees. There is no reasonable

basis to consider the dissipation of profits from price

fixing through competition on one side of the

platform to be an offsetting procompetitive benefit

from fixing the price on the other side.*’

There is substantively no economic difference

between the dissipation of profits from merchant fees

propped up by the Amex Restraints compared to the

dissipation of profits from price fixing. In addition,

the hypothetical collusive price fix among credit card

platforms would not be considered benign if the “net”

price were unchanged because of the full dissipation

of the cartel profits extracted from the merchant side

through competition on the cardholder side.28 The

outcome of “competition” with restraints such as

those imposed by Amex is little different from what

woul emerge from the collusion example — a non-

a7 Consider, for example, increased the product quality

that will be expected in a cartel that can control price more

effectively than product amenities. The increased product

quality does transfer some of the cartel profits from the cartel

members to the buyers. But this also results in a welfare loss,

in that the buyers’ value of the increased quality is lower than

that costs incurred to supply that higher quality (otherwise it

would be supplied under competition). Thus, society ends up

with lower overall welfare for two reasons — 1) higher prices

than the efficient level with the resulting loss of consumer

surplus, and 2) increased quality that costs more than its value.

2a That a price fix would likely be judged under a per se

standard is irrelevant to the economic point.

23

competitive price pair that may or may not alter the

sum of the prices.

Whether from collusion or from vertical

restraints on differential merchant pricing, Amex

and its competitors may benefit during the transition

to long-run equilibrium and the full dissipation of

profits earned from the supra-competitive merchant

prices. Those consumer cardholders that are

fortunate enough to meet the credit and income

requirements for high rewards cards may also

benefit. However, the merchants paying higher

prices to Amex, and the merchants’ customers using

other payment means, are harmed. More

importantly, economic efficiency is impeded as price

signals are distorted regarding choice of payment

means. Customers will be motivated to use their

rewards cards even when cash, debit, or check would

otherwise benefit them, and customers will be

motivated to take inefficient actions to qualify for the

high rewards cards.

The Court of Appeals decision would impose

on those challenging the Restraints the burden of

proving that the harm suffered is not outweighed by

any benefits to card holders, on the other side of the

platform. This requirement would have substantial

adverse impacts on antitrust enforcement. First, as

amici have emphasized, the effects on cardholders

should not be considered offsetting procompetitive

effects of anticompetitive restraints directed at

merchants. Any such benefits to cardholders flow

from the merchant restraints that support the supra-

competitive merchant fees. Foreclosure’ of

competition on some consumers should not be

justified by some of the excess profits being

24

dissipated to other consumers as a result of

competitive forces somewhere else.

Second, even if offsetting effects are

considered, the merchant restraints are imposed by

Amex, and Amex is the party best able to understand

and quantify any relevant offsetting competitive

benefits from its restraints. Perverse incentives

would be created if a platform could avoid antitrust

liability for harm to one side of the platform as long

as the victim could not prove that the spoils were not

fully passed on to the other side.

Third, the Court of Appeals overly broad

conception of two-sided markets could allow any

supplier suppressing retail competition through a

vertical restraint to point to the possibility of

incentives for higher quality, thus shifting the

burden from the defendant where it traditionally has

been placed, to the plaintiff. Because suppression of

price competition in any context carries the

possibility of increased competition in non-price

dimensions,” antitrust enforcement in these settings

would become needlessly complex, expensive, and

uncertain.

IV. CROSS-MARKET EXTERNALITIES

The Amex Restraints increase the prices that

merchants pay to Amex and thereby raise merchants’

costs.°° The Restraints prevent the merchants from

differentially raising retail prices to only Amex

29 See G. Stigler, Price and Non-Price Competition, 76 J.

Pol. Econ. 149, 149-54 (1968).

80 Pet. App. 51a..

25

cardholders to cover the cost increase those

cardholders impose on the merchan.s. The Amex

Restraints then motivate other credit card platforms

to raise their prices to merchants, further increasing

the merchants’ costs.*! Merchants incurring higher

costs in turn raise their prices to all their

customers.*? The result is higher retail prices to all

the merchants’ customers, including those who use

low-cost cash or debit cards. These customers, who

tend to have incomes or credit scores too low to

qualify for high rewards credit cards, will thus end

up subsidizing the rewards of more affluent

cardholders.*° This is further evidence of inefficient

pricing and adverse impacts on consumers that are

not Amex card users. These harms occur even if

Amex passed on all its high merchant fees to

cardholders through higher rewards, and they are

amplified when other credit card platforms increase

their merchant prices and cardholder benefits in

response to the Amex Restraints.

31 Dist. Ct. Opp., Pet. App. 207a-209a.

v2 The retail markets mainly impacted by the Amex

Restraints are competitive to a first approximation such that

the cost increase caused by higher merchant credit card fees

can be presumed to be fully, or nearly fully, passed on to

consumers. However, it is possible that in the full market

equilibrium, the final product prices could increase by more the

supra-competitive component of the Amex merchant fees. See

A. Dixit and J. Stiglitz, Monopolistic Competition and Optimum

Product Diversity, Am. Ec. Rev., (1997). The significant point is

that the analysis of the full effects on consumers of restraints

such as Amex’s is very complex and places an unreasonable

burden on plaintiffs if such full analysis is required.

* Id. at 210a-212a.

26

The Court of Appeals requires that analysis in

two-sided markets go beyond the direct effects on

competition on the side of the platform where a

restraint is imposed. If it is economically relevant to

consider the competitive impact beyond that side of

the platform, then proper analysis should also

consider effects beyond the platform itself, as the

restraint can have broad and_é arguably

anticompetitive effects on consumers who do not

participate on either side of the platform operated by

the firm that imposes the restraint.

However, there is a sound economic basis to

retain the established Rule of Reason analysis in a

two-sided market — a traditional analysis that

focuses on the direct effects of the restraint on the

targeted side of the platform. If the plaintiff

succeeds in demonstrating that the challenged

Restraints adversely impact competition among

platforms, then the plaintiffs initial burden should

be satisfied. With this initial burden satisfied, the

defendant should be required to demonstrate not

simply that the other side of the platform is affected,

but that there is a beneficial impact on competition

among platforms — that is, a procompetitive benefit.

In this regard, it is important to distinguish

reactive changes on the other side of the market from

procompetitive pricing to both sides. A two-sided

platform firm may offer a price pair that efficiently

subsidizes one side with price above marginal costs

to the other side. Consider the classic example of a

two-sided market — newspapers, which bring

together advertisers and readers. A newspaper may

compete by offering free papers to readers with

pricing to advertisers above the costs of running

27

additional ads, or might alternatively try a pricing

strategy of high subscription fees with little

advertising. With competing newspapers, the

market will “solve” efficient pricing, with readers and

advertisers choosing their preferred platform based

on the value offered.

A more complex situation occurs when the two

sides of the market directly interact, for example,

hotel booking platforms noted above. Assume one

booking service charges high fees to hotels, with low

fees to the travelers. When the traveler shows up at

the hotel, the service of the platform has been

consumed. Hence, when the traveler checks in, the

hotel can attempt to “undo” the platform pricing by

imposing a charge on the traveler. This describes a

classic free riding problem in which the hotels seek

to consume the valuable service of the platform that

assembles desirable travelers without paying for that

service.

Courts have significant experience in

addressing vertical restraints that may solve free-

riding problems. Consider resale price maintenance

discussed above. Once the plaintiff has shown that

the restraint causes higher retail prices, the

defendant can demonstrate that the RPM is

necessary to prevent free-riding that could make

point-of-sale services non-viable. And supporting

such point-of-sale services can allow the supplier

imposing the RPM to compete with other suppliers —

a procompetitive impact. Under the standard Rule of

Reason approach, the plaintiff can then counter that

the procompetitive effect does not outweigh the

anticompetitive effect.

28

In the context of the Restraints, a_ hotel

booking service charging high prices to hotels for its

services may argue that a restraint requiring that

hotels using its services charge the same price to

travelers even if booked through another service is

necessary for its viability. However, because of such

a restraint, alternative platform models with lower

prices to hotels become non-viable. The restraint

does not therefore increase competition, but rather

supports the business model of a_ particular

competitor, rather than the market deciding the

preferred business model.

Like hotels facing high fees from a travel

booking service, merchants facing high fees from

Amex deal directly with the other side of the market,

the cardholders. Absent the Amex Restraints,

merchants could attempt to reverse a high merchant

price—low cardholder price offering by imposing a

charge on the card users at the time of purchase for

use of the Amex card. However, it is not clear that

there can be any significant free-riding for most

merchants. In contrast with hotels, retail merchants

deal with many repeat customers. If the value of the

Amex services offered to cardholders exceeds any

benefit the merchant can offer for use of alternative

payment means, then the cardholders will likely,

henceforth, patronize merchants that do not levy

such charges. Thus, the market is likely to solve the

efficient pricing, and the Amex Restraints are

unnecessary. Regardless, Amex is certainly in the

best position to demonstrate whatever

procompetitive benefits it believes might result from

its Restraints.

29

In addition to not allowing the use of prices to

steer consumer choices, the Amex Restraints don’t

even allow the merchants to inform their customers

of the costs to the merchants from the customers’

choice among credit cards. Far from being the least

restrictive means of preventing any (unproven)

merchant free-riding, the Amex Restraints impede

price competition among platforms for’ the

merchants’ business. Such restraints that effectively

eliminate competition among horizontal competitors

are anticompetitive.4

V. MARKET POWER, MARKET DEFINITION

AND INSISTANCE

To cause competitive harm, a firm must have

market power. “[Mlarket power exists whenever

prices can be raised above the levels that would be

charged in a competitive market,”® or whenever a

seller can “require purchasers to accept burdensome

terms that could not be exacted in a completely

competitive market.”*6 Direct proof of competitive

harm demonstrates market power.®7

34 With the inability of merchants to steer, the credit card

merchant prices may continue to escalate to the point where

additional increases lead to merchants not accepting the cards.

This is analogous to a pure monopolist being constrained in

pricing because at high enough prices, customers will stop

buying the product.

36 Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2,

27, n. 46 (1984).

36 United States Steel Corp. v. Fortner Enterprises, Inc.

429 U.S. 610, 620 (1977).

a7 In Amex, the district court found that Amex both

possessed and exercised its ability to charge supra-competitive

30

If there is insufficient direct proof of

competitive harm, defining a relevant economic

market can be useful to identify the firms that

compete in providing the relevant products or

services to the affected customers. Having identified

the competitors, one can then assess whether a firm

or collection of firms controls enough of the market to

take actions detrimental to competition and to

customers.** Market definition is a fact-intensive

inquiry guided by economic analysis, and it turns on

the commercial realities facing customers who

purchase the product or service at issue from the

supplying firms. *9

Relevant market analysis properly focuses on

one level of an economic chain and includes the firms

that compete to sell the product or service to the

customers at that level.4° This established relevant

market analysis has been applied to firms that

operate as intermediaries in a distribution or

production chain. Retailers are intermediaries in

prices and impose burdensome terms on the merchants that it

would not be able to require in a competitive market. United

States v. Am. Express Co., 88 F. Supp. 3d 143, 187-207 (EDNY

2015). These findings, which were not disturbed by the Amex

panel, established market power.

38 See Jonathan Baker, Market Definition’ An Analytical

Overview, 74 Antitrust L.J. 129 (2007).

3e Eastman Kodak Co. v. Image Tech. Servs., Inc, 504

U.S. 451, 482 (1992); Todd v. Exxon Corp., 275 F.3d 191, 199-

200 (2d Cir. 2001) (Sotomayor, J.).

40 See Times-Picayune Pub. Co. v. United States, 345 U.S.

594, 610 (1953).

31

transactions between suppliers and consumers;

manufacturers are intermediaries in transactions

between input suppliers and goods purchasers. The

fact that such firms transact with two distinct groups

does not imply the need to define a two-sided market.

The anticompetitive harm that occurs, for example,

when retailers collusively raise the prices of final

goods sold to customers is properly assessed within

the relevant economic market defined by the close

substitutes for the final goods—a one-sided market.

Credit card companies compete for merchant

transactions and for cardholder usage. The

possibility that increased prices on one side (the

merchant side) may result in some benefits to the

other side should not alter the established market

definition analysis. Amex’s services to cardholders

are in no way substitutes for its services to

merchants. And while there may be some

interdependence in the pricing to cardholders and

merchants, such interdependence in pricing provides

no justification for including distinct and non-

substitutable services in the same relevant market.*!

Consider a firm offering complementary

services, for example, a pharmacy that dispenses

drugs and also offers in-store physician consultation

and prescriptions. If that pharmacy can set its

prescription drug prices at supra-competitive levels,

the demand for, and profit maximizing price of

physician services will fall. This is neither a pro-

41 See Phillip E. Areeda & Herbert Hovenkamp, Antitrust

Law: An Analysis of Antitrust Principles and Their Application

{ 565, at 104 (Supp. 2017) (“[Tlhe fact that a firm obtains its

profits from two different, non-substitutable groups does not

serve to place the two groups into the same market.”)

32

competitive offset to the supra-competitive pricing,

nor a reason to combine the services for the purposes

of assessing market power. If the pharmacy can set

supra-competitive prices for prescriptions, it has

market power and the sale of prescription drugs is a

relevant product market.*? Analogously, if Amex can

set supra-competitive prices to merchants, it has

market power, and the sale of credit card services to

merchants is a relevant product market.

Here, the Court of Appeals also confused the

source of Amex’s market power. Amex cardholders

may prefer to patronize merchants that offer them

effective discounts via high rewards. Indeed, some of

the cardholders will switch to competing merchants

if one merchant charges for the card use or declines

to accept Amex cards. Amex refers to this cardholder

loyalty as “insistence.”

To understand Amex’s market power, consider

Amex as a credit payment means supplier that

attracts buyers by promising discounts for purchases

from merchants. In expectation of such discounts

some of the buyers may elect to patronize only

merchants that accept Amex, and with only a small

42 The Court of Appeals confusion on this issue may result

from the fact that in two-sided markets with two-sided

externalities, a supra-competitive price to one side cannot be

identified without consideration of the other side. This fact,

however, does not justify including both sides in the same

market. Indeed, a very similar problem is present even when

both sides are considered — the combined net prices of the two-

sides being equal to combined marginal costs also is not

sufficient to imply that either price is competitive.

33

number of such insistent members,*® merchants will

be incentivized to accept Amex cards even though the

Amex fee is above the competitive level.‘

The implied threat of losing the Amex-

insistent card users allows Amex to charge the

merchant supra-competitive fees, fees in excess of

the costs of providing the payment functions, with

some of those fees being rebated via rewards

(discounts) to the card users. Amex does “attract

customer loyalty” by discounts to those customers,

and that loyalty then allows it to set fees to the

merchants above the amounts of the discounts to the

43 For example, assume a merchant is a typical grocery

store with a gross margin of 40%. If the differential costs of

accepting Amex or other high reward credit cards is 2%

compared to lower costs alternative payment means, then if

only 5% or more of Amex cardholders switch to an alternative

merchant in response to a 2% charge, the merchant would

accept the supra-competitive Amex fee.

44 The Second Circuit asserted that “[t/here is no

meaningful economic difference between ‘dropping American

Express’...and a decision not to accept American Express in the

first place.” Pet. App. at 46a-47a. This is incorrect as a matter

of economics. The initial participants in Amex will likely

increase their sales as some Amex cardholders switch their

purchases from non-participants to participants. As

participation becomes ubiquitous and nearly all merchants are

offering the discount to Amex cardholders, the cardholders will

re-sort themselves to their preferred stores independent of the

discount, with no change in the merchants sales. If all

competing merchants in a retail sector participate in Amex, the

end result is simply the perversion of the price signals related

to payment means choices, and price discrimination favoring

Amex (and other high reward) card users and disfavoring other

customers.

34

card users.45 Whether the supra-competitive portion

of the merchant fees is or is not subsequently

competed away on the cardholder side does not

change the fact that Amex is exercising market

power in its pricing to merchants. And the exercise

of that market power demonstrates the existence of a

relevant economic market for the pricing of Amex

services to merchants — a one-sided market.

VI. INCREASED AMEX OR CREDIT CARD

OUTPUT DOES NOT DEMONSTRATE THE

ABSENCE OF ANTICOMPETITIVE IMPACT

The Court of Appeals suggests that Amex

might justify its Restraints by a showing that Amex’s

“output” — that is, the volume of Amex transactions —

is increased because of the Restraints.*° However, if

the Amex Restraints have effects beyond the Amex

platform, as is the case here, then the Amex output

alone is not a proper indicator of the welfare effect of

the Restraints. A relevant analogy is an exclusive

dealing contract. An exclusive dealing vertical

restraint may increase the “output” of the firm

imposing this restraint, but at the expense of the

firm’s competitors and potential entrants — the

alternatives whose use is affected by the Restraints.

The proper measure of output would then be the size

of the market served by all competitors.

46 The Court of Appeals asserts that “Cardholder

insistence results not from market power, but instead from

competitive benefits on the cardholder side of the platform and

the concomitant competitive benefits to merchants who choose

to accept Amex cards.” Pet. App. 45a. Indeed, cardholder

insistence is not the result of market power, rather market

power is the result of insistence.

«“ Pet. App. 52a.

35

Here, the fundamental product at issue is

“payment.” If one desired to determine indirectly

through “output” the pro- or anti-competitive nature

of a credit card platform’s vertical price restraints,

the proper metric would be the output of all payment

methods that are impacted by the Amex Restraints.

The effect of the restraints is to increase the

subsidization of the users of credit card platforms by

customers paying by other means. With increased

cardholder benefits resulting from competition on the

cardholder side in the face of high credit card

merchant fees, customers will be motivated to switch

from other payment forms to credit cards. Thus, the

usage of credit cards may increase. But this is only

evidence of distortion in the competitive process, not

evidence that the restraints are procompetitive.‘’

For those customers switching to credit cards only

because of increased rewards, credit card use can be

presumed less efficient than the prior preferred

means of payment. As a consequence of the use . ~

less efficient means of payment, the cost wu

transacting will increase, and the output of all

payment means affected by the Amex Restraints —

the proper measure of output in this context — will be

expected to decline.

47 See, eg., Adan J. Levitin, Priceless? The Economic

Costs of Credit Card Merchant Restraints, 55 UCLA L. Rev.

1321, 1356 (2008) (“[Bly making credit cards appear relatively

cheaper to other payment systems, merchant restraints

encourage higher usage of credit cards than would otherwise

occur.”).

36

CONCLUSION

Based on the foregoing analyses, the amici

respectively ask the Court to reverse Second Circuit's

decision.

Respectfully submitted,

ANTHONY J. BOLOGNESE

Counsel of Record

BOLOGNESE & ASSOCIATES, LLC

1500 JFK Boulevard, Suite 320

Philadelphia, PA 19102

(215) 814-6750

ABolognese@Bolognese- Law.com

Counsel for Amici Curiae

Dated: December 14, 2017

APPENDIX

la

APPENDIX

Amicus John M. Connor is a Professor of Industrial

Economics emeritus at Purdue University. Professor

Connor specializes in research and teaching of

empirical industrial economics and antitrust policy.

His research has been cited in five court decisions

and more than 75 law review articles. Professor

Connor is also Senior Advisor to the American

Antitrust Institute. '

Amicus Martin Gaynor is a Professor of Economics

and the E.J. Barone Chair in Health Systems

Management at Carnegie Mellon University.

Professor Gaynor has consulted for the Federal

Trade Commission and the U.S. Department of

Justice, and also consulted for the Netherlands

Competition Authority on issues involving vertical

restraints and market definition.”

Amicus Daniel McFadden is a Nobel Laureate in

economics (2000) and the E. Morris Cox Professor

Emeritus of Economics and the Director of the

Econometrics Laboratory at the University of

California at Berkeley. Among his many honors,

Professor McFadden received the John Bates Clark

Medal from the American Economics Association

(1975) and the Frisch Medal from the Econometrics

Society (2000).*

! Purdue University, John M. Connor,

hitps://goo.e/ZaQdzl

. Martin Gaynor Curriculum Vitae, http://bit.ly/2fF TvQO

° University of California at Berkeley, Faculty Profiles,

McFadden, http://bit.ly/2eoJMk9

2a

Amicus Roger Noll is Professor of Economics

emeritus at Stanford University and a Senior Fellow

and member of the Advisory Board at the American

Antitrust Institute. Before coming to Stanford,

Professor Noll was a Senior Economist at the

President's Council of Economic Advisers, a Senior

Fellow at the Brookings Institution, and Institute

Professor of Social Science and Chair of the Division

of Humanities and Social Sciences at the California

Institute of Technology.‘

Amicus Jeffrey M. Perloff is a Professor in the

Department of Agricultural and Resource Economics

at the University of California at Berkeley. Professor

Perloff has authored textbooks on economics

including Modern Industrial Organization (with

Dennis Carlton), Microeconomics, Microeconomics:

Theory and Applications with Calculus, and

Estimating Market Power and Strategies (with

Larry Karp and Amos Golan).°

Amicus Joseph A. Stiglitz is a Nobel Laureate in

economics (2001) and is University Professor at

Columbia University. He is a former senior vice

president and chief economist of the World Bank and

a former member and chairman of the U.S.

President's Council of Economic Advisors. Based on

academic citations, Professor Stiglitz is the fourth-

most influential economist m the world today.®

‘ Stanford University Public Policy Program, Roger Noll

http://stanford.io/2fF UOiP

5 University of California at Berkeley Department of

Agricultural and Resource Economics, Jeffrey M. Perloff Brnef

Bio, Li

6 Columbia University, Brief Biography of Joseph E.

Stiglitz, https://goo.gl/s6tCkb

3a

Professor Stiglitz is currently consulting and is an

expert witness for a number of national supermarket

and drugstore chains that have challenged Amex’s,

Visa’s, and MasterCard’s restraints in related

lawsuits pending in the District Court.’

Amicus Lawrence J. White is Robert Kavesh

Professor of Economics and Deputy Chair of the

Economics Department at New York University's

Leonard N. Stern School of Business. He has served

on the Senior Staff of the President's Council of

Economic Advisers, and has also served as the

Director of the Economic Policy Office in the

Antitrust Division of the U.S. Department of Justice.

Professor White is also the General Editor of the

Review of Industrial Organization, a journal that

focuses on competition and monopoly in their many

forms and their effects on efficiency, innovation, and

social conditions.*

Amicus Ralph Winter holds the Canada Research

Chair in Business Economics and Policy at the

Sauder School of Business at the University of

British Columbia. * He was previously a Professor of

Economics at the University of Toronto, and has also

? Those merchants are: Ahold U.S.A., Inc.; Albertson's,

Inc; BI-LO, LLC; CVS Health, Inc; The Great Atlantic &

Pacific Tea Company, Inc.; H.E. Butt Grocery Co.; Hy-Vee, Inc.;

The Kroger Co.; Meijer, Inc.; Publix Super Markets, Inc.;

Raleys Inc.; Rite Aid HDQTRS Corp.; Safeway Inc.; Supervalu,

Inc.; and Walgreen Co.

a New York University Stern School of Business,

Lawrence J. White Biographical Summary,

9 University of British Columbia Sauder School of

Business, Ralph Winter, https://goo.gVennG6B

4a

served as President of the Canadian Economics

Association. Professor Winter has also consulted for

the U.S. Department of Justice and the Canadian

Competition Bureau. '°

10 University of British Columbia Sauder School of

Business, Ralph Winter Curriculum Vitae, http://bit.ly/2fPyrG9