Payment Card Networks Panel MaterialsPart 3 - 0

Case 1:96-cv-05238-JG-JO Document 1652 Filed 12/13/13 Page 1 of 55 PageID #: 10183
UNITED STATES DISTRICT COURT

EASTERN DISTRICT OF NEW YORK
IN RE PAYMENT CARD INTERCHANGE

FEE AND MERCHANT DISCOUNT
ANTITRUST LITIGATION
FOR PUBLICATION
MEMORANDUM
AND ORDER
05-MD-1720 (JG) (JO)
APPEARANCES
ROBBINS GELLER RUDMAN & DOWD LLP
655 West Broadway, Suite 1900
San Diego, CA 92101
KENNY NACHWALTER, P.A.

201 South Biscayne Boulevard, Suite 1100
Miami, FL 33131
ROBINS, KAPLAN, MILLER & CIRESI L.L.P

2800 LaSalle Plaza
800 LaSalle Avenue South
Minneapolis, MN 55402
SPERLING & SLATER

55 West Monroe Street, Suite 3200
Chicago, IL 60603
BERGER & MONTAGUE, P.C.

1622 Locust Street
Philadelphia, PA 19103
HANGLEY ARONCHICK SEGAL

PUDLIN & SCHILLER
30 North Third Street
Harrisburg, PA 17101
Co-Lead Counsel for Class Plaintiffs
Counsel for the Individual Plaintiffs
ARNOLD & PORTER LLP

399 Park Avenue
New York, NY 10022
WILLKIE FARR & GALLAGHER LLP

787 Seventh Avenue
New York, NY 10019
HOLWELL SHUSTER & GOLDBERG LLP

125 Broad Street, 39th Floor
New York, NY 10004
PAUL, WEISS, RIFKIND, WHARTON &

GARRISON LLP
2001 K Street, NW
Washington, DC 20006
Attorneys for Defendant Visa U.S.A. Inc.

Attorneys for Defendant MasterCard
International Incorporated
MORRISON & FOERSTER LLP

1290 Avenue of the Americas
New York, NY 10104
SHEARMAN & STERLING LLP

599 Lexington Avenue
New York, NY 10022
Attorneys for Defendant Bank of America
Attorneys for Defendant Barclays Bank
OMELVENY & MYERS LLP

Times Square Tower
7 Times Square
New York, NY 10036
SKADDEN, ARPS, SLATE, MEAGHER &

FLOM LLP
4 Times Square
New York, NY 10036
Attorneys for Defendant Capital One Bank

(USA)
Attorneys for Defendant JPMorgan Chase &

Co.
SIDLEY AUSTIN LLP

1 South Dearborn Street
Chicago, IL 60603
KEATING MUETHING & KLEKAMP PLL

1 East Fourth Street, Suite 1400
Cincinnati, OH 45202
Attorneys for Defendant Citibank, N.A.
Attorneys for Defendant Fifth Third Bancorp
KUTAK ROCK LLP

1650 Farnam Street
Omaha, NE 68102
WILMERHALE
7 World Trade Center, 250 Greenwich Street
New York, NY 10007
Attorneys for Defendant First National Bank

of Omaha
Attorneys for Defendant HSBC Finance

Corporation
JONES DAY
51 Louisiana Avenue, NW
PULLMAN & COMLEY, LLC

850 Main Street
Bridgeport, CT 06601
Attorneys for Defendant National City

Corporation
Attorneys for Defendant Texas Independent

Bancshares, Inc.
ALSTON & BIRD LLP

1201 West Peachtree Street
Atlanta, GA 30309
PATTERSON BELKNAP WEBB &

TYLER LLP
New York, NY 10036
Attorneys for Defendant SunTrust Banks,

Inc.
Attorneys for Defendant Wachovia Bank,

NA.
ii
FRIEDMAN LAW GROUP LLP

270 Lafayette Street
Suite 1410
New York, NY 10012
FREEDMAN BOYD DANIELS

HOLLANDER GOLDBERG & CLINE,
P.A.
P.O. Box 25326, 20 First Plaza, Suite 700
Albuquerque, NM 87102
Attorneys for Plaintiffs in No. 05-cv-5074

Attorneys for Plaintiffs in No. 05-cv-5075
CONSTANTINE CANNON LLP
335 Madison Avenue, 9th Floor
New York, NY 10017
QUINN EMANUEL
51 Madison Avenue, 22nd Floor
New York, NY 10010
Attorneys for Plaintiff NATSO, Inc.
Attorneys for Objector Home Depot U.S.A.,

Inc.
VORYS, SATER, SEYMOUR AND PEASE

LLP
52 East Gay Street
Columbus, OH 43215
EMERY, CELLI, BRINCKERHOFF &

ABADY LLP
75 Rockefeller Plaza, 20th Floor
New York, NY 10019
Attorneys for Objector Target Corporation
Attorneys for Objector National Retail

Federation
SCHLAM, STONE & DOLAN, LLP

26 Broadway, 19th Floor
New York, NY 10004
PERKINS COIE LLP

Four Embarcadero Center, Suite 2400
San Francisco, CA 94111
Attorneys for Amicus Curiae U.S. Public

Research Interest Group
Attorneys for Objector First Data

Corporation
BOIES, SCHILLER & FLEXNER LLP

575 Lexington Avenue, Seventh Floor
New York, NY 10022
KIRKLAND & ELLIS

153 East 53rd Street
New York, NY 10022
Attorneys for Objector American Express Co.
Attorneys for Objector Discover Financial

Services
iii
MILLER & CHEVALIER CHARTERED

655 Fifteenth Street, NW, Suite 900
THE LAW OFFICES OF JOHN J. PENTZ

19 Widow Rites Lane
Sudbury, MA 01776
Attorneys for Objector Blue Cross Blue Shield

Entities
Attorneys for Defendant Daviss Donuts and

Deli
LAW OFFICES OF EDWARD F. SIEGEL

705 South Alton Way #1C
Denver, CO 80247
THRASH LAW FIRM, P.A.

1101 Garland Street
Little Rock, AR 72201
Attorneys for Objector Vicente Consulting LLC
Attorneys for R&M Objectors
JOSHUA R. FURMAN LAW CORP.

15260 Ventura Boulevard, Suite 2250
Sherman Oaks, CA 91403
Attorneys for Objector Jon M. Zimmerman
OHIO ATTORNEY GENERAL MIKE DeWINE
150 East Gay Street, 23rd Floor
Columbus, OH 43215
Attorney for Objector State of Ohio
STATE OF CALIFORNIA DEPARTMENT

OF JUSTICE
OFFICE OF THE ATTORNEY GENERAL
455 Golden Gate Avenue, Suite 11000
San Francisco, CA 94102-7004
Attorney for Objector State of California
iv
CONTENTS
A.
Preliminary Statement......................................................................................................... 2
1.
Structure of a Credit Card Transaction; Interchange Fees .................................. 2
2.
The Default Interchange Rule; Honor-all-Cards Rules; Anti-Steering Rules .... 3
3.
The Course of the Litigation; Industry Changes Occurring During the

Litigation; and the Proposed Settlement ............................................................. 4
4.
Overview of Reasons for Approval .................................................................... 8
B.
The Claims in the Case ..................................................................................................... 13
C.
The Standard for Approving a Proposed Settlement ........................................................ 14

1.
Procedural Fairness ........................................................................................... 15
2.
Substantive Fairness.......................................................................................... 17
a.
The Complexity, Expense, and Likely Duration of the Litigation ............... 17
b.
The Reaction of the Class to the Settlement ................................................. 18
c.
The Stage of the Proceedings and the Amount of Discovery Completed..... 20
d.
The Risks of Establishing Liability and Damages, and of Maintaining the

Class Action through the Trial ...................................................................... 20
e.
The Ability of Defendants to Withstand a Greater Judgment....................... 29
f.
The Range of Reasonableness of the Settlement Fund in Light of the Possible

Recovery and Attendant Risks of Litigation ................................................. 29
D.
The Objections .................................................................................................................. 31

1.
E.
Rule Reforms .................................................................................................... 31

a.
The Elimination of the Networks No-Surcharge Rules ............................... 31
b.
The Buying Group Provision ........................................................................ 38
2.
The Releases ..................................................................................................... 39
3.
The Health Insurers Objections ....................................................................... 42
4.
Claims by States Acting in their Sovereign Capacity ....................................... 43
5.
Discovers Objection ........................................................................................ 45
6.
The Notice to Class ........................................................................................... 45
7.
Cohesiveness of the Rule 23(b)(2) Class; Adequacy of Class Plaintiffs .......... 46
The Plan of Allocation ...................................................................................................... 48
JOHN GLEESON, United States District Judge:

In this antitrust action, a putative class of approximately 12 million merchants
alleges that, among other things, defendants Visa U.S.A. Inc. (Visa) and MasterCard
International Incorporated (MasterCard), as well as issuing and acquiring banks (collectively
the defendants), conspired to fix interchange fees in violation of Section 1 of the Sherman Act.
Before me now is a motion by Class Plaintiffs,1 certain other plaintiffs who are
not members of the class (referred to throughout the case and in this opinion as the Individual
Plaintiffs2), and the defendants for final approval of a proposed settlement. In essence, the
settlement calls for (1) a cash recovery slightly in excess of $7 billion (before reductions for optouts) by members of a Rule 23(b)(3) class; and (2) certain reforms of the defendants rules and
practices to benefit the members of a Rule 23(b)(2) class. SA 33, 68.3 They also seek
1
Class Plaintiffs refers to proposed class representative merchants Photos Etc. Corp.; Traditions,
Ltd.; Capital Audio Electronics, Inc.; CHS Inc.; Crystal Rock LLC; Discount Optics, Inc.; Leons Transmission
Service, Inc.; Parkway Corp.; and Payless ShoeSource, Inc.
2
The Individual Plaintiffs are: Ahold U.S.A., Inc.; Albertsons LLC; Albertsons, Inc.; BI-LO, LLC;
Delhaize America, Inc.; Eckerd Corporation; The Great Atlantic & Pacific Tea Company; H.E. Butt Grocery
Company; Hy-Vee, Inc.; The Kroger Co.; Maxi Drug, Inc. (and doing business as Brooks Pharmacy); Meijer, Inc.;
Meijer Stores Limited Partnership; Pathmark Stores, Inc.; QVC, Inc.; Raleys; Rite Aid Corporation; Safeway Inc.;
SuperValu Inc.; and Walgreen Co.
3
Citations in the form SA __ refer to the paragraphs of the proposed settlement agreement,
which is titled Definitive Class Settlement Agreement. It is located at docket entry 1656-1 and referred to here as
the Settlement Agreement. The Rule 23(b)(3) class is defined at SA 2(a) as follows:
A Rule 23(b)(3) Settlement Class under Federal Rules of Civil Procedure

23(a) and (b)(3), from which exclusions shall be permitted, consisting of all
persons, businesses, and other entities that have accepted Visa-Branded Cards
and/or MasterCard-Branded Cards in the United States at any time from January
1, 2004 to the Settlement Preliminary Approval Date, except that this Class does
not include the named Defendants, their directors, officers, or members of their
families, financial institutions that have issued Visa- or MasterCard-Branded
Cards or acquired Visa- or MasterCard-Branded Card transactions at any time
from January 1, 2004 to the Settlement Preliminary Approval Date, or the
United States government.
The Rule 23(b)(2) class is defined at SA 2(b)as follows:
A Rule 23(b)(2) Settlement Class under Federal Rules of Civil Procedure
23(a) and (b)(2), from which exclusions shall not be permitted, consisting of all
persons, businesses, and other entities that as of the Settlement Preliminary
approval of the proposed plan of allocation of the settlement fund, and Class Counsel4 seeks
attorneys fees and costs.
I held a fairness hearing on September 12, 2013, at which there was extensive oral
argument in support of and in opposition to the proposed settlement.
For the reasons discussed below, I approve the proposed settlement and the plan
of allocation. The motion for fees and costs will be decided separately.
A. Preliminary Statement
1.
Structure of a Credit Card Transaction; Interchange Fees

A Visa or MasterCard credit card transaction involves five parties: (1) the
customer; (2) the merchant; (3) the acquiring bank; (4) the issuing bank; and (5) the network
itself, that is, Visa or MasterCard. The acquiring bank is the link between the network and the
merchant that accepts the card for payment. The issuing bank is the bank that issued the credit
card to the customer. When the cardholding customer presents a credit card to pay for goods or
services, the accepting merchant relays the transaction information to the acquiring bank. The
acquiring bank processes the information and transmits it to the network. The network relays the
information to the issuing bank, which approves the transaction if doing so is consistent with the
cardholders account status and credit limit. The approval is conveyed to the acquiring bank,
which in turn relays it to the merchant.
The issuing bank then transmits to the acquiring bank the amount of the purchase
price minus the interchange fee.5 The acquiring bank withholds an additional fee called the
Approval Date or in the future accept any Visa-Branded Cards and/or
MasterCard-Branded Cards in the United States, except that this Class shall not
include the named Defendants, their directors, officers, or members of their
families, financial institutions that have issued Visa- or MasterCard-Branded
Cards or acquired Visa- or MasterCard-Branded Card transactions at any time
since January 1, 2004, or do so in the future, or the United States government.
4
Class Counsel refers to the three firms appointed co-lead counsel for Class Plaintiffs: Robbins
Geller Rudman & Dowd LLP; Robins, Kaplan, Miller & Ciresi L.L.P; and Berger & Montague, P.C.
merchant discount fee for its processing services. Thus, the total amount the merchant
receives for the transaction is the purchase price minus the sum of the interchange fee and the
merchant discount fee.
Interchange fees vary based on factors that include the type of card used and the
type of merchant. Many Visa and MasterCard credit cards provide rewards to the cardholders.
Those rewards cost money, and thus these cards, referred to in the industry and here as premium
cards, are associated with higher interchange fees.
2.
The Default Interchange Rule; Honor-all-Cards Rules; Anti-Steering Rules

The competitive problem that gave rise to this case, according to the plaintiffs, is
the result of a combination of network rules. The Honor-all-Cards rules require merchants who
accept any Visa- or MasterCard-branded credit cards to accept all cards of that brand, no matter
what bank may have issued them and no matter the interchange fee. The Honor-all-Cards rules
created what the merchants term the hold-up problem. Unlike checks, which are redeemed by
the drawee banks at par, that is, without the drawee bank charging a fee for acceptance, the
issuing bank of a Visa or MasterCard credit card is free to demand whatever interchange fee it
chooses (hold-up) in order to accept the transaction from the merchant who is required to
accept the card.
As the merchants describe them, the default interchange rules are the networks
solution to the hold-up problem. Those rules establish mandatory interchange fees that apply
to every transaction on the network unless the merchant and the issuing bank have entered into a
bilateral interchange agreement. However, the merchants complain that the combination of the
Honor-all-Cards rules and the anti-steering rules, which are discussed further below, strips the
The interchange fee can be a flat fee, a percentage of the transaction price, or a combination of the
two.
issuing banks of any incentive to accept interchange fees lower than the default interchange fees.
And obviously merchants have no incentive to negotiate a higher interchange fee. Thus default
interchange, the merchants assert, becomes a fixed rate that applies to every credit card
transaction (with the narrow exception of transactions by very large merchants who have
sufficient volume that they can negotiate their own private interchange fees).
A linchpin to the problem, as far as the merchants are concerned, is the package
of anti-steering restraints that prohibit merchants from using price signals at the point of sale to
steer customers to less costly forms of payment. The no-surcharge rules prohibit merchants from
adding a surcharge to a transaction involving either of the networks credit cards. Thus, a
merchant who must pay a 2% interchange fee upon accepting a Visa or MasterCard credit card is
prohibited from adding a 2% surcharge (or any surcharge at all) to either discourage the use of
that card or to recoup the cost of acceptance. Similarly, until recently (as discussed further
below), no-discount rules prohibited merchants from offering price discounts at the point of sale.
There are other components to the networks anti-steering regimes, some of which are mentioned
below. In the aggregate, these essentially identical regimes prohibit merchants from informing
customers about higher-cost payment cards, incentivizing customers to use lower-cost cards or
other forms of payment, or recouping the acceptance costs of the cards the merchants are
required to honor.
3.
The Course of the Litigation; Industry Changes Occurring During the Litigation;
and the Proposed Settlement
This case has been extensively litigated for more than eight years. Discovery,
which began in 2005, included more than 400 depositions, the production and review of more
than 80 million pages of documents, the exchange of 17 expert reports, and a full 32 days of
expert deposition testimony.
4
While the case has been pending, there have been significant developments in the
industry, some of which are attributable in whole or in part to the case itself. The very structures
of Visa and MasterCard themselves changed; in 2008 and 2006, respectively, initial public
offerings (IPOs) converted each from a consortium of competitor banks into single-entity,
publicly traded companies with no bank governance.
In addition, federal legislation the so-called Durbin Amendment in 2010
removed the networks restrictions on discounting credit and debit cards at the network level.
Thus, Visa, MasterCard, American Express and Discover can no longer prohibit merchants from
discounting their cards.6 The Durbin Amendment did not change the networks rules prohibiting
surcharging.
In 2010, after an investigation assisted by the information developed by the
plaintiffs here, the Department of Justice (DOJ) filed lawsuits against Visa, MasterCard and
American Express. In consent decrees filed in 2011, Visa and MasterCard agreed to remove
their rules prohibiting merchants from product-level discounting of credit and debit cards.
Again, the resolution of that case against these defendants did not affect the surcharging
prohibitions.
By 2011, several significant motions had been briefed and were ripe for decision,
including a motion to certify classes,7 a motion to dismiss supplemental complaints arising out of
the IPOs, and cross-motions for summary judgment.
6
The Durbin Amendment was part of the Dodd-Frank Wall Street Reform and Consumer
Protection Act of 2010, Pub. L. 111-203, 124 Stat. 1376; it is codified at 15 U.S.C. 1693o-2(b)(3)(A)(i). The
Amendment did not allow for product-level discounting, by which a merchant could choose to discount a particular
card type.
7
The motion sought to certify an injunctive relief class (the (b)(2) class), and a damages class
(the (b)(3) class). See Fed. R. Civ. P. 23(b)(2) & (3).
Even as they litigated the case full-throttle, beginning in 2008 the parties engaged
in efforts to settle their disputes. They participated in mediation sessions with the Honorable
Edward A. Infante (Ret.) and, beginning in 2009, with Professor Eric D. Green as well. The
parties had dozens of meetings and conference calls with the mediators over the next several
years.
On November 2, 2011, I held oral argument on, inter alia, the parties crossmotions for summary judgment. See DE 1560.8 Even before that argument, however, the parties
requested through the mediators that any decision on those motions and on the motion for class
certification be withheld pending settlement discussions with the Court itself.9
On Friday and Saturday, December 2 and 3, 2011, at the express request of all
parties, Judge Orenstein and I engaged in lengthy settlement discussions with the parties and the
mediators. The ground rules were made clear and were agreed to on the record at the outset. In
an effort to foster as much candor and engagement as possible, all parties consented to ex parte,
off-the-record discussions about all of the various issues that required resolution before the case
could be settled. It was a process that had proven useful in a successful effort to settle a previous
antitrust class action against Visa and MasterCard. See In re Visa Check/Mastermoney Antitrust
Litig., 297 F. Supp. 2d 503 (E.D.N.Y. 2003), affd sub nom. Wal-Mart Stores, Inc. v. Visa U.S.A.,
Inc., 396 F.3d 96 (2d Cir. 2005). Indeed, several of the participants in the settlement discussions
in this case had been involved in those other settlement talks ten years earlier.
The settlement discussions on December 2 and 3, 2011 were productive. They
illuminated the parties positions and concerns regarding the merits of the plaintiffs claims, the
Citations in the form DE __ refer to docket entries in this case.

In the context of settlement discussions, my use of the term Court in this opinion refers to both
me and Magistrate Judge James Orenstein, who has worked tirelessly and effectively throughout this litigation and
was my partner in attempting to help the parties reach a consensual resolution of the case.
9
defendants defenses, the purposes and effects of the challenged rules, and the various ways in
which the rules might be changed by agreement to inject competition into the setting of
interchange fees for Visa and MasterCard credit cards.
After a third meeting with Judge Orenstein and me on December 15, 2011, the
parties resumed discussions with the mediators. On February 21, 2012, those discussions
resulted in an acceptance by all of the defendants and all of the named plaintiffs of the terms of a
mediators proposal. When the efforts to reduce the agreed-upon terms of that proposal to a
settlement agreement stalled, the parties once again sought and obtained the assistance of the
Court. On June 20 and 21, 2012, additional settlement discussions occurred in chambers
pursuant to the same terms and consent that had governed the discussions six months earlier. At
the conclusion of those two days of discussions, the parties informed the Court that they had
reached agreement on all of the issues necessary to settle the case.
On July 13, 2012, the parties filed a Memorandum of Understanding attaching a
document setting forth the terms of the settlement. See DE 1588. In October 2012, after
completing the drafting of related documents, including escrow agreements and a Plan of
Administration and Distribution, the parties executed the Settlement Agreement. See DE 1656.
On October 19, 2012, Class Counsel moved for preliminary approval of the proposed settlement,
which I granted on November 27, 2012. The order granting preliminary approval provisionally
certified classes under Rule 23(b)(2) and (b)(3). The Class Administrator notified class members
of the terms of the proposed settlement through a mailed notice and publication campaign that
included more than 20 million mailings and publication in more than 400 publications. The
notice plan was carried out between January 29, 2013 and February 22, 2013. Class members
had until May 28, 2013 to opt out of the damages class, object to the proposed settlement, or
both.
The proposed Settlement Agreement provides for, among other things:
The creation of two cash funds totaling up to an estimated $7.25 billion (before
reductions for opt-outs). SA 9-10, 11-13.
Visa and MasterCard rule modifications to permit merchants to surcharge on Visa- or
MasterCard-branded credit card transactions at both the brand and product levels.
SA 42, 55.
An obligation on the part of Visa and MasterCard to negotiate interchange fees in good
faith with merchant buying groups. SA 43, 56.
Authorization for merchants that operate multiple businesses under different trade
names or banners to accept Visa and/or MasterCard at fewer than all of its businesses.
SA 41, 54.
The locking-in of the reforms in the Durbin Amendment and the DOJ consent decree
with Visa and MasterCard, even if those reforms are repealed or otherwise undone. SA
40, 44, 53, 57, Appendix J.
4.
Overview of Reasons for Approval

As class action settlement proceedings go, this one has been anything but typical.
Some of the merchant plaintiffs who directly participated in the lengthy settlement discussions
and initially agreed to the terms of the settlement broke away and objected, as they were entitled
to do. Numerous other members of the merchant class have objected as well, on a variety of
grounds.
The behavior of a small number of objectors has threatened to undermine the
efforts of the others. Specifically, in their zeal to drum up objections and opt-outs by merchants
around the country, certain merchant groups established websites that spread false and
misleading information about the settlement and the merchants options. In April of this year I
had to order injunctive relief to remedy that situation; in May I came close to holding certain
entities in contempt of that injunction.
The oral presentations of the objectors at the fairness hearing were afflicted by
needless hyperbole. One of the merchant association principals who participated in the
settlement discussions and initially agreed to its terms argued that the members of his association
would be worse off if I approved the proposed settlement than they would be if they proceeded
all the way through trial and lost. Another likened the prospect of approval to the deprivation of
civil liberties in the aftermath of a terrorist attack, warning that only a slippery slope would
separate an order binding a large retailer to the proposed settlement and the government stripping
us of our houses and civil rights. A third cast Visa and MasterCard as modern-day Nazis, and
warned me not to assume the role of Neville Chamberlain.
If only the issues here were that simple. But in reality the vitriol and poor
behavior and feigned hysteria mask complex and difficult issues on which reasonable merchants
can and do disagree. Some of those issues stem from the fact that a lawsuit is an imperfect
vehicle for addressing the wrongs the plaintiffs allege in their complaint. For example, there are
forms of relief many objectors seek, such as the regulation of interchange fees, that this Court
could not order even if the plaintiffs obtained a complete victory on the merits. In addition, there
are features of the industry landscape, such as other credit card issuers with whom the defendants
compete, and laws in some states that prohibit merchants from surcharging the use of credit
cards, that are beyond the reach of this case but will undermine (at least in the near term) the
efficacy of the agreed-upon relief.
The proponents of the settlement disagree strongly with the objectors over the
economic value of the proposed settlement to the class members, and specifically over the
benefits of the proposed rules changes to the merchant class. As a result, I availed myself of the
provisions in Rule 706 of the Federal Rules of Evidence allowing for court-appointed expert
witnesses. On April 8, 2013, I appointed Dr. Alan O. Sykes of New York University School of
Law to advise the Court with respect to any economic issues that may arise in connection with
the forthcoming motion for final approval of the proposed settlement.10 Dr. Sykes filed a
memorandum with the Court on August 28, 2013.11 It has proved quite helpful, and he has the
gratitude of the Court.
Fortunately, well-established law provides a rubric for deciding the motion, and I
engage below in the prescribed multifactor inquiry. By way of overview, however, I conclude
that the proposed settlement secures both a significant damage award and meaningful injunctive
relief for a class of merchants that would face a substantial likelihood of securing no relief at all
if this case were to proceed.
Specifically, although the settlement either obtains or locks in place an array of
rules changes, at its heart is an important step forward: a rules change that will permit merchants
to surcharge credit cards at both the brand level (i.e., Visa or MasterCard) and at the product
level (i.e., different kinds of cards, such as consumer cards, commercial cards, premium cards,
etc.), subject to acceptance cost and limits imposed by other networks cards. For the first time,
merchants will be empowered to expose hidden bank fees to their customers, educate them about
those fees, and use that information to influence their customers choices of payment methods.
In short, the settlement gives merchants an opportunity at the point of sale to stimulate the sort of
network price competition that can exert the downward pressure on interchange fees they seek.
10
DE 2807. After consulting with the parties, see DE 5863, I issued a subsequent order establishing
procedures for the parties to provide information to Dr. Sykes and to respond to his advice to the Court. See DE
5873.
11
DE 5965.
10
Relief directed at the default interchange fees (even if such relief were available
from a court as opposed to a legislature) would not share that characteristic. Both Visa and
MasterCard have other elements to their fee structures. Relief aimed solely at the interchange
fees would leave them free to recoup any associated lost revenues by tinkering with those other
fees, and the merchant restraints would, in effect, place the merchants back where they started.
By focusing on those restraints, the proposed settlement seeks to empower merchants to steer
cardholders to lower-cost payment alternatives. The goal is to incentivize the networks to
compete for the merchants credit card volume through lower fees of all kinds, including
interchange fees, and to allow merchants to recoup their costs when their efforts to steer
customers to lower-cost means of payment do not succeed.
The objections to the proposed settlement are numerous, but in the main, they fail
for one of two reasons. First, the objectors complain about the failure of the proposed settlement
to eliminate other Visa and MasterCard rules, such as the default interchange and Honor-allCards rules. But those rules undeniably have significant procompetitive effects, and they lay at
the heart of Visas and MasterCards efforts to build the successful networks they now have.
Class Counsel have good reason to believe that even if the default interchange and the Honor-allCards rules are characterized as horizontal restraints (despite the IPOs), they will still receive
only Rule-of-Reason antitrust scrutiny, which they could quite easily withstand. Perhaps most
telling of all is that the Department of Justice, which recently conducted a thorough investigation
of these networks operating rules, declined even to challenge either rule. In short, it is hard to
persuasively challenge a compromise on the ground that it fails to eliminate rules that even a
complete success on the merits might not eliminate.
11
Second, the objectors complain about factors that they say will limit the
usefulness of their newfound ability to surcharge credit cards that carry costly interchange fees.
Those factors, which include the merchant restraints imposed by American Express and the laws
prohibiting surcharging in approximately ten states, are real, and they in fact undermine to an
extent the immediate utility of the rules reforms in the proposed settlement. A merchants ability
to surcharge a particularly expensive Visa card is less valuable if the merchant remains unable
because of American Expresss merchant restraints to steer customers away from even higherpriced American Express cards. Similarly, the option to surcharge that the proposed settlement
affords merchants obviously cannot be exercised in states that now have anti-consumer nosurcharge laws.
But the virulent objections based on those and other practical limitations on the
relief the proposed settlement affords fail to take sufficient account of the fact that, in the end
and despite its outsized proportions, this is just an antitrust lawsuit. Even if the plaintiffs spent
several years pursuing this unwieldy case to a successful conclusion (despite substantial odds
against such a result), this Court would be in no position to grant the sweeping relief the
objectors seek. It cannot regulate interchange fees or enjoin nonparties or preempt state laws or
reform network rules that do not violate the antitrust laws. The Sherman Act affords relief only
from certain proven anticompetitive business practices. And the agreed-upon relief here, which
has the potential to unleash a new competitive force on interchange fees, falls squarely in the
wheelhouse of what this lawsuit is all about.
The first Visa/MasterCard multi-district litigation broke the tie between credit and
debit cards. The Durbin Amendment removed discounting restrictions at the network level. The
consent decree in the governments case removed product-level discount restrictions. The IPOs
12
wrested control of Visa and MasterCard from the banks. This proposed settlement adds another
crucial reform the lifting of restrictions on network- and product-level surcharging. Even if the
objectors are right in contending that additional dominoes must fall before the alleged
anticompetitive behavior of Visa and MasterCard is eradicated, those dominoes will have to fall
in other forums. That does not alter the significance of the relief provided by the proposed
settlement.
B. The Claims in the Case
Beginning in June 2005, more than 40 class action complaints were filed by
merchants against the defendants. The class actions were consolidated in 2005, together with 19
individual actions. The Court appointed Class Counsel as co-lead counsel and a consolidated
amended complaint was filed on April 24, 2006. DE 317.12
As mentioned above, the plaintiffs allege that Visa and MasterCard adopted and
enforced rules and practices relating to payment cards that had the combined effect of
unreasonably restraining trade and injuring merchants. Those rules and practices include:
Rules regarding the setting of default interchange fees. See Second Cons. Am. Class
Action Compl.
A number of anti-steering rules including no surcharge rules, no discounting or
non-discrimination rules, and no minimum purchase rules that restrict merchants
from steering customers to lower-cost credit cards and/or forms of payment other than
Visa or MasterCard payment cards.
A number of exclusionary rules including all outlets rules, no bypass rules, and
no multi-issuer rules that restrict merchants in accepting and processing payments
made with Visa and MasterCard cards.
12
Also in 2005, the Individual Plaintiffs, a group of large retailer merchants, filed lawsuits against
Visa and MasterCard, making allegations narrower than those set forth in the class action complaints here. The
Individual Plaintiffs alleged that merchant restraints in the networks operating rules protected Visa and MasterCard
from interbrand competition on the merchant side (as opposed to the bank side) of the market. Both sets of lawsuits
the class actions and the Individual Plaintiffs complaints were consolidated before me in this multidistrict
litigation.
13
Honor-all-Cards rules, which required merchants to accept all the networks credit
cards or all the networks debit cards when proffered for payment, regardless of which
bank issued the card.
Class Plaintiffs allege that these rules insulate the Visa and MasterCard networks
from competition with each other, from other brands and from other forms of payment, allowing
Visa and MasterCard and the issuing banks to set supracompetitive default interchange fees.
Class Plaintiffs allege that these rules were adopted pursuant to unlawful
agreements among the banks and Visa, and among the banks and MasterCard. Specifically, they
allege that the defendant banks were members of Visa or MasterCard and were represented on
their boards, and thus determined the networks rules and practices. Class Plaintiffs allege that
the banks owned and effectively operated Visa and MasterCard, such that Visa and MasterCard
were unlawful structural conspiracies or walking conspiracies with respect to their network
rules and practices. Class Plaintiffs further allege that after the Visa and MasterCard IPOs, the
unlawful agreements among the banks and Visa, and among the banks and MasterCard,
continued.
Based on these allegations, Class Plaintiffs seek damages to compensate
merchants for supracompetitive default interchange fees in the past. They also seek injunctive
relief to restructure the networks rules and practices in the future.
C. The Standard for Approving a Proposed Settlement
Pursuant to Federal Rule of Civil Procedure 23(e), any settlement of a class action
requires court approval. A court may approve such a settlement if it is fair, adequate, and
reasonable, and not a product of collusion. Joel A. v. Giuliani, 218 F.3d 132, 138 (2d Cir.
2000). In so doing, the court must eschew any rubber stamp approval yet simultaneously stop
short of the detailed and thorough investigation that it would undertake if it were actually trying
14
the case. Detroit v. Grinnell Corp., 495 F.2d 448, 462 (2d Cir. 1974), abrogated on other
grounds by Goldberger v. Integrated Resources, Inc., 209 F.3d 43 (2d Cir. 2000). Judicial
discretion is informed by the general policy favoring settlement. See Weinberger v. Kendrick,
698 F.2d 61, 73 (2d Cir. 1982); see also Denney v. Jenkins & Gilchrist, 230 F.R.D. 317, 328
(S.D.N.Y. 2005) (There is a strong judicial policy in favor of settlements, particularly in the
class action context. The compromise of complex litigation is encouraged by the courts and
favored by public policy.) (footnotes, citations and internal quotation marks omitted), affd in
part and vacated in part, 443 F.3d 253 (2d Cir. 2006).
To evaluate whether a class settlement is fair, a district court examines (1) the
negotiations that led up to the settlement, and (2) the substantive terms of the settlement. See In
re Holocaust Victims Assets Litigation, 105 F. Supp. 2d 139, 145 (E.D.N.Y. 2000). In evaluating
procedural fairness, [t]he [negotiation] process must be examined in light of the experience of
counsel, the vigor with which the case was prosecuted, and the coercion or collusion that may
have marred the negotiations themselves. Id. at 145-46 (quoting Malchman v. Davis, 706 F.2d
426, 433 (2d Cir. 1983)). Factors relevant to the substantive fairness of a proposed settlement
include: (1) the complexity, expense, and likely duration of the litigation; (2) the reaction of the
class to the settlement; (3) the stage of the proceedings and the amount of discovery completed;
(4) the risks of establishing liability; (5) the risks of establishing damages; (6) the risks of
maintaining the class action through trial; (7) the ability of the defendant to withstand a greater
judgment; (8) the range of reasonableness of the settlement fund in light of the best possible
recovery; and (9) the range of reasonableness of the settlement fund to a possible recovery in
light of all the attendant risks of litigation. See Grinnell, 495 F.2d at 463.
1.
Procedural Fairness
15
The proposed settlement is the product of arms-length negotiations between

experienced and able counsel on all sides. The lawyers for the parties spent a great deal of time
in face-to-face, telephonic and written negotiations. Those negotiations were assisted by two
eminent mediators, Judge Infante and Professor Green, who have informed the Court that the
settlement negotiations were fair and conducted at arms length. My own participation in the
efforts to settle the case (along with Magistrate Judge Orenstein) confirmed those descriptions of
the negotiations, and nothing in the record suggests otherwise.
The objecting plaintiffs argue from the fact that a number of class members and
class representatives now oppose the settlement that the negotiations were not fair. I do not find
this argument persuasive. A number of objectors were deeply involved in the settlement
negotiations and mediation, and indeed accepted the mediators proposal that outlined the key
components of what became the Settlement Agreement. Wildfang Supp. Decl., 29, DE 5939-1.
Their current dissatisfaction with the terms of that agreement raises genuine issues of substantive
fairness, but it does nothing to undermine a record that demonstrates beyond any reasonable
doubt that the negotiations were adversarial and conducted at arms length by extremely capable
counsel.
Furthermore, there is no indication that the Settlement Agreement is the product
of collusion or that it confers upon the class representatives or any other subset of the class
improper[] . . . preferential treatment. In re NASDAQ Market-Makers Antitrust Litigation, 176
F.R.D. 99, 102 (S.D.N.Y. 1997). The objecting plaintiffs argue that because Visa and
MasterCard were able to negotiate jointly with Class Plaintiffs this may indicate collusion
specifically in regard to the revisions to the no-surcharging rules. Again, I am not persuaded.
Mediators and the Court were involved in the settlement negotiations. The Second Circuit has
16
recognized that the involvement of a mediator in pre-certification settlement negotiations helps

to ensure that the proceedings are free of collusion and undue pressure. DAmato v. Deutsche
Bank, 236 F.3d 78, 85 (2d Cir. 2001). In addition, as discussed more fully below, I conclude that
the injunctive relief at the heart of the settlement inures equally to the benefit of every single
member of the merchant class. Accordingly, I conclude that the negotiation process fairly
protected the interests of the settlement class.
2.
Substantive Fairness
a. The Complexity, Expense, and Likely Duration of the Litigation
The potential for this litigation to consume considerable additional time and
resources is great. The complexity of federal antitrust law is well known. See, e.g., Virgin Atl.
Airways Ltd. v. British Airways PLC, 257 F.3d 256, 263 (2d Cir. 2001) (noting the factual
complexities of antitrust cases); Weseley v. Spear, Leeds & Kellogg, 711 F.Supp. 713, 719
(E.D.N.Y. 1989) (antitrust class actions are notoriously complex, protracted, and bitterly
fought). Numerous motions remain pending before the Court, including motions to dismiss,
summary judgment motions, Daubert motions, and a motion to certify both a damages class
under Rule 23(b)(3) and an injunctive-relief class under Rule 23(b)(2). As to the class
certification motion, the losing party would likely seek interlocutory review by the Second
Circuit under Federal Rule of Civil Procedure 23(f) (review this Court would welcome and
encourage), delaying the case substantially.13 Class Counsel represent that a trial would take
several months, and I have no doubt they are correct. The losing parties would likely appeal any
adverse jury verdicts, thereby extending the duration of litigation. By contrast, the proposed
13
In the Wal-Mart case, twenty months elapsed between the order certifying the class and the
Second Circuits divided opinion affirming that decision. See In re Visa Check/Mastermoney Antitrust Litig., 192
F.R.D. 68 (E.D.N.Y. 2000), affd, In re Visa Check/MasterMoney Antitrust Litig., 280 F.3d 124, 129 (2d Cir. 2001).
17
settlement allows class members to take advantage of rules changes now those changes are
already in place and further provides for significant monetary compensation in the near future.
b. The Reaction of the Class to the Settlement
Class Counsel report that almost 21 million long-form notices were mailed.
Taking into account duplicate mailings, and in light of various other factors that make the precise
size of the class impossible to determine, Class Counsel estimate that approximately 12 million
merchants comprise the class. Only .05% of them have objected to the settlement, and 90% of
the objections are on boilerplate forms downloaded from websites that disseminated false and
misleading information for the precise purpose of drumming up objections and opt-outs. It is
thus difficult to ascribe significant weight to the bulk of the objections.
On the other hand, because the roster of objectors includes some of the nations
largest retailers, the objectors in the aggregate represent 19% of the total transaction volume.
Many of the named plaintiffs in the case are objectors. Indeed, the motion for final approval has
caused a rift among large United States retailers, all of whom agree that the current interchange
fees are too high due to anticompetitive practices.
The divisions among the major merchants run deep, but they are also nuanced.
For example, of the top 60 merchants (measured by transaction volume) to opt out of the (b)(3)
damages class, about half (27) have not objected to the settlement. Thus, those merchants, which
include almost all the major airlines, will seek to obtain a greater damage award from Visa and
MasterCard, but they apparently see value in the (b)(2) relief. And the conduct of the airlines is
significant; in other markets that allow the surcharging permitted by the proposed settlement,
airlines were among the first to adopt the practice, which has had the effect of moving
18
transactions to cheaper payment alternatives. Thus, not all of the opt-outs evidence
dissatisfaction with the rules changes in the proposed (b)(2) settlement.
Similarly, although some merchants have argued that the ability to surcharge is
useless to smaller retailers, such as grocery stores and convenience stores, Class Counsel report
that 15 of the top 25 convenience stores have not objected to the settlement, and 5 of those
neither objected nor opted out.
In short, the reaction of the class to the proposed settlement is a mixed bag. To
paraphrase the argument by counsel for the Individual Plaintiffs at the fairness hearing,
intelligent and thoughtful merchants with a common complaint, a common goal, and many other
things in common part company on the degree to which the proposed settlement obtains for the
merchant class what can reasonably be expected to be obtained in this case.
Given the transaction volume represented by the objectors, it would be facile to
conclude that the reaction of the class strongly favors approval of the settlement simply because
substantially less than one-tenth of one percent of the merchants have objected. But see Alba
Conte & Herbert Newberg, Newberg on Class Actions 11.41, at 108 (4th ed. 2002) ([A]
certain number of objections are to be expected in a class action with an extensive notice
campaign and a potentially large number of class members. If only a small number of objections
are received, that fact can be viewed as indicative of the adequacy of the settlement.); see also
DAmato v. Deutsche Bank, 236 F.3d 78, 86-87 (2d Cir. 2001) (holding that [t]he District Court
properly concluded that this small number of objections [18 out of 27,883 notices] weighed in
favor of settlement).
I also reject, however, the objectors argument that the transaction volume they
represent means that their objections must weigh heavily against approval. Rather, the reasons
19
advanced by the objectors require careful consideration in the ultimate determination of whether
the benefits offered by the settlement create a more attractive alternative to the merchant class
than incurring the risks of continued litigation. For reasons discussed throughout this
memorandum, I conclude that the objectors have failed adequately to acknowledge not only the
substantial impediments to succeeding on the merits in this case, but also the limitations on the
relief that would be available even if success were achieved. They have also underestimated the
significance of the Rule 23(b)(2) relief afforded by the settlement.
Accordingly, I conclude on balance that the reaction of the class favors approval
of the proposed settlement.
c. The Stage of the Proceedings and the Amount of Discovery Completed
This factor relates to whether Class Plaintiffs had sufficient information on the
merits of the case to enter into a settlement agreement, Cinelli v. MCS Claim Services, Inc., 236
F.R.D. 118, 121 (E.D.N.Y. 2006), and whether the Court has sufficient information to evaluate
such a settlement, Wal-Mart, 396 F.3d 96, 118.
Before the parties executed the Settlement Agreement, fact discovery had been
completed, the parties had exchanged expert reports and deposed the experts, and all dispositive
motions had been briefed and argued. Class Counsel thus had a more than adequate basis for
assessing the claims. See Wal-Mart, 396 F.3d at 118 (where several years of discovery,
summary judgment and mediation occurred prior to settlement, plaintiffs had a thorough
understanding of their case). This factor weighs in favor of approval.
d. The Risks of Establishing Liability and Damages, and of Maintaining the
Class Action through the Trial
20
In assessing the Settlement, the Court should balance the benefits afforded the
Class, including the immediacy and certainty of a recovery, against the continuing risks of
litigation. In re Top Tankers, Inc. Sec. Litig., 06 CIV. 13761 (CM), 2008 WL 2944620, at *4
(S.D.N.Y. July 31, 2008) (emphasis in original). In this case, the risks of establishing liability
and damages at trial, and of maintaining the class throughout the trial (the fourth, fifth, and sixth
factors bearing on substantive fairness, respectively) all militate in favor of approval. Indeed,
perhaps the most significant defect in the objectors collective presentation to the Court is the
abject failure to acknowledge the perils of not settling the case. Instead, the objectors appear to
proceed on the assumption that a complete victory on the merits is a foregone conclusion.
A wide range of outstanding issues affects Class Plaintiffs ultimate likelihood of
establishing liability, including the legal characterization of the challenged practices of Visa and
MasterCard, and whether those practices on balance would be deemed anticompetitive under the
Rule of Reason.14 Class Plaintiffs allege that a number of core network practices, including the
setting of default interchange fees, the Honor-all-Cards rule, and various network rules that
discourage merchants from encouraging or requiring consumers to use less expensive payment
mechanisms, have resulted in excessively high interchange fees. But proving that those rules
14
The Rule of Reason is the accepted standard for testing whether a practice restrains trade in
violation of Section 1 of the Sherman Act. Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877, 885
(2007). As described by the Supreme Court,
Under this rule, the factfinder weighs all of the circumstances of a case in
deciding whether a restrictive practice should be prohibited as imposing an
unreasonable restraint on competition. Appropriate factors to take into account
include specific information about the relevant business and the restraint's
history, nature, and effect. Whether the businesses involved have market power
is a further, significant consideration. In its design and function the rule
distinguishes between restraints with anticompetitive effect that are harmful to
the consumer and restraints stimulating competition that are in the consumer's
best interest.
Id. at 885-86 (internal quotation marks and citations omitted).
21
violate the antitrust laws is no mean feat. And even if that feat were accomplished, proving
damages is just as difficult.
i.
Illinois Brick
First, there is a threshold problem of standing. Defendants rely on Illinois Brick

Co. v. Illinois, 431 U.S. 720, 736 (1977), in which the Supreme Court held that indirect
purchasers may not recover antitrust damages. The Court reasoned that permitting suits by
these third parties would essentially transform treble-damages antitrust actions into efforts to
apportion the recovery among all potential plaintiffs that might have absorbed part of the illegal
overcharge, from direct purchasers to middlemen to ultimate consumers. However appealing an
effort to allocate the overcharge might seem in theory, it would add a dimension of complexity to
actions for antitrust damages that would seriously undermine their effectiveness. Id. at 737.
Illinois Brick was an action brought by the State of Illinois and local government
entities alleging that concrete block manufacturers had engaged in a price-fixing conspiracy. Id.
at 726-27. The government entities had contracted with general contractors, who in turn hired
masonry subcontractors who bought the concrete blocks for installation in government projects.
Id. at 735. The Supreme Court held that only a direct purchaser of the concrete blocks (i.e., a
masonry subcontractor) was a party injured in his business or property by the alleged antitrust
violation and thus entitled to sue under Section 1 of the Sherman Act. Id. at 729, 735-37.
The defendants here contend that the merchants lack standing because they are
indirect purchasers the acquiring banks being the direct purchasers with respect to the
interchange fees they allege were fixed. The objectors argue that the fees are effectively paid by
the merchants directly. The concern about costly and inefficient apportionment proceedings that
fueled Illinois Brick dissipates where contractual arrangements permit ready determinations of
22
how an anticompetitive overcharge is allocated along the distribution chain. Illinois Brick itself
acknowledged a narrow exception to its rule for cases in which the overcharge is passed along
via a pre-existing cost-plus contract. 431 U.S. at 736. The plaintiffs here might successfully
appeal to such policy considerations; the issuing banks interchange fees are identifiable and
separate from the total package of fees (including other network fees and the acquiring banks
merchant discount fees) that are passed along to the merchants.
On the other hand, the indirect purchaser doctrine is strictly applied, and the
exceptions are narrow.15 In the current posture of the case I need not resolve the issue, but I
think it clear that the indirect purchaser doctrine would be a source of significant uncertainty for
the plaintiffs if they sought to litigate their claims to a decision on the merits.
ii.
The Effect of the IPOs
As stated above, part of the core conduct the plaintiffs sought to address was
that Visa and MasterCard member banks [. . .] effectively control the decisions of both
Networks by setting rules and interchange fees for the networks to serve their collective
interest. First Consol. Am. Cl. Action Compl., 131-34, DE 317. However, after the filing of
that complaint, both Visa and MasterCard came out from under the control of their member
banks. The IPOs that accomplished that result strengthened the defendants argument that they
were no longer structural or walking conspiracies, and thus that the setting of interchange fees
cannot constitute horizontal price-fixing.
iii.
Default Interchange Rules
15
A recent Ninth Circuit opinion, In re ATM Fee Antitrust Litigation, 686 F.3d 741 (9th Cir. 2012),
presents a risk to the plaintiffs claims. In ATM Fee, the court granted summary judgment for the defendant on
Illinois Brick grounds and rejected three exceptions to the Illinois Brick rule that plaintiffs here have relied on in
opposition to the defendants motions for summary judgment: the co-conspirator exception; the ownership-andcontrol exception; and the exception for cases in which there is no realistic possibility that direct purchasers will
sue. Id. at 750-58. Though plaintiffs made arguments at summary judgment to distinguish the ATM Fee case, they
face a risk that this Court or a higher one would not be persuaded.
23
The vast majority of the objectors oppose final approval of the Settlement
Agreement because it does not require Visa and MasterCard to eliminate their default
interchange rules. But even assuming default interchange fees operate to increase the acceptance
costs of Visa and MasterCard credit cards, that does not mean they violate the antitrust laws.
The objectors assume that default interchange is inherently illegal, but in reality it is a very
complicated issue.
Defendants contend that without default interchange, the costs of Visa and
MasterCard credit card transactions would have been higher because of the costs associated with
the negotiation of individual interchange agreements. They further argue that their networks
default interchange rules are procompetitive because they enable issuers to improve card features
and rewards and reduce card finance charges and other costs. And default interchange benefits
merchants, the networks argue, by providing consumers greater purchasing incentives, thus
increasing consumer demand, which in turn increases merchant sales. Default interchange also
allows the banks that issue credit cards, rather than the merchants that accept them, to assume the
costs of fraud and non-payment.
The plaintiffs experts dont dispute these assertions as much as they argue that
they have become obsolete because the Visa and MasterCard networks have matured over
time.16 However, given that these practices are at the core of the defendants successful business
model, it would be difficult for plaintiffs to show that these practices have become antitrust
violations by virtue of industry maturation.
16
See, e.g., Report of Dr. Alan S. Frankel at 216; Report of Dr. Christopher A. Vellturo at 39;
Report of Dr. Joseph Stiglitz at 9-10; see also Obj. Pls. Br. at 56 (Visa and MasterCard have evolved into
mature and dominant payment system, and the suggestion that issuing banks need interchange to issue credit cards
on which most Americans have become dependent has been untenable for years, if not decades.)
24
No American court has ever held that Visa or MasterCards default interchange
rules violate the antitrust laws. In National Bancard Corporation, a district court concluded,
after a full trial on the merits, that the default interchange rule was of vital import to the
payment-card system because it relieved issuing and acquiring banks of the need to negotiate
potentially thousands of bilateral agreements, and it also assured universal acceptance. Natl
Bancard Corp. v. Visa U.S.A., Inc., 596 F. Supp. 1231, 1259-61 (S.D. Fla. 1984). The Eleventh
Circuit then upheld Visas default interchange rule. Natl Bancard Corp. v. Visa U.S.A., Inc.,
779 F.2d 592, 605 (11th Cir. 1986) (NaBanco). While it is true that the factual underpinning
of NaBanco that issuing banks need interchange fees to have adequate incentives to participate
in networks has eroded, a 2008 decision of the Ninth Circuit affirmed the dismissal of
challenges aimed at Visa and MasterCards default interchange rules. In Kendall v. Visa U.S.A.,
Inc., the court dismissed the allegation that the Banks conspired to fix the interchange fee on
the ground that merely charging, adopting or following the fees set by a Consortium is
insufficient as a matter of law to constitute a violation of Section 1 of the Sherman Act. 518
F.3d 1042, 1048 (9th Cir. 2008).
In short, the default interchange rules played an essential role in the construction
of the networks at issue here, and those networks provide substantial benefit to both merchants
and consumers. While the plaintiffs contend that the rules have outlived their procompetitive
effects now that the networks have matured, the setting of default interchange fees would almost
certainly be evaluated under the Rule of Reason, and the prospect that its anticompetitive effects
remain outweighed by its procompetitive ones is real. DOJs recent decision not to challenge the
default interchange rules despite the entreaties by Class Counsel that it do so further suggests
that the plaintiffs antitrust challenge to the rules could easily fail.
25
Professor Sykes has advised that the expected returns to continued litigation are
highly uncertain, and that plaintiffs[] face a substantial probability of securing little or no relief at
the conclusion of trial. Sykes Report at 3. Specifically, he states that the plaintiffs face
considerable difficulty in establishing . . . that the core practices at issue in the case and left in
place by the proposed settlement such as default interchange and [H]onor[-]all[-C]ards rules
cause anticompetitive harm that outweighs their pro-competitive benefits . . . . Id.
iv.
The Honor-all-Cards Rules
A number of objectors argue that the settlement should not be approved because it
does not eliminate the networks Honor-all-Cards rules. As discussed earlier, the Honor-allCards rules require merchants that accept Visa and/or MasterCard credit cards to accept all credit
cards issued on the same network, regardless of the issuer or the interchange fees associated with
the issuers card.
The Honor-all-Cards rules are closely interrelated in both their history and
rationale with the default interchange rules. The assurances that a networks cards will be
accepted wherever the networks logo is displayed is critical to customers desire to carry such
cards and to merchants willingness to accept them.
A number of courts and economists have found the Honor-all-Cards rule and
similar rules to be procompetitive under the Rule of Reason. The Second Circuit upheld a
blanket license a system analogous to the Honor-all-Cards rule in Buffalo Broadcasting
Co. v. ASCAP, 744 F.2d 917 (2d Cir. 1984). See also Benjamin Klein et al., Competition in TwoSided Markets: The Antitrust Economics of Payment Card Interchange Fees, 73 ANTITRUST L.J.
571, 592-93 (2006) (An honor-all-cards rule is the essence of a payment card system because it
26
assures each cardholder that his card will be accepted at all merchants that display the mark of
the card payment system.). As one antitrust practitioner put it:
[The Honor-all-Cards rule] is a classic example of a restraint that
was actually necessary for the functioning of the joint venture.
When Visa and MasterCard were formed think about this: You
have thousands of banks across the country issuing these cards,
thousands of banks acquiring merchants, millions of merchants
accepting these cards you need to have a seamless acceptance
experience. We all take it for granted, but you needed to have a
rule that ensured to you, as a consumer, that when you proffer the
Visa card, the merchant is going to take it. Its not going to say,
Ill take a Chase Visa card, but I dont like Citibank, so Im going
to turn that one down.
Panel Discussion II: Consumer Issues at 5-6 (Statement of Jeffrey Shinder) (Fordham Univ. Sch.
of Law 2008), Marth Decl., Ex. C.
In light of the procompetitive features of the Honor-all-Cards rules, it is no sure
thing, to put it mildly, that Class Plaintiffs will be able to prove they have anticompetitive effects
to such an extent that they violate the antitrust laws. The proposed settlement preserves the
integrity of the rules that made (and continue to make) the networks successful. At the same
time, by further relaxing merchant restraints regarding pricing, it provides for transparency and
competition at the point of sale. Merchants who choose to use the power the proposed rules
changes give them will be able to exercise control over (and perhaps reduce) their costs from
accepting Visa and MasterCard credit cards.
v.
Damages
Even if liability is established, Class Plaintiffs would still face the problems and
complexities inherent in proving damages to the jury. The plaintiffs theory of damages would
be hotly contested at trial. See, e.g., In re NASDAQ Market-Makers Antitrust Litig., 187 F.R.D.
465, 476 (S.D.N.Y. 1998) ([T]he history of antitrust litigation is replete with cases in which
27
antitrust plaintiffs succeeded at trial on liability, but recovered no damages, or only negligible
damages, at trial, or on appeal.). As Professor Sykes points out, it is very difficult to envision
the world that would exist but for default interchange fees and the Honor-all-Cards rules. It is
not likely that credit card interchange fees would then become zero, and given the greater total
costs to users of credit cards as compared to debit cards, it is also unlikely that credit card
interchange would be reduced by competition to the level of debit card interchange. Moreover,
there is reason to believe that lower interchange fees would make cards less attractive to
cardholders (e.g., through higher fees and charges, or reduced rewards). Thus, even if a damage
award measured by the reduction in interchange were possible, it might need to be offset by the
costs associated with the cards diminished value to the holders.
These damages-related issues may not be insurmountable, but they are
formidable. In addition, as I noted in the Wal-Mart case, the Class Plaintiffs would face a
practical problem at trial: They would be asking jurors who feel they had absorbed in the past
the cost of anticompetitive interchange fees through higher prices at the checkout counter to
award extravagant damages that they would absorb in the future through higher bank fees.
Jurors have great common sense. They would conclude with justification that they, not the
merchants, were the ultimate victims of any antitrust violations they found to have occurred, and
they would be naturally reluctant to victimize themselves yet again by awarding significant
damages against the networks and the defendant banks.
The various risks of establishing damages weigh strongly in favor of approving
the settlement.
vi.
Maintaining Class Status
28
Finally, Class Plaintiffs face the risk that the classes would not be certified or that
certification would be modified as the litigation continued. Class Plaintiffs motion to certify
Rule 23(b)(2) and (b)(3) classes is pending. A number of objecting class members voice the
hurdles that Class Plaintiffs would have to overcome to maintain class certification. Though
Class Plaintiffs have strong arguments that the classes should be certified, I note that the
certification of a similar class in the Wal-Mart case was affirmed only over a vigorous dissent,
see Wal-Mart, 280 F.3d at 147 (Jacobs, J., dissenting), and the legal landscape in which class
certification is litigated has deteriorated for plaintiffs in the intervening period. See, e.g.,
In re Initial Pub. Offerings Sec. Litig., 471 F.3d 24, 40 (2d Cir. 2006) (overruling the more
lenient some showing standard of Caridad v. MetroNorth Commuter Railroad, 191 F.3d 283,
293 (2d Cir. 1999)); see also Comcast Corp. v. Behrend, 133 S. Ct. 1426 (2013) (reversing class
certification in antitrust case based on an inadequate damages model). The risks associated with
the certification of classes weigh in favor of settlement.
e.
The Ability of Defendants to Withstand a Greater Judgment
Class Plaintiffs did not address this Grinnell factor, stating that the fact that the
defendants would be able to pay a substantial judgment does not counsel against approval of an
otherwise fair settlement. The objectors note that there is no risk that Visa and MasterCard and
the bank defendants would become insolvent. I agree that these defendants could withstand a
greater judgment, and thus this factor does not weigh in favor of approval.
f. The Range of Reasonableness of the Settlement Fund in Light of the Possible
Recovery and Attendant Risks of Litigation
The objectors argue that Class Plaintiffs should have insisted on a cash payment
closer to their experts damages projection. But the argument ignores the many factors that make
a jury award consistent with that projection nearly a trillion dollars after trebling
29
extraordinarily unlikely. The estimated $7.25 billion that defendants have agreed to pay
represents approximately 2.5% of total interchange fees paid by class members during the class
period, and thus 2.5% of the largest possible estimate of actual damage to merchants.
The damages provided by the proposed settlement are within the range of
reasonableness in light of the best possible recovery and in light of all the attendant risks of
litigation. The settlement proceeds establish a guaranteed fund worth approximately $7.25
billion (before reductions for opt-outs), the largest-ever cash settlement in an antitrust class
action. Class Plaintiffs expert estimates that the ability to surcharge could save merchants
between $26.4 and $62.8 billion in acceptance costs over the next decade. Frankel Decl., 6773, DE 2111-5. The agreement provides for substantial payments to the class members now
rather than leaving them with the prospect of uncertain relief later. The settlement amount
compares favorably with settlements reached in other antitrust class actions, especially given the
lack of prior governmental investigation and the disparity between the parties damages
calculations. See, e.g., In re Currency Conversion Fee Antitrust Litig., 263 F.R.D. 110, 123
(S.D.N.Y. 2009) (settlement of $336 million and injunctive relief represent an extraordinarily
significant recovery in light of the fact that Plaintiffs did not have the benefit of a Government
investigation), affd sub nom. Priceline.com, Inc. v. Silberman, 405 F. Appx 532 (2d Cir.
2010); NASDAQ, 187 F.R.D. at 478 (antitrust settlement of $1.027 billion approved where
plaintiffs and defendants damages estimates were vastly disparate).
The objectors complaint that the proposed settlement does not achieve everything
the plaintiffs sought at the outset of litigation does not tip this factor in their favor. As the
Second Circuit observed in Grinnell:
The fact that a proposed settlement may only amount to a fraction
of the potential recovery does not, in and of itself, mean that the
30
proposed settlement is grossly inadequate and should be

disapproved. In fact there is no reason, at least in theory, why a
satisfactory settlement could not amount to a hundredth or even a
thousandth part of a single percent of the potential recovery.
495 F.2d at 455 & n.2.
I conclude that the Settlement Agreement is both procedurally and substantively
fair.
D. The Objections
As noted, there are numerous objections to the settlement. I discuss the principal
ones below.
1. Rule Reforms
a.
The Elimination of the Networks No-Surcharge Rules
One of the principal accomplishments of the injunctive relief obtained by the

proposed settlement is the elimination, both at the network and product levels, of the rule
prohibiting surcharging by merchants. The proponents of the settlement tout this change as a
significant achievement; the objectors claim it is essentially worthless. The various facets of the
objectors position are addressed in detail below, but I find their position to be largely
unpersuasive. It is true that the value of surcharging to merchants is diminished by certain
factors, some of which are beyond the reach of this lawsuit. It is also true that many merchants,
for reasons sufficient to them, may choose not to avail themselves of the right to surcharge.
Nevertheless, I find that this rule change, which the Class Plaintiffs and the Individual Plaintiffs
fought very hard to obtain, is an indisputably procompetitive development that has the potential
to alter the very core of the problem this lawsuit was brought to challenge. At most, the
objectors arguments establish that the entire solution to that problem constitutes a mosaic, of
which the right to surcharge (like the other forms of relief in the proposed settlement) is but one
31
piece. But it is a central piece a critical accomplishment and the fact of the matter is that an
entire solution to the problem would be unattainable in this action even if it were litigated to its
final conclusion and (against considerable odds) plaintiffs prevailed on all of their claims.
The Class Plaintiffs, the Individual Plaintiffs, and the objectors agree that the
combined effect of the various rules challenged in this lawsuit is supracompetititve interchange
fees on Visa and MasterCard credit cards. The rules work together to prevent inter-brand
competition between each networks cards at the point of sale. If a customer presents a Visa card
for payment, the Honor-all-Cards rule requires the merchant to accept it (unless the merchant
chooses not to accept any Visa credit cards, an unlikely option given their ubiquity). The nosurcharge rule further prohibits the merchant from steering the customer to a competing
MasterCard credit card with a lower interchange rate by surcharging the higher-priced Visa card.
The allegedly supracompetitive interchange rate on the Visa card is hidden by the combined
effects of the rules and the merchant is forced to pay it.
The same is true with respect to different kinds of cards within each network.
Premium cards carry higher interchange fees. Merchants who want to steer customers away
from a higher-interchange MasterCard credit card to a different MasterCard card with a lower
rate by imposing a surcharge on the former are prohibited from doing so.
The proposed elimination of the no-surcharge rules finally would allow merchants
to make transparent and avoidable what has been opaque and inevitable. Customers can be told
what it costs the merchant to accept a particular card. Merchants can use their ability to disfavor
one network at the point of sale by surcharging. That power will incentivize both networks to
moderate or lower their interchange fees to avoid being disfavored. The customer can then
choose between using the premium card that is subject to a surcharge (she may value the rewards
32
of the high-cost card more than a lower purchase price) or being steered to a lower-cost card and
paying less for the transaction.
The ability to surcharge allows merchants to recoup the higher acceptance costs of
the expensive cards. It allows them to steer customers to less costly cards or to other payment
mechanisms, decreasing their card-acceptance costs. It allows market forces to operate on the
previously invisible (to customers) array of interchange fees, and will exert downward pressure
on those fees by injecting a form of competition the current rules have prohibited.
The objectors make a number of arguments in support of their overall claim that
the proposed rules changes that will allow surcharging is essentially worthless. Some are more
persuasive than others, but they do not in my view alter the fact that the rule change would be a
significant achievement for the class.
First, objectors contend that surcharging is prohibited by law in ten states. Thus,
merchants who do business only in those states17 contend that this aspect of the settlement will
be useless to them. I disagree for several reasons. One is that interchange fees are set on a
nationwide basis. Thus, surcharging (or the threat of surcharging) by merchants in the states
where it is permitted may well inure to the benefit of merchants in those ten states. Also, there is
reason to believe that at least some state laws are enforced in a manner that prohibits surcharging
only when the merchant fails to sufficiently disclose the increased prices for credit card use.
17
Certain national or multistate merchant objectors argue that they will not be able to surcharge
anywhere if they have even one store located in a state that prohibits surcharging. There is no support for this
contention. The Settlement Agreement does not contain any such prohibition and the Class Notice, which
defendants consented to, states the opposite:
. . . the fact that a merchants ability to surcharge may be restricted under the
laws of one or more states is not intended to limit the merchants ability under
the settlement to surcharge Visa or MasterCard credit cards where permitted by
state law.
Notice at 8.
33
More importantly, things change, and there is reason to believe that these statelaw impediments to a full deployment of the proposed relief will eventually be among them. Nosurcharge laws are not only anti-consumer, they are arguably irrational, and indeed one of them
New Yorks has bitten the dust in the brief interval since the fairness hearing. Expressions
Hair Design v. Schneiderman, 13 CIV. 3775 JSR, --- F.Supp.2d ---, 2013 WL 5477607
(S.D.N.Y. Oct. 3, 2013).
The plaintiffs in Expressions Hair Design were five New York retailers and their
principals. They challenged New York General Business Law 518, which prohibits a merchant
from imposing a surcharge when a consumer elects to pay with a credit card.18 As Judge Rakoff
observed, the proposed settlements elimination of the no-surcharge rule, which has already
become effective, gave the previously redundant state no-surcharge law renewed
importance. Id. at *4. The retailer plaintiffs wanted to surcharge credit card transactions
because of the two-to-three percent cost per transaction. They [did] not want to frame this price
difference as a cash discount because it wasnt, and calling it a discount would make the
advertised prices seem higher than they really were without making it transparent that the
higher price would be due solely to credit card transaction costs -- precisely the information
[they] wish to convey to [their] consumers. Id. at *5 (internal quotation marks omitted; some
alterations in original). Inspired by the rules change already in effect because of the proposed
settlement, the plaintiff merchants wanted to post a sign telling customers of a 3% surcharge due
18
The section reads in full as follows:

No seller in any sales transaction may impose a surcharge on a holder who elects
to use a credit card in lieu of payment by cash, check, or similar means.
Any seller who violates the provisions of this section shall be guilty of a
misdemeanor punishable by a fine not to exceed five hundred dollars or a term
of imprisonment up to one year, or both.
N.Y. Gen. Bus. Law 518.
34
to high interchange fees. However, fearful of prosecution under 518, they sought a preliminary
injunction precluding enforcement of 518 against them on First Amendment and vagueness
grounds.
In granting that relief, Judge Rakoff observed that Alice in Wonderland has
nothing on New Yorks no-surcharging law. In rejecting the states attempt to defend the statute
as pro-consumer, he wrote that
the statute actually perpetuates consumer confusion by preventing
sellers from using the most effective means at their disposal to
educate consumers about the true costs of credit-card usage. It
would be perverse to conclude that a statute that keeps consumers
in the dark about avoidable additional costs somehow directly
advances the goal of preventing consumer deception.
Expressions Hair Design, 2013 WL 5477607 at *11.
Citing the undeniable public interest in full access to the information about the
costs of credit card acceptance, the critical nature of that information to the millions of
payment decisions made each day, and the fact that no-surcharge rules have the effect of
artificially subsidizing credit at the expense of other payment methods, Judge Rakoff
preliminarily enjoined a surcharge ban indistinguishable from the bans that Visa and
MasterCard recently dropped from their retailer contracts as part of [the instant proposed]
antitrust settlement. Id. at *15, *14.
Nine other such state laws remain. The validity of those laws is not before me,
but I have no reason to doubt Judge Rakoffs assessment that they were enacted in the name of
consumer protection at the behest of the credit-card industry over the objection of consumer
advocates. Id. at *14. Will those laws diminish, at least in the near term, the efficacy of the
proposed relief here? Of course. But Class Counsel and counsel for the Individual Plaintiffs
have good reason to believe that further efforts, whether they take place in courts or legislatures,
35
will successfully address that problem. Those state laws, properly understood, hurt the very
consumers they were ostensibly enacted to protect by propping up high credit card acceptance
costs. They aid and abet a regime in which the poorest consumers subsidize the awards
conferred upon premium cardholders because merchants are prohibited from disfavoring those
premium cards through surcharging.
The objectors also contend that the elimination of the no-surcharge rule is
rendered useless by what all parties have termed the level-playing-field provision, which
conditions a merchants ability to surcharge a Visa or MasterCard credit card on a requirement
that it also surcharge other payment products of equal or greater cost of acceptance.19 Since
many merchants accept American Express, which carries an even higher cost acceptance, and the
American Express rules prohibit surcharging, most merchants will, as a practical matter, be
precluded from surcharging Visa and MasterCard products.
I note preliminarily that the mere fact that merchants may choose not to avail
themselves of the proposed relief (i.e., by continuing to accept American Express) does not
compel the conclusion that the indisputably procompetitive rules changes are not a valuable
achievement. But putting that aside, this objection, like many of the objectors claims, places in
19
Visas level-playing-field provisions reads as follows:

If a merchants ability to surcharge any Competitive Credit Card Brand that the
merchant accepts in a channel of commerce (either face-to-face or not face-toface) is limited in any manner by that Competitive Credit Card Brand, other than
by prohibiting a surcharge greater than the Competitive Credit Card Brands
Cost of Acceptance, then the merchant may surcharge Visa Credit Card
Transactions, consistent with the other terms of this Paragraph 42(a), only on
either the same conditions on which the merchant would be allowed to surcharge
transactions of that Competitive Credit Card Brand in the same channel of
commerce, or on the terms on which the merchant actually does surcharge
transactions of that Competitive Credit Card Brand in the same channel of
commerce, after accounting for any discounts or rebates offered at the point of
sale;
SA 42(a)(iv). MasterCards, id. 55(a)(iv), is substantially identical.
36
sharp relief the limited extent to which the problems merchants complain about in this industry
can be addressed in a single lawsuit. Just as anti-consumer state laws may, at least in the short
term, undermine the relief proposed here, so may the merchant restraints imposed by American
Express. The mere fact that American Express, through rules similar to those challenged here,
may continue to engage in what the objectors would describe as anticompetitive conduct is
hardly an indictment of the proposed settlement before me. The merchants who support the
proposed settlement are fully justified in the view that the American Express problem is not
only a problem that must be (and in fact is being) addressed elsewhere, but that isolating
American Express as the only remaining major network insulated from price competition will
assist in the effort to fix it. In any event, the problem posed by American Expresss merchant
restraints does not alter the fact that the essence of the injunctive relief obtained by the proposed
settlement will permit procompetitive actions by merchants at the point of sale, and that those
actions have the potential to ameliorate the precise anticompetitive effect supracompetitive
interchange fees that these lawsuits were brought to challenge. Tellingly, the objectors, for all
their reliance on the American Express problem in opposing the proposed settlement, have no
solution for that problem. That is no doubt because there could not be one in this case.
The objecting merchants complaints regarding the proposed settlements
requirement that merchants disclose surcharges at the point of entry, at the point of sale, and on
the sales or transaction receipt are without merit. Such disclosure requirements promote pricing
transparency by informing consumers of the charges they will face if they use a particular form
of payment. It will no longer be a hidden tax imposed on consumers. Merchants can now
educate them about of the costs of their different payment products.
37
Lastly, the objectors concerns regarding merchants ability to understand and

implement surcharging are misplaced. Essentially, the new surcharge system has four elements:
Merchants may surcharge the full average discount fee incurred (as determined by the
prior month or last 12 months);
Merchants may surcharge brand-wide (e.g., all Visa or MasterCard credit cards), or they
employ a more nuanced strategy and impose surcharges on one or more product groups
(e.g., Visa Signature cards or MasterCard World Elite cards, which carry higher fees for
many merchants);
Merchants must disclose to consumers that the surcharge does not exceed the merchants
cost of acceptance, and disclose the amount of the surcharge both before it is incurred and
on a receipt; and
If another more expensive network brand that the merchant accepts continues to restrict
surcharging, the merchant may not surcharge Visa and MasterCard without also
surcharging transactions on that competitor network. This is the level-playing-field
provision.
It is true that many merchants are unsophisticated, but I am confident that, with the assistance of
Class Counsel, they will be able to understand their newfound right to surcharge. Pls. Reply
Mem. in Supp. Mot. Final Approval 34, DE 5939.
b.
The Buying Group Provision
The objectors argue that the group-buying and all-outlets provisions will not
provide them a benefit because they will not take advantage of them, or that the relief is illusory
because Visa and MasterCard rules did not previously prohibit merchants from engaging in joint
negotiations. Though there were no rules that prohibited buying groups, it was the practice of
both Visa and MasterCard to refuse to negotiate over interchange fees with merchant buying
groups or other groups of merchants. Allowing groups of merchants to join together to negotiate
with Visa and MasterCard empowers merchants in their dealings with the networks by allowing
them to bargain collectively.
38
In sum, there is no history of negotiations between Visa and buying groups, or

between MasterCard and buying groups. Now there is an affirmative obligation on the part of
each network to negotiate in good faith with such groups. This feature of the proposed
settlement constitutes a meaningful reform that is favorable to merchants.
2. The Releases
The Settlement Agreement includes a release for each of the two classes. Set
forth in paragraphs 33 and 68 of the Settlement Agreement, respectively, the so-called (b)(3) and
(b)(2) releases have drawn numerous objections. The objectors assert that, among other
problems, the releases cover a virtually limitless range of claims, providing Visa and
MasterCard with immunity from all future merchant claims, including antitrust claims, and
therefore should be deemed void as against public policy. Obj. Plaintiffs Br. at 3-4, DE 1678.
Others have grounded similar objections in fundamental notions of due process. However,
because the releases cover only the claims that may properly be extinguished by the settlement of
a class action, I reject these objections.
Both releases cover claims that are or could have been alleged in this case. Such
releases are permissible, see Wal-Mart, 396 F.3d at 106-13, and indeed the ability to include
them in class action settlements is essential to providing defendants the litigation peace they
legitimately expect in return for the settlement of claims. The full array of future claims
embraced by such a release necessarily involves a measure of uncertainty, but the Second Circuit
has clearly established the rule of decision: The law is well established in this Circuit and
others that class action releases may include claims not presented and even those which could
not have been presented as long as the released conduct arises out of the identical factual
39
predicate as the settled conduct. Id. at 107 (quoting TBK Partners, Ltd. v. W. Union Corp., 675
F.2d 456, 460 (2d Cir. 1982)).
The most vehement of the objections to the release share two related flaws. First,
they assume the unlawfulness of both the default interchange and Honor-all-Cards rules, and
object to the release of future challenges to those rules. However, as discussed above, the
anticompetitive character of these rules is far from clear, and the Class Plaintiffs would have a
difficult time proving it here. Armed with both the Class Plaintiffs evidence and legal theories,
the Department of Justice chose not to challenge either rule when it investigated and proceeded
against Visa and MasterCard. In any event, because the illegality of those rules is at a minimum
an unsettled question, future challenges to them could properly be released even in the absence
of the alterations to the networks practices discussed in the following paragraph. See, e.g.,
Robertson v. Natl Basketball Assn, 556 F.2d 682, 686 (2d Cir. 1977).
Second, the objections fail to take sufficient account of the alterations
accomplished directly by the proposed settlement and indirectly during the course of this
litigation. As discussed above, in a post-settlement world, (1) merchants will be allowed to
surcharge (subject to the limitations discussed above) Visa and MasterCard credit cards at both
the brand and product levels; (2) merchants will be able to discount credit and debit transactions;
(3) the reforms achieved by the Durbin Amendment and of the Department of Justice lawsuit
will be preserved by agreement; and (4) Visa and MasterCard will be obligated to negotiate
interchange fees in good faith with merchant groups. Even if the objectors are correct that
default interchange and the Honor-all-Cards rule produce anticompetitive effects, the rules
reforms achieved by the settlement have the potential to meaningfully blunt those effects. In
exchange for a new, going-forward rules structure, the defendants are entitled to bargain for and
40
receive releases of claims that are or could have been alleged based on the identical factual
predicate of the claims in this case.
That is all these releases accomplish. They do not release the defendants from
liability for claims based on new rules or new conduct or a reversion to the pre-settlement rules.
They appropriately limit future damages claims based on the pre-settlement conduct of the
networks.
There is no due process right to opt out of the (b)(2) class, as some objectors
argue. Rule 23 contemplates binding settlements with no opt-out rights where the injunctive
relief achieved by the settlement is appropriate for the class as a whole. The (b)(2) settlement
here is limited to going-forward injunctive relief that changes the structure of the networks
practices. If merchants could opt out of the (b)(2) class, they would reap the benefits of that
relief anyway, as the injunctive relief is generally applicable to all merchants. To allow them to
opt out and pursue their own rules-based injunctive relief would eliminate the incentive to settle
that Rule 23(b)(2) was designed in part to create. The key to the (b)(2) class is the indivisible
nature of the injunctive or declaratory remedy warranted the notion that the conduct is such
that it can be enjoined or declared unlawful only as to all of the class members or as to none of
them. Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2557 (2011) (quoting Richard
Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U. L.REV. 97, at 132
(2009)). That is precisely the case with respect to the claims in this case that seek injunctive
relief from the bundle of network rules that result in according to the plaintiffs allegations
supracompetitive interchange fees in violation of the antitrust laws. Accordingly, the aspect of
the settlement that resolves those claims by neutralizing that bundle of rules is the proper subject
of a (b)(2) class from which no opt-outs are permitted.
41
That the release extends to future challenges to the reformulated rules or

substantially similar ones does not create the giant loophole multiple objectors fear. If the
networks make non-substantive changes to the post-release rules and related conduct, there will
be no reason for the release not to operate. Other changes are not subject to the release. Is there
room for litigation over whether future rules are substantially similar? Of course, but that is no
reason to prohibit the networks from making any rules changes, no matter how unrelated they
may be to the claims and conduct at issue here, on pain of losing the protection of an otherwise
lawful release.
I need not and cannot catalog here all the claims that fall within or without the
release. It suffices to say that the releases do not cover new, future anticompetitive conduct and
rules.
One of the many indicia of how ubiquitous the Visa and MasterCard credit
products have become is the fact that American Express, Discover and First Data Corporation
(which provides services, including equipment, to merchants), and Cardtronics (an ATM
operator) are also members of the merchant class. They object to the release of claims they may
have against Visa and MasterCard, which are among their competitors. These objectors seek to
make something of nothing; it is sufficiently clear from both the text and context of the releases
that these class members are releasing only claims that merchants have alleged or could have
alleged in this case in their capacity as merchants. The release does not bar claims that a class
member may have in its capacity as a payment-card competitor, an ATM operator, or any other
capacity other than as a merchant that accepts Visa and MasterCard credit cards.
3.
The Health Insurers Objections
42
Two health insurance objectors, The Wellpoint, Inc. and the Blue Cross Blue
Shield Entities object to the settlement on the basis that certain provisions of the Patient
Protection and Affordable Care Act of 2010 (Affordable Care Act) could cause the settlement
to affect them differently than other merchants. See Wellpoint Obj. 2-11, DE 2493-2; Blue
Cross Obj. 9-14, DE 2643. The provision that causes the health insurance objectors concern is
the medical loss ratio regulation. Specifically, they argue that a provision of the Affordable Care
Act that limits the amounts insurers can spend on non-health related activities could result in an
increased risk of having to pay certain rebates to customers. See Wellpoint Obj. 3; Blue Cross
Obj. 9-14. The regulation provides that for policies purchased by individuals as opposed to by
employers or groups, at least 80 percent of premium revenues have to be spent on either clinical
services or healthcare quality improvement activities. Tr. 160:6-10. If health insurers do not
meet this 80 percent payment, they are subject to a penalty (i.e., rebate payments). Payment card
fees, such as interchange fees, are neither clinical care nor healthcare quality improvement
activities. Therefore, the argument goes, paying those fees will count against health insurers as
they strive to comply with the 80 percent threshold requirement.
I agree with these objectors that no one thought of their unique concern in
formulating the settlement, but that is no reason not to approve it. The notion that interchange
fees may cause health insurers to cross to the wrong side of that 80% threshold is entirely
speculative.
4.
Claims by States Acting in their Sovereign Capacity

State Attorneys General object that because state governmental entities may
accept Visa or MasterCard cards for payment, the releases might bar claims that are uniquely
and exclusively claims belonging to the States as sovereigns, specifically state law
43
enforcement or parens patriae claims, including parens patriae claims for fines, civil, or other
penalties. State Obj. at 2, DE 2623.
As discussed above, the releases extend only to claims that are alleged or which
could have been alleged in this case. Thus, the settlement resolves claims made by persons,
businesses, and other entities that arise from or relate to their capacity as merchants that accept
Visa-Branded Cards and/or MasterCard-Branded Cards in the United States . . . . Revised Class
Notice at F2-12, DE 1740-2. Therefore, claims brought by States to vindicate interests in their
sovereign capacity are not barred by the release. Pls. Reply Mem. in Supp. Mot. Final Approval
84; Defs. Br. at 34.
Defendants have noted that while the releases do not extend to parens patriae
claims that States assert in their sovereign capacity, the releases bar claims that States may assert
in a representative capacity on behalf of state resident that are members of the Rule 23(b)(2) or
(b)(3) settlement class. States may assert parens patriae claims in a representative capacity for
injuries to the interests of state residents when specifically authorized by a federal or state
statute. See e.g., 15 U.S.C. 15c; Cal. Bus. Prof. Code 16760; 740 Ill. Comp. Stat. 10/7. In
such an action, the States claim is derivative of the state residents claim, and may be barred
where the residents claim is barred.
To resolve the concerns raised by state attorneys general, taking into
consideration the distinction between a state acting in its sovereign or quasi-sovereign capacity
and a state acting in its representative capacity, defendants have proposed the following
provision be added:
The Definitive Class Settlement Agreement and this Class
Settlement Order and Final Judgment do not bar an investigation or
action, whether denominated as parens patriae, law enforcement,
or regulatory, by a state, quasi-state, or local governmental entity
44
to vindicate sovereign or quasi-sovereign interests. The Definitive

Class Settlement Agreement and this Class Settlement Order and
Final Judgment bar a claim brought by a state, quasi-state, or local
governmental entitys proprietary interest (a) as a member of the
Rule 23(b)(2) Settlement Class, or (b) as a member of the Rule
23(b)(3) Settlement Class that has received or is entitled to receive
a financial recovery in this action. The Definitive Class Settlement
Agreement and this Class Settlement Order and Final Judgment
also bar a claim, whether denominated as seeking damages,
restitution unjust enrichment, or other monetary relief, brought by
a state, quasi-state, or local governmental entity for monetary harm
sustained by natural persons, businesses, or other non-state, nonquasi-state, and non-local governmental entities or private parties
who themselves (a) are members of the Rule 23(b)(2) Settlement
Class or (b) are members of the Rule 23(b)(3) Settlement Class.
Sept. 9, 2013 Letter from Defs. at 2 (August 21 Proviso), DE 5995.
To clarify the releases as they bear on states acting in their sovereign capacities, I
adopt that language, which shall be incorporated into the final settlement order and judgment.
5.
Discovers Objection
Discover argues that the level-playing-field provision that cabins merchants
ability to surcharge affects its network in unfair and competitively harmful ways. Specifically,
Discover requests that the Court reject the settlement because its Equal Treatment Rule will be
harmed by that provision. However, to the extent that Discover cards are lower-priced than Visa
and MasterCard products, Discover will not be affected by the level-playing-field provision. Its
claim that in some situations its Equal Treatment Rule may preclude surcharges against Visa
and MasterCard is not a reason to reject the Settlement Agreement.
6.
The Notice to Class

Certain objectors argue that the notice sent to the class contains false statements
that render it inadequate. These false statements include: (1) attributing certain rule changes,
namely the no-discounting and $10-minimum rules for credit transactions, to the proposed
settlement, when the changes predated the settlement; (2) the notice stated that the release covers
45
other fees and other rules, when it covers every merchant fee and all of Visas and
MasterCards rules and unwritten rules and practices; and (3) the notice fails to inform class
members that the surcharging relief will be unavailable to the majority of merchants and
transactions . . . . Obj. Pls. Br at 63.
The notice here meets the requirements of due process and notice standards. It
described the litigation, summarized the settlements terms, quoted the releases verbatim,
described the request for attorneys fees, expenses, and incentive awards for Class Plaintiffs, and
explained the deadline and procedure for filing objections to the settlement as well as opting out
of the case settlement class. The objectors complaints provide no reason to conclude that the
purposes and requirements of a notice to a class were not met here.
There is no showing that the notice did not meet this standard.
7.
Cohesiveness of the Rule 23(b)(2) Class; Adequacy of Class Plaintiffs

Some of the objectors contend that the Rule (b)(2) class is not cohesive. I
disagree.
A Rule 23(b)(2) class is warranted when the party opposing the class has acted
or refused to act on grounds that apply generally to the class, so that final injunctive relief . . . is
appropriate respecting the class as a whole. Fed. R. Civ. P. 23(b)(2). The network rules
regimes that gave rise to this case applied generally to every merchant accepting Visa or
MasterCard credit cards, and the injunctive relief in the proposed settlement does as well.
Specifically, all merchants have the same interest in being able to inform cardholders at the point
of sale of the acceptance costs of their credit cards and to either steer them to lower-cost
alternatives or recoup the cost of acceptance.
46
In fact, by focusing the settlement efforts on the merchant restraints, as opposed

to, for example, default interchange (which many objectors assert should have been the focus),
Class Plaintiffs have enhanced the cohesion of the class. Even if judicial regulation of default
interchange fees were within the reach of the plaintiffs in this action, any such regulation would
affect the class unequally. As mentioned above, default interchange operates only in the absence
of bilateral agreement, and some of the very large merchants have sufficient transaction volume
that they can actually negotiate for their own, lower interchange structures. By ending a specific
merchant restraint that prohibits point-of-sale, competition-enhancing actions, every single
merchant that elects to avail itself of the new rules changes will have received the same benefit.
And because that benefit redesigns the relationship between the each merchant and the networks
in precisely the same manner, the structural relief is generally applicable to the class in the
manner required by Rule 23(b)(2). The fact that some merchants may elect not to avail
themselves of the rule, or are prohibited by factors beyond the scope of this lawsuit from
surcharging, does not undermine my conclusion that the class is sufficiently cohesive to warrant
Rule 23(b)(2) relief in this case.
In addition, the Class Plaintiffs adequately represent both the (b)(2) and the (b)(3)
settlement classes. As discussed above, the interests of the Class Plaintiffs and the rest of the
(b)(2) class are not antagonistic. In addition, the Class Plaintiffs have been involved in this
action from the outset and have fulfilled all of the obligations associated with being class
representatives. They have participated in discovery, in mediation, in court sessions, in the
evaluation of the mediators proposals, and in the formulation of the Settlement Agreement.20
20
The other requirements for certification of settlement classes under Rule 23 have not been the
principal focus of objection, and they are easily met here. Both the (b)(2) and (b)(3) classes contain several million
merchants, satisfying numerosity. Commonality is satisfied because, for both classes, key questions of law and fact
the application of the antitrust laws to uniform Visa and MasterCard policies are common. Common answers to
47
E. The Plan of Allocation

As a general rule, the adequacy of an allocation plan turns on . . . whether the
proposed apportionment is fair and reasonable under the particular circumstances of the case.
In re PaineWebber Ltd. Partnerships Litig., 171 F.R.D. 104, 133 (S.D.N.Y. 1997), affd, 117
F.3d 721 (2d Cir. 1997). An allocation formula need only have a reasonable, rational basis,
particularly if recommended by experienced and competent class counsel. In re Am. Bank Note
Holographics, Inc., 127 F. Supp. 2d 418, 429-30 (S.D.N.Y. 2001) (internal quotation marks
omitted). Whether the allocation plan is equitable is squarely within the discretion of the
those questions are inevitable, see Dukes, 131 S.Ct. at 2551, since the questions focus on the application of law to
the defendants conduct (which was essentially the same toward all class members), not on the individual conduct of
many different plaintiffs. For similar reasons, typicality is satisfied, since the named Plaintiffs claims are for the
same type of injury under the same legal theory as the rest of the class. Reid v. SuperShuttle, International, Inc.,
2012 WL 3288816, at *4 (E.D.N.Y. Aug. 10, 2012). Adequacy, as discussed above, is also met here.
Each class also satisfies its respective subsection of Rule 23(b).
Because all the members of the injunctive relief class were subject to the same rules, and because the relief
afforded by that class is a change to those rules, the class satisfies the requirement that defendants have acted or
refused to act on grounds that apply generally to the class, so that final injunctive relief . . . is appropriate respecting
the class as a whole. Fed. R. Civ. P. 23(b)(2). As discussed above, the rules reforms created by the settlement in
particular, the ability for merchants to surcharge affect all (b)(2) class members equally.
The damages class satisfies Rule 23(b)(3)s predominance and superiority requirements. As discussed
under Rule 23(a), the key issues for the damages class relate to defendants conduct, and those issues vastly
outweigh that is, predominate over any relevant legal or factual determinations that vary among the plaintiffs. It
is true that individual differences exist in the class for example, the ability of a few large merchants to negotiate
their own interchange fees. But in the main, differences among the (b)(3) class members go to the conceivable
monetary recovery, not liability, and two aspects of the settlement mitigate concerns about individual damages
issues. First, those (b)(3) class members who believe they may do better on their own are permitted to opt out, and
many have done so. As discussed above, many of the opt-outs are very large, sophisticated businesses, and they are
competent to protect their own interests. The monetary settlement here does not harm them (and, to the extent it
establishes a precedent, probably helps them in their individual suits or negotiations). Second, for those class
members who have not opted out, the fact of the settlement is relevant, see Amchem Products, Inc. v. Windsor,
521 U.S. 591, 619 (1997), since it creates a single method and procedure for recovering monetary claims that might
otherwise be complex and individualized. And of course, it is well-established that [c]ommon issues may
predominate when liability can be determined on a class-wide basis, even when there are some individualized
damage issues. In re Visa Check/MasterMoney Antitrust Litig., 280 F.3d at 139.
The settlements relevance is obvious for the superiority determination as well. This settlement is not
perfect; as I noted above, one lawsuit cannot possibly address every partys concerns about interchange fees and the
networks rules. But for the purpose of remedying the plaintiffs damages claims flowing from the defendants
alleged antitrust violations, its hard to imagine a better alternative than a single, uniform claims procedure. Absent
certification, there would be no settlement, which would likely mean many fragmented lawsuits; that would risk
inconsistent results and place a huge litigation burden on the defendants. It is also likely that, for the many smaller
merchants that constitute the overwhelming majority of the class, the realistic alternative to certification is no case
at all, given the costs, risks, and range of recovery available in litigation. Cf. Carnegie v. Household Intl, Inc., 376
F.3d 656, 661 (7th Cir. 2004). In sum, common issues predominate and class treatment is superior because (b)(3)
class is sufficiently cohesive to warrant adjudication by representation. Amchem, 521 U.S. at 623.
Thus, both classes meet the requirements of Rule 23, and certification is proper.
48
district court. In re PaineWebber, 171 F.R.D. at 132. I find that the plan is both fair and
reasonable, and thus I approve it.
The plan works as follows. The Class Administrator will distribute the $6.05
billion Cash Fund to Authorized Cash Claimants, on a pro rata basis, depending on the amount of
actual or estimated interchange fees they paid during the class period. (Plan of Administration
and Distribution at SA Appendix I.) Payments to Authorized Interchange Claimants from the
estimated $1.2 billion Default Interchange Payments Fund will be made pro rata, and will be
based on one-tenth of one percent of the claimants Visa and MasterCard transactions during the
eight-month period as compared to total of all claim values for that fund. The amount of
interchange fees paid by each authorized cash claimant will be determined or estimated from
data obtained by Class Counsel from Visa, MasterCard, the bank defendants, non-defendant
acquiring banks and independent service organizations subpoenaed by Class Counsel, and from
the Authorized Cash Claimants themselves. (Id. at I-2).
There have been no substantive objections to the allocation plan except for a
speculative objection from Ace Hardware that I find to be without merit.
I conclude that this plan of allocation, which is recommended by experienced and
competent counsel, is fair, reasonable, and adequate. This conclusion is buttressed by the
relatively small number of opt-outs and absence of objections from class members. Accordingly,
I approve as final the allocation plan.
49
CONCLUSION
For the reasons stated above, the proposed settlement is approved with the minor
modification set forth above. A status conference regarding the next steps in the case shall be
held on January 10, 2014 at 3:00 P.M.
So ordered.
John Gleeson, U.S.D.J.

Dated: December 13, 2013
Brooklyn, New York
50
12-4671-CV(L)
Case: 12-4671
Document: 983
Page: 1
06/16/2014
1249672
105
12-4708-cv(CON), 12-4765-cv(CON), 13-4719-cv(CON), 13-4750-cv(CON), 13-4751-cv(CON),

13-4752-cv(CON), 14-32-cv(CON), 14-117-cv(CON), 14-119-cv(CON), 14-133-cv(CON), 14-157-cv(CON),
14-422-cv(CON), 14-443-cv(CON),14-480-cv(CON), 14-497-cv(CON), 14-530-cv(CON), 14-567-cv(CON),
14-584-cv(CON), 14-606-cv(CON), 14-663-cv(CON), 14-837-cv(CON)
IN THE
United States Court of Appeals

FOR THE SECOND CIRCUIT

On Appeal from the United States District Court
for the Eastern District of New York
JOINT PAGE-PROOF BRIEF FOR

OBJECTORS-APPELLANTS AND PLAINTIFFS-APPELLANTS
(MERCHANT APPELLANTS JOINT BRIEF)
Stephen R. Neuwirth*
Sanford I. Weisburst
Steig D. Olson
Cleland B. Welton II
QUINN EMANUEL URQUHART
& SULLIVAN, LLP
51 Madison Avenue, 22nd Floor
New York, New York 10010
212-849-7000
Jeffrey I. Shinder
Gary J. Malone
A. Owen Glist
335 Madison Avenue, 9th Floor
212-350-2700
Thomas C. Goldstein*
Eric F. Citron
GOLDSTEIN & RUSSELL P.C.
5225 Wisconsin Avenue, N.W., Suite 404
Washington, D.C. 20015
202-362-0636
Michael J. Canter
Robert N. Webner
Kenneth J. Rubin
VORYS, SATER, SEYMOUR AND PEASE LLP
52 East Gay Street
Columbus, Ohio 43215
614-464-6327
Gregory A. Clarick
CLARICK GUERON REISBAUM LLP
220 Fifth Avenue, 14th Floor
212-633-4310
(Parties listed on inside cover)
Case: 12-4671
Document: 983
Page: 2
06/16/2014
1249672
105
*Attorneys for Objector-Appellant

Home Depot U.S.A., Inc.
Attorneys for Plaintiffs-Appellants
Coborns Incorporated; DAgostino Supermarkets, Inc.; Jetro Holdings, LLC;

Affiliated Foods Midwest Cooperative, Inc.; National Association of Convenience
Stores (NACS); National Community Pharmacists Association (NCPA);
National Cooperative Grocers Association (NCGA); National Grocers
Association (NGA); National Restaurant Association (NRA); and NATSO Inc.;
And Objectors-Appellants
7-Eleven, Inc.; Academy, Ltd. d/b/a Academy Sports Outdoors;Aldo US Inc. d/b/a Aldo
and Call It Spring; Alon USA, LP (Alon Brands); Amazon.com, Inc.; American Eagle
Outfitters, Inc.; Barnes & Noble, Inc.; Barnes & Noble College Booksellers, LLC;
Best Buy Stores, L.P.; BJs Wholesale Club, Inc.; The William Carter Company (Carters);
Costco Wholesale Corporation; Crate & Barrel Holdings, Inc.; Darden Restaurants, Inc.;
Davids Bridal, Inc., DBD Inc. and Davids Bridal Canada Inc.; Dicks Sporting Goods, Inc.;
Dillards, Inc.; Family Dollar Stores, Inc.; Drury Hotels Company, LLC; Foot Locker, Inc.;
Gap Inc.; GNC Holdings, Inc. (General Nutrition Corporation); Genesco Inc.;
The Gymboree Corporation; HMSHost Corporation; IKEA North America Services, LLC;
J. Crew Group, Inc.; Kwik Trip, Inc.; Lowes Companies, Inc.; Marathon Petroleum LP;
Martins Super Markets, Inc.; Michaels Stores, Inc.; National Railroad Passenger
Corporation d/b/a Amtrak; Nike, Inc.; Panda Restaurant Group, Inc.;
Panera Bread Company; P.C. Richard & Son, Inc.; PETCO Animal Supplies, Inc.;
PetSmart, Inc.; RaceTrac Petroleum, Inc.; Recreational Equipment, Inc. (REI);
Roundys Supermarkets, Inc. d/b/a Pick N Save, Rainbow, Copps, Metro Market and
Marianos; Sears Holdings Corporation; Speedway LLC; Starbucks Corporation;
Stein Mart, Inc.; Thermo Fisher Scientific Inc.; The Wendys Company; The Wet Seal, Inc.;
Whole Foods Market, Inc.; Zappos.com, Inc.; Fleet Wholesale Supply Co., Inc.;
Mills Motor, Inc.; Mills Auto Enterprises, Inc.; Willmar Motors, LLC; Mills Auto Center, Inc.;
Fleet and Farm of Alexandria, Inc.; Fleet Wholesale Supply of Fergus Falls, Inc.; Fleet and
Farm of Green Bay, Inc.; Fleet and Farm of Menomonie, Inc.; Mills Fleet Farm, Inc.;
Fleet and Farm of Manitowoc, Inc.; Fleet and Farm of Plymouth, Inc.; Fleet and Farm
Supply Company of West Bend, Inc.; Fleet and Farm of Waupaca, Inc.; Mills E-Commerce
Enterprises, Inc.; Brainerd Lively Auto, LLC; Ashley Furniture Industries Inc.; Bealls, Inc.;
Boscovs, Inc.; The Buckle, Inc.; Buc-ees Ltd.; The Childrens Place Retail Stores, Inc.;
Cracker Barrel Old Country Store, Inc.; Cumberland Farms, Inc.; Express, LLC;
Family Express Corporation; New York & Company, Inc.; Republic Services, Inc.;
Swarovski U.S. Holding Limited; The Talbots, Inc.
Attorneys for Objectors-Appellants
Target Corporation; Macys, Inc.; Kohls Corporation; The TJX Companies, Inc.;
Staples, Inc.; J.C. Penney Corporation, Inc.; Office Depot, Inc.; L Brands, Inc.;
Big Lots Stores, Inc.; PNS Stores, Inc.; C.S. Ross Company; Closeout Distribution, Inc.;
Ascena Retail Group, Inc.; Abercrombie & Fitch Co.; OfficeMax Incorporated;
Saks Incorporated; The Bon-Ton Stores, Inc.; Chicos FAS, Inc.;
Luxottica U.S. Holdings Corp. and American Signature, Inc.
Case: 12-4671
Document: 983
Page: 3
06/16/2014
1249672
105
CORPORATE DISCLOSURE STATEMENT

The parties to this brief have submitted contemporaneously herewith a
Compendium of Corporate Disclosure Statements Pursuant to Rule 26.1 of the
Federal Rules of Appellate Procedure.
Case: 12-4671
Document: 983
Page: 4
06/16/2014
1249672
105
TABLE OF CONTENTS
Page
CORPORATE DISCLOSURE STATEMENT ..........................................................i
TABLE OF AUTHORITIES ..................................................................................... v
PRELIMINARY STATEMENT ............................................................................... 1
JURISDICTION......................................................................................................... 6
ISSUES PRESENTED............................................................................................... 6
STATEMENT OF THE CASE .................................................................................. 7
I.
The Underlying Anticompetitive Conduct ...................................................... 7
II.
The Settlement Negotiations And Agreement............................................... 10
III.
Reactions To The Settlement ......................................................................... 18
IV.
The Objections And Their Rejection By The District Court......................... 21

A.
Objections To The Release Of Monetary Claims ............................... 21
B.
Cohesion Objections ........................................................................... 22
C.
Objections To Adequacy Of Representation....................................... 25
D.
Objections To The Scope Of The Release .......................................... 26
STANDARD OF REVIEW ..................................................................................... 28

SUMMARY OF ARGUMENT ............................................................................... 28
ARGUMENT ........................................................................................................... 32
I.
The District Courts Judgment Impermissibly Extinguishes Class

Members Individualized Claims For Money Damages Without
Providing Opt-Out Rights. ............................................................................. 32
A.
Both the Due Process Clause and Rule 23 mandate that class
members have the right to opt out and pursue their
individualized legal claims. ................................................................. 32
ii
Case: 12-4671
B.
C.
II.
Page: 5
06/16/2014
1249672
105
The certification of the (b)(2) settlement class must be vacated

because it extinguishes merchants individualized legal claims
without providing an opt-out right. ..................................................... 38
1.
The settlement, on its face, releases individualized

monetary claims. ....................................................................... 38
2.
The settlement has the effect of releasing a host of

individualized legal claims. ...................................................... 39
There is no legitimate justification for denying merchants optout rights. ............................................................................................. 44

1.
There is no merit to the district courts theory that a nonopt-out class was permissible because the settlement
provides its members no monetary relief. ................................. 44
2.
Rule 23(b)(2) does not allow the certification of

monetary claims that arise in the future for the purpose of
creating litigation peace......................................................... 47
The Mandatory Class Lacked The Required Cohesion Of Interests. ............ 48

A.
The greatest degree of cohesion is required for mandatory

settlement classes. ............................................................................... 48
B.
The merchants bound to the (b)(2) class in this case were too
diverse for a single, indivisible injunction, and the settlement
does not treat those class members equally......................................... 52
C.
III.
Document: 983
1.
Class members had varying interests in the broad set of

claims that the settlement purported to release. ........................ 53
2.
Class members had varying interests in the one claim on

which limited relief was actually provided. .............................. 56
3.
The relief on the Complaints surcharging claim does not

constitute an indivisible injunction. .......................................... 61
The district courts cohesion analysis ignored these flaws. ................ 63
The Settlement Violates The Requirement Of Rule 23(a)(4) That The

Class Members Receive Adequate Representation. ...................................... 66
iii
Case: 12-4671
IV.
Document: 983
Page: 6
06/16/2014
1249672
105
A.
Adequate representation requires separate representatives and

counsel for subgroups with divergent interests. .................................. 66
B.
The (b)(2) and (b)(3) classes had antagonistic interests and

could not be adequately represented by common representatives
and counsel. ......................................................................................... 68
C.
There is no substitute for independent and adequate

representation. ..................................................................................... 74
1.
Overlap ...................................................................................... 75
2.
Results ....................................................................................... 78
By Releasing All Future Antitrust Claims, Including Claims That Far

Exceed The Scope Of The Complaint, The Settlement Violates
Controlling Precedent And Exceeds The Power Of A Federal Court. .......... 80
A.
The settlement unlawfully releases future antitrust claims. ................ 80
B.
The settlement unlawfully releases claims beyond the scope of

the present litigation. ........................................................................... 82
1.
The settlement improperly releases unripe future claims. ........ 84
2.
The settlement improperly releases present claims

beyond the scope of the case. .................................................... 87
CONCLUSION ........................................................................................................ 91
CERTIFICATE OF COMPLIANCE ....................................................................... 94
iv
Case: 12-4671
Document: 983
Page: 7
06/16/2014
1249672
105
TABLE OF AUTHORITIES
Page
Cases
Am. Express Co. v. Italian Colors Rest.,
133 S. Ct. 2304 (2013)............................................................................ 31, 81
In re Am. Intl Grp. Sec. Litig.,
689 F.3d 229 (2d Cir. 2012) ..........................................................................57
Am. Safety Equip. Corp. v. J. P. Maguire & Co.,
391 F.2d 821 (2d Cir. 1968) ..........................................................................81
Amchem Prods., Inc. v. Windsor,
521 U.S. 591 (1997)................................. 4, 35, 42, 45, 48, 49, 51, 52, 53, 57,
59, 64, 67, 68, 71, 73, 75, 77, 78
Authors Guild v. Google Inc.,
770 F. Supp. 2d 666 (S.D.N.Y. 2011) .............................................. 32, 87, 90
Barnes v. Am. Tobacco Co.,
161 F.3d 127 (3d Cir. 1998) ..........................................................................49
Casa Orlando Apartments, Ltd. v. Fed. Natl Mortgage Assn,
624 F.3d 185 (5th Cir. 2010) .........................................................................50
Charron v. Wiener,
731 F.3d 241 (2d Cir. 2013) ................................................................... 28, 37
E & L Consulting, Ltd. v. Doman Indus. Ltd.,
472 F.3d 23 (2d Cir. 2006) ............................................................................85
Eubank v. Pella Corp.,
2014 WL 2444388 (7th Cir. June 2, 2014)....................................................75
Fox Midwest Theatres, Inc. v. Means,
221 F.2d 173 (8th Cir. 1955) .........................................................................82
Gaines v. Carrollton Tobacco Bd. of Trade, Inc.,
386 F.2d 757 (6th Cir. 1967) .........................................................................82
Case: 12-4671
Document: 983
Page: 8
06/16/2014
1249672
105
Gerber v. MTC Elec. Techs. Co.,

329 F.3d 297 (2d Cir. 2003) ..........................................................................28
Hecht v. United Collection Bureau, Inc.,
691 F.3d 218 (2d Cir. 2012) ..........................................................................36
Jefferson v. Ingersoll Intl Inc.,
195 F.3d 894 (7th Cir. 1999) .........................................................................36
Joel A. v. Giuliani,
218 F.3d 132 (2d Cir. 2000) ..........................................................................37
In re Joint E. and S. Dist. Asbestos Litig.,
982 F.2d 721 (2d Cir. 1992) ..........................................................................75
Kartman v. State Farm Mut. Auto. Ins. Co.,
634 F.3d 883 (7th Cir. 2011) .........................................................................50
Lawlor v. Natl Screen Serv. Corp.,
349 U.S. 322 (1955).................................................................... 31, 81, 82, 84
Lemon v. Intl Union of Operating Engrs,
216 F.3d 577 (7th Cir. 2000) .........................................................................49
In re Literary Works in Elec. Databases Copyright Litig.,
654 F.3d 242 (2d Cir. 2011) .................... 2, 28, 37, 46, 52, 65, 68, 73, 74, 77,
78, 83, 85, 86
Logan v. Zimmerman Brush Co.,
455 U.S. 422 (1982).......................................................................................32
In re Masters Mates & Pilots Pension Plan & IRAP Litig.,
957 F.2d 1020 (2d Cir. 1992) ........................................................................28
M.D. ex rel. Stukenberg v. Perry,
675 F.3d 832 (5th Cir. 2012) .........................................................................50
Minn. Mining & Mfg. Co. v. Graham-Field, Inc.,
1997 WL 166497 (S.D.N.Y. Apr. 9, 1997) ...................................................82
Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc.,
473 U.S. 614 (1985)................................................................................ 81, 82
vi
Case: 12-4671
Document: 983
Page: 9
06/16/2014
1249672
105
Natl Super Spuds, Inc. v. N.Y. Mercantile Exch.,

660 F.2d 9 (2d Cir. 1981) ....................................................................... 83, 86
Ortiz v. Fibreboard Corp.,
527 U.S. 815 (1999)................................. 5, 31, 33, 36, 42, 43, 48, 51, 64, 65,
67, 71, 72, 73, 77
Phillips Petroleum Co. v. Shutts,
472 U.S. 797 (1985)................................ 12, 21, 29, 32, 33, 36, 37, 42, 43, 49
Prime Mgmt. Co., Inc. v. Steinegger,
904 F.2d 811 (2d Cir. 1990) ..........................................................................84
Robertson v. NBA,
556 F.2d 682 (2d Cir. 1977) ..........................................................................86
Robinson v. Metro-North Commuter R.R. Co.,
267 F.3d 147 (2d Cir. 2001) .................................................................... 48-49
In re St. Jude Med., Inc.,
425 F.3d 1116 (8th Cir. 2005) .......................................................................49
Sanjuan v. Am. Bd. of Psychiatry & Neurology, Inc.,
40 F.3d 247 (7th Cir. 1994) ...........................................................................82
Stephenson v. Dow Chem. Co.,
273 F.3d 249 (2d Cir. 2001),
vacated in part on other grounds,
539 U.S. 111 (2003).................................................................... 36, 42, 43, 77
TBK Partners, Ltd. v. Western Union Corp.,
675 F.2d 456 (2d Cir. 1982) ................................................................... 37, 83
Three Rivers Motor Co. v. Ford Motor Co.,
522 F.2d 885 (3d Cir. 1975) ..........................................................................82
United States v. Visa U.S.A., Inc.,
344 F.3d 229 (2d Cir. 2003) ..................................................................... 7, 40
In re Visa Check/MasterMoney Antitrust Litig.,
280 F.3d 124 (2d Cir. 2001) ......................................................... 2, 36, 52, 89
vii
Case: 12-4671
Document: 983
Page: 10
06/16/2014
1249672
105
Wal-Mart Stores, Inc. v. Dukes,

131 S. Ct. 2541 (2011)............................. 4, 12, 21, 29, 30, 33, 34, 35, 38, 39,
43, 45, 46, 49, 50, 55, 61, 63
Wal-Mart Stores, Inc. v. Visa U.S.A. Inc.,
396 F.3d 96 (2d Cir. 2005) ................................................................. 7, 46, 86
Weinberger v. Kendrick,
698 F.2d 61 (2d Cir. 1982) ............................................................................86
Constitution, Statutes, and Rules
U.S. Const. amend. V............................ 1, 2, 6, 21, 29, 32, 42, 43, 44, 49, 52, 64, 77
28 U.S.C. 1291 ........................................................................................................6
28 U.S.C. 1331 ........................................................................................................6
28 U.S.C. 1332 ........................................................................................................6
28 U.S.C. 1337 ........................................................................................................6
28 U.S.C. 2201 ........................................................................................................6
28 U.S.C. 2202 ........................................................................................................6
Fed. R. Civ. P. 23 ............................................................................................. passim
Fed. R. Civ. P. 23(a)(4) ............................................................. 66, 67, 71, 74, 75, 78
Fed. R. Civ. P. 23(b)(1)..................................................................................... 45, 47
Fed. R. Civ. P. 23(b)(2).................................................................................... passim
Fed. R. Civ. P. 23(b)(3).................................................................................... passim
Fed. R. Civ. P. 23(c)(2)(B) ......................................................................................34
Fed. R. Civ. P. 23(c)(4)(B) ......................................................................................71
Other Authorities
James Grimmelmann, Future Conduct and the Limits of Class-Action
Settlements, 91 N.C. L. Rev. 387 (2013) .................................................84, 85
viii
Case: 12-4671
Document: 983
Page: 11
06/16/2014
1249672
105
Benjamin Kaplan, Continuing Work of the Civil Committee: 1966

Amendments of the Federal Rules of Civil Procedure, 81 Harv. L.
Rev. 356 (1967) .............................................................................................45
Moores Federal Practice 23.25[5][e] ...................................................................67
ix
Case: 12-4671
Document: 983
Page: 12
06/16/2014
1249672
105
PRELIMINARY STATEMENT
Unless reversed, the district courts class-certification ruling in this case will
inaugurate a new and dangerous model for settlement class actions. The settlement
approved below forces a diverse collection of tens of millions of class members to
release a wide array of individualized monetary claims against the defendants.
Supreme Court precedent unambiguously bars that result; class members have the
right under both the Due Process Clause and Federal Rule of Civil Procedure 23 to
opt out of any settlement resolving monetary claims so that they may pursue those
claims individually. But the purpose and effect of this settlement is to evade that
established rule: It creates contrived opt-out and non-opt-out classes, represented
by the same lawyers and class representatives, in order to require all the members
of the larger, non-opt-out class to release all of their claims.
In approving the settlement, the district court allowed the defendants to pay
money to the opt-out class in exchange for a compulsory release from the non-optout class of claims against the defendants ongoing and future conductmoney
damages claims included. That trade violates the right of objecting class
memberslike the more than 200 objectors joining this Merchant Appellants
Joint Briefto litigate their own individualized claims and so to preserve their
only chance to stop conduct they believe is unlawful. The district courts judgment
approving this design should be reversed.
Case: 12-4671
Document: 983
Page: 13
06/16/2014
1249672
105
Here, in sum, are the facts: The plaintiffs are merchants that accept Visa and
MasterCard. They brought suit under the Sherman Act against Visa, MasterCard,
and certain of their member banks. The Complaint targeted specific
anticompetitive practices that defendants use to inflate the interchange fees
merchants pay for accepting their cards. The plaintiffs sought to proceed on behalf
of a massive collection of diverse merchants that take such cardsfrom the largest
chain stores to the smallest food trucks.
The plaintiffs lawyers then negotiated a settlement with the defendants.
Consistent with the Due Process Clause and Rule 23, the settlement could have
resolved the plaintiffs claims on an opt-out basis. See, e.g., In re Visa
Check/MasterMoney Antitrust Litig. (Visa Check), 280 F.3d 124, 147 (2d Cir.
2001) (Sotomayor, J.); In re Literary Works in Elec. Databases Copyright Litig.
(Literary Works), 654 F.3d, 242, 246 (2d Cir. 2011). But these defendants
conditioned a multi-billion dollar paymentand commensurate fee award to class
counselon a non-opt-out agreement that immunized the defendants from any
future challenge by any merchant to their ongoing conduct. The appellants joining
this brief are among the large proportion of merchants that objected to such a
settlement as not only substantively inadequate, but also a wrongful deprivation of
their fundamental right to protect their interests individually.
Case: 12-4671
Document: 983
Page: 14
06/16/2014
1249672
105
One set of counsel and representative plaintiffs negotiated the settlement for
both classes; the non-opt-out class, despite its divergent interests, was not afforded
independent representation. The unprecedented settlement they reached involves
two key elements.
First, the settlement defines a conventional opt-out class of merchants that
accepted Visa or MasterCard in the past. The settlement grants these merchants
cash as compensation for past damages, if they do not opt out.
Second, and critically, the settlement defines a non-opt-out class consisting
of all merchants that accept Visa or MasterCard at any time after November 28,
2012 (the date the district court granted the settlement preliminary approval). This
class includes all the members of the opt-out class who remain in businesseven
those who have actually opted outplus all the millions of merchants that will
ever be founded and accept credit cards at any point in the future. In substance, the
settlement grants these merchants limited prospective relief with respect to only
one challenged practice, while immunizing the defendants from suit regarding the
other practices challenged by the Complaint. Indeed, the immunity is substantially
broader than even that: From the date of preliminary approval, the settlement
forces all the non-opt-out class members to forever release their claims against the
defendants with respect to all the conduct challenged in the Complaint, all of
Case: 12-4671
Document: 983
Page: 15
06/16/2014
1249672
105
defendants other existing policies and practices, and any substantially similar
practices they ever adopt in the future.
This settlement scheme is unlawful. The Supreme Court recently and
unanimously reiterated that class-action judgments may not resolve individualized
monetary claims without an opt-out right. See Wal-Mart Stores, Inc. v. Dukes, 131
S. Ct. 2541, 2558-59 (2011). Yet this settlement does just that, granting
defendants sweeping prospective immunity from suitincluding suits for money
damages. Defendants retain that immunity forever, even if economic
circumstances change in a manner that exacerbates the anticompetitive effects of
their practices or creates new harms.
This settlement thus forces a group of motivated and well-equipped
commercial entitiesstanding ready and willing to litigate the unlawfulness of
defendants conductto surrender their high-value monetary claims forever. That
result inverts the core utility of class actions, which is to overcome the problem
that small recoveries do not provide the incentive for any individual to bring a solo
action prosecuting his or her rights. Amchem Prods., Inc. v. Windsor, 521 U.S.
591, 617 (1997) (citation and quotation omitted).
The settlement is plainly unlawful in other respects as well. It binds together
an astonishingly disparate class with tens of millions of membersessentially,
Case: 12-4671
Document: 983
Page: 16
06/16/2014
1249672
105
every imaginable merchanteach with radically different interests in the many

claims it releases.
Further, one set of class counsel and representatives bargained on behalf of
two classes with conflicting interests. The substantially larger non-opt-out class
including new, growing, and yet-to-be-created merchantsnaturally favored
forward-looking relief that would protect against future harms. The smaller, optout class necessarily had a relatively greater interest in retrospective reliefi.e.,
money damages. The class representatives and counsel thus had an incentive to
sacrifice the future-looking interests of the former for the money immediately
available to the latter. The Supreme Court has held that just such a design is
obvious[ly] unlawful. Ortiz v. Fibreboard Corp., 527 U.S. 815, 856 (1999).
This settlement structure is surely a boon to defendants, who can secure
permanent immunity for their ongoing practices. But Rule 23 does not exist to
strip objecting class members of the right to pursue their own legal claims. Those
claims vindicate not only private rights, but also the substantial public interest in
enforcement of the antitrust laws. If this settlement stands, class members will not
get their day in court, and practices that raise prices for everyone will be
immunized from any future challenge. Nothingand certainly not defendants
desire for litigation peace, SPA44can justify that result.
Case: 12-4671
Document: 983
Page: 17
06/16/2014
1249672
105
JURISDICTION
The district court had jurisdiction under 28 U.S.C. 1331, 1332, 1337,
2201, and 2202. This Court has jurisdiction under 28 U.S.C. 1291. The district
court entered a memorandum and order on December 13, 2013, SPA1, and a final
judgment on January 14, 2014, SPA73. The parties to this brief timely filed their
notices of appeal on December 13, 2013 and filed supplemental notices on January
21, 2014 and February 13, 2014 after the district court entered judgment.
JA[__]{DE6125; DE6126; DE6128; DE6212; DE6248; DE6249}.
ISSUES PRESENTED
1. Does the certification of a non-opt-out settlement class that extinguishes
class members individualized monetary claims violate the Due Process Clause and
Rule 23?
2. Does a non-opt-out settlement class lack cohesion when its millions of
existing and future members have inconsistent interests with respect to the
defendants many practices and policiesall of which are simultaneously
immunized from suit under the settlement?
3. Does a single set of class representatives and counsel provide inadequate
representation when they negotiate a settlement under which (i) one class receives
a large cash payment, while (ii) a second, non-opt-out class that receives only
Case: 12-4671
Document: 983
Page: 18
06/16/2014
1249672
105
modest injunctive relief is required to grant the defendants a sweeping release from
future liability?
4. Does this settlement unlawfully release future antitrust claims, unripe
claims against future conduct, and claims that exceed the scope of the Complaint?
STATEMENT OF THE CASE
This is an appeal from a judgment of the U.S. District Court for the Eastern
District of New York (Gleeson, J.), certifying settlement-only classes and
approving a final class-action settlement. The opinion is not yet reported but is
available at 2013 WL 6510737 (E.D.N.Y. Dec. 13, 2013).
I.
The Underlying Anticompetitive Conduct

Merchants are charged an interchange fee every time they accept a Visa or
MasterCard credit or debit card.1 These fees are lucrative: U.S. merchants alone
pay more than $40 billion per year. See JA[__]{DE-1533 (Plaintiffs Summ. J.
Opp. 20-21; Rebuttal Report of Alan S. Frankel, Ph.D. 216; Report of Robert H.
Topel at 22 n.52)}. The high price reflects the fact that interchange fees are set on
For further industry background, see Wal-Mart Stores, Inc. v. Visa U.S.A.
Inc., 396 F.3d 96, 101-02 (2d Cir. 2005), and United States v. Visa U.S.A., Inc.,
344 F.3d 229, 234-37 (2d Cir. 2003). Further detail is also provided in the
Merchant Trade Groups Brief, which focuses on the unfairness of the settlement.
These appellants join those arguments and incorporate them by reference, along
with the arguments advanced in the briefs of the Retailers and Merchants
Objectors, and U.S. PIRG and Consumer Reports.
7
Case: 12-4671
Document: 983
Page: 19
06/16/2014
1249672
105
a non-competitive, industry-wide basis through the Visa and MasterCard networks

that nominally exist to facilitate payments among merchants and banks.
This case arises from a consolidated class-action Complaint asserting
antitrust claims against Visa and MasterCard, as well as their member banks,
relating to these interchange fees. Visa and MasterCard have some 5,000 pages of
rules, spread over sixteen rulebooks, governing acceptance of Visa and MasterCard
transactions. See JA[__] (public rulebooks); JA[__]{Corrected 9/12/13 Tr. 100;
DE2605 (Amazon.com Obj. 14)}. But the Complaint challenged only a tiny
fraction of those rules and practices as restraining competition among banks over
interchange fees.2
The plaintiffs core allegation is that the defendants fix interchange fees by
adhering to published schedules of so-called default rates. See JA[__]
(Complaint 1, 443-68). These schedules provide a rate in the absence of a
bilateral agreement between a given bank and a particular merchant. In practice,
however, the default rate is the actual rate. Although banks and merchants can
theoretically negotiate individual agreements, that does not happen in reality. The
default ratestogether with other practiceseliminate any incentive for the banks
There were ultimately three complaints in the case, two of which were
addressed to the Visa and MasterCard IPOs. The operative complaint for present
purposes is the Second Consolidated Amended Class Action Complaint, and is
referred to as the Complaint. See JA[__].
8
Case: 12-4671
Document: 983
Page: 20
06/16/2014
1249672
105
to compete with each other and to negotiate with merchants over rates or terms of
acceptance. So Visa and MasterCard set the default fees at supra-competitive
levels, and the banks in turn apply those rates without risk that their competitors
will offer merchants a lower price.
The most important additional restraint supporting this anticompetitive
regime is the collection of rules known as Honor-All-Cards, which requires any
merchant that accepts any credit (or debit) card on the Visa or MasterCard network
to accept all credit (or debit) cards on that network. It makes no difference which
bank issued the card orcriticallywhat interchange fee applies to the card. So,
for example, if a merchant wants to take Visa credit cards, it must accept not only
basic Bank of America Visa cards but also Chases premium Sapphire Preferred
Visa cards, even if the merchant must pay a substantially higher interchange fee for
the latter. JA[__] (Complaint 8(m), 240, 244, 436). As a consequence, no bank
has an incentive to offer a merchant a lower interchange rate to accept any of its
cards: A merchant cannot reject any issuing banks Visa or MasterCard credit card
without dropping the entire network, including the less expensive cards of every
other bank.
Visa and MasterCard have other anti-steering rules that reinforce the
barriers to interchange competition among the banks. JA[__] (Complaint, 8(d)).
Although the settlement in this case does not provide any relief with respect to
Case: 12-4671
Document: 983
Page: 21
06/16/2014
1249672
105
default interchange or Honor-All-Cards, it does address one such restraint: the nosurcharging rule. Before the settlement, Visa and MasterCard barred merchants
from charging a customer any additional fee for using any kind of payment card.
If merchants could impose such surcharges, they could theoretically encourage
consumers to use lower-cost options. See, e.g., JA[__] (Complaint 8(d), 94, 97,
189-99). But many states prohibit surcharging by statute, making the networks
no-surcharging rules irrelevant in those parts of the country. See infra, at 22-23,
56-60.
II.
The Settlement Negotiations And Agreement

Class counsel and the defendants sought to negotiate a comprehensive
settlement that would bind every kind of merchant that accepts payment cards.
Their putative class has tens of millions of members and is breathtaking in scope.
It includes Amazons nationwide delivery service and the local pizza delivery
shop; big-box retailers and mobile food trucks; tech-savvy online sellers and local
corner stores; high-fashion retailers where almost everyone uses credit cards and
low-margin food marts where consumers routinely use debit, cash, or personal
checks. Indeed, the pervasive presence of payment cards stretches the class far
beyond recognizable retail merchants to include health insurers, state governments,
public utilities, and all other entities that accept Visa or MasterCard.
10
Case: 12-4671
Document: 983
Page: 22
06/16/2014
1249672
105
Class counsels stated goal in the litigation was to secure not only
compensation for the merchants past damages, but also lasting reform for the
future. As lead counsel explained: While th[e] action contained a damage
claim, and we certainly expected damages to be enormous, the primary goals were
to reform the market by eliminating the horizontal agreements among the banks to
fix the levels of interchange fees and enforce the rules that we were challenging.
JA[__]{DE2113-6 (Wildfang Decl. 24)}.
For their part, the defendants also had an overriding goal in settlement
negotiations: complete and permanent litigation peace extending well beyond the
limited structural changes they were willing to make to their practices going
forward. Because banks receive $40 billion annually in interchange fees, they
could easily afford to make a nominally large cash payment to the class, as well as
minor rules changes. But in return, they required assurances that they would never
face additional private suits by merchants relating to any of their policies or
practices. Throughout the negotiations, the two objectives were bound together.
As lead counsel further explained: The negotiations before the mediators were
alwaysone issue was monetary, the other issue was equitable relief. One was not
going to be reached without reaching the other. JA[__]{DE1732 (11/9/12 Tr. 9)}.
The interests of the negotiating parties culminated in the sweeping
settlement at issue in this appeal. It seeks to grant the defendants an expansive,
11
Case: 12-4671
Document: 983
Page: 23
06/16/2014
1249672
105
permanent immunity from suit by any merchant, including from legal claims for
money damages. But the settling parties faced the obstacle that Supreme Court
precedent unambiguously prohibits a mandatory class-action judgment that
resolves class members individualized monetary claims. See, e.g., Dukes, 131
S. Ct. at 2558-59; Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 811-12 (1985).
The settlement attempts to avoid that rule by defining two classes: (1) an opt-out
class that would receive money damages, and (2) a non-opt-out class that would
release all its claims prospectively as of the settlements preliminary approval.
First, the settlement creates an opt-out class certified under Rule 23(b)(3).
This class encompasses all merchants with damages claims arising before the date
of preliminary approval (November 28, 2012). SPA118 (Settlement 2(a)). Class
counsel estimated that this class contains more than 12 million members. SPA23.
Members of the (b)(3) class that did not opt out would receive payments from two
cash funds totaling up to an estimated $7.25 billion. SPA13.3 The participating
members of this class must release all existing and future claims against defendants
3
The cash payment was set at $6.05 billion, subject to reductions of up to

25% for opt-outs. SPA77-78 (Judgment 9(a)); SPA120-21, 125-26 (Settlement
9-11, 18-20). Because opt-outs exceeded 25% of the transaction volume at
issue, the $6.05 billion was reduced to about $4.5 billion. SPA77-78 (Judgment
8, 9(a)); JA[__]{DE5940 (Class Plaintiffs Fee Reply at 7)}. Those remaining in
the (b)(3) class will also receive an estimated ten-basis-point interchange fee
reduction for eight months. SPA54; SPA78 (Judgment 9(b)); SPA121-24
(Settlement 11-15). Together, this consideration is actually worth about $5.7
billion (or about $5.2 billion after deduction of counsels fees).
12
Case: 12-4671
Document: 983
Page: 24
06/16/2014
1249672
105
with respect to any of their current rules or practices, as well as any future rules or
practices that are substantially similar. SPA131-39 (Settlement 31-38). The
release granted by this class is non-mandatory, however, because merchants may
opt out.
The settlement also creates a second, non-opt-out class certified under Rule
23(b)(2). This class encompasses all merchants that have accepted Visa or
MasterCard since the date of preliminary approval or will accept either of them in
the future. SPA118 (Settlement 2(b)).4 The members of this class are defined by
their possession of claims arising any time after preliminary approvalincluding
individualized claims for money damages that accrue at any point in the future.
This class includes all the members of the opt-out class that remained in business
after preliminary approval (even if they opted out), plus tens of millions of
additional merchants that will subsequently open their doors and accept Visa or
MasterCard. In sum, because this class is mandatory, every merchant in the
country that now or in the future accepts Visa or MasterCard is bound by its terms
and barred from opting out.
Because acceptance of those brands is ubiquitous, and the settlement is only

relevant to merchants that do accept Visa or MasterCard, we use all merchants as
shorthand for all merchants to the extent they ever accept a Visa or MasterCard
transaction.
13
Case: 12-4671
Document: 983
Page: 25
06/16/2014
1249672
105
The settlement with the non-opt-out (b)(2) class leaves in placeand indeed
immunizes from any later suitthe default interchange schedules and Honor-AllCards rules that were the focus of the plaintiffs antitrust claims. Instead, Visa and
MasterCard agreed to three limited forms of prospective relief. See SPA85-87
(Judgment 13(c)-(f), (i)-(j)). These limited changes remain in place only until
July 20, 2021. SPA87 (Judgment 13(m)); SPA151, 164 (Settlement 45, 58).
First, the settlement permits merchants to accept Visa or MasterCard at some
outlets, but not others, if those outlets operate under separate trade names or
banners. SPA85 (Judgment 13(c)-(d)); SPA140-41, 153-54 (Settlement 41,
54). Visa and MasterCard never explicitly prohibited this practice, however. See
JA[__]{DE2448 (Costco Obj. 20); DE2644 (Wal-Mart Obj. 41)}. Further, this
relief is irrelevant to the vast majority of U.S. merchants, who operate exclusively
under one trade name.
Second, the settlement provides that Visa and MasterCard will negotiate in
good faith with merchant-organized buying groups. SPA86-87 (Judgment 13(i)(j)); SPA149-50, 163-64 (Settlement 43, 56). But here too, Visa and MasterCard
never expressly prohibited this practice before. SPA43. Further, the obligation is
only to negotiate; there is no enforceable duty to reach agreement. Finally, this
relief is of limited practical value for the many merchants who are unlikely to join
with competitors because of their size or business model.
14
Case: 12-4671
Document: 983
Page: 26
06/16/2014
1249672
105
Thirdand embodying the only form of relief the district court thought had
any material significancethe settlement provides that merchants may engage in
certain forms of surcharging. The settlement permits surcharging at the Brand
Level (i.e., all Visa and/or MasterCard transactions) or the Product Level (i.e.,
transactions on cards of the same type, such as all Visa Traditional Rewards cards
but not Visa Classic cards). SPA85-86 (Judgment 13(e)-(f)); SPA141-49, 154-63
(Settlement 42, 55). But while the settlement addresses defendants surcharging
bans, many class members will remain foreclosed from surcharging by substantial
legal, contractual, and practical barriers, including state law prohibitions and preexisting contracts. See infra, at 22-23, 56-60.
Whatever the value of these three forms of relief to individual class
members, the settlement mandates that, in exchange, the entire (b)(2) class grant
defendants a sweeping immunity from suitincluding suits for money damages.
All of the millions of existing and future (b)(2) class members are forced to release
their claims regarding defendants post-November 28, 2012 conduct. That release
covers all of Visas and MasterCards existing rules, all their unwritten practices,
and any future rules or practices that may in the future exist in the same or
substantially similar form. SPA169-72, 173-74 (Settlement 68, 71) (emphasis
added).
15
Case: 12-4671
Document: 983
Page: 27
06/16/2014
1249672
105
Unlike the changes described above, this broad release continues in

perpetuity. And it extends to new merchants that do not yet even exist. Merchants
that start accepting Visa and MasterCard transactions only after July 20, 2021
release their claims even though they will not receive any of the settlement relief at
all.
The (b)(2) release explicitly extends not only to claims in the Complaint on
which the class receives no relief, but also to claims that were notindeed, could
not have beenasserted in the Complaint. The settlement requires the (b)(2) class
to grant the defendants immunity from suit with respect to any other actual or
alleged Rule, SPA170 (Settlement 68(c)), defined to mean any rule, by-law,
policy, standard, guideline, operating regulation, practice, procedure, activity or
course of conduct relating to any Visa-Branded Card or any MasterCard-Branded
Card, SPA113 (Settlement 1(mm)) (emphasis added). It thus reaches the entirety
of defendants detailed rulebooks, as well as all of their unwritten rules, policies,
and practices.
The release applies to all manner of claims, including money damages
claims, even if they did not exist at the time the Complaint was filed. It also
encompasses claims that could only arise in the futurefor example, because
circumstances change to make a current policy unlawfully anticompetitive. The
release extinguishes any and all manner of claims including:
16
Case: 12-4671
Document: 983
Page: 28
06/16/2014
1249672
105
any form of declaratory, injunctive, or equitable relief, or any

damages or other monetary relief relating to the period after the date
of the Courts entry of Class Settlement Preliminary Approval that
any Rule 23(b)(2) Settlement Class Releasing Party now has, or
hereafter can, shall, or may in the future have.
SPA169 (Settlement 68) (emphasis added); see also SPA173-74 (Settlement 71).
These are claims that may not ripen for years, or decades, because the release
extends to the future effect of present rules or conduct, or substantially similar
rules or conduct, whether or not those practices have any current anticompetitive
effects. SPA171 (Settlement 68(g)-(h)).
For example, the settlement provides no relief from the default interchange
or Honor-All-Cards rules. But the mandatory (b)(2) release expressly bars all
claims arising out of or relating in any way to those specific practices. SPA16972 (Settlement 68). The release also specifically provides that it does not in any
way limit the ability of any Visa Defendant or MasterCard Defendant to set
interchange rates. SPA152, 166 (Settlement 51, 64).
The release also expressly includes damages claims against Visas Fixed
Acquirer Network Fee (FANF). SPA174 (Settlement 72(d)). But the plaintiffs
could not have asserted such a claim in the Complaint because Visa did not even
implement the FANF until after the close of summary judgment briefing.
17
Case: 12-4671
III.
Document: 983
Page: 29
06/16/2014
1249672
105
Reactions To The Settlement

When announced, the settlement was met with widespread opposition.
There were originally nineteen named plaintiffs in the case, including six major
trade associations. Tenthe majority, including all the trade associations acting in
the interest of their many membersobjected to the mandatory (b)(2) settlement.
See JA[___]{DE2447 (Coborns Obj. 7-12); DE2449 (DAgostino Obj. 8-11);
DE2459 (Jetro Obj. 6); DE2563 (Affiliated Foods Obj. 7); DE2561 (NACS Obj.
11-22); DE2619 (NCPA Obj. 13-18); DE2546 (NCGA Obj. 7); DE2475
(NGA Obj. 7-12); DE2464 (NRA Obj. 7-8); DE2461 (NATSO Obj. 6-11);
DE6006-1 (NACS Supp. Decl. 5-16); DE6006-2 (NCGA Supp. Decl. 5-18);
DE6006-3 (NCPA Supp. Decl. 6-13); DE6006-4 (NGA Supp. Decl. 5-17)}.
Lead class counsel reacted by dropping them as class representatives and excluding
them from all further negotiations, even though they would remain bound, as
members of the mandatory (b)(2) class, to the representation of class counsel and
the settlements broad release of claims.
In total, several thousand merchantslarge and smallobjected to the
mandatory (b)(2) class. [T]he roster of objectors include[d] some of the nations
largest retailers, representing almost 20% of all Visa and MasterCard U.S.
transaction volume. SPA23.
18
Case: 12-4671
Document: 983
Page: 30
06/16/2014
1249672
105
The over 200 appellants joining this brief are large and small merchants that
believe, for individualized reasons, that releasing all future claims against every
existing Visa or MasterCard practice is unacceptable. Somelike appellant
Amazonare large, growing, and particularly likely to engage in credit-card
transactions, and so have an especially keen interest in future-looking claims.
Otherslike appellant 105 Degreesare small merchants located exclusively in
states that prohibit surcharging, and so would have pursued a very different mix of
prospective relief. See Retailers and Merchants Br. 30. Yet others, like appellants
Wal-Mart, Target, and Home Depot, represent huge transactional volume, and
believe that the most important objective is ending the restraints that prevent them
from using that volume to negotiate better deals directly with card-issuing banks.
These well-equipped and motivated commercial entities stood ready,
willing, and able to litigate against defendants ongoing restraints; indeed, many
are now litigating those issues as opt-outs from the (b)(3) class. Nevertheless, the
mandatory (b)(2) settlement requires them to lay down their injunctive and
continuing monetary claims in exchange for relief they think does them no good.
Thus, the opt-out complaints that they have filed are limited to asserting damages
through November 28, 2012, even though the challenged conduct is ongoing,
because the settlement extinguishes their rights to seek any relief thereafter. See,
e.g., JA[__]{Second Amended Complaint, 7-Eleven, Inc. v. Visa Inc., Nos. 13-cv-
19
Case: 12-4671
Document: 983
Page: 31
06/16/2014
1249672
105
5746(JG)(JO), 14-md-1720(JG)(JO) (E.D.N.Y.), ECF No. 38; DE2495-2 (Target

Complaint)}.
The (b)(3) settlement similarly generated widespread opt-outs from
merchants, representing over 25% of transaction volume. JA[__]{DE5940 (Class
Plaintiffs Fee Reply at 7)}. These merchants elected to pursue their claims
individually, despite the fact that the value of those claims was substantially
limited by the mandatory release from the non-opt-out class binding every
merchant that remained in business. Indeed, the volume of opt-outs was so great
that it gave defendants the option to jettison the settlement entirely. SPA124-26,
190 (Settlement 17-20, 97). They declined to do so, however, preserving their
mandatory perpetual release.
The district court recognized that the motion for final approval caused a
rift among large United States retailers, and showed that divisions among the
major merchants run deep. SPA23. It noted that ten of the top twenty-five
convenience stores objected, and that many merchants regarded the (b)(2) relief as
essentially valueless to them. See, e.g., SPA24, 36, 38-43. Other merchants, such
as major airlines, opted out, believing they could obtain a larger cash recovery on
their own, but did not object because they apparently see value in the (b)(2)
relief. SPA23.
20
Case: 12-4671
IV.
Document: 983
Page: 32
06/16/2014
1249672
105
The Objections And Their Rejection By The District Court

Merchants and their representatives submitted four relevant categories of
objections to the district court, which rejected each of them.

A.
Objections To The Release Of Monetary Claims
Many merchants objected that the mandatory release imposed on the (b)(2)
class violated the Due Process Clause and Rule 23, both of which prohibit a classaction judgment from resolving claims for individualized relief, including claims
for money damages, without providing an opt-out right. See, e.g., JA[__]{DE2591
(Home Depot Obj. 15-30); DE2613 (1001 Property Solutions Obj. 5-10); DE2670
(Objecting Plaintiffs Obj. 21-24); DE2495-1 (Target Obj. 7-17); see also DE2427
(First Data Obj. 9-17)}. The objectors stressed that the unanimous portion of the
Supreme Courts recent decision in Dukes, 131 S. Ct. at 2558, as well as the
Courts prior opinion in Shutts, 472 U.S. at 797, held that individualized monetary
claims could not be resolved through a mandatory (b)(2) class.
The district court addressed these central objections in only one brief
paragraph, holding that [t]here is no due process right to opt out of the (b)(2)
class because [t]he (b)(2) settlement here is limited to going-forward injunctive
relief that changes the structure of the network practices. SPA46. Limiting its
analysis to the relief members of the non-opt-out class obtained, the court did not
address the far broader collection of individualized, monetary claims extinguished
21
Case: 12-4671
Document: 983
Page: 33
06/16/2014
1249672
105
by the (b)(2) release. See supra, at 15-17. The court also suggested, without
citation, that the mandatory (b)(2) class was lawful because it helped to ensure
litigation peace. SPA44.
B.
Cohesion Objections
Numerous appellants objected that the far-flung mandatory (b)(2) class

lacked sufficient cohesion. They explained that merchants did not share a common
interest in the Complaints allegations, as demonstrated by their unequal ability to
make use of the limited surcharging relief granted to the (b)(2) class. Further, they
had significantly varying interests in the many claims the class was required to
release. See, e.g., JA[__]{DE2591 (Home Depot Obj. 3-4, 19-27); DE2670
(Objecting Plaintiffs Obj. 24-27)}. For example, merchants that did not yet exist
were bound by the settlement and therefore required to release their claims against
defendants. But they received none of the cash settlement because they were
excluded from the (b)(3) class (since they were not in business before November
28, 2012). See, e.g., JA[__]{DE2591 (Home Depot Obj. 21); DE2670 (Objecting
Plaintiffs Obj. 38-39); see also DE2670-8 (Ex. 68, 16-17)}.
Objectors also explained that the class lacked cohesion because merchants
have significantly varying interests in obtaining relief against the no-surcharging
rule. For example, merchants located in ten states and Puerto Rico are prohibited
as a matter of law from engaging in surcharging. See SPA215-32 (state laws
22
Case: 12-4671
Document: 983
Page: 34
06/16/2014
1249672
105
barring surcharging). Also, because the settlements surcharging provisions

contain a most-favored-nation clause, merchants that accept American Express
(which effectively prohibits surcharging) cannot surcharge Visa and MasterCard
under the settlement. SPA41, 141-44, 154-57 (Settlement 42(a), 55(a));
JA[__]{DE5965 at 41-42}(court-appointed expert concluding that approximately
ninety percent of the (b)(2) class by volume would be unable to surcharge for this
reason). The district court itself acknowledged that, for these reasons, most
merchants will, as a practical matter, be precluded from surcharging Visa and
MasterCard products. SPA41.
For those merchants not subject to the foregoing blanket prohibitions, the
settlement still limits how they may surcharge. If a merchant wants to surcharge
Visa or MasterCard transactions, for example, the merchant must add the same
surcharge to all such transactions regardless of the cards issuer or product type.
SPA141, 154 (Settlement 42(a)(i), 55(a)(i)). This maintains the restraint on
inter-bank competition that was the principal target of this suit. And other
conditions further limit merchants ability to surcharge or explain to consumers the
defendants role in the higher prices they pay. See SPA148, 161-62 (Settlement
42(c)(iii)-(iv), 55(c)(iii)-(iv)).
Ajaypal Banga, MasterCards CEO, revealed that MasterCard insisted on
these restrictions to minimize any impact from surcharging:
23
Case: 12-4671
Document: 983
Page: 35
06/16/2014
1249672
105
We believe the best thing to do was looking at our experience of

surcharging in other markets where, frankly, it didnt really lead to a
great deal of actual surcharges being placed other than in a couple of
kinds of areas where cash isnt quite able to compete. So for example,
online airline bookings and the like. So when I think about that
here, in this agreement, we have also managed to get in some of those
protections, the declaration to the consumer with clarity, both on
the receipt and in the store, the level playing field concept that we
think weve got in there. All these were attempted as a way to sort of
try and box the issue while moving forward.... Thats kind of how I
approached it . And so it is friction. I dont like the friction but Im
trying to minimize it with as much lubricant as I can put in the system.
JA[__]{DE2670-8 (Ex. 94 (p. 370))}(emphasis added).
Many objectors also voiced unique concerns that had not been addressed in
any way by the settlement. For example, certain health insurers objected, noting
that the Affordable Care Act raised special regulatory concerns with interchange
fees and surcharging. JA[__]{DE2493-1; DE2643 (WellPoint and Blue Cross
Objections)}. A host of objectors noted that their individual circumstances made
surcharging relief valueless to them, or otherwise affected their perspective on the
mix of relief the case should have pursued. See, e.g., JA[__]{DE2411 (Boscovs
Obj. 3-4); DE2434 (Davids Bridal Obj. 9-23); DE2446 (Carters Obj. 718); DE2540 (Wawa Obj. 2-4); DE2458 (IKEA Obj. 16-33); DE2437 (Lowes
Obj. 14-29); DE2450 (Alon Obj. 15-33); DE2561 (NACS Obj. 23-37);
DE2644 (Wal-Mart Obj. 12-39); DE4640 (SIGMA Obj. 12-24)}.
Conceding that some of these concerns had not even been considered in the
negotiations, see, e.g., SPA48, the district court nonetheless addressed them only
24
Case: 12-4671
Document: 983
Page: 36
06/16/2014
1249672
105
briefly, concluding that the (b)(2) class was sufficiently cohesive because [t]he
network rules regimes that gave rise to this case applied generally to every
merchant accepting Visa or MasterCard credit cards, and the injunctive relief in the
proposed settlement does as well. SPA51. Although the court did address
arguments that surcharging relief was valueless to all, it did not address whether
the relief on that claim had different value to different class members. Nor did it
focus on class members different valuations of the claims the settlement released.
C.
Objections To Adequacy Of Representation
Objectors also raised the settlements failure to provide adequate

representation to the entire class. Numerous objectors explained that binding
future merchants that could not participate in the litigation and had no separate
representation could not be reconciled with Supreme Court precedent. See, e.g.,
JA[__]{DE2670 (Objecting Plaintiffs Obj. 28-36, 38-39); DE2592 (Dell Obj. 15);
DE2281 (Retailers and Merchants Obj. 11-20); DE3074 (Bridgestone Obj. 5-6);
DE4237 (Williams-Sonoma Obj. 5)}.
The district court did not analyze these objections with any particularity,
concluding merely that the interests of the Class Plaintiffs and the rest of the
(b)(2) class are not antagonistic. SPA52. The court did not address conflicts
created by the settlements release of claims by generations of future merchants
including merchants that will start operating after the structural changes in the
25
Case: 12-4671
Document: 983
Page: 37
06/16/2014
1249672
105
settlement sunset in 2021. Like the settling parties, the district court never
explained how it was permissible to create two separate classes without providing
each with separate representatives and separate counsel.
D.
Objections To The Scope Of The Release
Class members also objected to the scope of the release because, among
other things, it immunizes Visa and MasterCard from merchant lawsuits with
respect to all existing rules and policies and future versions thereof that are
substantially similar. Objectors emphasized that releasing ongoing and future
claims against default interchange and Honor-All-Cards would cement defendants
substantial market power. See JA[__]{DE2605 (Amazon.com Obj. 12); DE2444
(Amtrak Obj. 7-8, 24); DE2439 (Roundys Supermarkets Obj. 20); DE2451
(Barnes & Noble Obj. 23); DE2670 (Objecting Plaintiffs Obj. 28-36)}. They
also argued that it would protect defendants against competition from new payment
methods, such as payments from mobile devices. See, e.g., JA[__]{DE2279 (City
of Oakland Obj. 17); DE2598 (Consumers Union Obj. 8); DE2361 (U.S. PIRG
Obj. 5); DE2364 (Jo-Ann Stores Obj. 2); DE2435 (Dillards Obj. 27); DE2670
(Objecting Plaintiffs Obj. 33)}. Professor Sykes, the court-appointed expert,
echoed these concerns, stating that a release covering the future effects of all
existing or substantially similar conduct or rules raises a danger of adverse,
26
Case: 12-4671
Document: 983
Page: 38
06/16/2014
1249672
105
unintended consequences in a technologically dynamic industry, consequences that

are inevitably somewhat speculative at this time. JA[__]{DE5965 at 50}.
Visa and MasterCard confirmed these fears at the final fairness hearing.
Even though the Complaint did not concern mobile payments or emerging
technologies, they asserted that the settlement required merchants to release any
claim concerning the application of Visas or MasterCards Honor-All-Cards rules
to such technology. JA[__]{Corrected 9/12/13 Tr. 39}(A mobile phone
transaction, in my judgment, is clearly released.). The district court itself
expressed concern about this issue at the hearing.5 But its decision approving the
settlement was silent on the issue. The district court deemed it sufficient that the
settlement does not release the defendants from liability for claims based on new
rules or new conduct, and is therefore limited to claims that are or could have
been alleged on the identical factual predicate of the claims in this case. SPA46.
The court also held that immunizing defendants against future antitrust challenges
based on all present and substantially similar future conduct was permissible
because such conduct is not clearly illegal. SPA45-47.
See JA[__]{Corrected 9/12/13 Tr. 32}(I have a well-grounded concern

here that this release places the line of scrimmage in that future dispute as an
antitrust claim thats based on the application of those rules to a new technology,
places that line of scrimmage in the wrong spot.).
27
Case: 12-4671
Document: 983
Page: 39
06/16/2014
1249672
105
STANDARD OF REVIEW
Class certification and the approval of class-action settlements are generally
reviewed for abuse of discretion. Literary Works, 654 F.3d at 249; Charron v.
Wiener, 731 F.3d 241, 247 (2d Cir. 2013). However, this Court reviews the
decision de novo when, as here, the validity of the settlement rests on the
determination of novel issues of law. In re Masters Mates & Pilots Pension
Plan & IRAP Litig., 957 F.2d 1020, 1026 (2d Cir. 1992); see also Gerber v. MTC
Elec. Techs. Co., 329 F.3d 297, 302 (2d Cir. 2003). Moreover, where certification
rests on an error of law, the district court necessarily abuses its discretion.
Charron, 731 F.3d at 247.
SUMMARY OF ARGUMENT
The central feature of this settlement is the certification of a mandatory Rule
23(b)(2) class that is forced to release all claims against defendants ongoing and
future conduct. This structure was designed to permit a single set of class
representatives and counsel to provide defendants with a global, prospective
immunity from suitincluding suits for money damagesin exchange for a
substantial cash payment. Class members could opt out of receiving the money
(which went to the (b)(3) class), but could not save their claims from the allencompassing, forward-looking release (which came from the mandatory (b)(2)
class). This feature violates four separate doctrines designed to protect absent and
28
Case: 12-4671
Document: 983
Page: 40
06/16/2014
1249672
105
objecting class members from being bound against their will to a settlement that
benefits others at their expense.
First, this settlement expressly terminates the individualized monetary
claims of all the (b)(2) class members with no opt-out right. See SPA169
(Settlement 68). Indeed, the settlement pays billions of dollars to the (b)(3) class
on the exact same monetary claimsclaims that are distinguished only by the date
on which the damages accrue. Shutts held that such claims belong to individual
class members under the Due Process Clause and must be protected by the right to
opt out. Dukes unanimously held that Rule 23 channels all such claims to opt-out
classes certified under Rule 23(b)(3).
The district court nonetheless approved the settlement on the theory that the
relief the (b)(2) class obtained did not include money damages. But what matters
are the claims that are resolved by the settlementin particular, the claims that the
class has been forced to relinquish. It does not matter that the (b)(2) class received
no money, or that the defendants insisted on litigation peace. SPA44. The
courts certification of a mandatory (b)(2) class extinguishing individualized
monetary claims violated the Due Process Clause and Rule 23.
Second, the (b)(2) class was not cohesive, particularly under the heightened
standard that applies to mandatory classes. The (b)(2) class is massive, consisting
of millions of existing merchants of every possible variety and many millions more
29
Case: 12-4671
Document: 983
Page: 41
06/16/2014
1249672
105
that do not even exist yet. More than that, the (b)(2) class settlement resolves not
just one claim, nor just the claims in the Complaint, but essentially every possible
merchant claim against defendants existing rules and practices (and those that are
substantially similar) now and indefinitely into the future. Such a sprawling class,
resolving such a broad swath of claims, cannot be expected to bargain together for
a single, indivisible injunction benefitting all the members at once, as Rule
23(b)(2) requires. See Dukes, 131 S. Ct. at 2558. The best possible proof of that is
the deal that emerged: In exchange for releasing every other claimincluding the
claims that mattered most to many of the class membersthe (b)(2) class got relief
only on surcharging, even though class members in ten states are forbidden from
surcharging by law.
Third, the (b)(2) class was inadequately represented. The (b)(2) class was
limited to prospective relief and included millions of members (including recently
founded merchants, future businesses, and (b)(3) opt-outs) that had no interest in
the (b)(3) monetary relief at all. But the (b)(2) class had no lawyer and no class
representative whose role was solely to represent its predominantly future-looking
interests. Instead, both classes were represented by the same counsel and
representatives, who could not get their pecuniary reward through the (b)(3)
settlement without providing defendants with the global release they wanted from
the (b)(2) class.
30
Case: 12-4671
Document: 983
Page: 42
06/16/2014
1249672
105
This representation was structurally inadequate. As an initial matter, classes

with divergent interests need their own counsel and representatives. That is
especially so where, as here, parties cannot opt out and obtain their own
representation, even when they know that their putative representatives are not
protecting their interests. And it is triply true here, where there is a wellunderstood conflict of interest between the predominantly future-looking (b)(2)
class and the predominantly backwards-looking (b)(3) class. Class representatives
who can obtain immediate monetary relief have a recognized incentive to trade
away future-looking interests in return for more money now. The settling parties
decision to structure their deal to create past- and future-looking classes, while
providing those classes with no independent representation, thus embodies an
egregious and obvious violation of settled class-action precedents. See, e.g.,
Ortiz, 527 U.S. at 853, 856.
Finally, the (b)(2) release in this caseextinguishing essentially every
present and future challenge to defendants existing, and substantially similar,
practicesexceeds the permissible scope of class-action litigation. It
prospectively releases future conduct from antitrust attack, a result the Supreme
Court has condemned. See, e.g., Am. Express Co. v. Italian Colors Rest., 133 S.
Ct. 2304, 2310 (2013); Lawlor v. Natl Screen Serv. Corp., 349 U.S. 322, 328-29
(1955). It also extinguishes claims well beyond the scope of the Complaint,
31
Case: 12-4671
Document: 983
Page: 43
06/16/2014
1249672
105
including future claims that are not yet ripe and present claims that are utterly
unrelated to the facts at issue. Ultimately, the settlement represents an effort to use
class-action litigation to structure a regulatory solution for an entire industry, much
like the settlement Judge Chin rejected in Authors Guild v. Google Inc., 770 F.
Supp. 2d 666, 669 (S.D.N.Y. 2011) (Google Books). This is not the proper role
of federal litigation; Congress provided these appellants with a cause of action
under the Sherman Act, and they should be allowed to vindicate it as they see fit.
ARGUMENT
I.
The District Courts Judgment Impermissibly Extinguishes Class

Members Individualized Claims For Money Damages Without
Providing Opt-Out Rights.
A.
Both the Due Process Clause and Rule 23 mandate that class
members have the right to opt out and pursue their individualized
legal claims.
The Fifth Amendment prohibits the federal government from depriving

persons of their property without due process of law. U.S. Const. amend. V.
SPA207. That prohibition governs the entry of a judgment resolving a claim in
litigation. Because the claima chose in actionis a species of property
protected by the Due Process Clause, Logan v. Zimmerman Brush Co., 455
U.S. 422, 428 (1982), the individuals right to pursue the claim is a
constitutionally recognized property interest, Shutts, 472 U.S. at 807.
On that basis, the Supreme Court held in Shutts that if a court wishes to
bind an absent plaintiff concerning a claim for money damages or similar relief at
32
Case: 12-4671
Document: 983
Page: 44
06/16/2014
1249672
105
law, it must provide minimal procedural due process protection, including not
only the best-practicable notice but alsocriticallyan opportunity to remove
himself from the class. Id. at 811-12 (emphasis added). The right to object to the
settlement is not enough: Class members must have the right to remove
themselves from the judgment and pursue their claims on their own.
In the thirty years since Shutts, the Supreme Court has not once approved a
class-action judgment that purported to resolve individualized legal claims without
affording class members the right to opt out and pursue their personal claims as
they saw fit. Rather, the Court has reaffirmed that mandatory class actions
aggregating damages claims implicate the due process principle deep-rooted [in
our] historic tradition that everyone should have his own day in court. Ortiz, 527
U.S. at 846.
Federal Rule of Civil Procedure 23(b) embodies the same principles,
authorizing a non-opt-out, (b)(2) class only in unique circumstances where no
claims for individualized relief, such as individualized award[s] of monetary
damages, are at issue. Dukes, 131 S. Ct. at 2557. Instead, individualized
monetary claims belong in Rule 23(b)(3), the separate provision of the Rule that
guarantees absent class members the right to opt out. Id. at 2558. Under Rule 23,
class members individualized claims cannot be precluded by litigation they had
no power to hold themselves apart from. Id. at 2559. Instead, plaintiffs with
33
Case: 12-4671
Document: 983
Page: 45
06/16/2014
1249672
105
individual monetary claims [must] decide for themselves whether to tie their fates
to the class representatives or go it alonea choice Rule 23(b)(2) does not ensure
that they have. Id.
Indeed, as the Supreme Court explained, permitting a judgment to bind
members of a (b)(2) class with respect to their individualized monetary claims
would be inconsistent with the structure of Rule 23(b). Id. at 2558. Subsection
(b)(3) is designed for individualized legal claims in which class members may
have distinct interests. Accordingly, Rule 23(b)(3) permits class litigation
controlled by a representative only if common questions predominate over any
questions affecting only individual members and the class action is superior to
individual adjudication. Because those standards do permit the aggregation of
some individualized claims, Rule 23(b)(3) guarantees class members notice and the
opportunity to opt out. See Fed. R. Civ. P. 23(c)(2)(B).
By contrast, subsection (b)(2) contemplates a judgment binding the entire
class without notice and opt-out rights, and without regard to whether common
questions predominate, because it applies only when the case consists exclusively
of common claims in which the class has a single, indivisible interest. This
provision is never appropriate with respect to a class members individualized
claim for money. Dukes, 131 S. Ct. at 2558-59. In the Supreme Courts words,
[t]he key to the (b)(2) class is the indivisible nature of the injunctive or
34
Case: 12-4671
Document: 983
Page: 46
06/16/2014
1249672
105
declaratory remedy warrantedthe notion that the conduct is such that it can be
enjoined or declared unlawful only as to all of the class members or as to none of
them. Id. at 2257.
The prime examples of such situations are [c]ivil rights cases against
parties charged with unlawful, class-based discrimination. Amchem, 521 U.S. at
614. But Rule 23(b)(2) treatment is not even available for every claim seeking
only injunctive relief: Rule 23(b)(2) applies only when a single injunction or
declaratory judgment would provide relief to each member of the class, and is
limited to cases where the relief sought must perforce affect the entire class at
once. Dukes, 131 S. Ct. at 2557-58.
If counsel drafts a class complaint to include a truly common injunctive
claim alongside individualized legal claims, that of course does not strip class
members of their right to pursue the latter individually. The right to opt out cannot
be nullified whenever a plaintiff class, at its option, combines its monetary claims
with a requesteven a predominating requestfor an injunction. Id. at 2559.
If that were permissible, individual class members compensatory-damages
claims would be precluded by litigation they had no power to hold themselves
apart from. Id.6
In this regard, Dukes abrogated this Courts cases allowing monetary

claims to be certified in mandatory (b)(2) classes as long as injunctive claims
35
Case: 12-4671
Document: 983
Page: 47
06/16/2014
1249672
105
Applying these principles, this Court has consistently disapproved classaction judgments that purport to resolve individualized claims of class members
who did not get a chance to opt out. This Court has done so where the damages
claim already existed at the time the court entered the class-action judgment (as in
Shutts), if individual class members received inadequate notice of their opt-out
right. See Hecht, 691 F.3d at 222-23. It has also done so where the opt-out right
was ineffective because the precluded claim arose only after the court entered the
class-action judgment (as in Ortiz). See Stephenson v. Dow Chem. Co., 273 F.3d
249 (2d Cir. 2001), vacated in part on other grounds, 539 U.S. 111 (2003).
The district court identified no case approving the release of past, present, or
future individualized claimsespecially compensatory damages claimswithout
allowing class members to opt out. The precedents cited by the settling parties
below to justify using Rule 23(b)(2) to deprive merchants of their opt-out rights
only highlight that this settlement is unprecedented.
The most analogous decision, Visa Check, approved the certification of a
class of merchants under only Rule 23(b)(3), precisely to avoid the primary
concern about Rule 23(b)(2), i.e., the absence of mandatory notice and optout rights. Visa Check, 280 F.3d at 147 (Sotomayor, J.) (citing Jefferson v.
predominated. See Hecht v. United Collection Bureau, Inc., 691 F.3d 218, 22223 (2d Cir. 2012).
36
Case: 12-4671
Document: 983
Page: 48
06/16/2014
1249672
105
Ingersoll Intl Inc., 195 F.3d 894, 897 (7th Cir. 1999)). The recent settlement in
Literary Works released the defendants from future litigation over subsequent use
of the copyrighted works at issue. But, critically, it permitted class members to (1)
opt out of the settlement entirely or (2) opt out of the release for future use in
particular. 654 F.3d at 246-47. Even in the civil-rights cases at the core of Rule
23(b)(2), settlements in this Circuit have been carefully scrutinized to preserve
class members rights to pursue any individualized claims that might arise from the
defendants ongoing conduct. See, e.g., Joel A. v. Giuliani, 218 F.3d 132, 142 (2d
Cir. 2000) (settlement preserved the right of an individual plaintiff to sue for
damages or equitable relief tailored solely to the specific circumstances of that
individual plaintiff) (internal citation and quotations omitted); Charron, 731 F.3d
at 252 (noting that while (b)(2) settlement provided no relief on certain monetary
claims, it also d[id] not extinguish them).7
The only possible exception, TBK Partners, Ltd. v. Western Union Corp.,
675 F.2d 456 (2d Cir. 1982), predates the opt-out right announced in Shutts, and
the parties in that case did not dispute whether the class had been improperly
certified as a non-opt-out class. Id. at 460 n.4.
37
Case: 12-4671
B.
Document: 983
Page: 49
06/16/2014
1249672
105
The certification of the (b)(2) settlement class must be vacated

because it extinguishes merchants individualized legal claims
without providing an opt-out right.
1.
The settlement, on its face, releases individualized monetary

claims.
The judgment in this case presents precisely the scenario anticipatedand

unanimously forbiddenby the Supreme Court in Dukes. It disposes of every
class members claims whether individual or otherwise in nature, for any form
of damages or other monetary relief relating to the period after [November 28,
2012], regardless of when such claims accrue, in law or in equity. SPA169
(Settlement 68) (emphasis added); SPA90 (Judgment 16(c)). Indeed, it does so
in the most extreme way possible: It releases such claims entirely and for all time,
with no changes to the ongoing conduct that precipitated this case aside from
limited surcharging relief.
The settlement and release terminate the (b)(2) class members rights to
recover damages by artificially splitting the damages claims pertaining to
defendants ongoing conduct, permitting class members to pursue individually
only the subset of damages that accrued before November 28, 2012 (if they opted
out of the (b)(3) class), and forever extinguishing any right to recover the damages
they subsequently suffer from the same conduct. That release applies indefinitely
into the future, no matter how great the damages merchants incur; indeed, it applies
even if circumstances change and seriously exacerbate the anticompetitive effects
38
Case: 12-4671
Document: 983
Page: 50
06/16/2014
1249672
105
of the released rules and practices or their impact on individual class members.
SPA171 (Settlement 68(g)-(h)).
No one can foresee how the payment card industry will evolve, how
defendants present or future conduct might harm future competition, or how
heavily the economic harms might fall on particular class members given their
particular market circumstances. But under this settlement, it does not matter: In
direct contravention of Dukes, the settlement releases all those legal claims without
regard to individual merchants desires to preserve them for themselves. See 131
S. Ct. at 2557 ([A]t a minimum, claims for individualized relief do not satisfy
Rule 23(b)(2).).
2.
The settlement has the effect of releasing a host of

individualized legal claims.
It is easy to illustrate that the release extinguishes individualized legal

claims, not injunctive claims common to the class as a whole. This case looks
nothing like the civil-rights suits classically resolved under Rule 23(b)(2). The
released claims are individualized and monetarywhether the damages associated
with the challenged practices accrued in the past, are accruing today, or will accrue
in the futurebecause the entire dispute is over whether defendants practices
restrain competition and thereby raise prices or otherwise take dollars out of
merchants pockets. As the district court found, supracompetitive interchange
39
Case: 12-4671
Document: 983
Page: 51
06/16/2014
1249672
105
fees are the precise anticompetitive effect the claims here were brought to
challenge. SPA42.
In fact, the character of the (b)(2) classs claims is evident from the relief
provided to the separate (b)(3) class. While the relief the settlement provides to the
(b)(3) and (b)(2) classes is different (several billion dollars to the former; minor
rule changes to the latter), the claims that are settled on behalf of those classes are
identicaldistinguished only by the date on which the damages accrue.
The released claims themselves also illustrate the point. Take the HonorAll-Cards rules. See supra, at 9. Under this Courts holding in United States v.
Visa, 344 F.3d at 229, merchants allege that those rules are unlawful horizontal
restraints preventing competition among banks for merchant acceptance of their
cards. The harm that merchants suffer from those rules is the inflated fees they
pay, giving rise to a classic money damages claim under the Sherman and Clayton
Acts. JA[__] (Complaint 292-312, 371-84, 409-15, 443-56). The same is true of
default interchange rates, which plaintiffs have attacked as price fixing. Id. Yet,
going forward, the settlement expressly extinguishes all such claims. SPA169-72
(Settlement 68).
Also illustrative is the settlements release of claims regarding the Fixed
Acquirer Network Fee, which Visa charges merchants for attaching to its network.
The FANF notably was under investigation by the Justice Department at the time
40
Case: 12-4671
Document: 983
Page: 52
06/16/2014
1249672
105
of the settlement. See JA[__]{DE2670 (Objecting Plaintiffs Obj. 34 n.43);

DE2670-8 (Ex. 82 (p. 300))}. Although the settlement allows individuals to seek
future injunctive relief against the FANF, it extinguishes every merchants
individualized monetary claim for post-November 28, 2012 damages caused by
this practice, even if the merchant later prevails in proving that the fee is unlawful
and caused significant monetary harm. SPA93 (Judgment 16(g)(iv)); SPA174
(Settlement 72(d)). This is exactly the opposite of what Rule 23(b)(2) allows.
The structure of this settlement, if approved by this Court, would thus
eviscerate the opt-out right that Rule 23 protects. As Rule 23 has been consistently
construed, an individual considering whether to opt out from a (b)(3) class will
expect that, if she prevails in her individual suit, she will secure monetary relief
extending to the date of the judgment, as well as an injunction protecting her from
future injury. But under the model of this settlement, as of the date on which the
(b)(2) class is defined, opt-out claimants cannot recover for their ongoing damages
or obtain an injunction to prevent future harm.
The implications of approving such a mandatory class release are sweeping,
and startling. If this settlement is affirmed, virtually any class action implicating
ongoing conduct may be split into a backwards-looking (b)(3) class and a
mandatory, forward-looking (b)(2) class. Lead class-action plaintiffs will always
have an incentive to take this step because it provides them an enormous benefit to
41
Case: 12-4671
Document: 983
Page: 53
06/16/2014
1249672
105
offer the defendantscategorical immunity from civil liability to every class

member (including class members that do not even exist yet) for harmful conduct
the defendants want to continue. Other parties are already adopting this model as a
template for their settlement-only class actions. See, e.g., JA[__] (AmEx
Settlement). Unless this Court reverses the judgment below, this innovation will
almost certainly become the next stock device in the world of class-action
litigation and settlement. Cf. Amchem, 521 U.S. at 618.
Not surprisingly, precedents regarding claims for money damages that will
arise in the future have treated them as claims for legal relief to which the
procedural protections of Rule 23 and the Due Process Clause fully apply. In
particular, Ortiz regards the termination of monetary claims as a serious dueprocess problem even though a large segment of the disputed claims were by
future claimants who had no claim for damages at the time of the settlement.
Ortiz, 527 U.S. at 846.
In Stephenson, this Court likewise recognized that the Due Process Clause
does not permit a class-action judgment to release future damages claims without
adequate representation and an opportunity to opt out. 273 F.3d at 260 (citing
Shutts, 472 U.S. at 811-12). Stephenson thus refused to preclude a later-arising
damages claim for particular plaintiffs based on their absentee class-membership in
an earlier global settlement because, among other things, the plaintiffs likely
42
Case: 12-4671
Document: 983
Page: 54
06/16/2014
1249672
105
received inadequate notice of the class action and thus inadequate opportunity to
opt out. Id. at 261 n.8 (citing Shutts, 472 U.S. at 812). And that was true despite
the fact that the court that initially approved the Stephenson settlement was
virtually certain that these contingent monetary claims would not arise. Id. at 261.
These holdings confirm the commonsense point that claims accurately
described as claims for money damages that arise in the future are, of course,
claims for money damages and thus a species of individualized legal claim for
purposes of the Due Process Clause and the unanimous holding in Dukes. Indeed,
the fact that the settlement releases future damages creates an unavoidable
constitutional dilemma. As the Supreme Court has recognized, such releases
impermissibly negate the one process the Constitution has expressly provided for
the resolution of individualized legal claims: the jury trial. See Ortiz, 527 U.S. at
846 (By its nature, a mandatory settlement-only class action with legal issues
and future claimants compromises their Seventh Amendment rights without their
consent.) (emphasis added). The same is true, of course, for present objectors that
are actively withholding their consent and trying to preserve their jury trial rights.
43
Case: 12-4671
C.
Document: 983
Page: 55
06/16/2014
1249672
105
There is no legitimate justification for denying merchants opt-out

rights.
1.
There is no merit to the district courts theory that a non-optout class was permissible because the settlement provides its
members no monetary relief.
Although appellants argued at length below that the Due Process Clause and
Rule 23 guaranteed them the right to opt out of this settlement, the district court
dealt with those arguments in only one paragraph. The court held as a matter of
law that there is no due process right to opt out of the (b)(2) class because the
(b)(2) settlement here is limited to going-forward injunctive relief that changes
the structure of the networks practices. SPA46. In other words, the court
determined no opt-out right was necessary because the relief provided to the nonopt-out class was injunctive, rather than monetary. That reasoning fails for two
reasons.
First, whatever the character of the settlements relief to the (b)(2) class, it
still extinguishes individualized claimsincluding claims for money damages
with no opt-out right. Even the district court effectively acknowledged that the
released claims are inherently individualized. See, e.g., SPA52 (noting judicial
relief on interchange fee claims would affect the class unequally). The Due
Process Clause and Rule 23 guarantee the right to opt out with respect to the
resolution of those individualized claims, even if the class gets no relief. Indeed,
that right is especially important if the class has agreed to take nothing in exchange
44
Case: 12-4671
Document: 983
Page: 56
06/16/2014
1249672
105
for forever forfeiting its monetary claims. In such a case, a (b)(2) class has been
impermissibly certified with respect to the class members monetary claims, and
those claims have been precluded by litigation they had no power to hold
themselves apart fromthe exact thing Dukes forbids. See 131 S. Ct. at 2559.
The district courts analysis ignores altogether the monetary claims subject
to the mandatory release. The court noted that [t]o allow [merchants] to opt out
and pursue their own rules-based injunctive relief would eliminate the incentive to
settle that Rule 23(b)(2) was designed in part to create. SPA46 (emphasis added).
But the release extends well beyond claims for rules-based injunctive relief: It
also bars class members from pursuing compensation for any monetary injuries
those practicesor any others in the voluminous rulebookshave caused or will
cause at any point after November 28, 2012. The settlement thus expressly
releases all manner of claims whether individual or otherwise in nature, for
any damages or other monetary relief. SPA169 (Settlement 68); SPA90
(Judgment 16(c)).
Second, this (b)(2) class was an unnecessary and artificial contrivance that
inverted the design of Rule 23. Echoing Rule 23s drafters, the Supreme Court has
made clear that subsections (b)(1) and (b)(2) were intended to reflect existing,
standard practices in collective litigation, and that adventuresome innovations
were confined to Rule 23(b)(3). See, e.g., Amchem, 521 U.S. at 615; Benjamin
45
Case: 12-4671
Document: 983
Page: 57
06/16/2014
1249672
105
Kaplan, Continuing Work of the Civil Committee: 1966 Amendments of the

Federal Rules of Civil Procedure, 81 Harv. L. Rev. 356, 394 (1967). This
settlement, which splits a unitary claim in order to create a future-looking,
injunctive componentamenable to a global release for all time by a non-optout classis anything but standard. It thus belongs in Rule 23(b)(3). Indeed,
because the relief provided to (b)(3) classes frequently includes injunctions, there
is no need for the adventuresome innovation in claim-splitting this case seeks to
inaugurate. See, e.g., Wal-Mart, 396 F.3d at 112-13 (approving substantial
forward-looking relief in class action certified under Rule 23(b)(3)); Literary
Works, 654 F.3d at 249 (same).
In fact, the approval of this settlement would give Rule 23(b)(2) an entirely
new and dangerous function. The only possible purpose of including damages
claims against ongoing and future conduct in a (b)(2) settlement is to extinguish
them; attempting to dole out individualized damages to a (b)(2) class would be an
even more obvious violation of Dukes. Prohibiting the inclusion of such claims in
mandatory class settlements thus provides the only check against a very dubious
practice: allowing the settling plaintiffs to confer on defendants the right to injure
other class members in the future through conduct those other class members
would attack as unlawful if only they had the right to opt out and litigate on their
own. And that, in fact, is the central feature of this settlement: It enables Visa and
46
Case: 12-4671
Document: 983
Page: 58
06/16/2014
1249672
105
MasterCard to pay a feethe payment to the members of the (b)(3) classin

exchange for the unfettered right to continue practices that the appellants argue are
against the law, without any threat of future suit from any merchant.
2.
Rule 23(b)(2) does not allow the certification of monetary

claims that arise in the future for the purpose of creating
litigation peace.
Ultimately, the district court recognized that extinguishing all possible future
claims against defendants ongoing conduct was the sine qua non of the settlement.
It nonetheless believed that the ongoing and future damages claims of absent and
objecting merchants could be mandatorily sacrificed because it was essential to
providing defendants the litigation peace they legitimately expect[ed] in return for
the settlement of claims. SPA44. But global peace is not a prize that can be
bought over the objection of class members who prefer to preserve their
individualized legal claims for themselves.
Indeed, the Supreme Court has made perfectly clear thatcontrary to the
district courts suggestionthe mandatory sections of Rule 23 do not exist to
vindicate class-action defendants interest in achieving forward-looking global
peace. The only interest in global settlement the Supreme Court has even
suggested might justify the mandatory release of monetary claims arises in limited
fund cases under Rule 23(b)(1), where the resolved legal claims would be
terminated anyway because the available monies to pay them would be exhausted
47
Case: 12-4671
Document: 983
Page: 59
06/16/2014
1249672
105
(as in a bankruptcy). See, e.g., Ortiz, 527 U.S. at 839.8 In such cases, terminating
monetary claims is arguably justified because class-wide resolution would give
the class as a whole the best deal but would not give a defendant a better deal
than seriatim litigation would have produced. Id. But these considerations have
no application here. The only global interest the mandatory release accomplishes
here is to give defendants a better deal than seriatim litigation would produce
exactly the opposite of the type of necessity that could justify forcing plaintiffs like
appellants to give up the claims that belong to them by constitutional right.
II.
The Mandatory Class Lacked The Required Cohesion Of Interests.

The settlement is also invalid for the independent reason that it improperly
bound together, in a mandatory (b)(2) class, millions of diverse merchants with

conflicting interests in both the one claim on which they were granted relief and
the vastly broader collection of claims that the settlement resolved.
A.
The greatest degree of cohesion is required for mandatory

settlement classes.
Cohesion denotes the overarching requirement that any class defined

under any provision of Rule 23 contain a set of plaintiffs with sufficiently similar
interests to permit representative litigation. See, e.g., Amchem, 521 U.S. at 622-23;
Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 165 (2d Cir. 2001).
8
Notwithstanding the judicial crisis of asbestos claims, Amchem and Ortiz

insisted that the procedural and due process protections of Rule 23 not yield to
claims of exigency. See Ortiz, 527 U.S. at 864; Amchem, 521 U.S. at 605.
48
Case: 12-4671
Document: 983
Page: 60
06/16/2014
1249672
105
The requirement arises primarily from the constitutional interests that are directly
implicated when class-wide representation displaces the right of individuals to
pursue their own interests. See Amchem, 521 U.S. at 621 (Subdivisions (a) and
(b) focus court attention on whether a proposed class has sufficient unity so that
absent members can fairly be bound by decisions of class representatives.);
Shutts, 472 U.S. at 812 ([T]he Due Process Clause of course requires that the
named plaintiff at all times adequately represent the interests of the absent class
members.).
A courts inquiry into whether the class is cohesive is accordingly at its most
rigorous where class members are to be bound under Rule 23(b)(2) with no right to
opt out. See, e.g., In re St. Jude Med., Inc., 425 F.3d 1116, 1121 (8th Cir. 2005)
(Because unnamed members are bound by the action without the opportunity to
opt out of a Rule 23(b)(2) class, even greater cohesiveness generally is required
than in a Rule 23(b)(3) class.).9 Indeed, Rule 23(b)(2) does not provide for notice
or opt-out rightsand the Due Process Clause permits such a regimeonly
because, in a properly certified (b)(2) class, the interests of all class members are
so aligned that there is essentially no reason for them to litigate on their own. See
Dukes, 131 S. Ct. at 2559.
9
See also Robinson, 267 F.3d at 165 (similar); Lemon v. Intl Union of
Operating Engrs, 216 F.3d 577, 580 (7th Cir. 2000) (similar); Barnes v. Am.
Tobacco Co., 161 F.3d 127, 142-43 (3d Cir. 1998) (similar).
49
Case: 12-4671
Document: 983
Page: 61
06/16/2014
1249672
105
As Dukes explained, Rule 23(b)(2) applies only when a single injunction

would provide relief to each member of the class; it does not authorize class
certification when each individual class member would be entitled to a different
injunction or declaratory judgment against the defendant. 131 S. Ct. at 2557.
Thus, both before and after Dukes, courts have rejected class certification under
Rule 23(b)(2) where the class would not derive an indivisible, common benefit
from the injunctive relief being pursued. See, e.g., M.D. ex rel. Stukenberg v.
Perry, 675 F.3d 832, 846 (5th Cir. 2012) (vacating (b)(2) certification order
because district court erred in finding it irrelevant that some of the classs
requested relief would not apply to every class member); Kartman v. State Farm
Mut. Auto. Ins. Co., 634 F.3d 883, 893 (7th Cir. 2011) (Where a class is not
cohesive such that a uniform remedy will not redress the injuries of all plaintiffs,
class certification is typically not appropriate.); Casa Orlando Apartments, Ltd. v.
Fed. Natl Mortgage Assn, 624 F.3d 185, 200 (5th Cir. 2010) ([F]orty percent of
the class benefiting from an injunction is not sufficient to certify under (b)(2).).
The fact that defendants practices affect all class members is accordingly
insufficient to render the class cohesive, even if class members have certain
complaints in common about those practices. If individual class members would
want to litigate and redress their claims in different waysparticularly because the
case will resolve multiple claims, and the class members differing interests in
50
Case: 12-4671
Document: 983
Page: 62
06/16/2014
1249672
105
those claims might be traded off against each otherthen those claims are not
suitable for (b)(2) treatment. Instead, they are, at best, the kind of common
questions that may be certified only under Rule 23(b)(3). See, e.g., Ortiz, 521
U.S. at 854-58 (noting cohesion problem in mandatory class-action settlement
aggregating different kinds of claims).
Among (b)(2) classes, settlement-only classes present the very greatest
cohesion concerns. Divisions in the interests of class members may become
apparent during the course of litigation. By contrast, when the district court is
presented with proposed class definitions only as a part of an already completed
negotiation that will resolve the entire caseand decisions about which claims to
pursue and how some might be sacrificed to secure relief on others have not been
subject to the scrutiny of individual class members during the litigationthose
divisions are more likely to be obscured. As Amchem emphasized, heightened
scrutiny is required because a court asked to certify a settlement class will lack
the opportunity, present when a case is litigated, to adjust the class, informed by
the proceedings as they unfold. 521 U.S. at 620.
Settlements also create a special risk of trading off class members claims
against each other. Consider the classic example of a Rule 23(b)(2) class of
plaintiffs in a civil-rights suit. A hypothetical challenge to a males-only
admissions policy at a single military college could be pursued by a non-opt-out
51
Case: 12-4671
Document: 983
Page: 63
06/16/2014
1249672
105
class of female applicants. But imagine a broader challenge to all the genderspecific practices in every branch of the military. Even if such a claim could be
litigated on a class-wide basis, it is highly doubtful such a suit could be settled
through a mandatory (b)(2) class action, given that any effort to resolve the case
would inevitably require trading off some of the plaintiffs claims for others. A
lack of cohesion would arise from the competing interests of some class members
who sought, for example, greater gender integration in the Coast Guard and Air
Force, which might be traded for a release of all claims regarding the Army or
Marines, or bargained for different outcomes with respect to medical and combat
personnel. Binding class members on a mandatory basis to such a settlement
violates both Rule 23(b)(2) and the Due Process Clause because it puts the
common class representatives in the position of trading away the interests of one
subset of the class in return for relief for a different group.10
B.
The merchants bound to the (b)(2) class in this case were too
diverse for a single, indivisible injunction, and the settlement does
not treat those class members equally.
In Amchem, the Supreme Court described the lack of cohesion of the

asbestos-victim class by saying: No settlement class called to our attention is as
10
The failure of the proposed class to survive this heightened attention does
not mean that no class action is possible. The most common solutions to cohesion
problems are to form subclasses with separate representation (e.g., Literary Works,
654 F.3d at 256), to provide class members with opt-out rights (e.g., Visa Check,
280 F.3d at 147), and/or to narrow the claims involved (or the release granted) so
as to bring the interests of the class closer together.
52
Case: 12-4671
Document: 983
Page: 64
06/16/2014
1249672
105
sprawling as this one. 521 U.S. at 624. This case proves that statement outdated.
The two defining features of this (b)(2) settlement are the breadth of the claims
involved and the breadth of the class assembled. Each alone is unprecedented;
together, they make for a manifestly non-cohesive class.
1.
Class members had varying interests in the broad set of claims

that the settlement purported to release.
This appears to be the broadest commercial class ever assembled. While

the precise size of the class [is] impossible to determine, Class Counsel estimate
that approximately 12 million merchants comprise the class. SPA23. But those
are only the merchants that existed as of the date of preliminary approval. It does
not count the tens of millions of future merchants that will come into existence
later. All those millions of merchants are likewise bound to the settlement, and
they include every imaginable type of merchant: anyone in the country who sells
any kind of thing in any kind of way or ever might sell anything you can think of
in any way you can conceiveso long as they accept credit cards, as almost every
merchant will. Indeed, since the release lasts forever, the range of merchants
captured by this class is endless.
Accordingly, even the district court acknowledged that members of the
(b)(2) class had different interests in one of the core issues in the casethe setting
of default interchange rates. As the court found, the claims asserted on behalf of
the mandatory class seek injunctive relief from the bundle of network rules that
53
Case: 12-4671
Document: 983
Page: 65
06/16/2014
1249672
105
result inaccording to the plaintiffs allegationssupracompetitive interchange

fees in violation of the antitrust laws. SPA46. And yet, as the court expressly
recognized, any judicial regulation of default interchange fees would affect the
class unequally. SPA52 (emphasis added). The court explained that members of
the class had unequal interests in obtaining such relief because default interchange
operates only in the absence of bilateral agreement, and some of the very large
merchants have sufficient transaction volume that they can actually negotiate for
their own, lower interchange structures. Id. Thus, even under the district courts
formulation, different members of the class would have differing interests in a
central claim in the case, depending on their size and business models.
Further, this caseat least as settledconcerned not just one of the
defendants practices, nor even several of their rules that relate to interchange fees
identified in the Complaint, but all of the express policies in defendants massive
rulebooks, their unwritten policies and practices, and any future rules, policies, or
practices that are substantially similar. As counsel for defendants stated at the
fairness hearing, the releases are designed to encompass all of the rules and
policies and conduct of the defendants to the extent they adversely affect
merchants that accept MasterCard and Visa. JA[__]{Corrected 9/12/13 Tr. 3738}; see also JA[__]{DE2670-8 (Ex. 66 (p. 43))}(Visa General Counsel stating to
54
Case: 12-4671
Document: 983
Page: 66
06/16/2014
1249672
105
investors that the release covers all of Visa and MasterCards rules in existence as
of the time of approval.).
This expands exponentially the lack of cohesion in this settlement-only
class: An already maximally diverse group of merchants will have even more
conflicting interests in determining which rules and practices harm them the most
and should be the subject of any negotiated relief. For that reason, these class
members would certainly not seek an indivisible bargain with respect to all those
claims at once. Dukes, 131 S. Ct. at 2558. The best evidence of that is the
bargain they got: The class representatives primarily secured relief on one claim
(surcharging) while forever abandoning every other claim of every class member
(such as the core challenges to Honor-All-Cards and default interchange).
Merchants diverse interests in the indescribably broad collection of other
claims released by the settlement (some of which have little or nothing to do with
interchange) are likewise illustrative. For example, the FANF rate varies with the
number of merchant locations, so large merchants would be more concerned about
that issue than small ones. JA[__]{DE2670-8 (Ex. 84 (pp. 318-21))}. Some
merchants operate in industries that are so competitive that surcharging is highly
unlikely. Some merchants may be well-suited to rolling out mobile-payment
technology, and would be much more concerned with releasing such claims than
their competitors. In a settlement that concerns every written and unwritten
55
Case: 12-4671
Document: 983
Page: 67
06/16/2014
1249672
105
network practice and a class of every conceivable type of merchant, this list goes
on and on and on.
In sum, the cohesion analysis must take account not only of the variations
among the members with respect to the relief they obtained, but also of every claim
they have given up, and the balance struck by the settlement of the case as a whole.
From that vantage, it is hard to imagine a class less cohesive than this one.
2.
Class members had varying interests in the one claim on which

limited relief was actually provided.
The variance among the interests of the (b)(2) class members is best
illustrated by the sole claim on which they obtained any material relief
surcharging. See supra, at 14-15, 22-23. There is no dispute that the class
members have conflicting interests in surcharging: The district court itself
acknowledged it.
The district court recognized that many merchants that do want to surcharge
cannotthis settlement notwithstanding. The court found that laws ban merchants
from surcharging in at least ten states. SPA40.11 The only injunctive relief that
is even arguably material thus does not apply to the merchants in those states at
allit is as if the settlement had an explicit clause excluding them. The district
11
The district court cited nine, including some of the nations largest
California, Florida, and Texas. New Yorks surcharging ban has been struck
down, but that decision is currently on appeal to this Court. JA[__]{appellate
docket}. Utah has subsequently enacted a surcharging ban. SPA231.
56
Case: 12-4671
Document: 983
Page: 68
06/16/2014
1249672
105
court accordingly recognized, with significant understatement, that those laws

would diminish, at least in the near term, the efficacy of the proposed relief for
some of the class. Id.
The clearest possible example of a class lacking common interest in a claim
is whereas hererelief on that claim will not apply to certain members as a
matter of law. As this Court has held, when variations in state law might cause
class members interests to diverge, a district court should pay particular
attention to Rule 23s requirements designed to protect absentees by blocking
unwarranted or overbroad class definitions. In re Am. Intl Grp. Sec. Litig., 689
F.3d 229, 243 (2d Cir. 2012); see also Amchem, 521 U.S. at 624 (noting that
[d]ifferences in state law undermine class cohesion).
The classs lack of cohesion is equally demonstrated by the settlements
most-favored-nation provision. Some merchants accept only Visa and MasterCard
credit cards. But nearly 70% of merchants accept American Express, which
separately prohibits surcharging, and those merchants comprise over 90% of
credit-card transaction volume nationwide. JA__{DE2111-1 at 48-49}; JA__
{DE2670-5, 65}. As the district court and its appointed expert recognized, all
those merchants would be prohibited from surcharging Visa and MasterCard under
the terms of the settlement and their contracts with American Express. See SPA42
(finding that, merchant restraints imposed by American Express would, like
57
Case: 12-4671
Document: 983
Page: 69
06/16/2014
1249672
105
state laws, also undermine the [surcharging] relief); SPA41 (because of

existing agreements, most merchants will, as a practical matter, be precluded from
surcharging Visa and MasterCard products.). Those merchants obviously have
much less interest in how the surcharging claim is resolved than the subset of
members that accept only Visa and MasterCard.
The value of the surcharging relief also varies among merchants with
different business models. The court explained that many merchants, for reasons
sufficient to them, may choose not to avail themselves of the right to surcharge.
SPA36. By contrast, the major airlines seemed to be more sanguine about the
value of surcharging than other classes of merchants, including smaller retailers,
such as grocery stores and convenience stores, which were more likely to have
objected or opted out. SPA23-24. The district court took this as evidence that the
surcharging relief had some value for purposes of assessing the fairness of the
settlement and the reaction of the class. Id. But it failed to recognize the point that
matters under Rule 23(b)(2): The different class members divergent valuations
demonstrated the classs lack of cohesion. Even if surcharging did have some
value to certain members of the class, forcing airlines and grocery stores to accept
the same bargain, negotiated by a single set of representatives, far exceeded what
Rule 23(b)(2) allows.
58
Case: 12-4671
Document: 983
Page: 70
06/16/2014
1249672
105
Despite recognizing the barriers to surcharging that many class members

faced, the district court held that the fact that some merchants may elect not to
avail themselves of the rule, or are prohibited by factors beyond the scope of this
lawsuit from surcharging, does not undermine my conclusion that the class is
sufficiently cohesive. SPA 52. This reasoning is manifestly incorrect.
Discounting preexisting factors affecting the interests of class members as
beyond the scope of this lawsuit simply erases all content from the cohesion
analysis. In applying the heightened attention required for settlement-only class
certification, a court must of course consider factors such as [d]ifferences in state
law and class members different, pre-existing circumstances that might
undermin[e] class cohesion. Amchem, 521 U.S. at 620, 624. In fact, the only
way to determine cohesion is to ask whether such factors create different interests
among the class with respect to the relief that the lawsuit does control. If the
cohesion inquiry merely asked whether the class members received the same relief,
without regard to preexisting factors affecting its value, then virtually every
settlement class would be cohesive, no matter how disparately that relief might
apply. And the district courts analysis simply ignores that class members with
different interests in surcharging would not bargain for the same surcharging relief,
and so cannot be forced into a single class with respect to this Complaint.
59
Case: 12-4671
Document: 983
Page: 71
06/16/2014
1249672
105
Surcharging also illustrates that class counsel were not even aware of the
divergent interests of their sprawling collection of millions of clients. Once the
settlement was disclosed, pharmacies objected that they had no realistic ability to
surcharge because of restrictive Medicare regulations and prohibitions on
surcharging in their contracts with health insurers and pharmacy benefit managers.
JA[__]{DE 2619 (NCPA Obj. 22)}. Similarly, health insurers objected because
the Affordable Care Act requires them to spend a certain portion of premium
revenues on medical services and thus leaves them differently situated from other
(b)(2) class members with respect to surcharging. While the district court
agree[d] with these objectors that no one thought of their unique concern in
formulating the settlement, it viewed that as no reason not to approve it,
because the insurers objections were speculative. SPA48. But this misses the
point: In this representative litigation, no one thought of their unique concern,
and therefore no one protected their interestsand because they had no opt-out
rights, they were powerless to protect themselves. A settlement class that is so
sprawling that it does not even recognize the interests it affects obviously fails the
heightened cohesion requirement for certification under Rule 23(b)(2).
For all these reasons, the members of the (b)(2) class were differently
situated and would have ascribed very different value to the Complaints
underlying surcharging allegations for purposes of negotiating a global settlement
60
Case: 12-4671
Document: 983
Page: 72
06/16/2014
1249672
105
of claims. This settlement plainly violated Dukes holding that Rule 23(b)(2)
applies only when a single injunction would provide relief to each member of
the class. 131 S. Ct. at 2257. The district courts conclusion that the class is
nonetheless cohesive for purposes of Rule 23(b)(2) is wrong as a matter of law.12
3.
The relief on the Complaints surcharging claim does not

constitute an indivisible injunction.
The relief on the surcharging claim is furthermore forbidden by Rule

23(b)(2) because it is not an indivisible injunction benefitting all [class] members
at once. Dukes, 131 S. Ct. at 2558. Indeed, the relief provided on this claim is
divisible on its face. The settlement expressly provides that [n]othing in this
Class Settlement Agreement shall prevent the Defendants from contracting with
merchants not to surcharge. SPA149, 162-63 (Settlement 42(f), 55(f)). In other
words, defendants remain free to balkanize the class after the fact, and exploit
bargaining leverage (which will surely be greater as to some merchants than
12
Another minor rules change in the settlement further demonstrates the

district courts inattention to the classs lack of cohesion. The settlement
authorizes merchants that operate multiple businesses under different trade
names to accept Visa and MasterCard on a trade-name basis. SPA13; SPA14041, 153-54 (Settlement 41, 54). This relief is useless to small businesses that
operate under one name and so represents another unsurprising divergence in a
class that includes everything from YUM! Brands (KFC, Taco Bell, Pizza Hut) to
the local pizza shop. But it also creates tensions even among large-volume
merchants: Gap Inc. operates six brands with very different business models; The
Home Depot conducts the vast majority of its business under one banner. The
court did not even address whether this relief will benefit each member of the
class, and it obviously will not.
61
Case: 12-4671
Document: 983
Page: 73
06/16/2014
1249672
105
others) to create a system in which the allegedly common practice that underlies
(b)(2) certification becomes uncommon once again.
This provision of the settlement leaves defendants free to bargain
individually with strategic merchants whose national prominence might actually
allow their surcharging practices to pose a threat to defendants inflated rates.
Those merchants can be offered a private, individualized deal to avoid the classwide benefit that the district court repeatedly invoked. See, e.g., SPA38
(speculating that surcharging might reduce interchange rates on a nationwide
basis). For example, if a large merchant who is an industry leader in a segment
(say, McDonalds) decides to pursue surcharging, defendants can offer that
merchant a break on its interchange rates in exchange for its agreement not to
surcharge. For competitive reasons, smaller merchants that vie with McDonalds
for customers would then be discouraged from surcharging. This may bring down
the rate for McDonalds, but certainly not for the whole industry, let alone the
whole class of millions of merchants. Whatever benefits surcharging may produce,
it is the isolated merchants that may be able to surcharge who will realize the
greatest savings. See, e.g., JA[__]{DE2111-5 (Frankel Decl. 68)}. This result is
not an indivisible injunction providing a common benefit to the class as a whole.
62
Case: 12-4671
C.
Document: 983
Page: 74
06/16/2014
1249672
105
The district courts cohesion analysis ignored these flaws.
The district courts opinion addressed the cohesiveness of the (b)(2) class in
only a single page. The court concluded that the class was cohesive because [t]he
network rules regimes that gave rise to this case applied generally to every
merchant accepting Visa and MasterCard credit cards. SPA51. But that is true
only in a superficial wayin the same irrelevant sense in which, for example, the
set of all Wal-Mart employee policies and practices could be said to apply
generally to every Wal-Mart employee. See Dukes, 131 S. Ct. at 2551-57. The
network rules consist of thousands of pages of different policies, and countless
more unwritten practices, with greatly varying impacts on the different merchants
based on, for example, business models, market conditions, and state and federal
laws. In such a wide-ranging settlement, there are inevitably innumerable issues
including contractual and regulatory obligationsthat affect particular class
members abilities to derive value from relief on particular claims, and yet will
escape the attention of class counsel and representatives.
The district court also thought that, by focusing the settlement efforts on the
merchant restraints [i.e., surcharging], as opposed to, for example, default
interchange, Class Plaintiffs have enhanced the cohesion of the class because
regulation of default interchange would affect the class unequally. SPA52. The
problem with that reasoning is that the settlement efforts were focused on default
63
Case: 12-4671
Document: 983
Page: 75
06/16/2014
1249672
105
interchangethe district court approved a class settlement that released all claims
forever regarding default interchange, along with everything else in the Visa and
MasterCard rulebooks. The requirements of the Due Process Clause and Rule 23
exist to protect those claims from being resolved without adequate representation
and effective consent, whether or not the class obtains any relief. Indeed, those
protections are especially relevant if the class will not receive any relief on the
claim being resolved. Accordingly, the settlement efforts were no more focused
on the merchant restraints than they were on default interchange; the only
difference was who got relief and who gave up (or, more accurately, was forced to
give up) the respective kinds of claims.
The district courts error in this regard is similar to one the Supreme Court
identified in Ortiz. There, the plaintiffs attempted to establish cohesion by
demonstrating a shared interest in the common fund created by the settlement. But
the Supreme Court held that the determination whether proposed classes are
sufficiently cohesive to warrant adjudication must focus on questions that
preexist any settlement. Ortiz, 527 U.S. at 858 (quoting Amchem, 521 U.S. at
622-23). The same is true here: The cohesiveness of the class must be determined
by reference to the entire set of claims that the class could have pursued, not just
the particular relief that the settlement provided in the end.
64
Case: 12-4671
Document: 983
Page: 76
06/16/2014
1249672
105
Similarly, the Court in Ortiz found that conflicting interests of present and
future claimants were not resolved by giving them equal rights to a settlement fund
because their different interests required different treatmentsome claims were
more valuable than others. As the Court held: The very decision to treat them all
the same is itself an allocation decision with results almost certainly different from
the results that those with immediate injuries or claims of indemnified liability
would have chosen. 527 U.S. at 857; see also Literary Works, 654 F.3d at 253
(similar). Here, too, the decision to grant all merchants a limited right to surcharge
is a result[] almost certainly different from the results that merchants that are
precluded from surcharging would have chosenespecially in light of the
different value many different members in this class of millions would ascribe to
the claims being given up in exchange.
In short, this case does not involve a class with a claim that calls for a single,
class-wide remedy necessarily benefitting each member of the class at once.
Instead, whatever possible benefits injunctive relief might have here would fall
(and do fall) unequally on class members, while the settlement simultaneously
releases a host of claims with different values to those same members. For this
reason, many different members of the class would have pursued a radically
different settlement. In such a case, while it might be possible to certify a class
65
Case: 12-4671
Document: 983
Page: 77
06/16/2014
1249672
105
under Rule 23(b)(3), there is no permissible way to force the settlement on

objecting class members under Rule 23(b)(2).
III.
The Settlement Violates The Requirement Of Rule 23(a)(4) That The

Class Members Receive Adequate Representation.
This Court should also hold that the settlement is unlawful because the
structure of the settlement negotiations deprived the class members of adequate

representation. Most obviously, the (b)(2) class that received so little in exchange
for its sweeping release was represented by those who stood to reap colossal
financial benefits by negotiating a larger recovery for the (b)(3) class. Class
members bound to the (b)(2) settlement were not represented by a separate class
representative or lawyer with their interests solely in mind. In reality, the (b)(2)
class was merely the means to an endobtaining the mandatory release defendants
demanded from all the merchants that might ever sue them as the price of a large
cash payment for the (b)(3) class.
A.
Adequate representation requires separate representatives and

counsel for subgroups with divergent interests.
Multiple classes in a single case that have different interests in the outcome
must have separate class representatives and counsel to avoid structural conflicts.
Rule 23(a)(4) allows certification only if the representative parties will fairly and
adequately protect the interests of the class. This requirement is related tobut
distinct fromthe cohesion requirement discussed in Part II, because an adequate
66
Case: 12-4671
Document: 983
Page: 78
06/16/2014
1249672
105
representative must in fact protect the interests of every class member. Thus, as
with cohesion, an even higher degree of scrutiny is required where individual class
members have no opportunity to opt out and protect those interests for themselves.
See supra, at 49-51.
The principal role of the adequacy requirement is to prevent parties from
determining the rights of absent class members through a global compromise with
no structural assurance of fair and adequate representation for the diverse groups
and individuals affected. Amchem, 521 U.S. at 627 (emphasis added).
Accordingly, [t]he adequacy inquiry under Rule 23(a)(4) serves to uncover
conflicts of interest between named parties and the class they seek to represent.
Id. at 625.
While the Rules text focuses on the named plaintiffs, the Supreme Court
has recognized that the adequacy of representation enquiry is also concerned with
the competency and conflicts of class counsel. Ortiz, 527 U.S. at 856-57 & n.31
(quoting Amchem, 521 U.S. at 626 n.20). For that reason, an attorney who
represents another class against the same defendant may not serve as class
counsel. Id. at 856 (citing Moores Federal Practice 23.25[5][e]). Indeed,
following Ortiz and Amchem, the clear rule is that, in any case involving subgroups
with diverse or antagonistic interests, [o]nly the creation of subclasses, and the
advocacy of an attorney representing each subclass, can ensure that the interests of
67
Case: 12-4671
Document: 983
Page: 79
06/16/2014
1249672
105
that particular subgroup are in fact adequately represented. Literary Works, 654
F.3d at 252.
B.
The (b)(2) and (b)(3) classes had antagonistic interests and could
not be adequately represented by common representatives and
counsel.
The merchants whose claims are bound to this settlement have substantially
different interests in the nature of the relief they might receive. Some merchants
have very little interest in prospective rules changes: They may have already
ceased operating, or be shrinking in volume, so that they place far greater emphasis
on obtaining money for past claims now. The opposite is true for merchants that
do not yet exist or growing companies, which are much more concerned with
achieving lasting changes to the networks practices as opposed to getting more
money for past injuries. Cf. Amchem, 521 U.S. at 626 ([F]or the currently injured,
the critical goal is generous immediate payments. That goal tugs against the
interest of exposure-only plaintiffs in ensuring ample [relief] for the future.).
The negotiation of this settlement brought those conflicting interests into
stark relief. Both the district court and the settling parties recognized that the
defendants would provide a large cash payment for retrospective damages only if
they prospectively received overarching litigation peacethat, in particular, any
deal would be contingent on a forward-looking release from the mandatory (b)(2)
class that would extinguish all possible future claims. As Duncan MacDonald,
68
Case: 12-4671
Document: 983
Page: 80
06/16/2014
1249672
105
independent consultant and former General Counsel for Citigroups North

American and European card businesses, candidly explained: The release is a
breathtaking success for the bankcard industry. It is about as comprehensive as
any Ive ever seen. It should end the industrys antitrust wars for years to come.
JA[__]{DE2670-8 (Ex. 67 (p. 52))}(emphasis added).
In other words, defendants would agree to pay money now in exchange for
freedom from future threats of interference with their rules and practices. More
money for the (b)(3) class would buy more peace for defendants. As lead counsel
for the Class Plaintiffs acknowledged at the preliminary approval hearing: The
negotiations before the mediators were alwaysone issue was monetary, the other
issue was equitable relief. One was not going to be reached without reaching the
other. JA[__]{DE1732 (11/9/12 Tr. at 9)}(emphasis added).
But, although the settlement in this case created two classesone with
greater retrospective interests and another that could only receive prospective
reliefboth had the same class representatives and counsel. This was entirely a
problem of the settlement proponents own creation. Rather than addressing
tensions among class members in ways required by precedentproviding
members the right to opt out; creating independently represented subclasses for
merchants with different interests; or narrowing the claims the case would
resolvethe representatives created two classes defined by kinds of relief. That
69
Case: 12-4671
Document: 983
Page: 81
06/16/2014
1249672
105
was their strategy to avoid the bar to resolving individualized monetary claims
without an opt-out right. But this strategically crafted structure only gave rise to
another fatal flaw. It created two classes with opposing interests: one seeking a
large monetary award and willing to cede claims arising in the future; and the other
seeking injunctive relief only. Having manufactured this structure, the proponents
of the settlement at the very least were required to ensure that each class received
its own champion and place at the bargaining table.
But there was no named plaintiff with solely the interests of the (b)(2) class
in mind. Instead, the class representatives consisted exclusively of established
merchants with claims on the (b)(3) class settlement, and they represented both
classes in common. Worse still, when a majority of the original named plaintiffs,
including the six trade associations (which themselves had de minimis damages
claims) objected to the balance that settlement negotiations had struck, they were
dropped as representatives with respect to both classes. In other words, in their
role representing the (b)(2) class, class counsel fired their clients and restructured
the representation so that all that remained were class representatives committed to
a deal that gave the (b)(3) class money in exchange for a broad release from the
(b)(2) class. Every single representative that expressed a desire to prioritize the
mandatory (b)(2) settlement over the money was removed from the process.
70
Case: 12-4671
Document: 983
Page: 82
06/16/2014
1249672
105
The representatives that remained naturally prioritized their interest in

monetary relief. In fact, all were ineligible for the surcharging relief that
supposedly justifies the (b)(2) settlement because they operate in states that
prohibit surcharging or accept American Express. JA[__]{DE2670 at 39-40}.
That such merchants approved the settlement, while the trade associations with de
minimis damages and predominating interests in future relief objected, is powerful
evidence of a conflict between past and future claims requiring separate
representation.
Indeed, this structure created the all-too-predictable possibility that the
representatives would trade away the future-looking interests of the mandatory
(b)(2) class in return for more money todaya very nearly zero-sum affair. That
is the exact kind of conflict that Rule 23(a)(4) prohibits. As Ortiz put it: [I]t is
obvious after Amchem that a class divided between holders of present and future
claims (some of the latter attributable to claimants not yet born) requires
division into homogeneous subclasses under Rule 23(c)(4)(B), with separate
representation to eliminate conflicting interests of counsel. 527 U.S. at 856
(emphasis added).
The dilemma was structural. Representatives with present interests simply
cannot fight for the best possible relief for future-looking claims. Even if they did,
they would then fail in their obligation to class members interested in greater
71
Case: 12-4671
Document: 983
Page: 83
06/16/2014
1249672
105
monetary relief. There is no representative that can adequately represent two client
groups with such directly conflicting interests, especially when the members of one
cannot protect themselves by opting out.
The potential for conflict was, if anything, worse for class counsel, whose
stake in the (b)(3) classs multi-billion dollar recovery had no parallel when it
came to fighting for the (b)(2) class. Indeed, the settlement left in place one of the
best-understood conflicts for counsel in class-action law: letting lawyers with
everything to gain from a monetary settlement on behalf of present claimants
bargain on behalf of future claimants as well. In a closely related context, Ortiz
discussed this problem at length and condemned it as egregious:
In this case, at least some of the same lawyers representing
plaintiffs and the class had also negotiated the separate settlement of
45,000 pending claims, the full payment of which was contingent on a
successful Global Settlement Agreement. Class counsel thus had
great incentive to reach any agreement in the global settlement
negotiations that they thought might survive a Rule 23(e) fairness
hearing, rather than the best possible arrangement for the substantially
unidentified global settlement class. The resulting incentive to favor
the known plaintiffs was, indeed, an egregious example of the
conflict noted in Amchem resulting from divergent interests of the
presently injured and future claimants.
Id. at 852-53 (internal citations and quotations omitted). This case is
indistinguishable: With a fee request representing the largest share of a multibillion dollar fundand with that fund dependent on reaching an agreement that
provided a mandatory release from the (b)(2) classclass counsel faced the natural
72
Case: 12-4671
Document: 983
Page: 84
06/16/2014
1249672
105
incentive to be sure that the deal got done. No class counsel could be expected to
scuttle a multi-billion dollar settlement based on the conviction that the (b)(2) class
was releasing too much in exchange for too little injunctive relief. As Ortiz makes
clear, the law cannot indulge assumptions so contrary to human nature, at least in
any class action settlement with the potential for gigantic fees. Id. at 852.
As precedent demonstrates, the absence of separate and adequate
representation for the unique interests of the future-looking (b)(2) class is fatal. In
Amchem and Ortiz, the Supreme Court considered efforts to create global
settlements regarding asbestos-related injuries. A key problem identified in both
cases was the tension between present claimants who had already developed
symptoms from asbestos exposure, and exposure-only plaintiffs who were in
jeopardy, but as yet had no injury. The former (like the (b)(3) class here) wanted
the greatest possible relief for existing claims; the latter (like the (b)(2) class here)
wanted the greatest possible protection for future claimants. In both cases, the
Supreme Court insisted on sub-classes with separate representation to eliminate
conflicting interests of counsel, and condemned the settlements because [n]o
such procedure was employed [t]here. Ortiz, 527 U.S. at 856; see also Amchem,
521 F.3d at 620. No such procedure was employed here either.
Perhaps the closest case on point is this Courts recent rejection of the
settlement in Literary Works on grounds of inadequate representation. There,
73
Case: 12-4671
Document: 983
Page: 85
06/16/2014
1249672
105
digital database owners wanted a global settlement with copyright holders about
the works in the database. The settlement recognized that three different kinds of
copyright claims with different values were at issue, and so it provided different
relief for the three subgroups. It did not, however, provide them separate
representatives in bargaining for that outcome.
Noting that [o]nly the creation of subclasses, and the advocacy of an
attorney representing each subclass, can ensure that the interests of that particular
subgroup are in fact adequately represented, this Court rejected the settlement.
Literary Works, 654 F.3d at 252 (emphasis added). Even if class representatives
themselves belonged to each group, and so had some incentive to look out for each
group of claims, [t]he selling out of one category of claim for another [wa]s not
improbable. Id. To avoid that risk, each subgroup needed its own representative
and its own separate counsel, so there was always someone who advanced the
strongest arguments in favor of [each categorys] recovery. Id. at 253. And yet
that feature is as absent here as it was in Literary Works.
C.
There is no substitute for independent and adequate

representation.
The settlement proponents and the district court offered two responses to the
failure to provide independent representation to the (b)(2) class here. Both are
unpersuasive and confirm that this settlement fails to satisfy Rule 23(a)(4).
74
Case: 12-4671
1.
Document: 983
Page: 86
06/16/2014
1249672
105
Overlap
The district court evidently believed that the class representatives could
represent both classes because they were members of both classes. See SPA52
(The Class Plaintiffs adequately represent both the (b)(2) and the (b)(3) settlement
classes.). This Court has already rejected that interpretation of Rule 23(a)(4) in a
discussion the Supreme Court adopted wholesale in Amchem:
[W]here differences among members of a class are such that
subclasses must be established, we know of no authority that permits a
court to approve a settlement without creating subclasses on the basis
of consents by members of a unitary class, some of whom happen to
be members of the distinct subgroups. The class representatives may
well have thought that the Settlement serves the aggregate interests of
the entire class. But the adversity among subgroups requires that the
members of each subgroup cannot be bound to a settlement except by
consents given by those who understand that their role is to represent
solely the members of their respective subgroups.
521 U.S. at 627 (quoting In re Joint E. and S. Dist. Asbestos Litig., 982 F.2d 721,
743 (2d Cir. 1992)) (emphasis added); see also Eubank v. Pella Corp., 2014 WL
2444388, *3 (7th Cir. June 2, 2014) (calling a single set of representatives agreeing
to nationwide settlement on behalf of two subclasses a red flag[]). It is thus
dispositive that in the negotiation of this settlement, there has never been any class
counsel or class representative who has bargained solely for the benefit of the
(b)(2) class.
Moreover, the overlap between the (b)(2) and (b)(3) classes should not be
misunderstood or overstated. As an initial matter, the membership is not the same:
75
Case: 12-4671
Document: 983
Page: 87
06/16/2014
1249672
105
There are tens of thousands of now-defunct merchants belonging only to the (b)(3)
class; and there are millions of other merchants that have been newly created, will
be created in the future, or opted out, and thus belong only to the (b)(2) class.13
The problem is exemplified by future merchants that start their businesses
after 2021, when the settlement relief expires. Such merchants will have no right
to complain about any Visa or MasterCard practice that is substantially similar to
any aspect of todays rulebook or current unwritten practices. Those unlucky
future merchants do not even exist yet, but this settlement has already deprived
them of all the claims they might ever have against Visa and MasterCard. What
will they have received in exchange? Literally nothing. Meanwhile, existing
merchants founded after November 2012 may be able to surchargefor a few
years, if they operate in the right state, and they dont take American Expressbut
unlike the class representatives who purported to act on their behalf, they have no
claim whatsoever on defendants settlement fund.
Failing to provide separate representation to this enormous class of future
merchants is an egregious version, on an even-shorter time frame, of the
13
It is easy to miss the size of the future merchant class whose interests the
settlement completely ignored. The number of new firms founded in the United
States each year is in the hundreds of thousands. Accordingly, the class of future
merchants launching after the settlement datemerchants who would have no
interest in a monetary settlementwill likely include many millions of members.
That class alone would be among the most sprawling commercial classes ever
certified.
76
Case: 12-4671
Document: 983
Page: 88
06/16/2014
1249672
105
representation problem identified in Stephenson. There, Agent Orange claimants

who did not discover their injuries until after 1994 were left uncovered by a 1984
settlement. See 273 F.3d at 260-61. This Court found that, because they lacked
effective representation in the 1984 settlement, the Due Process Clause prohibited
applying the judgment to them. That a similar class of millions here received no
independent consideration or voice in the settlement makes absolutely clear that
the representation afforded to the (b)(2) class did not measure up to the necessary
standard according to the four-square holdings of precedents like Stephenson,
Literary Works, Amchem, and Ortiz. Representatives focused exclusively on the
(b)(2) classs interests would not have left such future interests out in the cold.
Those who opted out of the (b)(3) class here likewise deserve special
attention because they demonstrate the deep structural anomaly of this case. In a
typical (b)(3) case, there is little concern regarding class counsels representation
of those plaintiffs who have opted out, at least once they have left the case: Their
claims are no longer subject to adjudication by representation, and they can hire
their own lawyers to protect their own interests in opt-out litigation. But here,
those that opted out of the (b)(3) class remained bound as members of the (b)(2)
class to the very attorneys and class representatives who negotiated the settlement
the opt-outs were rejecting as members of the (b)(3) class. In fact, those
representatives were exclusively parties and lawyers who, unlike the (b)(3) opt-
77
Case: 12-4671
Document: 983
Page: 89
06/16/2014
1249672
105
outs, remained interested in the money that could be obtained by trading away the
rights of the (b)(2) class for a global settlement. Accordingly, as members of the
(b)(3) class opted out but remained bound to the (b)(2) class, the misalignment of
their interests with those of their representatives only became worse.
2.
Results
Although both class counsel and the district court touted the magnitude of
the monetary settlement, the results of the negotiations in no way suggest that the
(b)(2) class received adequate representation. Positive results, even if fair to all
members, are no substitute for adequate representation. See, e.g., Literary Works,
654 F.3d at 253 (The rationale is simple: how can the value of any subgroup of
claims be properly assessed without independent counsel pressing its most
compelling case?). In any event, the results here are unfair, which furnishes
additional evidence of the (b)(2) classs inadequate representation. See id. at 25254 (noting that Amchem permits courts to find Rule 23(a)(4) violations based on
settlement results).
Other briefs discuss the fairness of the settlement as such, but it suffices for
these purposes to focus on just one comparison. As detailed above, supra, at 1517, 22-23, the (b)(2) class obtained (at best) some relatively inconsequential rule
changes in exchange for a dramatic release of claims. The (b)(3) class, meanwhile,
got billions of dollars.
78
Case: 12-4671
Document: 983
Page: 90
06/16/2014
1249672
105
This comparison reflects poorly on the representation of the (b)(2) class, and
eviscerates the district courts rationale for concluding that the settlement was fair
with respect to those class members. For example, the district court repeatedly
emphasized its belief that the class claims faced hurdles to success on the merits.
But that reasoning cannot explain the stark contrast between the size of the cash
settlement for the (b)(3) class and the narrowness of the relief for the (b)(2) class.
Each classs claims would have similar merit, as they are distinguished only by the
date on which the damages accrue.
Ultimately, it is clear that the (b)(2) class was a means to the broader end of
global resolution, not an end in itself with rights that received independent and
adequate representation. The settlement fund in this case is large14 and as the
outcome fully attests, there was simply no structural assurance that class
representatives and counsel with a claim against that sum had a sufficient incentive
to prevent the (b)(2) classs interests from being sacrificed in its pursuit. If
anything, that sums size simply shows the value to defendants created by the
mismatch between the (b)(2) classs broad and mandatory release of future claims
and the meager future-looking relief that the (b)(2) class secured.
14
To the banks, of course, the amount is relatively inconsequentialless than

two months worth of interchange fees.
79
Case: 12-4671
IV.
Document: 983
Page: 91
06/16/2014
1249672
105
By Releasing All Future Antitrust Claims, Including Claims That Far

Exceed The Scope Of The Complaint, The Settlement Violates
Controlling Precedent And Exceeds The Power Of A Federal Court.
The settlement separately fails because the (b)(2) release exceeds the
permissible scope of a class-action settlement. The effect of that release is to

immunize Visa and MasterCard from future antitrust challenges, as well as an
array of other claims that could not have been resolved by the Complaint. That
release extends broadly in both time and subject mattersweeping aside
essentially all present and future claims that merchants may have, but granting only
limited reliefwhether or not the extinguished claims are ripe or have anything to
do with the core complaints in the case. Putting to the side that this is bad antitrust
policy, it is not the proper business of settling parties and federal courts.
A.
The settlement unlawfully releases future antitrust claims.
This Court has long recognized that antitrust claims are uniquely laden with
public concerns. In American Safety, for example, this Court explained that [a]
claim under the antitrust laws is not merely a private matter. The Sherman Act is
designed to promote the national interest in a competitive economy; thus, the
plaintiff asserting his rights under the Act has been likened to a private attorneygeneral who protects the publics interest. Am. Safety Equip. Corp. v. J. P.
Maguire & Co., 391 F.2d 821, 826 (2d Cir. 1968). That is so because [a]ntitrust
violations can affect hundreds of thousandsperhaps millionsof people and
80
Case: 12-4671
Document: 983
Page: 92
06/16/2014
1249672
105
inflict staggering economic damage. Id. The Supreme Court has thus affirmed
that private agreements cannot waive future antitrust claims without violating
public policy. See, e.g., Italian Colors, 133 S. Ct. at 2310; Mitsubishi Motors
Corp. v. Soler Chrysler-Plymouth, Inc., 473 U.S. 614, 637 n.19 (1985) (If a private
agreement operated as a prospective waiver of a partys right to pursue
statutory remedies for antitrust violations, we would have little hesitation in
condemning the agreement as against public policy.).
This doctrine is fully applicable to settlement agreements that purport to
resolve future antitrust claims under the auspices of a federal court. In Lawlor, 349
U.S. at 328-29, the Supreme Court stated that extinguishing claims which did not
even then exist and which could not possibly have been sued upon in the previous
case would in effect confer on [defendants] a partial immunity from civil
liability for future violations. The Court then held that, given the public interest
in vigilant enforcement of the antitrust laws through the instrumentality of the
private treble-damage action, conferring such a partial immunity from civil
liability for future violations is consistent with neither the antitrust laws nor the
doctrine of res judicata. Id. at 329. Accordingly, even though the suit in Lawlor
was brought by the very same plaintiffs who had earlier settled a case brought on
the same antitrust theory, the Court reversed a decision giving that settlement
preclusive effect as to the same conduct undertaken after the settlement date. Id.
81
Case: 12-4671
Document: 983
Page: 93
06/16/2014
1249672
105
Federal courts since Lawlor have consistently disapproved settlements that

purported to absolve parties from liability for future violations of the antitrust laws.
See, e.g., Sanjuan v. Am. Bd. of Psychiatry & Neurology, Inc., 40 F.3d 247, 250
(7th Cir. 1994); Three Rivers Motor Co. v. Ford Motor Co., 522 F.2d 885, 896
n.27 (3d Cir. 1975); Gaines v. Carrollton Tobacco Bd. of Trade, Inc., 386 F.2d
757, 759 (6th Cir. 1967); Fox Midwest Theatres, Inc. v. Means, 221 F.2d 173, 180
(8th Cir. 1955); Minn. Mining & Mfg. Co. v. Graham-Field, Inc., 1997 WL
166497, *3 (S.D.N.Y. Apr. 9, 1997).
The (b)(2) release in this case violates Lawlor and Soler by waiving antitrust
claims with respect to future conduct. As discussed, it applies to all of defendants
present rules and unwritten practices, as well as any new rules or conduct that are
substantially similar. SPA171 (Settlement 68(g)-(h)). The release also
expressly bars claims concerning the future effect of that conduct, even if
competitive conditions change dramatically. Id.; SPA173-74 (Settlement 71);
supra, at 16-17. From defendants perspective, this was the whole point. Supra, at
69.
B.
The settlement unlawfully releases claims beyond the scope of the

present litigation.
Because the settlement is a private agreement conferring antitrust immunity,

the Court could simply stop there and invalidate it under Lawlor and Soler. But
that would severely understate the scope of the problem. This is not merely a
82
Case: 12-4671
Document: 983
Page: 94
06/16/2014
1249672
105
private agreement by one party not to sue another in the future for a violation
affected with the public interest; instead, it is a mandatory settlement that prevents
merchantsthe parties most susceptible to defendants market powerfrom
challenging anticompetitive conduct forever. That converts the settlement from a
private agreement that violates the antitrust policy of the United States into one
that actively tries to make industrial policy for the whole United States credit-card
market. The implications for consumers in the form of higher credit card fees and
higher prices for goods are obvious. Settled principles of class-action law prevent
such a broad, future-looking release of claims.
This Court has held that any release in a class-action settlement is limited to
the claims that could be precluded by a judgment against the class following a trial.
See Natl Super Spuds, Inc. v. N.Y. Mercantile Exch., 660 F.2d 9, 16-18 (2d Cir.
1981) (Friendly, J.) (If a judgment after trial cannot extinguish claims not asserted
in the class action complaint, a judgment approving a settlement in such an action
ordinarily should not be able to do so either.); Literary Works, 654 F.3d at 247
(class release may not extend beyond claims that were or could have been pled).
This limitation is referred to as the identical factual predicate doctrine, and it
limits class-action releases to only those claims that were pled or could have been
pled on the precise facts before the court. See, e.g., TBK, 675 F.2d at 460.
83
Case: 12-4671
Document: 983
Page: 95
06/16/2014
1249672
105
This doctrine imposes two kinds of limitations. First are limits in time: A
settlement cannot release future claims based on later arising facts not yet before
the court. Second are limits in scope: A settlement cannot release kinds of claims
beyond those properly at issue in the case. The settlement in this case contravenes
both limitations.
1.
The settlement improperly releases unripe future claims.
Claims related to future conduct are unripethe conduct has not happened
yetand so fall outside the jurisdiction of the court and the factual predicates of
the class-action case before it. See, e.g., Prime Mgmt. Co., Inc. v. Steinegger, 904
F.2d 811, 816 (2d Cir. 1990) (While a previous judgment may preclude litigation
of claims that arose prior to its entry, it cannot be given the effect of extinguishing
claims which did not even then exist and which could not possibly have been sued
upon in the previous case.) (quoting Lawlor, 349 U.S. at 328). Indeed, releasing
future conduct that occurs in an unknown factual context poses severe dangers to
class membersespecially where, as here, they have no opportunity to protect
themselves by opting out. See generally James Grimmelmann, Future Conduct
and the Limits of Class-Action Settlements, 91 N.C. L. Rev. 387 (2013) (explaining
dangers associated with allowing class-action settlement agreements to release
future conduct).
84
Case: 12-4671
Document: 983
Page: 96
06/16/2014
1249672
105
In Literary Works, this Court suggested that an opt-out class-action

settlement could release future copyright-infringement claims because a trial in
that case would have resolved whether the defendants could legally continue to
sell and license the works at issue. See 654 F.3d at 248. But that doctrine cannot
be extended to this cases damages release for three reasons.
First, unlike the injunctive claim regarding future use in Literary Works,
which by definition was forward-looking, the varied antitrust damages claims
released here depend on market conditions and competitive impacts that can only
be assessed based on existing or past factual circumstancescircumstances that
will likely change in the future and could not possibly have been litigated in this
case. Restraints of trade evaluated under the rule of reason do not ripen into
antitrust violations until anticompetitive effects are shown. See, e.g., E & L
Consulting, Ltd. v. Doman Indus. Ltd., 472 F.3d 23, 29 (2d Cir. 2006).
Second, the claims that were released in Literary Works are always released
in a future-looking way. Copyright claims are amenable to licensing; indeed, the
settlement was conceived of as a continuing license. See 654 F.3d at 247.
Unlike antitrust law, the very essence of copyright is the power to release future
infringement claims in exchange for present consideration. See, e.g.,
Grimmelmann, 91 N.C. L. Rev. at 409-10.
85
Case: 12-4671
Document: 983
Page: 97
06/16/2014
1249672
105
Finally, Literary Workslike other cases applying the identical factual

predicate doctrine, even outside the context of future claimswas not a
mandatory settlement.15 Not only did it allow class members to opt out, but it
specifically allowed them to withhold a future-looking license and preserve their
statutory right to bar future use of their copyrighted works. 654 F.3d at 247.
Citing Robertson v. NBA, 556 F.2d 682, 686 (2d Cir. 1977), the district court
held that future claims could be released because the legality of defendants rules
was an unsettled question. SPA45. But Robertson is inapposite. It addresses a
different doctrine that would condemn even a settlement confined to the matters
properly before the court if it allowed the perpetuation of clearly illegal behavior.
See Robertson, 556 F.2d at 686. Robertson, which predates Super Spuds, in no
way suggests that a class-action settlement can release claims regarding future
conduct not before the court so long as the conduct is arguably kosher. As Judge
Friendly recognized in Super Spuds, the issue is not just the legality of the conduct
released, but the extent to which that conduct is subject to the power of the court
15
See, e.g., Wal-Mart, 396 F.3d at 112 (agreeing with analysis in prior case
that release was not problematic because, inter alia, it provided class members
the opportunity to opt out); Weinberger v. Kendrick, 698 F.2d 61, 77 (2d Cir.
1982) (distinguishing Super Spuds on grounds that released claims were added
before class certification, and settlement afforded an opportunity to opt out);
Super Spuds, 660 F.2d at 19 (distinguishing prior case with broad release because
class members could opt out).
86
Case: 12-4671
Document: 983
Page: 98
06/16/2014
1249672
105
and the class representatives, which are both limited to resolving the particular case
or controversy implicated by the facts and claims at bar.
To be sure, courts will frequently have the instincteven the correct view
that approving a prospective regime agreed to in a global settlement will achieve
more for the parties or public than disapproving it. But especially in a commercial
case with no opt-out rights, a settlement that attempts to implement a forwardlooking business arrangement for an entire industry, without permission of the
[class members], simply goes too far. Google Books, 770 F. Supp. 2d at 669.
Enforcing the established bars on releasing future antitrust claims not properly
before the court respects the limited role of federal litigation and ensures that class
actions remain a respected tool in the service of proper goals. Conversely,
allowing class actions to release claims against future conduct by all prospective
plaintiffsas this one doesis an invitation for private parties to engage in
industry-wide regulation that is more likely than not to prioritize parochial goals
over the public good.
2.
The settlement improperly releases present claims beyond the

scope of the case.
Even the present claims that the settlement releases extend well beyond the
identical factual predicate of the claims that were actually brought. The district
court recognized the cases proper scope: it concerned four categories of allegedly
anticompetitive network rulesdefault interchange rules, certain anti-steering
87
Case: 12-4671
Document: 983
Page: 99
06/16/2014
1249672
105
rules, certain exclusionary rules, and Honor-All-Cards ruleswhich have

allow[ed] Visa and MasterCard and the issuing banks to set supracompetitive
default interchange fees. SPA18-19. At a minimum, the identical factual
predicate doctrine would limit a class-action release to such claims alone.
But the releases are in fact much broader. They unambiguously bar claims
based on any of defendants current rules and practicesas well as substantially
similar future conductnot just claims based on the four categories of allegedly
anticompetitive rules and the resultant default interchange fees. Supra, at 15-17.
The releases even purport to bar damages claims concerning the FANF, which
appears nowhere in the Complaint because it was implemented in April 2012, three
years after merits discovery had closed.
The over-breadth of the Rule 23(b)(2) release is exacerbated by the
settlements definition of Rule 23(b)(2) Settlement Class Releasing Parties as
including subsidiaries of any class member, without geographic limitation.
SPA166 (Settlement 66); SPA88 (Judgment 16(a)). This definition facially
encompasses British supermarket chain ASDAa subsidiary of appellant WalMart Stores, Inc.which is now litigating claims in European Union courts
seeking substantial monetary relief for anticompetitive conduct abroad. See
JA[__]{DE2644 (Wal-Mart Obj. 56-58)}. Thus, ASDA, which cannot be a
member of either class and receives nothing from the settlement, could face the
88
Case: 12-4671
Document: 983
Page: 100
06/16/2014
1249672
105
argument that its extra-territorial claims, which are well outside this cases factual
predicate, are subject to the (b)(2) release as well.
The problem goes even deeper. The settlement proponents and the district
court incorrectly portrayed the scope of the release as limited to the factual
predicates of the case because it covers only existing rules and those that are
substantially similar. SPA44-47. This fails to recognize that a rule could have
very different effects in different factual contexts.
The history of the Honor-All-Cards rules, and defendants exploitation of
them to suppress competition, illustrates the point. Honor-All-Cards rules that
were introduced in the 1960s and applied solely to credit cards were utilized by
Visa and MasterCard in the 1990s to extend their market power into the emerging
market for debit transactionstying practices that resulted in the Visa Check
settlement, which this Court approved. JA[__]{DE455-4 4; DE455-5
4}(settlement provisions requiring revisions to Honor-All-Cards to untie debit
from credit).
Here, counsel for defendants made clear that Visa and MasterCard have
every intention to use their Honor-All-Cards rules again as a tying devicethis
time by linking mobile payments to payments made with traditional payment
cards. See supra, at 26-27. To the extent they do so, such claims would depend on
future facts, including the extent to which the application of the Honor-All-Cards
89
Case: 12-4671
Document: 983
Page: 101
06/16/2014
1249672
105
rules to mobile payments suppresses competition in those technologies. And this is

only one of innumerable future factual scenarios in which the application of an
existing or substantially similar rule would result in future conduct with
radically different, anticompetitive effects.
The effect of a mandatory release of claims of such breadth for eternity is to
replace the statutory rights and remedies that Congress has provided with a
privately negotiated, quasi-regulatory regime for the credit-card industry going
forward. As Judge Chin recently recognized, this is not properly the business of
settling parties and federal courts. In Google Books, the court recognized that a
large part of the settlement was directed to future and ongoing arrangements
[that] would release Google (and others) from liability for certain future acts. 770
F. Supp. 2d at 676-77. The court rightly concluded that this second part of the
[settlement] contemplates an arrangement that exceeds what the Court may permit
under Rule 23, because it is an attempt to use the class action mechanism to
implement forward-looking business arrangements that go far beyond the dispute
before the Court in this litigation. Id. at 677. Such matters, the court noted, are
for Congress, and not the federal courts. The same is true here. Indeed, the result
here looks even more like legislation, because objectors to the Google Books
regime for the future of that industry at least had the right to opt out.
90
Case: 12-4671
Document: 983
Page: 102
06/16/2014
1249672
CONCLUSION
The judgment should be reversed.
Dated: New York, New York
June 16, 2014
Respectfully submitted,
/s/ Thomas C. Goldstein
Stephen R. Neuwirth*
Sanford I. Weisburst
Steig D. Olson
Cleland B. Welton II
QUINN EMANUEL URQUHART &
SULLIVAN, LLP
51 Madison Avenue
22nd Floor
New York, NY 10010
Jeffrey I. Shinder
Gary J. Malone
A. Owen Glist
335 Madison Avenue
9th Floor
New York, NY 10017
Thomas C. Goldstein*
Eric F. Citron
GOLDSTEIN & RUSSELL, P.C.
5225 Wisconsin Avenue, N.W.
Suite 404
Michael J. Canter
Robert N. Webner
Kenneth J. Rubin
VORYS, SATER, SEYMOUR
AND PEASE LLP
52 East Gay Street
Columbus, OH 43215
Gregory A. Clarick
CLARICK GUERON REISBAUM LLP
40 West 25th Street
12th Floor
New York, NY 10010
(Parties listed on following pages.)
91
105
Case: 12-4671
Document: 983
Page: 103
06/16/2014
1249672
105
Attorneys for Objector-Appellant Home Depot U.S.A., Inc.
Attorneys for Plaintiffs-Appellants

Coborns Incorporated; DAgostino Supermarkets, Inc.; Jetro Holdings, LLC;
Affiliated Foods Midwest Cooperative, Inc.; National Association of Convenience
Stores; National Community Pharmacists Association; National Cooperative
Grocers Association (NCGA); National Grocers Association; National
Restaurant Association; and NATSO Inc.;
And for Objectors-Appellants
7-Eleven, Inc.; Academy, Ltd. d/b/a Academy Sports Outdoors; Aldo US Inc.
d/b/a Aldo and Call It Spring; Alon USA, LP (Alon Brands); Amazon.com, Inc.;
American Eagle Outfitters, Inc.; Barnes & Noble, Inc.; Barnes & Noble College
Booksellers, LLC; Best Buy Stores, L.P.; BJs Wholesale Club, Inc.; The William
Carter Company (Carters); Costco Wholesale Corporation; Crate & Barrel
Holdings, Inc.; Darden Restaurants, Inc.; Davids Bridal, Inc., DBD Inc. and
Davids Bridal Canada Inc.; Dicks Sporting Goods, Inc.; Dillards, Inc.; Family
Dollar Stores, Inc.; Drury Hotels Company, LLC; Foot Locker, Inc.; Gap Inc.;
GNC Holdings, Inc. (General Nutrition Corporation); Genesco Inc.; The
Gymboree Corporation; HMSHost Corporation; IKEA North America Services,
LLC; J. Crew Group, Inc.; Kwik Trip, Inc.; Lowes Companies, Inc.; Marathon
Petroleum LP; Martins Super Markets, Inc.; Michaels Stores, Inc.; National
Railroad Passenger Corporation d/b/a Amtrak; Nike, Inc.; Panda Restaurant
Group, Inc.; Panera Bread Company; P.C. Richard & Son, Inc.; PETCO Animal
Supplies, Inc.; PetSmart, Inc.; RaceTrac Petroleum, Inc.; Recreational
Equipment, Inc. (REI); Roundys Supermarkets, Inc. d/b/a Pick N Save,
Rainbow, Copps, Metro Market and Marianos; Sears Holdings Corporation;
Speedway LLC; Starbucks Corporation; Stein Mart, Inc.; Thermo Fisher
Scientific Inc.; The Wendys Company; The Wet Seal, Inc.; Whole Foods Market,
Inc.; Zappos.com, Inc.; Fleet Wholesale Supply Co., Inc.; Mills Motor, Inc.;
Mills Auto Enterprises, Inc.; Willmar Motors, LLC; Mills Auto Center, Inc.;
Fleet and Farm of Alexandria, Inc.; Fleet Wholesale Supply of Fergus Falls,
Inc.; Fleet and Farm of Green Bay, Inc.; Fleet and Farm of Menomonie, Inc.;
Mills Fleet Farm, Inc.; Fleet and Farm of Manitowoc, Inc.; Fleet and Farm of
Plymouth, Inc.; Fleet and Farm Supply Company of West Bend, Inc.; Fleet and
Farm of Waupaca, Inc.; Mills E-Commerce Enterprises, Inc.; Brainerd Lively
Auto, LLC; Ashley Furniture Industries Inc.; Bealls, Inc.; Boscovs, Inc.; The
Buckle, Inc.; Buc-ees Ltd.; The Childrens Place Retail Stores, Inc.; Cracker
Barrel Old Country Store, Inc.; Cumberland Farms, Inc.; Express, LLC; Family
Express Corporation; New York & Company, Inc.; Republic Services, Inc.;
Swarovski U.S. Holding Limited; and The Talbots, Inc.
92
Case: 12-4671
Document: 983
Page: 104
06/16/2014
1249672
105
Attorneys for Objectors-Appellants

Target Corporation; Macys, Inc.; Kohls Corporation; The TJX Companies,
Inc.; Staples, Inc.; J.C. Penney Corporation, Inc.; Office Depot, Inc.; L Brands,
Inc.; Big Lots Stores, Inc.; PNS Stores, Inc.; C.S. Ross Company; Closeout
Distribution, Inc.; Ascena Retail Group, Inc.; Abercrombie & Fitch Co.;
OfficeMax Incorporated; Saks Incorporated; The Bon-Ton Stores, Inc.; Chicos
FAS, Inc.; Luxottica U.S. Holdings Corp. and American Signature, Inc.
93
Case: 12-4671
Document: 983
Page: 105
06/16/2014
1249672
105
CERTIFICATE OF COMPLIANCE
This brief complies with this Courts order dated May 27, 2014 (ECF No.
936), granting leave to file an oversized brief, because it contains 20,992 words,
excluding the parts of the brief exempted by FRAP 32(a)(7)(B)(iii).
This brief complies with the typeface requirements of FRAP 32(a)(5) and
the type-style requirements of FRAP 32(a)(6) because it has been prepared in a
proportionately spaced typeface using Microsoft Word 2007 in 14-point Times
New Roman font.
/s/ Thomas C. Goldstein
94
Case: 12-4671
Document: 1124
Page: 1
10/15/2014
1345377
101
12-4671(L)
12-4708(con), 12-4765(con), 13-4719(con), 13-4750(con), 13-4751(con), 13-4752(con),
14-32(con), 14-117(con), 14-119(con), 14-133(con), 14-157(con), 14-159(con), 14-192(con),
14-197(con), 14-219(con), 14-225(con), 14-241(con), 14-250(con), 14-266(con), 14-303(con),
14-331(con), 14-349(con), 14-404(con), 14-422(con), 14-443(con),14-480(con), 14-497(con),
14-530(con), 14-567(con), 14-584(con), 14-606(con), 14-663(con), 14-837(con)
In The United States Court Of Appeals

For The
Second Circuit
________________
________________
for the Eastern District of New York
________________
PAGE-PROOF BRIEF FOR PLAINTIFFS-APPELLEES
________________
K. CRAIG WILDFANG
PAUL D. CLEMENT
THOMAS J. UNDLIN
Counsel of Record
RYAN W. MARTH
JEFFREY M. HARRIS
BERNARD PERSKY
CANDICE C. WONG
ROBINS, KAPLAN, MILLER &
BANCROFT PLLC
CIRESI L.L.P.
1919 M Street NW
800 LaSalle Avenue
Suite 470
(202) 234-0090
pclement@bancroftpllc.com
Counsel for Plaintiffs-Appellees
(Additional Counsel Listed on Inside Cover)
October 15, 2014
Case: 12-4671
Document: 1124
H. LADDIE MONTAGUE
MERRILL G. DAVIDOFF
MICHAEL J. KANE
1622 Locust Street
Page: 2
10/15/2014
1345377
101
JOSEPH GOLDBERG
FREEDMAN BOYD GOLDBERG
URIAS & WARD, P.A.
20 First Plaza
Suite 700
BONNY E. SWEENEY
JOSEPH DAVID DALEY
ALEXANDRA SENYA BERNAY
ROBBINS GELLER RUDMAN &
DOWD LLP
655 West Broadway
Suite 1900
San Diego, CA 92101
Case: 12-4671
Document: 1124
Page: 3
10/15/2014
1345377
101

The parties to this brief complied with Rule 26.1 of the Federal Rules of
Appellate Procedure by submitting a Compendium of Corporate Disclosure
Statements on June 16, 2014. See D.E. 988.
Case: 12-4671
Document: 1124
Page: 4
10/15/2014
1345377
101
TABLE OF CONTENTS
CORPORATE DISCLOSURE STATEMENT .......................................................... i
TABLE OF AUTHORITIES .................................................................................... iv
INTRODUCTION .................................................................................................... 1
STATEMENT OF THE ISSUES .............................................................................. 3
STATEMENT OF THE CASE.................................................................................. 4
A.
The Payment Card Industry.................................................................. 4
B.
Eight Years of Hard-Fought Litigation................................................. 6
C.
Industry Reforms During, and Due to, the Litigation .......................... 8
D.
Mediation and Settlement....................................................................11
E.
Settlement Review and Approval ....................................................... 16

STANDARD OF REVIEW .................................................................................... 25
ARGUMENT .......................................................................................................... 26
I.
The Structure Of The Settlement Classes Conforms Precisely To

Bedrock Certification Requirements Under Rule 23. .................................. 26
A.
The Rule 23(b)(2) Class Unquestionably Brought Claims For

Indivisible Relief............................................................................. 27
B.
The Rule 23(b)(2) Class Required No Opt-Out Rights; All

Claims for Individualized Monetary Relief Were Separately
and Properly Certified Under Rule 23(b)(3). ..................................... 36
C.
There Is Nothing Improper About Having a Rule 23(b)(2)

Class Settlement Foreclose Possible Future Claims Seeking
Damages. ............................................................................................ 40
D.
Both the Rule 23(b)(2) Class and Rule 23(b)(3) Class Were
Adequately Represented. .................................................................... 47
ii
Case: 12-4671
II.
Document: 1124
Page: 5
10/15/2014
1345377
101
The District Court Acted Well Within Its Broad Discretion In

Finding The Overall Settlement Fair, Reasonable, and Adequate
Under Rule 23(e). ......................................................................................... 56
A.
B.
The Relief Obtained by the Class is Outstanding. ............................. 57

1.
The $7.25 Billion Damages Fund is the Largest-Ever

Cash Relief in an Antitrust Class Action Settlement. .............. 57
2.
The Removal of Restraints on Surcharging Offers

Valuable Relief. ........................................................................ 60
3.
The Other Injunctive Reforms Offer Valuable Relief. ............. 67
4.
The Settlement Is Not Unreasonable Merely Because It

Does Not Include All of the Relief Sought by
Objectors. ................................................................................. 70
The Standard Release Conforms With All Applicable Law. .............. 72
III.

Finding The Fee Award Reasonable. ............................................................ 78
IV.

Finding The Settlement Notice Reasonable. ................................................ 82
CONCLUSION ....................................................................................................... 85
CERTIFICATE OF SERVICE
iii
Case: 12-4671
Document: 1124
Page: 6
10/15/2014
1345377
101
Cases
Amchem Prods. v. Windsor,
521 U.S. 591 (1997) .................................................................................... passim
Armstrong v. Bd. of Sch. Dirs.,
616 F.2d 305 (7th Cir. 1980) ...............................................................................77
Assn For Disabled Ams. v. Amoco,
211 F.R.D. 457 (S.D. Fla. 2002) .........................................................................42
Barnes v. Am. Tobacco,
161 F.3d 127 (3d Cir. 1998) ................................................................................33
Bendix v. Midwesco,
486 U.S. 888 (1988) ............................................................................................55
Bristol v. Louisiana-Pacific,
916 F. Supp. 2d 357 (W.D.N.Y. 2013) ................................................................38
Charron v. Pinnacle Grp.,
874 F. Supp. 2d 179 (S.D.N.Y. 2012) .................................................................59
Charron v. Wiener,
731 F.3d 241 (2d Cir. 2013) ................................................................... 25, 47, 55
DAmato v. Deutsche Bank,
236 F.3d 78 (2d Cir. 2001) ....................................................................... 2, 25, 48
Denney v. Deutsche Bank,
443 F.3d 253 (2d Cir. 2006) ................................................................................47
Detroit v. Grinnell Corp.,
495 F.2d 462 (2d Cir. 1974) ............................................................. 17, 23, 56, 70
Dewey v. Volkswagen,
681 F.3d 170 (3d Cir. 2012) ......................................................................... 54, 55
Easterling v. Dept of Corr.,
278 F.R.D. 41 (D. Conn. 2011) ...........................................................................38
iv
Case: 12-4671
Document: 1124
Page: 7
10/15/2014
1345377
101
Eubank v. Pella,
753 F.3d 718 (7th Cir. 2014) ...............................................................................52
Expressions Hair Design v. Schneiderman,
975 F. Supp. 2d 430 (S.D.N.Y. 2013) .................................................................66
Goldberger v. Integrated Resources,
209 F.3d 43 (2d Cir. 2000) .......................................................................... passim
Gooch v. Life Investors Ins.,
672 F.3d 402 (6th Cir. 2012) ........................................................................ 38, 55
Handschu v. Special Servs. Div.,
605 F. Supp. 1384 (S.D.N.Y. 1985) ............................................................. 63, 70
Handschu v. Special Servs. Div.,
787 F.2d 828 (2d Cir. 1986) ................................................................... 31, 57, 83
Hecht v. United Collection Bureau,
691 F.3d 218 (2d Cir. 2012) ................................................................................43
Huyer v. Wells Fargo,
295 F.R.D. 332 (S.D. Iowa 2013) .......................................................................38
In re AIG Sec. Litig.,
689 F.3d 229 (2d Cir. 2012) ................................................................................34
In re Enron Sec., Derivative & ERISA Litig.,
586 F. Supp. 2d 732 (S.D. Tex. 2008).................................................................81
In re IKO Roofing Prods. Litig.,
757 F.3d 599 (7th Cir. 2014) ...............................................................................30
In re IPO Sec. Litig.,
471 F.3d 24 (2d Cir. 2006) ..................................................................................26
In re Johns-Manville,
759 F.3d 206 (2d Cir. 2014) ......................................................................... 77, 78
In re Literary Works Copyright Litig.,
654 F.3d 242 (2d Cir. 2011) ........................................................................ passim
Case: 12-4671
Document: 1124
Page: 8
10/15/2014
1345377
101
In re Managed Care Litig.,

2010 WL 6532985 (S.D. Fla. 2010) ...................................................................76
In re Motor Fuel Sales Practices Litig.,
271 F.R.D. 263 (D. Kan. 2010) ...........................................................................65
In re NASDAQ Antitrust Litig.,
169 F.R.D. 493 (S.D.N.Y. 1996) .........................................................................38
In re Payment Card Interchange Fee Antitrust Litig.,
986 F. Supp. 2d 207 (E.D.N.Y. 2013) .................................................................16
In re Payment Card Interchange Fee Antitrust Litig.,
991 F. Supp. 2d 437 (E.D.N.Y. 2014) .................................................................20
In re Salomon Analyst Metromedia Litig.,
544 F.3d 474 (2d Cir. 2008) ................................................................................26
In re St. Jude Medical,
425 F.3d 1116 (8th Cir. 2005) .............................................................................32
280 F.3d 124 (2d Cir. 2001) ................................................................................43
297 F. Supp. 2d 503 (E.D.N.Y. 2003) .................................................... 58, 73, 82
In re Vitamin C Antitrust Litig.,
2012 WL 5289514 (E.D.N.Y. 2012) ...................................................................82
In re Vitamin C Antitrust Litig.,
279 F.R.D. 90 (E.D.N.Y. 2012)...........................................................................38
In re WorldCom Sec. Litig.,
388 F. Supp. 2d 319 (S.D.N.Y. 2005) .................................................................81
Jefferson v. Ingersoll,
195 F.3d 894 (7th Cir. 1999) ...............................................................................36
Jermyn v. Best Buy,
276 F.R.D. 167 (S.D.N.Y. 2011) .........................................................................38
vi
Case: 12-4671
Document: 1124
Page: 9
10/15/2014
1345377
101
218 F.3d 132 (2d Cir. 2000) ................................................................... 26, 51, 57
Kartman v. State Farm,
634 F.3d 883 (7th Cir. 2011) ...............................................................................39
LaGarde v. Support.com,
2013 WL 1283325 (N.D. Cal. 2013) ........................................................... 63, 68
Lawlor v. National Screen Service,
349 U.S. 322 (1955) ............................................................................................76
Masters v. Wilhelmina Model Agency,
473 F.3d 423 (2d Cir. 2007) ................................................................................26
Maywalt v. Parker & Parsley Petrol.,
67 F.3d 1072 (2d Cir. 1995) ................................................................................58
McBean v. New York,
233 F.R.D. 377 (S.D.N.Y. 2006) .........................................................................58
McReynolds v. Richards-Cantave,
588 F.3d 790 (2d Cir. 2009) ............................................................................2, 45
Natl Super Spuds v. N.Y. Mercantile Exch.,
660 F.2d 9 (2d Cir. 1981) ....................................................................................74
New England Carpenters Fund v. First DataBank,
602 F. Supp. 2d 277 (D. Mass. 2009) .................................................................59
Nottingham Partners v. Trans-Lux,
925 F.2d 29 (1st Cir. 1991) .................................................................................42
Ortiz v. Fibreboard,
527 U.S. 815 (1999) ..................................................................................... 46, 52
Parsons v. Ryan,
754 F.3d 657 (9th Cir. 2014) ...............................................................................28
Petrovic v. Amoco,
200 F.3d 1140 (8th Cir. 1999) .............................................................................56
vii
Case: 12-4671
Document: 1124
Page: 10
10/15/2014
1345377
101
Phillips Petroleum v. Shutts,

472 U.S. 797 (1985) ............................................................................................42
Rich v. Martin Marietta,
522 F.2d 333 (10th Cir. 1975) .............................................................................28
Robertson v. NBA,
556 F.2d 682 (2d Cir. 1977) ................................................................................77
Robinson v. Metro-North,
267 F.3d 147 (2d Cir. 2001) ................................................................... 31, 39, 44
San Diego Police Officers Assn v. San Diego City Emps. Ret. Sys.,
568 F.3d 725 (9th Cir. 2009) ...............................................................................42
Scarver v. Litscher,
371 F. Supp. 2d 986 (W.D. Wis. 2005) ...............................................................42
Schwarz v. Dall. Cowboys,
2001 WL 1689714 (E.D. Pa. 2001) ....................................................................76
Shady Grove Orthopedic v. Allstate,
559 U.S. 393 (2010) ............................................................................................46
Soberal-Perez v. Heckler,
717 F.2d 36 (2d Cir. 1983) ........................................................................... 25, 82
Stephenson v. Dow Chemical,
273 F.3d 249 (2d Cir. 2001) ......................................................................... 42, 54
Stinson v. City of N.Y.,
282 F.R.D. 360 (S.D.N.Y. 2012) .........................................................................38
Suffolk Cnty. v. Long Island Lighting,
907 F.2d 1295 (2d Cir. 1990) ..............................................................................48
Thompson v. Metro. Life,
216 F.R.D. 55 (S.D.N.Y. 2003) .................................................................... 58, 85
United States v. E. I. du Pont de Nemours & Co.,
366 U.S. 316 (1961) ..............................................................................................9
viii
Case: 12-4671
Document: 1124
Page: 11
10/15/2014
1345377
101
United States v. Local 359, United Seafood Workers,

55 F.3d 64 (2d Cir. 1995) ....................................................................................75
Velez v. Novartis,
244 F.R.D. 243 (S.D.N.Y. 2007) .........................................................................38
VKK Corp. v. NFL,
244 F.3d 114 (2d Cir. 2001) ................................................................................76
W. Alton Jones Found. v. Chevron,
97 F.3d 29 (2d Cir. 1996) ....................................................................................74
Wal-Mart v. Dukes,
131 S. Ct. 2541 (2011) ................................................................................ passim
Wal-Mart v. Visa,
396 F.3d 96 (2d Cir. 2005) .......................................................................... passim
698 F.2d 61 (2d Cir. 1982) ..................................................................................82
Williams v. G.E. Capital,
159 F.3d 266 (7th Cir. 1998) ...............................................................................76
Statutes
2010 Dodd-Frank Wall Street Reform and Consumer Protection Act,
Pub. L. No. 111-203, 124 Stat. 1376 (2010) .......................................................10
15 U.S.C. 1693o-2.................................................................................................10
28 U.S.C. 2072(b) ................................................................................................ 45
Rules
Fed. R. Civ. P. 23(a) .................................................................................................47
Fed. R. Civ. P. 23(b) ......................................................................................... passim
Fed. R. Civ. P. 23(e) ......................................................................................... passim
Fed. R. Civ. P. 23(h) ................................................................................................ 78
Fed. R. Civ. P. 23 advisory committee notes (1966)......................................... 28, 76
ix
Case: 12-4671
Document: 1124
Page: 12
10/15/2014
1345377
101
Fed. R. Civ. P. 23 advisory committee notes (1996)................................................30

Fed. R. Civ. P. 23 advisory committee notes (2003)................................... 37, 49, 50
Other Authorities
Credit Card Interchange Fees: Antitrust Concerns? Hearing Before S.
Comm. on the Judiciary, 109th Cong. (2006) ....................................................61
McLaughlin on Class Actions (10th ed.)..................................................... 32, 41, 85
Newberg on Class Actions (4th ed. 2002) ......................................................... 34, 83
Wright & Miller, Federal Practice and Procedure (3d ed. 1998) .............. 27, 28, 29
Case: 12-4671
Document: 1124
Page: 13
10/15/2014
1345377
101
INTRODUCTION
This settlement was the culmination of eight years of arms-length, hard-fought
litigation before a highly experienced judge, including four years of adversarial
mediation before two distinguished independent mediators. The settlement secures
up to $7.25 billion in damages, as well as unprecedented, immediately effective
structural reforms of the payment card networks that may very likely exceed the
value of the monetary relief in the long run. SPA67. The district court, with its
extensive experience with the relevant issues and knowledge of the litigation,
correctly deemed that settlement a significant success. SPA61.
A small group of objectorsmany of whom supported a near-identical
version of the settlement submitted to the district court three months before its
executionclaims to be dissatisfied with the scope of structural reforms achieved
as a result of this litigation. Objectors have channeled those frustrations into various
attacks on the settlement classes, suggesting that certification of an injunctive-relief
class alongside a damages class in this action gave rise to cohesion, due process,
and adequacy problems. But many of objectors complaints call into question
common and unobjectionable practices, such as litigating Rule 23(b)(2) and 23(b)(3)
classes simultaneously and foreclosing future challenges to agreed-upon, going-
forward conduct. There is absolutely nothing improper about those prototypical uses
Case: 12-4671
Document: 1124
Page: 14
10/15/2014
1345377
101
of Rule 23, and the district court considered and properly rejected each of objectors
arguments.
Objectors stated concerns about the structure of the classes ultimately
devolve into complaints about the overall fairness of the settlement. But the district
court, after exhaustive evaluation, approved the settlement as both procedurally and
substantively fair. That fact-intensive determination is well-supported by the robust
structural assurances of fair and adequate representation of all interests in the
proceedings overseen by the two mediators who wholeheartedly endorsed the
settlement. It accords with the settlements massive and meaningful relief,
particularly when measured against the likely complexity, expense, delay, and risk
of proceeding to trial. SPA15, 61. And the district courts holding is bolstered by
the strong judicial policy in favor of settlements, particularly in the class action
context. McReynolds v. Richards-Cantave, 588 F.3d 790, 803 (2d Cir. 2009).
Objectors criticisms of the scope of the settlement relief and release do not
come close to a clear showing that the District Court has abused its discretion.
DAmato v. Deutsche Bank, 236 F.3d 78, 85 (2d Cir. 2001). Objectors dramatically
understate the importance of the rule reforms obtained. Those reforms permit
merchants, for the first time ever, to engage in surcharginga tool by which
merchants can inform customers about the costs of credit card acceptance, direct
customers toward less costly payment methods, and recoup acceptance costs.
Case: 12-4671
Document: 1124
Page: 15
10/15/2014
1345377
101
Conversely, objectors dramatically overstate the scope of the settlements release of

future claims. That release employs standard language that courts have repeatedly
approved, and the Due Process Clause provides the ultimate assurance that claims
will notindeed, cannotbe impermissibly released. Objectors further challenges
to the reasonableness of the attorneys fee award and the settlement notice are
equally meritless.
In short, the settlement reflects a remarkable accomplishment, providing
monetary and injunctive relief of unprecedented value and scope. The district courts
determinations that the settlement, fee award, and notice were appropriate lie well
within its broad discretion and should be affirmed.
STATEMENT OF THE ISSUES
1.
Whether the district court acted within its broad discretion in
concluding that the Rule 23(b)(2) injunctive-relief and Rule 23(b)(3) damages
classes conform with their respective class certification requirements.
2.
Whether the district court acted within its broad discretion in approving
the settlementwhich included $7.25 billion in money damages and unprecedented

injunctive reliefas fair, reasonable, and adequate under Rule 23(e).
3.
Whether the district court acted within its broad discretion in
calculating and approving the attorneys fee award as reasonable.
Case: 12-4671
4.
Document: 1124
Page: 16
10/15/2014
1345377
101
Whether the district court acted within its broad discretion in approving
the settlement notice as reasonable.

A.
The Payment Card Industry
Visa and MasterCard were created as bank-owned joint ventures a half-

century ago, and were subsequently owned and controlled by the largest banks in the
United States. The banks were members of Visa and MasterCard, owned stock in
Visa and MasterCard, placed representatives on the Visa and MasterCard boards of
directors and committees, and issued Visa and MasterCard payment cards. Through
those influential positions, the member banks predictably established and enforced
network rules that produced ever-increasing interchange feesthe fees that a
merchant must pay to a card-issuing bank in order to accept Visa or MasterCard-
branded cards. See D.E. 1153 (Second Consolidated Amended Class Action
Complaint) 94-100.
Visa and MasterCard have grown exponentially over the last four decades. In
1970, only 16% of U.S. families had a credit card, and less than one million U.S.
merchantsapproximately 20%accepted payment cards. By 2006, 77% of U.S.
adults had at least one credit card, and merchant acceptance was ubiquitous. In 2007
alone, transaction volume on bank-issued credit cards topped two trillion dollars,
and Visa and MasterCard transactions accounted for roughly 75% of that volume.
D.E. 1153 126, 274-276.
4
Case: 12-4671
Document: 1124
Page: 17
10/15/2014
1345377
101
As their market power grew, the interchange fees Visa and MasterCard levied
on merchants rose as well. Merchants faced a schedule of default interchange fees
for every transaction on the network. In practice, because private agreements to
deviate from the default rates were exceptionally rare, these were not just default
fees, but actual fees. Along with payroll and rent, interchange fees became one of
the largest operating costs for many businesses. By the mid-2000s, card-issuing
banks were reaping more than $30 billion annually in interchange fees. D.E. 2113-
6 (Wildfang Decl.) 17.
Interchange fees were not the only anticompetitive restraint adopted by the
networks and the banks that controlled them. They also adopted various anti-
steering restraints as network rules that every card-accepting merchant was required
to follow. Coupled with merchants obligations to pay default interchange fees and
to Honor-all-Cards bearing the Visa or MasterCard brand regardless of the issuing
bank or the amount of the interchange fee, these anti-steering restraints enabled
banks to demand exorbitant sums at the expense of merchants that had no alternative
but to continue paying the fees.
The foremost example of an anti-steering restraint was the anti-surcharging
rule, which prohibited merchants from adding a surcharge at the point of sale to alert
customers to the costs of payment options and recoup the costs associated with credit
card usage. D.E. 1165 (Pls. Mot. for Class Cert.) at 3. This rule prevented
Case: 12-4671
Document: 1124
Page: 18
10/15/2014
1345377
101
merchantswho were unable, as a practical matter, to stop accepting Visa and

MasterCardfrom using surcharges to steer customers towards less-expensive
payment methods that did not carry the same fees. Customers, in turn, were left in
the dark about those costs that they were indirectly bearing in the form of higher
prices. This disabled both merchants and customers from exerting downward
competitive pressures on interchange fees. D.E. 1153 189-199.
Similarly, Visas and MasterCards anti-minimum purchase rules prevented
merchants from requiring a minimum cost before they would accept Visa and
MasterCard cards. In addition, their anti-discounting rules (or anti-discrimination
rules) prohibited merchants from offering discounts and other benefits for purchases
made with non-Visa and non-MasterCard products. See D.E. 1153 8 (defining
Anti-Steering Restraints as the No-Surcharge Rule; the No-Minimum Purchase
Rule; and the Networks so-called anti-discrimination rules).
B.
Eight Years of Hard-Fought Litigation
In June 2005, merchants filed the first of over 40 class-action complaints

alleging that Visa and MasterCard and their member banks conspired, inter alia, to
impose and fix the price of interchange fees in violation of the Sherman Act. The
class actions were consolidated later that year along with 19 individual cases before
Judge John Gleeson in the Eastern District of New York.
Case: 12-4671
Document: 1124
Page: 19
10/15/2014
1345377
101
From the outset, plaintiffs requested certification of both a damages class

under Rule 23(b)(3) and an injunctive-relief class under Rule 23(b)(2). See D.E.
317 (First Consolidated Amended Class Action Complaint) 97; D.E. 1153 108.
Plaintiffs sought both monetary damages to compensate them for the overcharges
caused by th[e] illegal conspiracy and equitable relief to protect themselves
against continuing and future harm. D.E. 317 at 1. Both types of relief were
premised on identical facts and evidence about the defendants uniform course of
anticompetitive conduct. And both types of relief were equally important in
ensuring an effective remedy. D.E. 1165 at 39.
The settlement that ultimately emerged was the result of eight years of hard-
fought litigation that consumed enormous resources and was closely monitored by
the district court. The parties went toe-to-toe over discovery, which commenced in
2005 and took more than five years. That process entailed more than 400
depositions, the production and review of more than 80 million pages of documents,
the exchange of 17 expert reports, and a full 32 days of expert deposition testimony.
SPA9. The parties expert reports covered nearly 5,000 pages.
The parties also vigorously disputed class certification. Plaintiffs moved for
certification of the Rule 23(b)(2) and 23(b)(3) classes in 2008. Defendants opposed
certification and challenged the opinions of plaintiffs class certification expert. The
Case: 12-4671
Document: 1124
Page: 20
10/15/2014
1345377
101
district court received briefing over nearly fourteen months and held oral argument
on class certification in November 2009.
The parties additionally engaged in multiple rounds of briefing on motions to
dismiss. After one round of briefing and argument on plaintiffs First Supplemental
Class Action Complaint (post-IPO period), the district court partially dismissed with
leave to re-plead. Defendants then launched a second round, moving to dismiss the
First Amended Supplemental Class Action Complaint, Second Consolidated
Amended Class Action Complaint, and Second Supplemental Class Action
Complaint. The district court heard oral argument on these motions in November
2009.
The parties also exchanged cross-motions for summary judgment. In 2011,
plaintiffs sought partial summary judgment and defendants sought judgment as to
the entirety of the case. The district court received briefing over five months and
held oral argument on the cross-motions in November 2011. The parties also filed
eight Daubert motions in conjunction with the summary judgment briefing.
C.
Industry Reforms During, and Due to, the Litigation
Plaintiffs litigation efforts brought public (and Department of Justice)

attention to the networks anticompetitive practices and spurred major industry
developments. Indeed, several reforms sought by plaintiffs were achieved outside
the courtroom, often as a direct result of this litigation. While those reforms marked
Case: 12-4671
Document: 1124
Page: 21
10/15/2014
1345377
101
a great accomplishment for merchants, they also injected delay and uncertainty into
the litigation process.
Corporate Restructurings. Shortly after the first class action was filed, and in
part to avoid ruinous antitrust liability, the networks abandoned their longstanding
joint venture structures. D.E. 1152 (First Amended Supplemental Class Action
Complaint) 149(d). MasterCarda joint venture for over four decades
restructured in 2006. Visaa joint venture since 1970followed in early 2008.
Both became publicly owned and operated, with the banks divesting their ownership
and relinquishing their board memberships and voting control over network rules.
This formal separation of the networks and banks, whereby the banks no longer
control the business decisions of Visa and MasterCard, marked a tectonic industry
shift. Cf. United States v. E. I. du Pont de Nemours & Co., 366 U.S. 316, 326 (1961)
(divestiture is the most drastic of antitrust remedies).
Durbin Amendment. Beginning in 2009, class counsel also became
significantly involved in developing and drafting legislative reforms of the networks
practices. Class counsel recommended to leading merchant groups that, instead of
seeking public-utility-like regulation of interchange fees, they focus on limiting fees
on debit card transactions to give merchants a low-cost alternative to which
merchants could eventually steer customers. The new strategy worked, and
Case: 12-4671
Document: 1124
Page: 22
10/15/2014
1345377
101
culminated in the Durbin Amendment to the 2010 Dodd-Frank Wall Street Reform
and Consumer Protection Act, Pub. L. No. 111-203, 124 Stat. 1376 (2010).
The Durbin Amendment limited the networks anti-discounting rules,
restricting their ability to inhibit the ability of any person to provide a discount or
in-kind incentive to steer consumers at the point of sale toward less-expensive
forms of payment. 15 U.S.C. 1693o-2(b)(2)(A). It also limited the networks anti-
minimum purchase rules, preventing them from inhibit[ing] the ability of any
person to set a minimum dollar value for the acceptance by that person of credit
cards. Id. 1693o-2(b)(3)(A). And, critically, the Durbin Amendment authorized
the Federal Reserve Board to cap interchange fees on Visa and MasterCard debit
card transactions to ensure that they are reasonable and proportional to the cost
incurred by the issuer. Id. 1693o-2(a)(3)(A). By limiting debit interchange fees,
Congress rendered debit cards a substantially lower-priced form of payment other
than cash to which merchants could steer consumers.
Department of Justice Consent Judgment. As early as 2006, DOJ and several
state attorneys general contacted class counsel and expressed interest in the
litigation. In 2008, at the urging of class counsel, DOJ opened an antitrust
investigation into Visas and MasterCards anti-steering restraints. After it issued a
Civil Investigative Demand on plaintiffs seeking all products of discovery relating
to the Anti-Steering Rules, class counsel granted extraordinary access to their
10
Case: 12-4671
Document: 1124
Page: 23
10/15/2014
1345377
101
document and deposition databases, work product, and advice over the next three
years. D.E. 2113-6 119. That assistance directly resulted in DOJ filing its own suit
against Visa and MasterCard.
As the district court later observed, the plaintiffs did not piggyback on
previous government actionindeed, the government piggybacked on their efforts.
SPA59. In 2011, DOJ entered a Consent Judgment against Visa and MasterCard that
secured critical modifications to the anti-discounting rules, enabling merchants to
offer discounts at both the brand level (e.g., discounts on cards other than Visa credit
cards) and product level (e.g., discounts on cards other than higher-cost Visa
Signature cards or MasterCard World Elite cards). In the Consent Judgment, Visa
and MasterCard also committed to providing free services to help merchants
determine the costs of accepting particular cards.
D.
Mediation and Settlement
1.
In the wake of these significant industry reforms, the completion of
discovery, and briefing and argument on class certification, dismissal, and summary
judgment, the parties agreed at an advanced stage of the litigation to settle their
claims. The settlement was the culmination of a painstakingly thorough and
inclusive four-year mediation process spearheaded by two of the most experienced
mediators in the country, Retired Chief Magistrate Judge Edward Infante and
Professor Eric Green.
11
Case: 12-4671
Document: 1124
Page: 24
10/15/2014
1345377
101
The parties jointly selected Judge Infante and Professor Green after months-
long negotiations. D.E. 2113-6 177. Between April 2008 and December 2011, the
parties met, either jointly or separately, with one or both of the mediators
approximately 45 times, and exchanged hundreds of calls and e-mails in an attempt
to make progress toward settlement. D.E. 2113-6 175, 181. The district court
(with assistance from the mediators) also held several multi-day settlement
conferences. Proposed class representativesincluding many current-objectors
were repeatedly invited to participate. They did participate, and were heard at length
by the mediators and the court. D.E. 1111-2 (Infante Decl.) 7; D.E. 1111-3 (Green
Decl.) 22-28.
The mediators presented a settlement proposal in December 2011. After
weeks of discussions, the partiesincluding many current-objectorsaccepted that
proposal. By then, all parties were intimately familiar with the strengths and
weaknesses of their claims. Class counsel concluded that accepting the mediators
proposal was preferable to the only alternative, which was many more years of
litigation while merchants continued to be hamstrung by the no surcharge rules of
Visa and MasterCard and remaining anti-steering rules. D.E. 2113-6 185. In class
counsels view, settlement was particularly attractive when compared to what
was reasonably likely to be obtained by injunction in a trial before Judge Gleeson.
Id. Class counsel recognized that even after many more years of delay and further
12
Case: 12-4671
Document: 1124
Page: 25
10/15/2014
1345377
101
litigation, victory was by no means assured; plaintiffs faced a number of potential

risks in their efforts to establish liability and damages and maintain a class action.
By February 2012, the partiesagain, including many current-objectors
agreed to negotiate toward a final settlement through the process laid out by the
mediators and the court. D.E. 1111-2 8; D.E. 1111-3 29. The mediators guided
months of careful debate over nearly every settlement term, with all parties
recognizing the far-reaching impact the settlement would have on the costs and
mechanics of payment card acceptance.
In July 2012, the partiesstill including many current-objectorsreached a
final settlement and agreed to set[] out the parties binding obligation to enter into
the terms outlined in a Memorandum of Understanding. D.E. 1588. The district
court then tabled all pending motions for relief (including motions concerning
discovery, class certification, dismissal, summary judgment, and the preclusion of
expert opinion testimony). 7/17/2012 Order.
After the Memorandum of Understanding was filed, however, in the midst of
an aggressive objection campaign, several then-class representatives (now-
objectors) reversed course. Class counsel emphatically did not fire[] their clients,
Merchant Appellants Br. (MA-Br.) 70, as objectors contend. To the contrary, class
counsel moved to withdraw as their counsel only after those objectors withdrew their
binding support for the settlement and often only after new counsel appeared on
13
Case: 12-4671
Document: 1124
Page: 26
10/15/2014
1345377
101
their behalf. The final settlement, virtually identical to the July 2012 agreement, was
executed in October 2012.
2.
The settlement secures substantial relief for plaintiffs. It achieves the
largest-ever cash relief in an antitrust casetwo funds totaling an estimated $7.25

billion1 (before opt-out reductions)to compensate all persons, businesses, and
other entities that have accepted Visa or MasterCard cards in the United States from
January 1, 2004 to the preliminary settlement approval date (November 27, 2012).
SPA118 (Settlement 2(a)). It also achieves an unprecedented bundle of network
rule reforms for all merchants that accept Visa and MasterCard cards in the United
States as of November 27, 2012 or after. SPA118 (Settlement 2(b)).
Chief among the rule reforms is the lifting of the networks prohibition on
surcharging, SPA141-49 (Settlement 42), which had been the most potent anti-
steering restraint. See D.E. 317 238 (The Anti-Steering Restraints (and
particularly the No-Surcharge Rule) are anticompetitive vertical restraints.)
(emphasis added). In class counsels view, [w]inning the surcharging tool is the
most consequential and empowering development yet in the long battle U.S.
merchants have waged to counter the anticompetitive practices and legacies in the
The $7.25 billion is composed of two funds: a $6.05 billion cash settlement and
an estimated $1.2 billion fund based on a holdback of ten basis points in interchange
fee payments by class members during an eight-month period after the opt-out
period. SPA121-23 (Settlement 10-13).
14
Case: 12-4671
Document: 1124
Page: 27
10/15/2014
1345377
101
credit-card industry. D.E. 2113-6 195. For the first time, the networks cannot
prevent merchants from imposing surcharges on Visa and MasterCard transactions
at the brand or product level to steer customers towards lower-cost payment
methods. Because the amount of any surcharge is disclosed before it is incurred (and
on the receipt after purchase), customers are also educated about the costs of credit
card acceptance and may in turn opt for cheaper payment methods. Indeed, even the
threat of surcharging helps incentivize networks to moderate or lower their fees to
stay competitive.
Additional rule reforms further enhance the surcharging tool. The networks
are now required to negotiate in good faith for better rates with bona fide merchant
buying groups; permit merchants to use different acceptance strategies at different
outlets; and lock in the discounting, minimum price, and other reforms of the Durbin
Amendment and DOJ Consent Judgment until July 2021. SPA140-50, 153-64
(Settlement 41, 42(g), 43, 54, 55(g), 56).
Both mediators attested that settlement negotiations were fair, adversarial,
and always conducted at arms-length. D.E. 1111-2 12; accord D.E. 1111-3 33.
Both lauded counsel for zealously represent[ing] the interests of their clients, D.E.
1111-2 12, and giving their best professional effort, D.E. 1111-3 33. And both
declared that the settlement terms were fair, reasonable, and adequate, taking into
account the risks, strengths and weaknesses of [the parties] respective positions on
15
Case: 12-4671
Document: 1124
Page: 28
10/15/2014
1345377
101
the substantive issues of the case and the risks and costs of continued litigation.
D.E. 1111-3 33; accord D.E. 1111-2 13.
E.
Settlement Review and Approval
The district court approved the settlement after an exhaustive review of both
the settlement process and the substantive terms. The court held a preliminary
approval hearing at which it heard from objectors and reviewed supporting evidence.
After granting preliminary approval and provisionally certifying the (b)(2) and (b)(3)
settlement classes, the court invited objectors to file written submissions and appear
at the final approval hearing. D.E. 1745 21. The court also appointed Dr. Alan
Sykes of New York University School of Law to offer independent analysis of the
economic issues raised by the settlement. And, at the final approval hearing, the
court heard again from objectors.
The district court issued a 55-page decision granting final approval of the
settlement. See In re Payment Card Interchange Fee Antitrust Litig., 986 F. Supp.
2d 207 (E.D.N.Y. 2013). The court first deemed the settlement procedurally fair,
concluding that the record demonstrates beyond any reasonable doubt that the
negotiations were adversarial and conducted at arms length by extremely capable
counsel. SPA21 (emphasis added). It emphasized that there is no indication that
the Settlement Agreement is the product of collusion, and that the negotiation
process fairly protected the interests of the settlement class. SPA21-22. Even many
16
Case: 12-4671
Document: 1124
Page: 29
10/15/2014
1345377
101
objectors were deeply involved in the settlement negotiations and mediation, and
indeed accepted the mediators proposal that outlined the key components of what
became the Settlement. SPA21.
The district court then found that the settlement fell well within the range of
reasonableness under the multi-factor standard of Detroit v. Grinnell Corp., 495
F.2d 462, 463 (2d Cir. 1974). The court noted that the advanced stage of litigation
gave all involved a more than adequate basis for assessing the claims after more
than eight years of full-throttle litigation. SPA9, 11, 25. It emphasized that
further litigation would entail lengthy delays. Numerous motions remain[ed]
pending, the class certification motion would inevitably entail interlocutory
review by the Second Circuit, a trial would take several months, and the losing
parties would likely appeal any adverse jury verdicts. SPA22. And, even after
many more years of litigation, the class would still face the prospect of uncertain
relief. SPA35. By contrast, the settlement affords a certain and largely immediate
recovery, enabling class members to take advantage of rules changes now and
receive significant monetary compensation in the near future. SPA23.
The district court thoroughly considered and rejected the concerns of
objectors, who comprised 0.05%, or one-twentieth of 1%, of the estimated class.
The court lamented that [t]he behavior of a small number of objectors has
threatened to undermine the efforts of the others with needless hyperbole,
17
Case: 12-4671
Document: 1124
Page: 30
10/15/2014
1345377
101
vitriol, and the improper use of websites that disseminated false and misleading
information. SPA13-14, 23.2 The objectors, moreover, appeared to assum[e] that
a complete victory on the merits is a foregone conclusion. SPA26. In fact, a wide
range of outstanding issues created substantial risks and uncertainties if litigation
were to proceed. SPA25-26. For instance, objectors assume that default
interchange is inherently illegal, but in reality it is a very complicated issue. SPA29.
Objectors similarly assumed the illegality of the Honor-all-Cards rule, even though
there was record evidence and analogous caselaw suggesting that this rule would be
found procompetitive under the Rule of Reason. SPA31. And plaintiffs faced
additional complexities in proving damages to the jury and risks associated
with class certification. SPA32-34.
Objectors also underestimated the significance of the Rule 23(b)(2) relief.
SPA25. The district court declared surcharging an indisputably procompetitive
development that has the potential to alter the very core of the problem this lawsuit
was brought to challenge. SPA35-36. For merchants, the surcharging relief
removes a central piece of the problem. SPA37. Merchants can now provide
The district court observed that 90% of the objections [were generated] on
boilerplate forms downloaded from the websites, which were established for the
precise purpose of drumming up objections and opt-outs and misled merchants
about their options. SPA13, 23. The court had previously been forced to issue
injunctive relief to address this misinformation campaign and came close to holding
certain entities in contempt. SPA14.
18
Case: 12-4671
Document: 1124
Page: 31
10/15/2014
1345377
101
customers with clear signals of what it costs the merchant to accept a particular
card, giving them valuable leverage to encourage the use of cheaper payment
methods and brands. SPA37. Customers, meanwhile, have a choice between
surcharged payment methods and lower-cost payment methods. And networks will
face incentives to moderate or lower their interchange fees to avoid being
disfavored. SPA37. As the court explained, [e]ven if the objectors are right in
contending that additional dominoes must fall before the alleged anticompetitive
behavior of Visa and MasterCard is eradicatedwhether independent constraints
on surcharging posed by state laws or third partiesthose dominoes will have to
fall in other forums. SPA18.
The district court similarly dismissed criticisms of the release. Consistent
with precedent, the settlement releases only claims that are or could have been
alleged based on the identical factual predicate of the claims in this case. SPA46.
Though the release appropriately limit[s] future damages claims based on the pre-
settlement conduct of the networks, it does not release claims based on new rules
or new conduct or a reversion to the pre-settlement rules. SPA46. And, again,
objectors underestimated the significance of the (b)(2) relief by painting the release
as giving away valuable claims for nothing. SPA45.
The district court squarely rejected objectors contention that the (b)(2) class
should have afforded members a due process right to opt out. SPA46. The (b)(2)
19
Case: 12-4671
Document: 1124
Page: 32
10/15/2014
1345377
101
claims sought injunctive relief from [a] bundle of network rules that are
precisely the proper subject of a (b)(2) class from which no opt outs are
permitted. SPA46. The court also rejected criticisms of the settlement notice.
Separately, the district court issued a 17-page decision approving a $544.8
million attorneys fee9.56% of the damages fund, after opt-out reductionsas a
reasonable overall fee in light of the unique size, duration, complexity, and
relief of this case. SPA69-70. See In re Payment Card Interchange Fee Antitrust
Litig., 991 F. Supp. 2d 437 (E.D.N.Y. 2014). Applying the multi-factor standard of
Goldberger v. Integrated Resources, 209 F.3d 43, 4748 (2d Cir. 2000), the court
again emphasized that the substantial injunctive relief may very likely exceed the
value of the monetary relief in the long run. SPA67. This far-reaching relief
confirmed the courts judgment that class counsel litigated the case with skill and
tenacity and that the settlement would not exist but for counsels assumption of
risk and extraordinary efforts. SPA59, 61.
To calculate the fee, the court used a sliding scale that awarded counsel
diminishing percentages of the settlement fund as the fund increased. SPA69. It
further confirmed that the lodestar multiplier was comparable to multipliers in other
large, complex cases. SPA70.
20
Case: 12-4671
Document: 1124
Page: 33
10/15/2014
1345377
101
SUMMARY OF ARGUMENT
I.
The structure of the settlement classes, including the uncontroversial
coexistence of Rule 23(b)(2) and (b)(3) classes in a single case, fully satisfied
Rule 23. The (b)(2) class was a paradigmatic (b)(2) class seeking exclusively
injunctive, indivisible reliefnamely, the elimination or modification of nationwide
network rules that apply generally to all card-accepting merchants. Precisely
because that class sought indivisible relief, no opt-outs from the class were feasible,
let alone necessary. The (b)(3) class, in contrast, sought and obtained substantial
monetary relief, and fully comported with the (b)(3) opt-out and notice requirements.
Although objectors seek to characterize the inclusion of a non-opt-out (b)(2)
class and opt-out (b)(3) class in one action as anomalous, such arrangements are
common in situationslike this onewhere both injunctive relief and damages are
needed to provide an adequate remedy. Far from evading any legal requirements of
(b)(2) and (b)(3), the class structure hewed carefully to each subsection and fully
heeded the dictates of Wal-Mart v. Dukes, 131 S. Ct. 2541 (2011). Class certification
was therefore proper under a straightforward application of Rule 23.
Objectors efforts to conjure up a due process problem from that class
structure are wholly without merit. Objectors concede that the (b)(2) claims for relief
are exclusively injunctive, but argue that opt-out rights must be provided because
the settlement bars certain future damages claims by (b)(2) class members against
21
Case: 12-4671
Document: 1124
Page: 34
10/15/2014
1345377
101
defendants. But that is neither unusual nor legally problematic. The propriety of
(b)(2) certification turns only on the claims for relief, which are exclusively
injunctive and fully compliant with (b)(2). That a settlement releases future
challenges to the agreed-upon, going-forward conductincluding hypothetical and
uncertain future damages claimsis not surprising. Such releases are common
inducements for defendants to settle, and appropriate injunctive relief should limit
future damages claims based on the new regime, since it is designed to address the
challenged conduct prospectively. And, of course, the propriety of both the
injunctive relief and any associated releases will be properly considered as part of
the analysis of the settlements fairness under Rule 23(e).
There is absolutely no basis for a novel bright-line rule that classes certified
under (b)(2) can never, in a settlement release, foreclose a hypothetical future claim
seeking damages. Such a rule would prove unworkable and run afoul of the Rules
Enabling Act. Unsurprisingly, objectors cite no caselaw whatsoever for the
proposition that a (b)(2) class is no longer a (b)(2) class solely because potential
future challenges to the agreed-upon conduct are foreclosed as part of a
comprehensive settlement.
Objectors efforts to manufacture an adequacy-of-representation problem fare
no better. As the district court correctly concluded, class representatives and counsel
were adequate representatives of both the (b)(2) and (b)(3) classes. The structural
22
Case: 12-4671
Document: 1124
Page: 35
10/15/2014
1345377
101
assurances of fair and adequate representation in the settlement negotiations were at

their apex given the involvement of Judge Gleeson, Magistrate Judge Orenstein, and
the two highly-regarded mediators. There is, moreover, no conflict of interest, much
less a fundamental conflict going to the heart of the litigation, between the (b)(2) and
(b)(3) classes. To the contrary, the class memberships overlap almost entirely; they
exist as two classes by virtue of Rule 23, and not by virtue of different claims,
different facts, or antagonistic interests. The substantial (b)(2) relief and massive
(b)(3) damages fund confirm only that neither class interest was left out in the cold,
and both classes reaped significant benefits from class counsels zealous advocacy.
II.
The district courts determination that the settlement as a whole fell
within a range of reasonableness, Grinnell, 495 F.2d at 463, was an appropriate

exercise of its broad discretion. The notion that the largest-ever cash relief in an
antitrust class action settlement was reasonable hardly requires extended
comment. And, as the district court recognized, the injunctive reforms of the
networks longstanding merchant restraints may prove even more valuable in the
long run.
Chief among these reforms is the removal of anti-surcharging restraints,
which for decades prevented merchants from imposing surcharges on Visa and
MasterCard transactions to steer customers to less-costly methods of payment and
incentivize the networks to lower their interchange fees. Other reforms, such as the
23
Case: 12-4671
Document: 1124
Page: 36
10/15/2014
1345377
101
buying group provision and all-outlets provision, will only enhance merchants new
surcharging opportunities and exert further downward pressures on interchange fees.
The release in exchange for that injunctive relief, moreover, is a standard-
form release that courts have repeatedly approved. Objectors insinuate that the
settlement releases certain future challenges in violation of due process or public
policy. But as the district court found and as the parties agreed below, the releases
merely foreclose challenges to the going-forward regime agreed upon in the
settlement and, in any event, must be interpreted to reflect, not violate, due process
limits. Objectors seek no less than an improper advisory opinion to predetermine
issues that the release does not even raise.
III. The district courts determination that an attorneys fee of 9.56% of the
damages fund was reasonable under the circumstances, Goldberger, 209 F.3d at
47, was likewise an appropriate exercise of its broad discretion. The district court
grounded its analysis in the unique size, duration, complexity, and relief of
this case, and appropriately lauded class counsel and the nearly 60 additional law
firms that worked on this case over eight years for their skill and tenacity in
achieving the significant success of the settlement. SPA56, 61. Objectors dispute
the district courts assessment of counsels performance and the settlement relief,
and quibble with the specific percentages used in its sliding-scale percentage
24
Case: 12-4671
Document: 1124
Page: 37
10/15/2014
1345377
101
calculation of the fee amount. But those arguments merely rehash their meritless
criticisms of the settlement as substantively unreasonable.
IV. Finally, the district courts determination that the settlement notice was
reasonable[], Soberal-Perez v. Heckler, 717 F.2d 36, 43 (2d Cir. 1983), was also
an appropriate exercise of its broad discretion. The notice need only fairly apprise
prospective class members of the terms of the settlement and of options available to
them moving forward.
The notice herewhich described the litigation and
settlement terms, quoted the releases, described the request for attorneys fees, and
explained the procedures for filing objections and opting outreadily met that
standard. Objectors posit various misstatements and omissions in the notice, but the
district court considered and rejected those arguments several times. The district
courts judgment should be affirmed in all respects.
STANDARD OF REVIEW
This Court will disturb a judicially-approved settlement only when an
objector has made a clear showing that the district court has abused its discretion.
DAmato, 236 F.3d at 85 (emphasis added). A court abuses its discretion only when
its decision rests on an error of law or a clearly erroneous factual finding, or when
its decision cannot be located within the range of permissible decisions. Charron
v. Wiener, 731 F.3d 241, 247 (2d Cir. 2013). This considerable deference is rooted
in a recognition that the district court is uniquely exposed to the litigants, and their
25
Case: 12-4671
Document: 1124
Page: 38
10/15/2014
1345377
101
strategies, positions and proofs. Joel A. v. Giuliani, 218 F.3d 132, 139 (2d Cir.
2000).
Similarly, this Court will disturb a district courts determination on class
certification, as well as its rulings that individual Rule 23 requirements have been
met, only upon an abuse of discretion. In re IPO Sec. Litig., 471 F.3d 24, 3132 (2d
Cir. 2006). Where the district court has granted class certification, this Court accords
noticeably more deference than when we review a denial. In re Salomon Analyst
Metromedia Litig., 544 F.3d 474, 480 (2d Cir. 2008).
This Court likewise reviews attorneys fee awards, and the form and content
of notice to class members, for an abuse of discretion. See Goldberger, 209 F.3d at
47-48; Masters v. Wilhelmina Model Agency, 473 F.3d 423, 438 (2d Cir. 2007).
ARGUMENT
I.
The Structure Of The Settlement Classes Conforms Precisely To Bedrock

Certification Requirements Under Rule 23.
From the inception of this litigationwith the support of all class
representatives, including many current-objectorsplaintiffs consistently sought

certification of two classes: a class seeking prospective injunctive relief under Rule
23(b)(2), and a class seeking retrospective monetary relief under Rule 23(b)(3).
Those classes readily satisfied their respective subsections of Rule 23. The (b)(2)
class sought exclusively indivisible, generally applicable injunctive relief and fully
complied with all requirements for a (b)(2) class. The (b)(3) class, in turn, sought
26
Case: 12-4671
Document: 1124
Page: 39
10/15/2014
1345377
101
monetary relief, and fully complied with all requirements of Rule 23(b)(3), including
notice to the class and an opportunity to opt-out. The district court thus correctly
concluded that certification of these classes was proper.
Objectors attempt to channel their substantive objections to the settlement
issues that are properly addressed in a review of the settlements fairness under Rule
23(e), see infra Part IIinto various attacks on the structure of the classes. But
as Rule 23(a) and (b) arguments, the objections are simply misplaced. The district
court considered and rejected all of objectors arguments and correctly concluded
that this case involves a textbook case for certification under Rule 23.
A.
The Rule 23(b)(2) Class Unquestionably Brought Claims For

Indivisible Relief.
1.
Rule 23(b)(2) allows class treatment when the alleged wrongdoer has
acted or refused to act on grounds that apply generally to the class, so that final
injunctive relief or corresponding declaratory relief is appropriate respecting the
class as a whole. Fed. R. Civ. P. 23(b)(2); see also 7AA Wright & Miller, Federal
Practice and Procedure 1775 (3d ed. 1998) ([T]wo basic factors must be
present (1) the opposing partys conduct or refusal to act must be generally
applicable to the class and (2) final injunctive or corresponding declaratory relief
must be requested for the class). It is well-settled that the key to the (b)(2) class
is the indivisible nature of the injunctive or declaratory remedy warrantedthe
27
Case: 12-4671
Document: 1124
Page: 40
10/15/2014
1345377
101
notion that the conduct is such that it can be enjoined or declared unlawful only as
to all of the class members or as to none of them. Dukes, 131 S. Ct. at 2557.
That characteristic does not require class members to be in an identical
situation, Rich v. Martin Marietta, 522 F.2d 333, 340 (10th Cir. 1975), or have
suffered identical injuries, Parsons v. Ryan, 754 F.3d 657, 688 (9th Cir. 2014), or
indeed even be aggrieved by or desire to challenge defendants conduct. Wright
& Miller 1775; see also Fed. R. Civ. P. 23 advisory committee notes (1966)
(Action or inaction is directed to a class within the meaning of this subdivision even
if it has taken effect or is threatened only as to one or a few members of the class).
Instead, all that is required to proceed as a (b)(2) class is that the relief sought must
perforce affect the entire class. Dukes, 131 S. Ct. at 2558.
As the Supreme Court has underscored, [t]he procedural protections
attending the (b)(3) classpredominance, superiority, mandatory notice, and the
right to opt outare unnecessary to a (b)(2) class precisely because [w]hen a
class seeks an indivisible injunction benefitting all its members at once, there is no
reason to undertake a case-specific inquiry into whether class issues predominate or
whether class action is a superior method of adjudicating the dispute. Id. at 2558.
Those characteristics are simply self-evident. Indeed, even an individual suit
enjoining defendants nationwide operations would affect all members of the class;
28
Case: 12-4671
Document: 1124
Page: 41
10/15/2014
1345377
101
thus the requirements of Rule 23(a) and (b)(2) provide class members with additional
protection, and the (b)(3) factors are unnecessary.
This case presents a paradigmatic use of Rule 23(b)(2). The (b)(2) class
consists of merchants that accept any Visa-Branded Cards and/or MasterCard-
Branded Cards in the United States. SPA118. Those merchants challenged and
sought to change nationwide network rules that apply to all merchants that accept
Visa and MasterCard. See Wright & Miller 1775 (What is necessary is that the
challenged conduct be premised on a ground that is applicable to the entire
class.). The analysis of the legality of those network rules under the Sherman Act
and Clayton Act is exactly the same for each and every merchant that is subject to
the rules. In short, plaintiffs antitrust claims allege that Visa and MasterCard have
acted or refused to act on grounds that apply generally to the class. Fed. R. Civ.
P. 23(b)(2).
Moreover, the (b)(2) class sought injunctive reliefstructural rule reforms
that was indivisible and generally applicable to all merchants. When the challenged
nationwide network rules changed based on the agreed-upon injunction, the relief
necessarily affected all members of the class. Among other relief, the plaintiffs
sought to eliminate or modify the anti-surcharging rules, the default interchange
rules, and the Honor-all-Cards rules for all merchants that accept Visa and
MasterCard. A complaint that seeks to enjoin uniform, nationwide merchant
29
Case: 12-4671
Document: 1124
Page: 42
10/15/2014
1345377
101
restraints necessarily redesigns the relationship between each merchant and the
networks in precisely the same manner. SPA52.
Because this case involves challenges to nationwide, generally applicable
network rules, it is far afield from a scenario in which each individual class member
would be entitled to a different injunction against the defendant. Dukes, 131 S.
Ct. at 2557; see also In re IKO Roofing Prods. Litig., 757 F.3d 599, 602 (7th Cir.
2014) (unlike in Dukes, [i]n a suit alleging a defect common to all instances of a
consumer product the conduct does not differ). Because Visa and MasterCard
are nationwide networks each with uniform rules, it would have been impossible to
have a patchwork of injunctions that enjoined the challenged network rules only with
respect to certain merchants. The district court thus had little difficulty concluding
that this is precisely the kind of case for which Rule 23(b)(2) was intended.
SPA46; see also Fed. R. Civ. P. 23 advisory committee notes (1996) (citing example
of a (b)(2) class of purchasers, say retailers of a given description, against a seller
alleged to have undertaken to sell to that class at [discriminatory] prices).
Making the application of (b)(2) even more straightforward, the (b)(2) claims
here consist exclusively of claims for injunctive relief. In Dukes, the Supreme Court
made clear that claims for monetary relief cannot be litigated in a (b)(2) class
alongside bona fide claims for injunctive or declaratory relief, at least where the
monetary relief is not incidental to the injunctive or declaratory relief. 131 S. Ct.
30
Case: 12-4671
Document: 1124
Page: 43
10/15/2014
1345377
101
at 2557. Because the Supreme Court expressly left open the possibility that some
truly incidental monetary relief might still be allowed in a (b)(2) action, objectors
overstate matters by asserting, based on Dukes, that Rule 23(b)(2) applies only
when the case consists exclusively of common claims. MA-Br. 34. But in all
events, the (b)(2) class at issue here seeks only indivisible injunctive relief, which
Dukes confirmed unquestionably belongs in Rule 23(b)(2).
2.
Objectors assertion that the (b)(2) class is not cohesive is badly
misplaced. MA-Br. 44-66. As an initial matter, class cohesion is not expressly

required by the text of Rule 23(b)(2) or any Supreme Court precedent interpreting
that rule. But even if cohesion is required, it is shorthand for the need for generally
applicable, class-wide injunctive relief, which is amply satisfied here. This Court
has observed, in the few instances where it used the term in passing, that a Rule
23(b)(2) class seeking declaratory and injunctive relief is cohesive by nature.
Handschu v. Special Servs. Div., 787 F.2d 828, 833 (2d Cir. 1986). That is, [w]here
class-wide injunctive or declaratory relief is sought in a (b)(2) class action for an
alleged group harm, there is a presumption of cohesion and unity between absent
class members and the class representatives. Robinson v. Metro-North, 267 F.3d
147, 165 (2d Cir. 2001).
Objectors do not come close to overcoming this presumption of cohesion.
To the contrary, this effort to reform the practices employed in a nationwide network
31
Case: 12-4671
Document: 1124
Page: 44
10/15/2014
1345377
101
is the very paradigm of a (b)(2) case. Objectors core argument is that the (b)(2)
class was not cohesive because it was very large, and included merchants that had
varying interests in the relief ultimately obtained. MA-Br. 53-61. But that is
wrong for a number of reasons. Whatever differences there may be among class
members in terms of their size[s] and business models, MA-Br. 53-54, they are
all subject to the defendants generally applicable network rules and practices, which
will be impacted on a nationwide basis by the going-forward relief. That is all the
cohesion Rule 23(b)(2) would demand.
Objectors cite a smattering of decisions from other jurisdictions rejecting
(b)(2) classes as non-cohesive, but the facts of those cases only underscore the lack
of any cohesion issue here. Most of those decisions arose in the mass tort context,
where courts employ[] Rule 23(b)(2) sparingly because factual differences
among individual class members may affect critical elements of plaintiffs claims,
such as proximate causation, reliance and defendants affirmative defenses.
1 McLaughlin on Class Actions 5:15 (10th ed.) (emphasis added). For example,
the Eighth Circuit decertified a (b)(2) class of artificial heart valve patients seeking
a medical-monitoring injunction because each plaintiffs need (or lack of need) for
medical monitoring is highly individualized, based on the patients medical
history and risk factors, the different elements triggering culpability, and other
considerations. In re St. Jude Medical, 425 F.3d 1116, 1122 (8th Cir. 2005); see also
32
Case: 12-4671
Document: 1124
Page: 45
10/15/2014
1345377
101
Barnes v. Am. Tobacco, 161 F.3d 127, 146 (3d Cir. 1998) (denying certification of
medical-monitoring class in suit against cigarette manufacturers).
Here, in contrast, the injunctive reforms sought are generally applicable, and
the same rules and rule modifications apply to all merchants. And objectors identify
no way in which the purported differences among class members would affect the
elements of a claim for that injunctive relief under the Sherman Act and Clayton Act.
The identity of the particular merchant-plaintiff would have no bearing whatsoever
on whether the challenged network rules were the product of unlawful restraints of
trade.
Objectors also claim that the (b)(2) class was not cohesive because some states
would prohibit merchants from surcharging credit card transactions even if
surcharging were permitted by the terms of the settlement. In support of that
argument, objectors cite Amchem Prods. v. Windsor, 521 U.S. 591 (1997), for the
proposition that variations in state law can undermine (b)(2) cohesion. But
Amchem involved a class seeking monetary relief under Rule 23(b)(3), and the
differing state laws went to the very availability of plaintiffs causes of action (and
thus whether common issues predominated among the class). As the Court
explained, state law varied widely on such critical issues as viability of [exposure-
33
Case: 12-4671
Document: 1124
Page: 46
10/15/2014
1345377
101
only] claims [and] availability of causes of action for medical monitoring. Id. at
609-10.3
There is no comparable concern in this case or in (b)(2) classes more broadly.
State anti-surcharging laws hardly immunize defendants conduct under the federal
antitrust laws, and at most limit the extent to which some class members will benefit
from the relief obtained. But that does not change the reality that defendants rules
and conduct will change on a nationwide basis in a manner applicable to all class
members. The fact that some class members may not be able to take full advantage
of the change because of independent state-law obstacles to changing the class
members conduct does not create a Rule 23(b)(2) problem. See 2 Newberg on Class
Actions 6:15 (4th ed. 2002) (That not all members of the class may seek or desire
the same relief, or may otherwise have disparate interests, will not bar the
certifying of a class action seeking injunctive and declaratory relief).4
Objectors also cite In re AIG Sec. Litig., 689 F.3d 229, 243 (2d Cir. 2012), but
that case merely cited Amchem and noted that the district court should address in the
first instance whether variations in state law might cause class members interests
to diverge. Here, Judge Gleeson fully considered such issues and concluded that
any state-by-state variation in surcharging rules was insufficient to defeat class
certification. SPA38-41.
4
Related speculation that health insurers might be actually or effectively

prevented from surcharging by federal and state regulationsincluding wholly
hypothetical effects of the Affordable Care Acts Medical Loss Ratio ruleslikewise
has no bearing on Rule 23(b)(2) cohesion in the relevant sense. See Blue Cross Br.
18-21.
34
Case: 12-4671
Document: 1124
Page: 47
10/15/2014
1345377
101
Objectors, moreover, confuse an expectation of some minimal cohesion with

a standard of equal benefit or after-the-fact satisfaction from the relief obtained.
MA-Br. 58-66; cf. Amchem, 521 U.S. at 623 (noting that (b)(3) predominance inquiry
trains on the legal or factual questions that preexist any settlement) (emphasis
added). There is absolutely no support for such an impossible-to-satisfy standard.
A (b)(2) class does not unravel merely because certain class members are more
sanguine about, or realize greate[r] savings from, the relief obtained. MA-Br.
58, 62. At bottom, objectors so-called cohesion arguments are just challenges to
the substantive fairness of the settlement rather than the propriety of class
certification. Any concerns about whether certain class members received enough
relief from the ultimate settlement, see MA-Br. 58-61, are certainly relevant to
whether the settlement was fair, reasonable, and adequate under Rule 23(e). But
those concerns do not change the fact that Visa and MasterCard applied the
challenged network rules to all merchants, and that any modifications to the network
rules effectuated by this litigation would apply to each and every merchant that
accepts Visa and MasterCard.
* * *
In short, this class falls in the heartland of Rule 23(b)(2). [A]ll the members
of the injunctive relief class were subject to the same rules, the relief afforded by
that class is a change to those rules, and all members of the (b)(2) class sought
35
Case: 12-4671
Document: 1124
Page: 48
10/15/2014
1345377
101
injunctive relief and injunctive relief alone. SPA53 n.20. The district court did not
abuse its discretion in finding the criteria for Rule 23(b)(2) certification satisfied.
B.
The Rule 23(b)(2) Class Required No Opt-Out Rights; All Claims

for Individualized Monetary Relief Were Separately and Properly
Certified Under Rule 23(b)(3).
1.
Precisely because the hallmark of a (b)(2) class is a claim for indivisible
injunctive or declaratory relief, (b)(2) classes provide no opportunity for ... class
members to opt out, and do[] not even oblige the district court to afford them notice
of the action. Dukes, 131 S. Ct. at 2558. Since the relief obtained through a (b)(2)
class action isby definitiongenerally applicable to all class members, an
individual cannot request exclusion and pursue relief individually. As the district
court explained, [i]f merchants could opt out of the (b)(2) class, they would reap
the benefits of that relief anyway. SPA46; see also Jefferson v. Ingersoll, 195 F.3d
894, 897 (7th Cir. 1999) (individual suits would confound the interest of other
plaintiffs when an injunction affects everyone alike).
Indeed, [t]he procedural protections attending the (b)(3) class are
inapplicable precisely because [w]hen a class seeks an indivisible injunction
benefitting all its members at once, there is no reason to undertake a case-specific
inquiry into whether class issues predominate or whether class action is a superior
method of adjudicating the dispute. Dukes, 131 S. Ct. at 2558. As the Rule 23
Advisory Committee explained, given the characteristics of the [(b)(2)] class,
36
Case: 12-4671
Document: 1124
Page: 49
10/15/2014
1345377
101
there is no right to request exclusion. Fed. R. Civ. P. 23 advisory committee notes

(2003). Requiring notice would not only be pointless but counterproductive,
creating the risk that notice costs may deter the pursuit of class relief and thereby
crippl[ing] actions that do not seek damages. Id. Whereas the opt-out right in
(b)(3) classes follows directly from the ability to pursue individual claims for
damages, asserting an opt-out right in a (b)(2) class is a non sequitur.
2.
The district court fully understood these basic principles and certified
both a valid (b)(2) mandatory class and an equally valid (b)(3) opt-out class.5 As the
court easily concluded, [e]ach class satisfies its respective subsection of Rule
23(b) and certification is proper. SPA53 n.20.
Including two valid classes in a single action is neither remarkable nor
problematic. See Fed. R. Civ. P. 23 advisory committee notes (2003) (noting notice
requirements for the (b)(3) class only where a Rule 23(b)(3) class is certified in
conjunction with a (b)(2) class). Two rights do not somehow make a wrong.
[W]here injunctive relief and damages are both important components of the relief
requested, court[s] have regularly certified an injunctive class under Rule 23(b)(2)
and a damages class under Rule 23(b)(3) in the same action. In re NASDAQ
A number of merchants chose to exercise their right to opt-out of the (b)(3) class.
Those plaintiffs individual claims for money damages remain pending before the
district court.
37
Case: 12-4671
Document: 1124
Page: 50
10/15/2014
1345377
101
Antitrust Litig., 169 F.R.D. 493, 515 (S.D.N.Y. 1996); see also Gooch v. Life
Investors Ins., 672 F.3d 402, 428-29 (6th Cir. 2012) (appropriate to certify distinct
(b)(2) and (b)(3) classes, where (b)(2) relief is a separable and distinct type of
relief).6
Here, from the outset of the litigation, plaintiffs requested certification of both
a (b)(3) class for monetary damages to compensate them for overcharges and
a (b)(2) class for equitable relief to protect themselves against continuing and future
harm. D.E. 317 at 1. The damages provided relief for excessive interchange fees
already paid, while the injunctive relief addressed defendants anticompetitive
practices going forward. Plaintiffs consistently maintainedlong before settlement
even entered the picturethat the injunctive relief was as important as the
damages. D.E. 1165 (Pls. Mot. for Class Cert.) at 39; see also Velez v. Novartis,
244 F.R.D. 243, 271 (S.D.N.Y. 2007) (If Plaintiffs prevail on the merits it would
serve little purpose to award money damages for discrimination without addressing
the institutional structure that perpetuates it.).
E.g., Huyer v. Wells Fargo, 295 F.R.D. 332, 345 (S.D. Iowa 2013); Bristol v.
Louisiana-Pacific, 916 F. Supp. 2d 357, 370 (W.D.N.Y. 2013); Stinson v. City of N.Y.,
282 F.R.D. 360, 381 (S.D.N.Y. 2012); In re Vitamin C Antitrust Litig., 279 F.R.D. 90
(E.D.N.Y. 2012); Easterling v. Dept of Corr., 278 F.R.D. 41 (D. Conn. 2011);
Jermyn v. Best Buy, 276 F.R.D. 167 (S.D.N.Y. 2011).
38
Case: 12-4671
Document: 1124
Page: 51
10/15/2014
1345377
101
The district court agreed, suggesting that a damages-only award would permit
the networks merchant restraints [to], in effect, place the merchants back where
they started, thus allowing the networks to simply recoup[] any associated lost
revenues by tinkering with other fees. SPA16. Injunctive relief targeted at those
merchant restraints was therefore crucial because it would avert such
circumvention and permit merchants themselves to exert competitive pressures on
interchange fees. SPA18.
Objectors are thus flatly wrong to characterize the structure of the classes as
an artificial contrivance that inverted the design of Rule 23. MA-Br. 45. There is
nothing at all anomalous or unusual about pairing a non-opt-out (b)(2) class and an
opt-out (b)(3) class in the same case. The existence of both classes was a necessary
byproduct of the distinct subsections of Rule 23(b) and plaintiffs position that
defendants had violated the antitrust laws (thus, the (b)(3) class) and would continue
to do so absent injunctive relief (thus, the (b)(2) class).
3.
Given the need for both backward-looking monetary relief and
forward-looking injunctive relief, the (b)(2) class was not in any way a sham
request[] for injunctive relief [to] provide cover for (b)(2) certification of claims that
are brought essentially for monetary recovery. Robinson, 267 F.3d at 164; cf.
Kartman v. State Farm, 634 F.3d 883, 889 (7th Cir. 2011) (noting technique of
39
Case: 12-4671
Document: 1124
Page: 52
10/15/2014
1345377
101
recasting a straightforward claim for damages as a claim for damages and injunctive
relief to make [a] case more amenable to class certification).
For that reason, objectors efforts to shoehorn this case into a Dukes scenario
are unavailing. In Dukes, the plaintiffs sought injunctive relief, declaratory relief,
and backpay, all in a single (b)(2) class. The Supreme Court rejected that maneuver,
holding that because the individualized backpay claims were more than incidental
to the injunctive or declaratory relief, they instead belong[ed] in Rule 23(b)(3).
131 S. Ct. at 2557-58.
The (b)(2) and (b)(3) classes in this case are consistent with both the letter and
spirit of Dukes. This is not an attempt to smuggle damages claims into a (b)(2) class,
as the existence of the parallel (b)(3) class attests. There was no such parallel (b)(3)
class in Dukes, which suffices to distinguish it. That the (b)(2) settlement foreclosed
future efforts to obtain damages based on the agreed-upon, going-forward conduct
is an entirely distinct issue and entirely unobjectionable, as explained next.
C.
There Is Nothing Improper About Having a Rule 23(b)(2) Class

Settlement Foreclose Possible Future Claims Seeking Damages.
1.
The (b)(2) class here sought injunctive relief and injunctive relief alone.
Thus, objectors claim to an opt-out right depends not on the claims for relief of the
(b)(2) class, which are wholly unobjectionable, but entirely on the fact that the
settlement featured a release that foreclosed future challenges to the networks
40
Case: 12-4671
Document: 1124
Page: 53
10/15/2014
1345377
101
agreed-upon, going-forward conduct, including unknown future claims seeking

damages based upon that conduct.
That argument mixes apples and oranges. It wholly rewrites the established
certification criteria, under which [t]he dispositive factor that must be assessed in
determining whether a class may be certified under Rule 23(b)(2) is the type of relief
the plaintiffs actually seeknot the type of future claims released by the settlement.
1 McLaughlin on Class Actions 5:15. As the Supreme Court explained in Dukes,
(b)(2) certification turns on the relief sought, requests for relief, claims for
relief, remedy warranted, and relief entitled to. 131 S. Ct. at 2557. The
relief sought by the (b)(2) class herechanges to defendants network ruleswas
classic injunctive relief that unquestionably belonged in (b)(2).
Objectors so-called due process complaint about the release of hypothetical
and uncertain future damages claims as part of the settlement of a non-opt-out class
is untethered to the specifics of this case. They are essentially advancing a bright-
line rule under which (b)(2) classes can never release a future damages claim in
settlement. Tellingly, however, objectors fail to identify even a single case, and to
our knowledge there is none, that has ever based the propriety of (b)(2) class
certification on the nature of future claims foreclosed by a settlement release, as
opposed to the existing claims for which relief is sought.
41
Case: 12-4671
Document: 1124
Page: 54
10/15/2014
1345377
101
To the contrary, numerous (b)(2) settlement classes have released future

claims seeking relief for going-forward conduct. See, e.g., San Diego Police
Officers Assn v. San Diego City Emps. Ret. Sys., 568 F.3d 725, 734-36 & n.7 (9th
Cir. 2009); Nottingham Partners v. Trans-Lux, 925 F.2d 29, 32-34 (1st Cir. 1991);
TBK Partners v. W. Union, 675 F.2d 456, 459-60 (2d Cir. 1982); Scarver v. Litscher,
371 F. Supp. 2d 986, 997 (W.D. Wis. 2005); Assn For Disabled Ams. v. Amoco, 211
F.R.D. 457, 472 (S.D. Fla. 2002). More often, (b)(2) certification cases simply do
not address the released claims, which go more to the settlements fairness under
Rule 23(e) than the Rule 23(b)(2) criteria.
Each of the cases cited by objectors in support of their purported opt-out
rights, see MA-Br. 32-43, is readily distinguishable. Objectors primarily rely on
cases involving classes certified under (b)(3). For instance, they cite the Supreme
Courts statement in Phillips Petroleum v. Shutts that if a court wishes to bind an
absent plaintiff concerning a claim for money damages or similar relief at law, it
must provide minimal procedural due process protection, including an opportunity
to remove himself from the class. 472 U.S. 797, 811-12 (1985) (emphasis added).
But Shutts involved claims for money damages certified under the state-law
equivalent of a (b)(3) class.
Objectors also note this Courts quotation of Shutts in Stephenson v. Dow
Chemical, 273 F.3d 249, 258 (2d Cir. 2001), but Stephenson likewise involved a
42
Case: 12-4671
Document: 1124
Page: 55
10/15/2014
1345377
101
(b)(3) class with retrospective damages claims. And objectors mistakenly rely on In
re Visa Check/MasterMoney Antitrust Litig., 280 F.3d 124 (2d Cir. 2001), claiming
that this Court affirmed the certification of a (b)(3) class to avoid the primary
concern about Rule 23(b)(2), i.e., the absence of mandatory notice and opt-out
rights. MA-Br. 36. In fact, the Court stated in Visa Check that the primary
concern about certifying a class with significant damages under Rule 23(b)(2) is the
absence of mandatory notice and opt-out rights. 280 F.3d at 147 (emphasis added).
This Courts precise concern was with certifying significant damages claims in a
non-opt-out class, and it did not reach the propriety of (b)(2) certification.
The other cases objectors cite involved individualized monetary claims that,
although improperly certified under a different subsection, manifestly belonged in
(b)(3). For instance, Dukes noted the importance of opt-out rights, but in the context
of already-accrued claims for individualized monetary backpay, which the Court
held should have been certified under (b)(3). Hecht v. United Collection Bureau,
691 F.3d 218 (2d Cir. 2012), similarly underscored the importance of opt-out rights,
but in an even-more-extreme circumstance: the class, though somehow certified
under (b)(2), had only sought damages claims in its complaint.
It is no coincidence that objectors opt-out authorities all involve either (b)(3)
actions or cases that should have been (b)(3) actions. Courts are justifiably
concerned about efforts to evade the protections of (b)(3) by shoehorning damages
43
Case: 12-4671
Document: 1124
Page: 56
10/15/2014
1345377
101
claims into a (b)(2) class. But that is manifestly not the problem in a case like this,
where the damages claims are placed into a separate (b)(3) class with full opt-out
rights. Cf. Robinson, 267 F.3d at 165 (Where class-wide injunctive or declaratory
relief is sought in a (b)(2) class action adequate representation will generally
safeguard absent class members interests and thereby satisfy the strictures of due
process.).
Moreover, in a (b)(2) settlement, a release of hypothetical and uncertain future
damages claims based on the injunction-modified, going-forward conduct is an
unobjectionable feature. Particularly in a (b)(2) class that includes substantial
prospective injunctive relief designed to address future violations, the future claims
released may well not exist if the injunction works as intended and helps restore
competitive conditions. Thus, objectors concerns amount to no less than a sweeping
contention that Rule 23(b)(2)which does contemplate such settlements but does
not provide an opt-out rightis unconstitutional on its face. There is simply no
basis for arguing that due process, Rule 23, or anything else supports a per se rule
against foreclosing hypothetical future damages claims in a (b)(2)-certified class.
2.
Objectors proposed bright-line rulethat a hypothetical future claim
seeking damages can never be foreclosed in the resolution of a (b)(2) class action
is not only wholly unsupported, but unworkable. It would mean that (b)(2) classes
could be unproblematically certified for litigation, but become impossible to settle.
44
Case: 12-4671
Document: 1124
Page: 57
10/15/2014
1345377
101
No rational defendant would enter a settlement by which it commits to restructuring

its practices without some assurance of peace from claims based on the newly-
modified practices. It would be cold comfort indeed if defendants were to make the
changes sought by the (b)(2) class, but then face suits for damages based on those
agreed-upon changes. As this Court has noted, [c]lass action settlements simply
will not occur if the parties cannot set definitive limits on defendants liability. Wal-
Mart v. Visa, 396 F.3d 96, 106 (2d Cir. 2005).
Objectors rule would thus create a world where class actions are easier to
certify for litigation than for settlementa worst-of-all-worlds scenario that not
even the most strident of class action skeptics would support. That would be entirely
impractical and antithetical to the strong judicial policy in favor of settlements,
particularly in the class action context. McReynolds, 588 F.3d at 803.
Objectors approach, moreover, would run directly afoul of the Rules
Enabling Act of Rule 23. The Act mandates that rules of procedure shall not
abridge, enlarge or modify any substantive right. 28 U.S.C. 2072(b). In other
words, Rule 23 is supposed to be a procedural device, not transform the substance
of the underlying claim or limit or expand the available relief.
There can be no dispute that an individual merchant would be free to foreclose
future damages claims in exchange for injunctive relief that will obviate future
violations (and thus render future litigation unnecessary). Yet objectors would deem
45
Case: 12-4671
Document: 1124
Page: 58
10/15/2014
1345377
101
a similar settlement of the same underlying substantive claim unlawful if it were

negotiated by a class of merchants. Using the limited procedural device of Rule 23
to put class action plaintiffs on an inferior footing to individual plaintiffs in this
manner would squarely abridge substantive rights. Cf. Ortiz v. Fibreboard, 527
U.S. 815, 845 (1999) (noting potential Rules Enabling Act problem based on
tension between the limited fund class actions pro rata distribution in equity and
the rights of individual tort victims at law). As the Supreme Court has stated, [a]
class action merely enables a federal court to adjudicate claims of multiple parties
at once, instead of in separate suits, and leaves the parties legal rights and duties
intact and the rules of decision unchanged. Shady Grove Orthopedic v. Allstate,
559 U.S. 393, 406-08 (2010) (plurality opinion).
In the end, objectors fail to avoid the commonsense conclusion that this (b)(2)
class seeking only injunctive relief is a proper (b)(2) class. Though objectors have
concerns about the breadth of the relief and release, those arguments ultimately have
little to do with class certification and everything to do with the fairness of the
settlement. E.g., MA-Br. 38 (settlement releases such claims entirely [for] no
changes aside from limited surcharging relief). Such concerns certainly do not
support a bright-line rule that class certification under Rule 23(b)(2) categorically
prohibits foreclosing future damages claims. None of this is to say that the release
46
Case: 12-4671
Document: 1124
Page: 59
10/15/2014
1345377
101
of future claims should escape judicial scrutiny altogether, but it is to say that the
proper place for that analysis is in Rule 23(e).
D.
Both the Rule 23(b)(2) Class and Rule 23(b)(3) Class Were
Adequately Represented.
1.
Litigating (b)(2) and (b)(3) classes in tandem avoids problems by
protecting the opt-out rights of the members of the (b)(3) class; it does not remotely
introduce an adequacy-of-representation problem under Rule 23(a)(4). Adequacy
requires that the representative parties will fairly and adequately protect the
interests of the class. Fed. R. Civ. P. 23(a)(4). [D]istrict courts must make sure
that the members of the class possess the same interests, and that no fundamental
conflicts exist among the members. Charron, 731 F.3d at 249. As is well-settled
in this Court, [a] conflict or potential conflict alone will not necessarily defeat
class certificationthe conflict must be fundamental, Denney v. Deutsche Bank,
443 F.3d 253, 268 (2d Cir. 2006), and go to the very heart of the litigation.
Charron, 731 F.3d at 250.
Based on its deep familiarity with the litigation and parties, the district court
did not abuse its discretion in concluding that class representatives and counsel
adequately represente[d] both the (b)(2) and the (b)(3) settlement classes. SPA52.
Over an eight-year period, experienced and able class counsel litigated the case
with skill and tenacity, expending 500,000 hours of work on the case. SPA21, 61.
They reviewed more than 50 million pages of documents in discovery and deposed
47
Case: 12-4671
Document: 1124
Page: 60
10/15/2014
1345377
101
more than 400 witnesses. SA103. And both the district court and the mediators
concluded that the intensive settlement negotiations were fair, adversarial, and
conducted at arms length. SPA21.
Indeed, this Court has said repeatedly that the inclusion of independent
mediators helps to ensure that [settlement] proceedings were free of collusion and
undue pressure. DAmato, 236 F.3d at 85; Suffolk Cnty. v. Long Island Lighting,
907 F.2d 1295, 1323 (2d Cir. 1990). Here, counsel for the parties met jointly or
separately with one or both mediatorsmediators the parties jointly selectedon
approximately 45 occasions. D.E. 2113-6 175, 181. And those structural
assurance[s] of fair and adequate representation were further enhanced by the
participation of Judges Gleeson and Orenstein, pursuant to the request of all parties,
near the end of the process. Amchem, 521 U.S. at 627. Class representatives,
meanwhile, participated in discovery, in mediation, in court sessions, in the
evaluation of the mediators proposals, and in the formulation of the Settlement
Agreement, readily fulfill[ing] all of the obligations associated with being class
representatives. SPA52.
Beyond that, the settlement as a whole provides no evidence of prejudice to
the interests of a subset of plaintiffs. In re Literary Works Copyright Litig., 654
F.3d 242, 252 (2d Cir. 2011). This was not a pre-packaged settlement that was
indicative of collusion between the lead plaintiffs and the defendants. Class counsel
48
Case: 12-4671
Document: 1124
Page: 61
10/15/2014
1345377
101
recommended settlement to the court only after years of hard-fought litigation.

Discovery was complete, motions for dismissal, summary judgment, and class
certification had been fully briefed and argued, and the parties had engaged in an
arduous four-year mediation process. Nor was this a settlement conceived of by
class counsel in collusion with the defendants; it was the mediators proposal that
outlined the key components of what became the Settlement (notably, a proposal
accepted by many current-objectors). SPA21. The mediators were not somehow
complicit in an effort to disadvantage an underrepresented subset of the class.
Nor was this a lowball settlement; it secures the largest-ever cash recovery in
an antitrust class action settlement and historic reforms of decades-old network rules.
The class representatives properly discharged their obligation to represent the best
interests of the class as a whole, rather than any individual members of it, Fed.
R. Civ. P. 23 advisory committee notes (2003), achieving results that were not merely
adequate, but outstanding. See infra Part II.
2.
Objectors nonetheless insist that there was a fundamental conflict of
interest between the (b)(2) and (b)(3) classes that rendered class representatives and
counsel incapable of adequately representing (b)(2) class members.7 Objectors
Objectors assertion that class counsel fired their clients so that all that
remained were class representatives committed to a deal that gave the (b)(3) class
money in exchange for a broad release from the (b)(2) class misstates the record.
MA-Br. 70. The objectors that were formerly class representatives all agreed to the
mediators settlement proposal and participated in settlement conferences. Of those
49
Case: 12-4671
Document: 1124
Page: 62
10/15/2014
1345377
101
primarily attempt to analogize this case to Amchem, Ortiz, and other cases involving
subgroups with antagonistic interests. MA-Br. 67. In those cases, objectors
contend, the Supreme Court emphasized the importance of creating subclasses and
appointing separate counsel for such antagonistic subgroups.
But objectors
overread Amchem and Ortiz and, in search of a real conflict, imagine tradeoffs here
between retrospective damages and prospective injunctive relief.
Amchem and Ortiz do not remotely impose a per se requirement of separately
represented subclasses whenever there is tension among class members. Instead, the
Supreme Court required that intra-class conflicts be addressed by structural
assurance[s] of fair and adequate representation, of which subclasses are but one
example. See Amchem, 521 U.S. at 627 (structure of the negotiations matters as
well). Here, Judge Infante, Professor Green, Judge Gleeson, and Magistrate Judge
Orenstein all confirmed a record that demonstrates beyond any reasonable doubt
that the negotiations were adversarial and conducted at arms length. SPA21
(emphasis added).
former class representatives, all but one agreed to the July 2012 Memorandum of
Understanding committing to the final settlement. Only after those then-class
representatives changed their minds, and often only after new counsel of record
appeared on their behalf, did class counsel move to withdraw as their counsel. See
Fed. R. Civ. P. 23 advisory committee notes (2003) (class representatives do not
have an unfettered right to fire class counsel and cannot command class counsel
to accept or reject a settlement proposal).
50
Case: 12-4671
Document: 1124
Page: 63
10/15/2014
1345377
101
Moreover, Amchem and Ortiz involved parties mov[ing] jointly for

conditional class certification and approval of a settlement agreement where [t]he
district court granted the motion without any litigation. Joel A., 218 F.3d at 139
(distinguishing Amchem on this ground).
Here, by contrast, the adequacy of
representation was demonstrated through years of contested litigation. Imposing a

separate representation requirement at the tail end would be wholly impractical, and
contrary to the interests of both classes by making simultaneous settlements of
damages and injunctive relief claims virtually impossible.
More fundamentally, the (b)(2) and (b)(3) classes here are not subgroups
with antagonistic interests. Indeed, they are not subgroups at allthey are
largely one and the same group.
Any merchant that was in business before
November 27, 2012 and continues to operate going forward will receive monetary
relief through the (b)(3) class (subject to an opt-out) and will benefit from the rule
changes obtained through the (b)(2) class.
Objectors note that the overlap between the two classes is not total. MA-Br.
75-76. But objectors cite no cases applying that stringent of a standard in evaluating
adequacy. Indeed, objectors primarily rely on Amchem, which involved a conflict
between two mutually exclusive groupsplaintiffs that were currently injured by
asbestos, and exposure-only plaintiffs that faced only potential future injuries. 521
U.S. at 595. There was thus a far more serious risk that counsel would be unable to
51
Case: 12-4671
Document: 1124
Page: 64
10/15/2014
1345377
101
simultaneously represent the interests of both groups. Id.; see also Ortiz, 527 U.S.
at 857 (presently injured versus future claimants; Pre-1959 claimants versus
post-1959 claimants); Eubank v. Pella, 753 F.3d 718, 721 (7th Cir. 2014)
(customers who had already replaced or repaired their defective windows versus
those who hadnt).
Nor does this case remotely resemble In re Literary Works, 654 F.3d 242, in
which three categories of claims (A, B, and C) vied for an allocation of funds from
a fixed sum. Any increase in Cs allocation required a corresponding decrease in A
and Bs allocation. And if the claims exceeded the fixed sum, Category C claims
exclusively bore the brunt of any necessary reductions. In that context, this Court
reasonably found that Category C-only plaintiffsthe largest contingent of class
membershad diverging interests as to the distribution of that recovery from
plaintiffs with Category A and B claims. Id. at 252, 254. Here, those distribution
and tradeoff concerns have no bearing at all. The vast majority of plaintiffs had both
(b)(2) and (b)(3) claims, and both categories of plaintiffs sought distinct types of
relief that were not capped.
The near-total overlap between the (b)(2) and (b)(3) classes, moreover,
reinforces their complementary, not antagonistic, relationship. Objectors posit a
structural dilemma in (b)(2) and (b)(3) relief, whereby class representatives are
inherently driven to trade future-looking (b)(2) interests for present (b)(3) benefits.
52
Case: 12-4671
Document: 1124
Page: 65
10/15/2014
1345377
101
See MA-Br. 71 (Representatives with present interests simply cannot fight for the
best possible relief for future-looking claims.). But (b)(2) and (b)(3) have
peacefully co-existed for decades, and litigants routinely combine them when
seeking both prospective and retrospective relief. See supra n.6. There is simply no
support for objectors suggestion that these two standard forms of class actions
somehow become volatile when combined either in this case or more generally.
Here, the district court correctly recognized that the two types of relief worked
as essential and complementary components of one fair, reasonable, and adequate
remedy. And, precisely because (b)(2) and (b)(3) interests are typically aligned and
rooted in common claims, objectors fail to cite a single case citing a conflict of
interest between a (b)(2) and (b)(3) class that rendered class counsel unable to
represent both groups simultaneously. They exist as two classes not because of
factual differences among members of a class such that subclasses must be
established, Amchem, 521 U.S. at 627, but by virtue of Rule 23(b).8
Certain objectors contend that, although they are members of the merchant
class, they were not adequately represented because they do additional business in
a non-merchant capacity. See American Express Br. 15-24; First Data Br. 25-34. As
the district court correctly held, [t]hese objectors seek to make something of
nothing. SPA47. The settlement unambiguously does not bar claims that a class
member may have in its capacity as a payment-card competitor, an ATM operator,
or any other capacity other than as a merchant that accepts Visa and MasterCard
credit cards in the United States. SPA47 (emphasis added).
53
Case: 12-4671
Document: 1124
Page: 66
10/15/2014
1345377
101
Nor is there any conflict of interest between present and future merchants
within the (b)(2) class. Objectors note that in Stephenson, certain future claimants
were deemed inadequately represented by class representatives who previously
negotiated a (b)(3) class settlement. But those future claimants had retrospective
damages claims for Agent Orange-related injuries sustained in the 1960s and 1970s
in Vietnam. This Court simply held that the veterans who became aware of their
Agent Orange-related injuries after 1994when the (b)(3) settlement damages fund
was deplet[ed], leaving no more reliefwere inadequately represented in
negotiations for that relief. 273 F.3d at 258.
Here, by contrast, the (b)(2) relief is prospective and will indivisibly benefit
all present and future merchants. Far from there being some radical asymmetry
between present merchants and those objectors who purport to be predominantly
concerned with future injuries, Blue Cross Br. 23, all merchants similarly benefit
from the rule reforms. Indeed, the vast majority of the class is composed of ongoing
merchants with as much of a concern with future injuries as any future merchants
that do not yet exist. See, e.g., Dewey v. Volkswagen, 681 F.3d 170, 185-86 (3d Cir.
2012) (a past claimant[] can continue to suffer leakage into the future to the same
extent as a future claimant).
3.
In the end, objectors cannot seriously dispute that class representatives
and counsel shared their interest in (b)(2) relief. Objectors instead lament that class
54
Case: 12-4671
Document: 1124
Page: 67
10/15/2014
1345377
101
representatives had less interest and put greater emphasis on (b)(3) relief. MABr. 68 (emphasis added). And so the alleged fundamental conflict ultimately boils
down to a difference in the degree of emphasis between two types of relief that
virtually all class members were jointly pursuing. That does not come close to an
adequacy defect necessitating separate counsel. All class settlements strike
compromises, and if compromises automatically created subclasses that required
separate representation, the class action procedure would become even more
cumbersome. Charron, 731 F.3d at 253-54; see also Dewey, 681 F.3d at 186-87
(To hold that differing valuations [of class-wide relief] by themselves render the
representative plaintiff inadequate would all but eviscerate the class action device.);
Gooch, 672 F.3d at 429.
In all events, there was no difference in emphasis between the two types of
relief. To be sure, comparing monetary to injunctive success is an imperfect
exercise, cf. Bendix v. Midwesco, 486 U.S. 888, 897 (1988) (Scalia, J., concurring)
(like judging whether a particular line is longer than a particular rock is heavy),
and would mean, perversely, that the greater the representatives monetary
achievement, the less adequate their representation. But as the district court properly
found, the value of the injunctive relief here may very likely exceed the value of the
monetary relief in the long run. SPA67 (emphasis added).
55
Case: 12-4671
Document: 1124
Page: 68
10/15/2014
1345377
101
Of course, objectors disagree with that assessment and think that the class
representatives and counsel should have held out longer for a better deal. But again,
that merely underscores that objectors driving concerns (which are meritless, see
infra Part II) relate to the reasonableness of the settlement relief and release. Again,
those issues do not escape judicial scrutiny altogether. But any such concerns should
be directed to Rule 23(e) and its analysis of the settlements overall fairness rather
than repackaged as an adequacy defect under Rule 23(a)(4). See, e.g., Petrovic v.
Amoco, 200 F.3d 1140, 1146 (8th Cir. 1999) (objectors challenge [to] the propriety
of the award of compensation was more properly directed to the objectors
contention that the settlement was not fair, adequate, and reasonable than
adequacy).
II.
The District Court Acted Well Within Its Broad Discretion In Finding
The Overall Settlement Fair, Reasonable, and Adequate Under Rule
23(e).
Objectors various attacks on class certification in the guise of cohesion, due
process, and adequacy issues are really just flawed efforts to repackage unpersuasive
challenges to the overall fairness of the settlement. Those arguments fare no better
under the Rule 23(e) rubric where they belong. Rule 23(e)(2) does not demand that
the settlement be perfect; it need only fall within a range of reasonableness.
Grinnell, 495 F.2d at 463.
56
Case: 12-4671
Document: 1124
Page: 69
10/15/2014
1345377
101
The district courts judgment that this settlement fell within that range of
reasonablenessrooted in the courts unique expos[ure] to the litigants, and their
strategies, positions and proofswarrants considerable deference. Joel A., 218
F.3d at 139. Indeed, the district court is owed heightened deference where, as
here, experience has imparted to the judge a particularly high degree of
knowledge. Id. The courts discretionary judgment, moreover, was shared by the
two independent mediators who steered the negotiations and proposed the
parameters of the eventual agreement. D.E. 1111-2 12; D.E. 1111-3 33; see also
Wal-Mart, 396 F.3d at 116 (applying presumption of fairness to class settlement
reached in arms-length negotiations between experienced, capable counsel after
meaningful discovery). Those in the best position to evaluate whether the
settlement constitutes a reasonable compromise, Handschu, 787 F.2d at 833,
pronounced this settlement a more than reasonable resolution.
A.
The Relief Obtained by the Class is Outstanding.
1. The $7.25 Billion Damages Fund is the Largest-Ever Cash Relief in

an Antitrust Class Action Settlement.
Against the prospect of uncertain relief years down the line, the settlement
secures significant monetary compensation in the near future in the form of an
estimated $7.25 billion damages fund. SPA23. That historic sum represents the
largest-ever cash relief in an antitrust class action settlement, and is more than double
the recovery in any previous private antitrust action. It is also the third-largest class
57
Case: 12-4671
Document: 1124
Page: 70
10/15/2014
1345377
101
action settlement in history. The notion that this massive monetary recovery could
somehow be inadequate beggars belief. Cf. Wal-Mart, 396 F.3d at 119 (describing
$3 billion settlement paid over ten years as staggering); In re Visa
Check/MasterMoney Antitrust Litig., 297 F. Supp. 2d 503, 512 (E.D.N.Y. 2003)
(approving $3 billion settlement paid over ten years).
Rather than attack the adequacy of this massive recovery, objectors ignore it
altogether. Objectors focus singularly on the (b)(2) injunctive relief, which they
denigrate as [l]iterally nothing. MA-Br. 76. But it is well-settled that the
reasonableness of a settlement must be taken as a whole. Maywalt v. Parker &
Parsley Petrol., 67 F.3d 1072, 1079 (2d Cir. 1995). [I]t is not the Courts
prerogative to pick and choose terms of the settlement, [or] redact portions of the
agreement. McBean v. New York, 233 F.R.D. 377, 382 (S.D.N.Y. 2006). Thus,
courts look to the non-exhaustive Grinnell factors with a recognition that not every
factor must weigh in favor of settlement; rather the court should consider the totality
of these factors in light of the particular circumstances. Thompson v. Metro. Life,
216 F.R.D. 55, 61 (S.D.N.Y. 2003).
Objectors even go so far as to fault the district court for not conducting a
stand-alone Grinnell analysis exclusively for the (b)(2) relief. Merchant Trade
Groups Br. (MTG-Br.) 33-36. Objectors are mistaken about the value of the
(b)(2) relief, see infra Part II.A-2-4, and ignore that the district court specifically
58
Case: 12-4671
Document: 1124
Page: 71
10/15/2014
1345377
101
addressed and affirmed that the (b)(2) relief was reasonable and reasonably justified
the release. SPA36-47. Their formalistic insistence on two separate multi-factor
Grinnell analyseswhen the (b)(2) and (b)(3) analyses overlapped almost
entirelyis meritless and directly at odds with the holistic Rule 23(e) inquiry.
Unsurprisingly, courts addressing global settlements involving (b)(2) and (b)(3)
classes have not applied the Grinnell factors to each class in isolation. See, e.g.,
Charron v. Pinnacle Grp., 874 F. Supp. 2d 179, 196 (S.D.N.Y. 2012) (the
Settlement offers them redress for past injuries, while affording significant
systemic benefits (protocols, monitoring, lease audit, injunction)); New England
Carpenters Fund v. First DataBank, 602 F. Supp. 2d 277, 281 (D. Mass. 2009) ($2.7
million cash payment combined with the AWP rollback provisions constitutes a
reasonable settlement). Indeed, objectors fail to cite a single example of what they
demand: a stand-alone (b)(2) Grinnell analysis focus[ed] exclusively on the
value of the rules changes, the impact of the mandatory release, and the risks of
litigating the injunctive claims. MTG-Br. 34.
In any event, there is simply no brushing aside the massive damages fund
that promises millions of merchants long-awaited compensation for long-
accumulated overcharges. SPA61. When considered together with the valuable
injunctive reforms, the relief achieved was far more than fair, reasonable, and
adequate.
59
Case: 12-4671
Document: 1124
Page: 72
10/15/2014
1345377
101
2. The Removal of Restraints on Surcharging Offers Valuable Relief.

The damages fund standing alone, while massive, would risk allowing
merchant restraints [to], in effect, place the merchants back where they started,
permitting the networks to simply recoup any associated lost revenues by tinkering
with other fees. SPA16. Thus, the settlement also secures groundbreaking
injunctive reforms of several long-entrenched merchant restraints that had allowed
the networks and banks to charge excessive interchange feesreforms that the
district court found may very likely exceed the value of the monetary relief in the
long run. SPA67 (emphasis added). These immediately effective rule reforms,
which class members [could] take advantage of now, further counseled against
the alternative of pursuing many more years of litigation while merchants continued
to be hamstrung by the no surcharge rules of Visa and MasterCard and remaining
anti-steering rules. D.E. 2113-6 185.
The heart of injunctive relief was the lifting of the networks bans on
surcharging. SPA22. Plaintiffs pushed very hard to obtain this change, and
defendants steadfastly resisted it. SPA36. From the inception of the litigation, the
anti-surcharging rules were viewed as a linchpin to the problem, as far as the
merchants [were] concerned. SPA9; see also D.E. 1165 at 3 (foremost example
60
Case: 12-4671
Document: 1124
Page: 73
10/15/2014
1345377
101
of an anti-steering restraint). Even numerous objectors publicly touted the value of

surcharging relief.9
For the first time since the networks were created a half-century ago,
merchants can impose a surcharge on Visa and MasterCard credit card transactions
at the point of sale to recover the full costs of acceptance of such transactions and
steer customers to less costly payment methods and brands. SPA15. This marks a
sea change in the payment industry. Merchants gain valuable leverage from being
able to educate customers about the costs of accepting credit cards. In the district
courts words, surcharges can finally make transparent and avoidable what has been
opaque and inevitable. SPA37. Levying a surcharge on credit card payments
enables merchants to steer customers towards using lower-cost and non-surcharged
payment methods or brands. D.E. 2111-5 (Frankel Decl.) 68-69.
A surcharge is even more effective than a discount in this regard because
customers react more strongly to losses from perceived penalties (such as a
surcharge) than perceived rewards. Id. 48. Cash, check, and debit card
customers will no longer be forced to subsidize the additional costs of serving credit
See Credit Card Interchange Fees: Antitrust Concerns? Hearing Before S.

Comm. on the Judiciary, 109th Cong. 41 (2006) (NACS witness) (anti-surcharging
rule is part of the reason why this is a broken market and should not exist); id.
at 28 (Merchants Payments Coalition) (anti-surcharging rule is part of their
anticompetitive scheme to fix interchange fees that reinforces price fixing
efforts).
61
Case: 12-4671
Document: 1124
Page: 74
10/15/2014
1345377
101
card customers. Id. 35. As customers opt for cheaper payment methods and brands,
meanwhile, merchants may benefit from decreased card-acceptance costs and
increased revenues from the surcharges. Id. 68-69. In the long run, merchants
may lower their posted retail prices (further benefitting customers) and boost their
total sales.
Even the threat of surcharging benefits all merchants regardless of whether
they surcharge. The networks will face an incentive to lower or moderate their
interchange fees because they will lose more transactions if they maintain high
interchange fees with surcharging than without. The district court specifically noted
expert estimates that, in all, surcharging may save merchants $26.4 to $62.8 billion
in acceptance costs over the next decade. SPA35-36. It represents an indisputably
procompetitive development that has the potential to alter the very core of the
problem this lawsuit was brought to challenge. SPA35.
Objectors briefs are replete with references to the illusory and limited
surcharging relief. MA-Br. 38; MTG-Br. 54. But objectors do not deny, nor could
they, the pro-competitive effects of surcharging. Instead, drawing on various facts
external to the litigation and the settlement, they complain that lifting prohibitions
on surcharging does not guarantee that every merchant will begin affirmatively
surcharging. The district court considered and rejected each of those arguments as
unpersuasive. SPA36.
62
Case: 12-4671
Document: 1124
Page: 75
10/15/2014
1345377
101
First, objectors suggest various reasons why some merchants might choose
not to surcharge. For example, they note that some merchants operate in industries
that are so competitive that surcharging is highly unlikely. MA-Br. 55. As the
district court reasoned, the mere fact that merchants may choose not to avail
themselves of the proposed relief does not compel the conclusion that the
indisputably procompetitive rules changes are not a valuable achievement. SPA41;
see also LaGarde v. Support.com, 2013 WL 1283325, at *6 (N.D. Cal. 2013) (while
it is unknown as to how many class members will actually take advantage of relief,
these deficiencies do not weigh against a finding of fairness and adequacy);
Handschu v. Special Servs. Div., 605 F. Supp. 1384, 1417 (S.D.N.Y. 1985), affd,
787 F.2d 828 (2d Cir. 1986) (The settlement does not achieve everything they wish
for. Few settlements do. But insisting on everything disregards the limitations
arising out of present law.).
Objectors also argue that American Expresss separate anti-surcharging rule
diminishes the value of surcharging relief. They contend that, because American
Express generally carries higher acceptance costs than Visa and MasterCard,
merchants will not surcharge Visa or MasterCard cards if it drives customers to
American Express. Thus, objectors reason, merchants who choose to maintain their
relationships with American Express would not be able to take advantage of
surcharging opportunities immediately.
63
Case: 12-4671
Document: 1124
Page: 76
10/15/2014
1345377
101
But, again, neither the court nor class counsel can control a merchants choice
to continue accepting cards issued by American Express or any other entity not a
party to this lawsuit. As the district court noted, objectors have no solution for that
[American Express] problem because there could not be one in this case,
SPA42even if plaintiffs had proceeded to trial and prevailed. But that does not
mean that the American Express problem is intractable, just that it is the subject
of a different lawsuit. In fact, American Express subsequently agreed, as part of a
settlement in a different class action, to allow merchant surcharging in certain
circumstances. See In re Am. Express Anti-Steering Rules Antitrust Litig. (II), No.
11-md-2221, D.E. 306-2 8(e) (E.D.N.Y. Jan. 7, 2014). Thus, the settlement of this
case provided substantial relief vis--vis the defendants here, which is all that can
realistically be expected of this lawsuit, and the value of that relief became magnified
by external events.
Next, objectors point to laws on the books in approximately ten states that
would impede merchants from surcharging in those jurisdictions. Merchants in
those states previously faced two independent obstacles to surcharging
prohibitions from the networks and from the statesand now face only one.
Objectors complain that the remaining state-law obstacle limits the value of the relief
for merchants in those states. But removing state-law obstacles again goes well
beyond the scope of th[e] lawsuit. SPA52. As the district court concluded, [e]ven
64
Case: 12-4671
Document: 1124
Page: 77
10/15/2014
1345377
101
if the objectors are right in contending that additional dominoes must fall before the
alleged anticompetitive behavior of Visa and MasterCard is eradicated, those
dominoes will have to fall in other forums. SPA18.
In all events, objectors overstate the extent of the state-law obstacles and
understate the extent to which the relief obtained here will itself cause other
dominoes to fall. The district court noted that even in the ten states that limit
surcharging, at least some state laws are enforced in a manner that prohibits
surcharging only when the merchant fails to sufficiently disclose the increased prices
for credit card use. SPA38. Even if merchants in those states forgo surcharging
altogether, the court added, the fact that interchange fees are set on a nationwide
basis means that surcharging in other statesor even the threat of surcharging
will exert downward competitive pressures on interchange fees to the benefit of all
merchants nation-wide. SPA38; see also In re Motor Fuel Sales Practices Litig.,
271 F.R.D. 263, 289 & n.36 (D. Kan. 2010) (all class members benefit; class
members from non-conversion states have a right to purchase ATC fuel from
Costco in conversion states).
Indeed, the relief imposed here puts undeniable pressure on those state laws.
It is one thing for state law to reinforce the uniform practice of Visa, MasterCard,
and American Express. It is quite another thing for state laws to remain as the only
obstacle to more-transparent pricing after contractual surcharging prohibitions have
65
Case: 12-4671
Document: 1124
Page: 78
10/15/2014
1345377
101
been eliminated as part of antitrust settlements. In fact, just as the American Express
issue was addressed in separate litigation, these state laws are also under attack in
separate litigation. As the district court recognized, a recent decision barring
enforcement of one such law on constitutional grounds, see Expressions Hair Design
v. Schneiderman, 975 F. Supp. 2d 430 (S.D.N.Y. 2013) (appeal pending), indicates
that the dominoes may already be falling, and that independent events will only
magnify the already-substantial value of the surcharging relief.
Finally, objectors point to the potential for networks to enter bilateral
agreements with merchants and speculate that they could offer[] [a] merchant a
break on its interchange rates in exchange for its agreement not to surcharge. MA-
Br. 62. But far from swallow[ing] the Settlements surcharging relief, MTG-Br.
55, this underscores the broader benefits of that relief. As a direct result of the threat
to surcharge, networks may be pressured to moderate and make concessions on
interchange fees. See D.E. 2113-6 196 (after Australia rescinded anti-surcharging
rules in 2003, merchants used threat of surcharging to negotiate significantly lower
American Express fees).
In the end, all merchants have an interest in lifting the anti-surcharging
restraints because all merchants have the same interest in being able to inform
cardholders at the point of sale of the acceptance costs of their credit cards and to
either steer them to lower-cost alternatives or recoup the cost of acceptance.
66
Case: 12-4671
Document: 1124
Page: 79
10/15/2014
1345377
101
SPA51. The elimination of decades-old prohibitions may not guarantee that every
merchant will promptly begin surcharging, much less eradicate anticompetitive
practices in the payment industry nationwide in one stroke. But none of objectors
criticisms of the limits of the surcharging relief detracts from the fact that this was a
critical accomplishment and significant step forward in exposing interchange
fees to competitive market forces. SPA15, 37.
3. The Other Injunctive Reforms Offer Valuable Relief.
The surcharging relief does not stand alone. The settlement contains
numerous other immediately effective meaningful programmatic reforms that
complement and enhance the effectiveness of merchants new surcharging
opportunities and will further exert downward pressure on interchange fees. SPA44.
Previously, Visa and MasterCard had an unbroken practice of refusing to
negotiate over interchange fees with merchant buying groups. The settlements
buying group provision now ensures that if a group of merchants makes a proposal
to Visa or MasterCard, the network cannot turn a blind eye. Instead, Visa and
MasterCard have an affirmative duty to negotiate in good faith and, if the proposal
provides commercially reasonable benefits, exercise good faith in deciding
whether to accept or reject the proposal. SPA149-50, 163-64 (Settlement 43, 56).
Smaller merchants are now better equipped to use collective bargaining power to
obtain the scale economies, organizational efficiencies and negotiating ability of
67
Case: 12-4671
Document: 1124
Page: 80
10/15/2014
1345377
101
large merchants. D.E. 2111-5 56. Objectors do not deny that the reform is pro-
competitive; they merely question the extent to which it will be utilized. But again,
that unknown do[es] not weigh against a finding of fairness and adequacy.
LaGarde, 2013 WL 1283325, at *6. As plaintiffs expert attested, even modest
competitive pressures on interchange fees produced by buying group efforts may
produce substantial savings. D.E. 2111-5 55-57, 66.
Meanwhile, merchants were previously compelled in practice to accept Visa
or MasterCard cards at all their outlets and banners (brands) because the networks
made volume discounts on interchange fees contingent on that across-the-board
acceptance. The settlements all-outlets provision eliminates that practice.
Merchants may now accept Visa or MasterCard cards at some, but not all, of their
businesses without being penalized with the volume discounts. SPA140-41, 153-54
(Settlement 41, 54). Thus, for instance, a merchant can now decline to accept the
networks cards at its discount store banner to keep costs and prices as low as
possible, yet continue to accept the cards in its other stores. Because the higher the
fees, the more likely the merchant is to refuse acceptance of a card brand, this rule
change will further increase the elasticity of demand with respect to merchant fees,
and thus intensify the competitive constraints facing the Networks over the level of
their merchant fees. D.E. 2111-5 54.
68
Case: 12-4671
Document: 1124
Page: 81
10/15/2014
1345377
101
Finally, plaintiffs challenged Visas and MasterCards anti-discounting and

anti-minimum-price rules, which prevented merchants from offering discounts and
banned minimum purchase amounts for credit card usage. The Durbin Amendment
and DOJ Consent Judgmentdevelopments that piggybacked on [plaintiffs]
efforts, SPA59took superseding steps toward dismantling those rules. They
enabled merchants to offer discounts, rebates, and other in-kind incentives and set
minimum purchase amounts, and required Visa and MasterCard to provide, at no
cost, services to help merchants determine the acceptance costs of Visa and
MasterCard credit cards.
The discounting, minimum-purchase, and cost information provisions of the
settlement now lock in, until 2021, the dismantling of those anti-steering restraints.
SPA139-40, 150-51, 153, 164 (Settlement 40, 44, 53, 57).
The Durbin
Amendment and DOJ Consent Judgment, while achieving crucial tools, were also
subject to the vagaries of modification and repeal. The settlement firmly shields
those gains from erosion by the whims of public opinion or Rule 60(b) assertions of
changed circumstances, ensuring that they remain valuable enhancements to
merchants newfound ability to surcharge. Merchants are now assured of their
ability to offer discounts and minimum purchase rules in addition to, or in lieu of,
surchargesan empowering toolbox with which they can steer customers toward
69
Case: 12-4671
Document: 1124
Page: 82
10/15/2014
1345377
101
more cost-effective payment methods, brands, and products, and incentivize

networks to keep interchange fees in check.
4. The Settlement Is Not Unreasonable Merely Because It Does Not
Include All of the Relief Sought by Objectors.
Based on its extensive knowledge of the litigation, the district court concluded
that rule reforms achieved would meaningfully blunt any lingering
anticompetitive effects of the Honor-all-Cards and default-interchange rules.
SPA45, 61. Objectors, however, lament the fact that this settlement does not obtain
the wholesale rescission of the Honor-all-Cards and default-interchange rules as
well. MA-Br. 14, 17. That objectors (like any plaintiff) would have preferred even
more relief is understandable. But as this Court has said time and again: Each side
gives up a number of things. This is the way settlements usually work. Wal-Mart,
396 F.3d at 113. The fact that a proposed settlement may only amount to a fraction
of the potential recovery does not, in and of itself, mean that the proposed settlement
is grossly inadequate and should be disapproved. Grinnell, 495 F.2d at 455 & n.2;
see also Handschu, 605 F. Supp. at 1385 (It is beside the point for objectors to
criticize the settlement because it falls short of a state of law they devoutly desire
but have not yet achieved).
As they did below, objectors fail to grapple with the limitations on the relief
that would be available even if success were achieved and assum[e] that a
complete victory on the merits is a foregone conclusion. SPA25-26. But continuing
70
Case: 12-4671
Document: 1124
Page: 83
10/15/2014
1345377
101
with litigation was not without serious risks. See Defendants-Appellees Br. Part I.A2. In fact, the district court noted that plaintiffs faced an uphill battle on their
challenges to the default interchange and Honor-all-Cards rules. The district court
pointedly criticized objectors for assum[ing] that default interchange is inherently
illegal, [when] in reality it is a very complicated issue. SPA29. It noted that no
court had ever held that Visa or MasterCards default interchange rules violate the
antitrust laws, and that the practices had procompetitive effects for consumers that
may have outweighed any anticompetitive harm. SPA30.
The district court further questioned whether a court could even permissibly
engage in the regulation of interchange fees if the plaintiffs obtained a complete
victory on the merits. SPA14, 16. Likewise, the court noted that plaintiffs would
have to confront adverse caselaw implicating the Honor-all-Cards rule that made it
no sure thing that Class Plaintiffs will be able to prove they have anticompetitive
effects to such an extent that they violate the antitrust laws. SPA32. The district
court also discussed risks that plaintiffs may have faced in establishing damages and
maintaining class status.
In light of the substantial delays and uncertainties that extending the litigation
for many more years would entail, it was eminently reasonable to conclude that the
settlement reliefmost of which was immediately effectivewas the best possible
outcome for plaintiffs. Under the totality of the circumstances, the settlement easily
71
Case: 12-4671
Document: 1124
Page: 84
10/15/2014
1345377
101
falls within a range of reasonableness. That the settlement does not obtain all of
the relief that some objectors would have preferred does not in any way take it
outside that realm of reasonableness.
B.
The Standard Release Conforms With All Applicable Law.
In exchange for the substantial damages and restructured network practices,

the settlement releases all claims that are alleged or which could have been alleged
by plaintiffs in this litigation.10 SPA169 (Settlement 68). That is a standard form
of release that courts have repeatedly approved in class settlements. It is a form of
release, moreover, that this Court has recognized is often pivotal to achieve
comprehensive settlement of class actions. In re Literary Works, 654 F.3d at 247-
48; see also Wal-Mart, 396 F.3d at 106 ([c]lass action settlements simply will not
occur if the parties cannot set definitive limits on defendants liability).
Objectors complain about the release of hypothetical and uncertain future
claims seeking damages, but that is really just a reprise of their mistaken criticisms
of the settlement reforms as [l]iterally nothing. MA-Br. 76. The structural reforms
are, as the district court found, designed to provide substantial and meaningful
10
See SPA169-70 (Settlement 68) (releasing claims arising out of or relating in

any way to any conduct, acts, transactions, events, occurrences, statements,
omissions, or failures to act of any Rule 23(b)(2) Settlement Class Released Party
that are alleged or which could have been alleged from the beginning of time until
the date of the Courts entry of the Class Settlement Preliminary Approval Order)
(emphasis added).
72
Case: 12-4671
Document: 1124
Page: 85
10/15/2014
1345377
101
relief. SPA15. There is nothing unusual about foreclosing hypothetical future claims
based on conduct addressed prospectively by a valid injunction. The injunctions
immediately-effective, meaningful reformscoupled with other industry reforms
triggered by this litigation, such as the separation of the payment networks from the
banks, the DOJ Consent Judgment, and the Durbin Amendmentdramatically
change the landscape going forward, such that the value of any foreclosed future
claims will likely be de minimis.
Moreover, as the district court correctly concluded, the release covers only
the claims that may properly be extinguished by the settlement of a class action.
SPA44. It is of no moment that the release covers rules and practices that were not
expressly challenged in this action. The four corners of a complaint have never
delineated the outer bounds of a release. To the contrary, it is well-settled that class
action releases may include claims not presented and even those which could not
have been presented as long as the released conduct arises out of the identical
factual predicate as the settled conduct. Wal-Mart, 396 F.3d at 107. Time and
again, this Court has approved nearly identical releases as consistent with the
identical factual predicate test. Cf. Visa Check, 297 F. Supp. 2d at 512 (claims
which have been asserted or could have been asserted); In re Literary Works, 654
F.3d at 247 (claims that were or could have been pled).
73
Case: 12-4671
Document: 1124
Page: 86
10/15/2014
1345377
101
That is all that this straightforward release does. As the district court
determined, the settlement simply releases claims that are or could have been
alleged based on the identical factual predicate of the claims in this case. SPA46.
At the final approval hearing, defendants agreed unequivocallyas they reiterate
here, Defendants-Appellees Br. Part IIthat the release is limited by the Identical
Factual Predicate Doctrine which is the law of the Second Circuit. D.E. 6094 at
36 (emphasis added); see also id. (Defs: Nobody is proposing that the release be
construed beyond the Identical Factual Predicate Doctrine.).
Claims about the rules and conduct that enabled the networks to maintain
supra-competitive default interchange fees, their IPOs, or their status as structural
conspiracies by virtue of their rules, are thus released. Claims about any new rules
and conduct are not released.11 Claims about any reversion to the pre-settlement
rules are likewise not released. See id. at 228-29 (defendants agreeing with
plaintiffs list of conduct not covered by the release). The district courts findings
regarding the parties intentions will be respected on appeal unless they are clearly
erroneous. W. Alton Jones Found. v. Chevron, 97 F.3d 29, 33 (2d Cir. 1996).
11
This release is thus far afield from the release rejected in Natl Super Spuds v.
N.Y. Mercantile Exch., 660 F.2d 9 (2d Cir. 1981). There, the release of claims based
on unliquidated contracts was deemed improper when the claims, the complaint,
the class certification opinion, and the settlement notice all exclusively concerned
liquidated contracts. Id. at 16-17. Even class action plaintiffs did not purport to
represent anyone with claims based on unliquidated contracts. Id. at 17.
74
Case: 12-4671
Document: 1124
Page: 87
10/15/2014
1345377
101
Indeed, [f]ew persons are in a better position to understand the meaning of a

[settlement] than the district judge who oversaw and approved it. United States v.
Local 359, United Seafood Workers, 55 F.3d 64, 68 (2d Cir. 1995).
For claritys sake, the settlement notes that the released claims include future
claims based on the networks continued adherence to (1) rules or conduct left
unmodified by the settlement that are challenged or could have been challenged, (2)
rules or conduct modified by the settlement, and (3) rules or conduct substantially
similar to (1) or (2). SPA171 (Settlement 68(g)-(h)). That unremarkable provision
merely bars collateral attacks on continued adherence to the practices agreed upon
in the settlement. See In re Literary Works, 654 F.3d at 248 (release of claims
regarding future infringements is not improper).12
As the district court noted, there may well be room for litigation over whether
future rules are substantially similar, but the limitation ensures that only non-
substantive changes to the agreed-upon, going-forward rules and conduct are
released. SPA47. As the court reasonably explained, that limitation appropriately
cabins the release, and there is no need for an advisory opinion that would catalog
here all the claims that fall within or without the release. SPA47. That accords
with the settled principle that the court conducting the action cannot predetermine
12
Contrary to objectors insinuations, see MA-Br. 76, the release would not apply
if defendants were, after 2021, to revert to their previous rules.
75
Case: 12-4671
Document: 1124
Page: 88
10/15/2014
1345377
101
the res judicata effect of the judgment; this can be tested only in a subsequent
action. Fed. R. Civ. P. 23 advisory committee notes (1966).
The substantially similar limitation likewise readily distinguishes the
various cases, cited by objectors, in which courts have found releases of future
claims to violate public policy. In Lawlor v. National Screen Service, 349 U.S. 322
(1955), for instance, the Supreme Court suggested that a partial immunity from
civil liability for future violations would be consistent with neither the antitrust
laws nor the doctrine of res judicata. Id. at 329. The released claims, however,
involved conduct that was all subsequent to the judgment, did not even then
exist [at the time of settlement] and could not possibly have been sued upon in
the previous case. Id. at 328. This release, by contrast, bars only claims that are
alleged or which could have been alleged in this case, including future claims
arising out of the practices sanctioned in the structural reforms embraced by the
district court.13
13
See also Williams v. G.E. Capital, 159 F.3d 266, 274 (7th Cir. 1998) (enforcing
release of claims that even if not ripe were closely enough related to the
[released] disclosure claims that everything could be resolved in the settlement); In
re Managed Care Litig., 2010 WL 6532985, at *12 (S.D. Fla. 2010) (enforcing
release barring lawsuit based on continuation of pre-release conduct); Schwarz v.
Dall. Cowboys, 2001 WL 1689714, at *1 (E.D. Pa. 2001) (approving release of a
continuation of such policies, practices, contracts, conduct or provisions); see also
VKK Corp. v. NFL, 244 F.3d 114, 126 (2d Cir. 2001) (It is not uncommon for a
release to prevent the releasor from bringing suit against the releasee for engaging
in a conspiracy that is later alleged to have continued after the releases execution.).
76
Case: 12-4671
Document: 1124
Page: 89
10/15/2014
1345377
101
Nor can this release be said to grant antitrust immunity, when the challenged
conduct has not been held by the courts to be clearly illegal under the antitrust
laws. Robertson v. NBA, 556 F.2d 682, 686 (2d Cir. 1977). It is well-settled that a
court should not reject a settlement on grounds that it authorizes illegality if the
alleged illegality is not a legal certainty; that would in effect try the case by
deciding unsettled legal questions. Id.; see also Armstrong v. Bd. of Sch. Dirs., 616
F.2d 305, 321 (7th Cir. 1980) (before a settlement may be rejected because it
initiates or authorizes a clearly illegal or unconstitutional practice, prior judicial
decisions must have found that practice to be illegal or unconstitutional as a general
rule).
More generally, there is no basis to adopt objectors maximalist interpretation
of the release to conjure up a due process or public policy problem. Under bedrock
canons of contract construction and constitutional avoidance, this Court need only
read the release consistent with its standard terms, rather than impute an intent to
invite objectors parade of horribles. See In re Johns-Manville, 759 F.3d 206, 216
(2d Cir. 2014) (common canons of contract construction call upon us to reject
an interpretation that assumes order bound entities without constitutionally
sufficient notice). Indeed, the Due Process Clause is the ultimate backstop.
Precisely because a release cannot release claims in a manner that deprives future
litigants of their due process rights, releases are interpreted to reflect, not violate,
77
Case: 12-4671
Document: 1124
Page: 90
10/15/2014
1345377
101
those limits. See id. (interpreting order to bar claims only by those parties that
received constitutionally sufficient notice). Any future concerns that those limits
are being crossed can be addressed if and when such issues actually arise.
In short, the (b)(2) release here is a standard provision, fully consistent with
due process, that reflects the importance of the structural relief and the practical
necessity of giving defendants legal peace in exchange for the substantial relief
obtained. The district courts considered judgment that the release is a proper
component of a fair, reasonable, and adequate settlement warrants deference and
should be affirmed.
III.
The Fee Award Reasonable.
In a certified class action, the court may award reasonable attorneys fees.
Fed. R. Civ. P. 23(h). The key consideration is what is reasonable under the
circumstances. Goldberger, 209 F.3d at 47. The circumstances here include nearly
a decade of hard-fought litigation, the largest antitrust class action settlement award
in history, and injunctive relief that likely will prove more valuable still. The district
court approved a $544.8 million attorneys feeapproximately 9.56% of the net
cash fund, after opt-out reductionsas a reasonable award under these unique facts
and circumstances of the settlement. SPA69. That determination, laid out in a
dedicated fees opinion with painstaking transparency and detail, falls comfortably
within the district courts ample discretion and should be affirmed.
78
Case: 12-4671
Document: 1124
Page: 91
10/15/2014
1345377
101
The district court here grounded its analysis in the unique size, duration,
complexity, and relief of this case. SPA56. Class counsel and the nearly 60
additional law firms that worked on the case went toe-to-toe with a group of the
nations largest financial institutions and their talented counsel over an eight-year
period. They devoted, by conservative estimates, 500,000 hours of time to the case
without assurance of any compensation. The litigation was of singular size and
complexity, raising a plethora of difficult issues that went to the heart of how the
payment card industry has operated since its inception. SPA62. Each class
members share of that award was well below that which any class member would
have paid to prosecute this action and below what private plaintiffs typically pay.
See D.E. 2113-5 (Silver Decl.) at 25-34.
A handful of objectors take issue with the courts assessment of individual
factors under the traditional six-factor framework of Goldberger, 209 F.3d at 50
(factors include (1) the time and labor expended by counsel; (2) the magnitude and
complexities of the litigation; (3) the risk of the litigation ; (4) the quality of
representation; (5) the requested fee in relation to the settlement; and (6) public
policy considerations). But none take issue with the most important Goldberger
factorthe risk of the litigationwhich indisputably weighed in favor of a sizeable
fee. SPA59. As the district court explained, [i]f not for the attorneys willingness
79
Case: 12-4671
Document: 1124
Page: 92
10/15/2014
1345377
101
to endure for many years the risk that their extraordinary efforts would go
uncompensated, the settlement would not exist. SPA59.
Objectors instead appear to quibble with the district courts assessment of the
quality of representation and the requested fee in relation to the settlement,
rehashing their objections to the settlement itself. E.g., Unlimited Vacations Br. 9
(this is a negative value settlement where [c]lass members would be better off
with no settlement at all). Those hyperbolic criticisms are as unavailing in the fee
context as in the settlement context. As the district court found, the settlement
secured not only a massive damages award, but crucial programmatic reforms of
great value that together constitute a significant success. SPA60-61. Nothing
objectors say warrants disturbing that determination, much less overturning the
courts weighing of the Goldberger factors as a whole.
Objectors criticisms of the district courts graduated fee schedule similarly
rest on those mistaken premises. To calculate the fee, the district court adopted a
sliding scale that fixed the percentage of the fund to which counsel was entitled
through a declining schedule, thus addressing the worry that it is not ten times as
difficult to prepare, and try or settle a 10 million dollar case as it is to try a 1 million
dollar case. Goldberger, 209 F.3d at 52. Thus, the court awarded counsel 33% (a
common contingency fee arrangement in less complex class actions) of the first $10
million of the fund, 30% of the next $40 million, 25% of the next $50 million, and
80
Case: 12-4671
Document: 1124
Page: 93
10/15/2014
1345377
101
so forth, with percentages declining as the fund increased. The schedule was based
on empirical studies of federal class action settlements in recent years and the
unique facts and circumstances of the settlement. SPA64, 67, 69.
Objectors would like the scale to slide more steeply, rehashing their
complaints about the underlying settlement. Unlimited Vacations Br. 17; see id. at
19 (urging percentages of 4%, 3% and 2%, instead of Judge Gleesons 10%, 8%
and 6%). As explained, however, class counsel more than adequately represented
both the (b)(2) and (b)(3) classes. And in all events, the district court hardly exited
the realm of reasonableness in using one set of numbers rather than objectors
preferred figures.
Finally, some objectors appear to view any increase above the lodestar amount
as an undeserved windfall. In fact, courts regularly approve fees that reflect a
multiplier from the lodestar, and the multiplier here3.41is squarely in the range
of previously approved multipliers. As the district court explained, 3.41 is
comparable to (indeed, nearly identical to) the one I awarded in the WalMart case
ten years ago, and it is also comparable to multipliers in other large, complex cases.
SPA70; cf. In re WorldCom Sec. Litig., 388 F. Supp. 2d 319, 35459 (S.D.N.Y.
2005) (approving multiplier of 4.0 in $3.5 billion complex securities case); In re
Enron Sec., Derivative & ERISA Litig., 586 F. Supp. 2d 732, 803 (S.D. Tex. 2008)
([A] multiplier of 5.2 is warranted, given the unmatched size of the recovery,
81
Case: 12-4671
Document: 1124
Page: 94
10/15/2014
1345377
101
the obstacles and risks faced by [counsel] from the beginning, and the skill and
commitment exhibited by counsel.); In re Vitamin C Antitrust Litig., 2012 WL
5289514, at *10 (E.D.N.Y. 2012) (lodestar multiples of between 3 and 4.5 had
become common). And it reflects a smaller multiplier than that approved in Visa
Check, 297 F. Supp. 2d at 524, which this same district court presided over and
regarded as substantially less challenging. SPA63.
At bottom, there is no basis to disturb the district courts fee award as an abuse
of discretion.
IV.
The Settlement Notice Reasonable.
The standard for an adequate settlement notice, whether analyzed under the
Due Process Clause or Rule 23, is one of reasonableness. See SoberalPerez, 717
F.2d at 43; Fed. R. Civ. P. 23(e)(1). There are no rigid rules to determine whether
a settlement notice to the class satisfies constitutional or Rule 23(e) requirements.
Wal-Mart, 396 F.3d at 114. This Court has said that a notice need only fairly
apprise the prospective members of the class of the terms of the proposed settlement
and of the options that are open to them in connection with [the] proceedings.
Weinberger v. Kendrick, 698 F.2d 61, 70 (2d Cir. 1982). Accordingly, [n]umerous
decisions, no doubt recognizing that notices to class members can practicably
contain only a limited amount of information, have approved very general
description[s] of the proposed settlement. Id.
82
Case: 12-4671
Document: 1124
Page: 95
10/15/2014
1345377
101
The district court found that the notice supplied here easily met that standard:
It described the litigation, summarized the settlements terms, quoted the releases
verbatim, described the request for attorneys fees, expenses, and incentive awards
for Class Plaintiffs, and explained the deadline and procedure for filing objections
to the settlement as well as opting out of the case settlement class. SPA51. The
notice also notified class members of how they could obtain more information from
class counsel or the Class Administrator though a toll-free number, a website, and
traditional channels including mail and telephone. To ensure accessibility to the
average class member, 4 Newberg on Class Actions 11:53, class counsel even
consulted a plain language expert in the drafting, and made the notice and website
available in eight languages. D.E. 2111-7 (Azari Decl. 9); D.E. 2111-6 (Hamann
Decl. 31).
To maximize notice circulation, class counsel worked closely with the Class
Administrator to identify class members and compile a database of 19,874,922
unique mailing addresses, later supplemented by an additional 969,970. See D.E.
2111-6 12, 18-19. In all, the notice and publication campaign included more
than 20 million mailings and publication in more than 400 publications. SPA12.
Moreover, the notice prompt[ed] widespread reaction from class members, further
demonstrating that it had served its due process purpose. Handschu, 787 F.2d at
833; see D.E. 2111-6 26 (over 93,000 calls to toll-free number from December 2012
83
Case: 12-4671
Document: 1124
Page: 96
10/15/2014
1345377
101
through April 2013); id. 29, 32 (over 3.743 million website visitors in two
months).
The district court considered and rejected certain objectors claims that the
notice contained false statements. Objectors raised these arguments repeatedly, at
preliminary approval, at final approval, and at proceedings regarding the misleading
websites some objectors had created (where the district court nearly held several
objectors in contempt, SPA14). Each time, the district court rejected the challenges
to the settlement notice as meritless. Nonetheless, objectors reiterate their thrice-
rejected claim that the notice was misleading because any changes to the anti-
discounting network rules stem from the DOJ Consent Judgment, not from the
settlement. As the district court correctly held, the notice was fully accurate. It
appropriately refers to anti-discounting rule changes because the settlement creates
an affirmative network obligation to permit discounting, independent of the Consent
Judgment. Indeed, the settlement locks in the discounting reforms of the Consent
Judgment until 2021, ensuring that the changes will be unaffected even if the
Consent Judgment is vacated.
Objectors further contend that the notice contains no disclosure of the size of
the class, the aggregate damages suffered, the average loss per class member
quantified as a percentage of the class members sales to customers using Visa and
MasterCard, and the percentage of the aggregate damages that comprise the
84
Case: 12-4671
Document: 1124
Page: 97
10/15/2014
1345377
101
settlement benefit. Optical Etc. Br. 3. But those arguments fare no better. Neither
Rule 23 nor due process requires that the notice report the estimated value of
damages. Thompson, 216 F.R.D. at 67; see also id. (rejecting criticism that the
notice failed to detail class members individual benefits).
In the end, there is absolutely no basis to disturb the district courts
determination that the settlement notice was the best practicable, reasonably
calculated, under all the circumstances, to apprise interested parties of the pendency
of the action and afford them an opportunity to present their objections. Hecht,
691 F.3d at 224. By conveying enough information about the settlement and its
implications for participants to enable class members to make an informed decision
about whether to be heard concerning the settlement or, if allowed, to opt-out, 2
McLaughlin on Class Actions 6:17, the notice plainly suffices.
CONCLUSION
This class action resulted in one of the largest class recoveries ever realized.
It also resulted in wide-ranging changes to the way the payment card industry
operates, both directly through the injunctive relief realized, and indirectly through
the federal legislation and enforcement actions that this litigation prompted. The
monetary recovery and injunctive relief provided by the settlement grant significant
benefits to all class members, including compensation for past harm and protection
against future injury. In the judgment of the district court, and the highly
85
Case: 12-4671
Document: 1124
Page: 98
10/15/2014
1345377
101
experienced mediators who assisted the court in facilitating the resolution of these
complex and challenging claims, the settlement was not only appropriate, but
compelling.
On this record, the district court did not abuse its considerable discretion in
approving the settlement, awarding fees to class counsel, and in approving notice of
the settlement to class members. The decision of the district court should be affirmed
in all respects.
86
Case: 12-4671
Document: 1124
Page: 99
10/15/2014
1345377
101
Respectfully submitted,
s/Paul D. Clement
PAUL D. CLEMENT
Counsel of Record
JEFFREY M. HARRIS
CANDICE C. WONG
BANCROFT PLLC
1919 M Street NW
Suite 470
(202) 234-0090
pclement@bancroftpllc.com
K. CRAIG WILDFANG
THOMAS J. UNDLIN
RYAN W. MARTH
BERNARD PERSKY
ROBINS, KAPLAN, MILLER &
CIRESI L.L.P.
800 LaSalle Avenue
H. LADDIE MONTAGUE
MERRILL G. DAVIDOFF
MICHAEL J. KANE
1622 Locust Street
JOSEPH GOLDBERG
FREEDMAN BOYD GOLDBERG
URIAS & WARD, P.A.
20 First Plaza
Suite 700
BONNY E. SWEENEY
JOSEPH DAVID DALEY
ALEXANDRA SENYA BERNAY
ROBBINS GELLER RUDMAN &
DOWD LLP
655 West Broadway
Suite 1900
San Diego, CA 92101

October 15, 2014
87
Case: 12-4671
Document: 1124
Page: 100
10/15/2014
1345377
101
WITH TYPE-VOLUME LIMITATION
I hereby certify that:
1. This brief complies with the type-volume limitation of Fed. R. App. P.
32(a)(7)(B) and the order issued on October 2, 2014 granting Plaintiffs-Appellees
motion for leave to file a brief of up to 20,000 words because it contains 19,976
words, excluding the parts of the brief exempted by Fed. R. App. P. 32(a)(7)(B)(iii).
2. This Brief complies with the typeface requirements of Fed. R. App. P.
32(a)(5) and the typestyle requirements of Fed. R. App. P. 32(a)(6) because it has
been prepared in a proportionally spaced typeface using Microsoft Word 2013 in 14point font.
October 15, 2014
s/Candice C. Wong
Candice C. Wong
Case: 12-4671
Document: 1124
Page: 101
10/15/2014
1345377
101
I hereby certify that, on October 15, 2014, an electronic copy of this Brief for
Plaintiffs-Appellees was filed with the Clerk of Court using the ECF system, thereby
serving all counsel of record.
s/Paul D. Clement
Paul D. Clement
Case: 12-4671
Document: 1123
Page: 1
10/15/2014
1345294
98
12-4671- cv(L)

14-0584-cv(CON), 14-0606-cv(CON), 14-0663-cv(CON), 14-0837-cv(CON)
United States Court of Appeals

for the
Second Circuit
IN RE PAYMENT CARD INTERCHANGE FEE AND
MERCHANT DISCOUNT ANTITRUST LITIGATION
___________________________
for the Eastern District of New York (Gleeson, J.)
JOINT PAGE PROOF BRIEF OF

DEFENDANTS-APPELLEES
BENJAMIN R. NAGIN
EAMON P. JOYCE
MARK D. TATICCHI
SIDLEY AUSTIN LLP
787 Seventh Avenue
New York, NY 10019
CARTER G. PHILLIPS
SIDLEY AUSTIN LLP
1501 K Street, NW
(202) 736-8000
cphillips@sidley.com
DAVID F. GRAHAM
ROBERT N. HOCHMAN
SIDLEY AUSTIN LLP
One South Dearborn Street
Chicago, IL 60603
Attorneys for Defendants-Appellees Citigroup Inc., Citibank, N.A., and Citicorp

(See Inside Cover for Additional Counsel and Parties)
Case: 12-4671
Document: 1123
ROBERT C. MASON
ARNOLD & PORTER LLP
399 Park Avenue
New York, NY 10022
(212) 715-1000
robert.mason@aporter.com
ROBERT J. VIZAS
ARNOLD & PORTER LLP
Three Embarcadero Center, 10th Floor
San Francisco, CA 94111
MARK R. MERLEY
MATTHEW A. EISENSTEIN
ARNOLD & PORTER LLP
555 12th Street, NW
RICHARD J. HOLWELL
MICHAEL S. SHUSTER
DEMIAN A. ORDWAY
New York, NY 10004
(646) 837-5153
mshuster@hsgllp.com
Page: 2
10/15/2014
1345294
98
KEILA D. RAVELO
WESLEY R. POWELL
787 Seventh Avenue
New York, NY 10019
KENNETH A. GALLO
PAUL, WEISS, RIFKIND, WHARTON
& GARRISON LLP
2001 K Street, NW
(202) 223-7300
gallo@paulweiss.com
GARY R. CARNEY
PAUL, WEISS, RIFKIND, WHARTON
& GARRISON LLP
New York, NY 10019
Attorneys for Defendants-Appellees
MasterCard Incorporated and
MasterCard International Incorporated

Visa Inc., Visa U.S.A. Inc., and
Visa International Service Association *
(See Next Page for Additional Counsel and Parties)
Arnold & Porter LLP is counsel to the Visa Defendants-Appellees except as to Objectors-Appellants
Barnes & Noble, Inc., Barnes & Noble College Booksellers LLC, J.C. Penney Corporation, and The TJX
Companies, Inc. and related entities.
Case: 12-4671
Document: 1123
Page: 3
MARK P. LADNER
MICHAEL B. MILLER
250 West 55th Street
New York, NY 10019
(212) 468-8000
mladner@mofo.com
Attorneys for Defendants-Appellees Bank
of America, N.A., BA Merchant Services
LLC (f/k/a National Processing, Inc.),
Bank of America Corporation, and FIA
Card Services, N.A. (f/k/a MBNA
America Bank, N.A. and Bank of
America, N.A. (USA))
ANDREW J. FRACKMAN
ABBY F. RUDZIN
Times Square Tower
7 Times Square
New York, NY 10036
(212) 326-2000
afrackman@omm.com
10/15/2014
1345294
98
JAMES P. TALLON
New York, NY 10022
(212) 848-4000
jtallon@shearman.com
Barclays Bank plc (in its individual
capacity and as successor in interest to
Barclays Financial Corp.) and Barclays
Bank Delaware
RICHARD L. CREIGHTON
DREW M. HICKS
One East Fourth Street, Suite 1400
(513) 579-6400
rcreighton@kmklaw.com
Attorneys for Defendant-Appellee
Fifth Third Bancorp
Attorneys for Defendants-Appellees Capital

One Bank (USA), N.A., Capital One
F.S.B., and Capital One Financial
Corporation
(See Next Page for Additional Counsel and Parties)
Case: 12-4671
Document: 1123
Page: 4
JOHN P. PASSARELLI
JAMES M. SULENTIC
KUTAK ROCK LLP
The Omaha Building
1650 Farnam Street
Omaha, NE 68102
(402) 346-6000
John.Passarelli@KutakRock.com
First National Bank of Omaha
PETER E. GREENE
BORIS BERSHTEYN
PETER S. JULIAN
SKADDEN, ARPS, SLATE, MEAGHER
& FLOM LLP
Four Times Square
New York, NY 10036
(212) 735-3000
peter.greene@skadden.com
JONATHAN S. MASSEY
LEONARD A. GAIL
MASSEY & GAIL, LLP
1325 G STREET NW
Suite 500
(202) 652-4511
jmassey@masseygail.com
10/15/2014
1345294
98
ALI M. STOEPPELWERTH
WILMER CUTLER PICKERING HALE
AND DORR LLP
1875 Pennsylvania Avenue, NW
(202) 663-6589
ali.stoeppelwerth@wilmerhale.com
HSBC Finance Corporation and
HSBC North America Holdings Inc.
JOHN M. MAJORAS
JOSEPH W. CLARK
JONES DAY
(202) 879-3939
jwclark@jonesday.com
National City Corporation and
National City Bank of Kentucky

JPMorgan Chase & Co., JPMorgan
Chase Bank, N.A., Chase Bank USA,
N.A., Chase Manhattan Bank USA, N.A.,
Chase Paymentech Solutions, LLC, Bank
One Corporation, Bank One Delaware,
N.A., and JPMorgan Chase Bank, N.A.,
as acquirer of certain assets and
liabilities of Washington Mutual Bank
(See Last Page of Cover for Additional Counsel and Parties)
Skadden, Arps, Slate, Meagher & Flom LLP is counsel to the Chase Defendants-Appellees except as to
Objectors-Appellants American Express Co., American Express Travel Related Services Company, Inc.,
American Express Publishing Corp., Serve Virtual Enterprises, Inc., ANCA 7 LLC d/b/a Vente Privee,
USA, AMEX Assurance Company, Accertify, Inc., Wal-Mart, Inc., Alon USA, LP, Amazon.com,
Zappos.com, Foot Locker, Inc., and J.C. Penney Corporation, Inc. and related entities.
Case: 12-4671
Document: 1123
Page: 5
TERESA T. BONDER
VALARIE C. WILLIAMS
KARA F. KENNEDY
ALSTON & BIRD LLP
One Atlantic Center
1201 W. Peachtree Street, NW
Atlanta, GA 30309
(404) 881-7000
teresa.bonder@alston.com
SunTrust Banks, Inc. and SunTrust Bank
ROBERT P. LOBUE
WILLIAM F. CAVANAUGH
PATTERSON BELKNAP WEBB
& TYLER LLP
New York, NY 10036
(212) 336-2000
rplobue@pbwt.com
Wachovia Bank, N.A., Wachovia
Corporation, and Wells Fargo
& Company
10/15/2014
1345294
JONATHAN B. ORLEANS
ADAM S. MOCCIOLO
850 Main Street
Bridgeport, CT 06601
(203) 330-2000
jborleans@pullcom.com
Texas Independent Bancshares, Inc.
98
Case: 12-4671
Document: 1123
Page: 6
10/15/2014
1345294
98

Consistent with this Courts Order of May 27, 2014, see ECF Docket Entry
935 in No. 12-4671(L), the parties to this brief complied with Rule 26.1 of the
Federal Rules of Appellate Procedure by submitting a Compendium of Corporate
Disclosure Statements on June 16, 2014, see Docket Entry 988.
Case: 12-4671
Document: 1123
Page: 7
10/15/2014
1345294
98
TABLE OF CONTENTS
CORPORATE DISCLOSURE STATEMENT ..........................................................i
TABLE OF AUTHORITIES ....................................................................................iv
COUNTERSTATEMENT OF THE ISSUES............................................................ 1
STATEMENT OF THE CASE .................................................................................. 1
A.
Factual Background .................................................................... 3
B.
Commencement And Litigation of Plaintiffs Claims ................ 8
C.
Settlement.................................................................................. 13
1.
The Settlements Terms .................................................. 14
2.
Bases for Approval ......................................................... 17

ARGUMENT ........................................................................................................... 24
I.
JUDGE GLEESON PROPERLY APPROVED THE CLASS

SETTLEMENT. .................................................................................. 26
A.
The District Court Did Not Abuse Its Discretion

In Approving The Settlement.................................................... 26
1.
The Settlement Was Procedurally Fair. .......................... 27
2.
The Settlement Was Substantively Fair In Light Of

The Substantial Legal Defenses Plaintiffs Faced. .......... 28
3.
a.
Plaintiffs Faced Substantial Hurdles In

Proving A Conspiracy And Anticompetitive
Restraints. ............................................................. 31
b.
Even If Plaintiffs Could Establish Liability,

It Was Doubtful That They Would Obtain
Their Desired Remedies. ...................................... 39
The Settlement Was Substantively Fair In Light Of

The Relief Provided To The Class. ................................ 46
ii
Case: 12-4671
B.
II.
III.
Document: 1123
Page: 8
10/15/2014
1345294
98
The District Court Did Not Abuse Its Discretion In

Certifying The (b)(2) And (b)(3) Settlement Classes In
This Case, Based On The Specific Facts In This Record. ........ 50
1.
The Non-Opt-Out (b)(2) Class Settlement Properly

Resolved Existing Claims For Injunctive Relief,
Not Those For Existing Monetary Damages. ................. 51
2.
The (b)(2) Class Was Not Improperly Certified For

Settlement Because It Released Claims For Future
Liability Stemming From The Post-Settlement
Rules. .............................................................................. 56
a.
Dukes Does Not Preclude Certification Of

The 23(b)(2) Settlement Class. ............................ 60
b.
Shutts Also Does Not Support The

Argument That The Release Rendered
The (b)(2) Class Improper. ................................... 63
c.
This Courts Cases Also Do Not Lead To

A Different Outcome. ........................................... 63
THE SCOPE OF THE RELEASES PROVIDED BY THE

23(b)(2) CLASS IS LAWFUL. ........................................................... 66
A.
The Identical Factual Predicate Doctrine Permits

A Broad Release Of Claims. ..................................................... 67
B.
The Releases Here Are Tailored To The Identical

Factual Predicate Doctrine. ..................................................... 68
THE COMPETITORS OBJECTIONS ARE EQUALLY

MERITLESS. ...................................................................................... 72
A.
The Competitors Claims, As Competitors, Are Not

Released. ................................................................................... 73
B.
Discovers Group-Boycott Claim Lacks Merit......................... 75
CONCLUSION ........................................................................................................ 78
CERTIFICATE OF COMPLIANCE WITH FEDERAL RULE OF
APPELLATE PROCEDURE 32(a) .............................................................. 83
CERTIFICATE OF SERVICE ................................................................................ 84
iii
Case: 12-4671
Document: 1123
Page: 9
10/15/2014
1345294
98
Page
CASES
In re Am. Express Merchants Litig.,
634 F.3d 187 (2d Cir. 2011) ...............................................................................70
In re Am. Intl Grp., Inc. Sec. Litig.,
689 F.3d 229 (2d Cir. 2012) ...................................................................26, 50, 52
Amchem Prods., Inc. v. Windsor,
521 U.S. 591 (1997)......................................................................................49, 65
Armstrong v. Bd. of Sch. Dirs.,
616 F.2d 305 (7th Cir. 1980), overruled on other grounds by
Felzen v. Andreas, 134 F.3d 873 (7th Cir. 1998) ...............................................72
In re ATM Fee Antitrust Litig.,
554 F. Supp. 2d 1003 (N.D. Cal. 2008)..............................................................34
In re ATM Fee Antitrust Litig,
686 F.3d 741 (9th Cir. 2012), cert. denied sub nom.
Brennan v. Concord, EFS, Inc., 134 S. Ct. 257 (2013) ......................................41
Baby Neal ex rel. Kanter v. Casey,
43 F.3d 48 (3d Cir. 1994) ...................................................................................49
Barnes v. Am. Tobacco Co.,
161 F.3d 127 (3d Cir. 1998) ...............................................................................53
Bd. of Trade v. United States,
246 U.S. 231 (1918)............................................................................................31
Bishop v. Gainer,
272 F.3d 1009 (7th Cir. 2001) ............................................................................55
Blue Cross & Blue Shield United of Wis. v. Marshfield Clinic,
65 F.3d 1406 (7th Cir. 1995) ..............................................................................77
Bristol Vill., Inc. v. La.-Pac. Corp.,
916 F. Supp. 2d 357 (W.D.N.Y. 2013)...............................................................62
iv
Case: 12-4671
Document: 1123
Page: 10
10/15/2014
1345294
98
Buffalo Broad. Co. v. ASCAP,

744 F.2d 917 (2d Cir. 1984) ...............................................................................38
Charron v. Wiener,
731 F.3d 241 (2d Cir. 2013), cert. denied sub nom.
Suarez v. Charron, 134 S. Ct. 1941 (2014) ........................................................64
City of Detroit v. Grinnell Corp.,
495 F.2d 448 (2d Cir. 1974), abrogated on other grounds by
Goldberger v. Integrated Res., Inc., 209 F.3d 43 (2d Cir. 2000) .................28, 31
Cnty. of Suffolk v. Long Island Lighting Co.,
907 F.2d 1295 (2d Cir. 1990) .............................................................................25
Cooper v. Fed. Reserve Bank of Richmond,
467 U.S. 867 (1984)...................................................................................... 48-49
DAmato v. Deutsche Bank,
236 F.3d 78 (2d Cir. 2001) .................................................................................27
Gooch v. Life Investors Ins. Co. of Am.,
672 F.3d 402 (6th Cir. 2012) ..............................................................................62
Hansberry v. Lee,
311 U.S. 32 (1940)..............................................................................................49
691 F.3d 218 (2d Cir. 2012) ...............................................................................64
Huyer v. Wells Fargo & Co.,
295 F.R.D. 332 (S.D. Iowa 2013).......................................................................62
Ill. Brick Co. v. Illinois,
431 U.S. 720 (1977).....................................................................................passim
Image Technical Servs., Inc. v. Eastman Kodak Co.,
125 F.3d 1195 (9th Cir. 1997) ............................................................................45
218 F.3d 132 (2d Cir. 2000) ...................................................................2425, 75
Johnson v. Meriter Health Servs. Emp. Ret. Plan,
702 F.3d 364 (7th Cir. 2012) ..............................................................................62
v
Case: 12-4671
Document: 1123
Page: 11
10/15/2014
1345294
98
Kansas v. UtiliCorp United, Inc.,

497 U.S. 199 (1990)......................................................................................4041
Kartell v. Blue Shield of Mass., Inc.,
749 F.2d 922 (1st Cir. 1984)...............................................................................33
Kendall v. Visa U.S.A., Inc.,
518 F.3d 1042 (9th Cir. 2008) ............................................................................37
La. High Sch. Athletic Assn v. St. Augustine High Sch.,
396 F.2d 224 (5th Cir. 1968) ..............................................................................55
Lawlor v. Natl Screen Serv. Corp.,
349 U.S. 322 (1955)......................................................................................7071
In re Literary Works in Elec. Databases Copyright Litig.,
654 F.3d 242 (2d Cir. 2011) ........................................................................passim
Marisol A. v. Giuliani,
126 F.3d 372 (2d Cir. 1997) .........................................................................53, 54
Marcera v. Chinlund,
595 F.2d 1231 (2d Cir.), vacated on other grounds sub nom.
Lombard v. Marcera, 442 U.S. 915 (1979) ........................................................49
Matsushita Elec. Indus. Co. v. Epstein,
516 U.S. 367 (1996)............................................................................................48
In re Nassau Cnty. Strip Search Cases,
461 F.3d 219 (2d Cir. 2006) ...............................................................................54
Natl Bancard, Inc. (NaBanco) v. VISA U.S.A., Inc.,
596 F. Supp. 1231 (S.D. Fla. 1984), affd,
779 F.2d 592 (11th Cir. 1986) .....................................................................passim
Natl Bancard, Inc. (NaBanco) v. VISA U.S.A., Inc.,
779 F.2d 592 (11th Cir. 1986) .................................................................... passim
Norton v. Sams Club,
145 F.3d 114 (2d Cir. 1998) ...............................................................................69
Nottingham Partners v. Trans-Lux Corp.,
925 F.2d 29 (1st Cir. 1991)...........................................................................5758
vi
Case: 12-4671
Document: 1123
Page: 12
10/15/2014
1345294
98
N.Y. Cent. R.R. v. White,

243 U.S. 188 (1917)............................................................................................63
Ocean State Physicians Health Plan, Inc. v. Blue Cross &
Blue Shield of R.I., 883 F.2d 1101 (1st Cir. 1989) .............................................77
Ortiz v. Fibreboard Corp.,
527 U.S. 815 (1999)............................................................................................65
Parsons v. Ryan,
754 F.3d 657 (9th Cir. 2014) ..............................................................................55
Paycom Billing Servs., Inc. v. MasterCard Intl, Inc.,
467 F.3d 283 (2d Cir. 2006) ...............................................................................41
In re Payment Card Interchange Fee & Merchant Disc.
Antitrust Litig., 398 F. Supp. 2d 1356 (J.P.M.L. 2005)........................................9
In re Payment Card Interchange Fee & Merchant Disc.
Antitrust Litig.Opt Out Cases,
No. 1:14-md-1720-JG (E.D.N.Y. July 18, 2014) ...............................................32
Phillips Petroleum Co. v. Shutts,
472 U.S. 797 (1985)............................................................................................63
Reyns Pasta Bella, LLC v. Visa USA, Inc.,
442 F.3d 741 (9th Cir. 2006) ..................................................................13, 7172
Robertson v. NBA,
556 F.2d 682 (2d Cir. 1977) .........................................................................72, 76
Robertson v. NBA,
622 F.2d 34 (2d Cir. 1980) ...........................................................................67, 72
Robinson v. Metro-N. Commuter R.R.,
267 F.3d 147 (2d Cir. 2001), abrogated on other grounds by
691 F.3d 218 (2d Cir. 2012) ...................................................................49, 53, 54
San Diego Police Officers Assn v. San Diego City Emps. Ret. Sys.,
568 F.3d 725 (9th Cir. 2009) ..............................................................................57
vii
Case: 12-4671
Document: 1123
Page: 13
10/15/2014
1345294
98
SEC v. First Jersey Sec., Inc.,

101 F.3d 1450 (2d Cir. 1996) .............................................................................48
Stephenson v. Dow Chem. Co.,
273 F.3d 249 (2d Cir. 2001),
affd in part by an equally divided Court and vacated in part on other
grounds, 539 U.S. 111 (2003) ............................................................................64
Sullivan v. DB Invs., Inc.,
667 F.3d 273 (3d Cir. 2011) (en banc) ...............................................................60
Sykes v. Mel Harris & Assocs., LLC,
285 F.R.D. 279(S.D.N.Y. 2012) .........................................................................62
TBK Partners, Ltd. v. W. Union Corp.,
675 F.2d 456 (2d Cir. 1982) .........................................................................59, 69
Tennessean Truckstop, Inc. v. NTS, Inc.,
875 F.2d 86 (6th Cir. 1989) ................................................................................33
United States v. Am. Express Co.,
No. 10-cv-4496-NGG (E.D.N.Y. filed Oct. 4, 2010) .........................................11
In re Visa/MasterCard Antitrust Litig.,
295 F. Supp. 2d 1379 (J.P.M.L. 2003) .................................................................9
Wal-Mart Stores, Inc. v. Dukes,
131 S. Ct. 2541 (2011)................................................................................ passim
Wal-Mart Stores, Inc. v. Visa U.S.A. Inc. (In re Visa Check/Mastermoney
Antitrust Litig.), 280 F.3d 124 (2d Cir. 2001),
overruled on other grounds, Miles v. Merrill Lynch & Co. (In re
Initial Pub. Offerings Sec. Litig.), 471 F.3d 24 (2d Cir. 2006) ........13, 24, 6364
Wal-Mart Stores, Inc. v. Visa U.S.A. Inc.,
396 F.3d 96 (2d Cir. 2005) ..........................................................................passim
698 F.2d 61 (2d Cir. 1982) .....................................................................26, 2930
West Virginia v. Chas. Pfizer & Co.,
440 F.2d 1079 (2d Cir. 1971) .............................................................................30
viii
Case: 12-4671
Document: 1123
Page: 14
10/15/2014
1345294
98
Williams v. Gen. Elec. Capital Auto Lease, Inc.,

159 F.3d 266 (7th Cir. 1998) ..............................................................................66
STATUTE
15 U.S.C. 1693o-2.................................................................................................10
RULE
Fed. R. Civ. P. 23 ..............................................................................................passim
ix
Case: 12-4671
Document: 1123
Page: 15
10/15/2014
1345294
98
COUNTERSTATEMENT OF THE ISSUES

Did the District Court abuse its discretion in certifying separate settlement
classes under Rules 23(b)(2) and 23(b)(3) of the Federal Rules of Civil Procedure
and approving as fair, adequate, and reasonable the settlement entered into by
those classes?
Nine years ago, various merchants brought federal antitrust claims against
Visa, MasterCard and various member banks (collectively, Defendants)
challenging as anti-competitive the core structures undergirding the Visa and
MasterCard networks. Those networks permit widespread issuance and acceptance
of bank-issued payment cards, which enhances consumer purchasing power and
increases overall demand for merchant goods and services, while practically
eliminating the risk of non-payment for the millions of merchants who accept Visa
and/or MasterCard.
Facing dispositive motions presenting substantial defenses to their claims,
and facing additional uncertainty regarding their ability to obtain the remedies they
sought even if they prevailed on liability, the Class Plaintiffs settled with
Defendants. The settlement resulted from four years of negotiations, in which
Judge Gleeson, Magistrate Judge Orenstein, and two distinguished mediators
played significant hands-on roles.
Case: 12-4671
Document: 1123
Page: 16
10/15/2014
1345294
98
The court-approved settlement provides substantial monetary reliefup to

an estimated $7.25 billion (subject to reduction for opt-outs)to a Rule 23(b)(3)
class that had sought damages for alleged antitrust violations, and going-forward
relief to a Rule 23(b)(2) class. Only the going-forward relief is challenged here.
That relief makes a number of changes to the rules governing the Visa and
MasterCard networks that were sought by plaintiffs. Having agreed to make and
retain these changes to the networks, Defendants asked for, and the Class Plaintiffs
agreed to, releases cover[ing] claims that are or could have been alleged in this
case, SPA44, releases that protect Defendants against endless litigation about the
lawfulness of the networks rules to which plaintiffs had agreed. The court
explained that [i]n exchange for a new, going-forward rules structure, the
defendants are entitled to bargain for and receive releases of claims that are or
could have been alleged based on the identical factual predicate of the claims in
this case, and stated [t]hat is all these releases accomplish. SPA45SPA46.
Given the strength of Defendants legal and evidentiary hand, the possibility
that the court would uphold the lawfulness of the networks as they existed at the
time of the settlement was far more than theoreticala point underscored by the
court-appointed expert. After balancing the substantial monetary and injunctive
relief provided in the settlement against plaintiffs likelihood of success in the
litigation, the District Court properly concluded that the settlement was just, fair,
Case: 12-4671
Document: 1123
Page: 17
10/15/2014
1345294
98
and reasonable, and approved it. The result of that ruling is peace from further
challenges to the existing rules of the Visa and MasterCard networks, which were
the subject of the litigation, as modified pursuant to settlement. As explained
below, Objectors show no error in the District Courts approval of the settlement,
much less an abuse of discretion. The judgment should be affirmed.
A.
Factual Background
Consumers take for granted that a merchant in the Visa and MasterCard
networks will accept all versions of those cards as payment for the merchants
goods or services. It does not matter which bank issues the consumers Visa- or
MasterCard-branded card, and it does not matter which particular card issued by
that bank (e.g., cards earning miles, or cards offered to individuals just starting to
build their credit history) the consumer uses. This is so because the Visa and
MasterCard networks have been structured to ensure widespread acceptance,
which benefits consumers and merchants.
Visas and MasterCards networks are built to complete a complex multiparty transaction with no friction. As illustrated below, a typical transaction on the
Visa or MasterCard network involves five participants: cardholder, merchant, card
issuer (usually a bank), acquirer (which, again, is usually a bank, and which
contracts with the merchant and pays it promptly following a transaction), and the
payment card network itself. See SPA7
Case: 12-4671
Document: 1123
Page: 18
10/15/2014
1345294
98
McCormack Rep. 17, fig.4 (lodged with D.E. 2088).

When a consumer swipes a card or enters card information for online
purchases, the merchant collects the payment-card information and sends the
details of the transaction to the merchants bank (the acquiring bank), which then
forwards that information to the appropriate network. SPA7. The network, in turn,
relays the transaction data to the bank that issued the customers card (the issuing
bank), which confirms, among other things, that the customer has sufficient credit
(or sufficient funds in the debit card context) to cover the purchase. Id. If all is in
order, the issuing bank so advises the acquiring bank, which transmits that
Case: 12-4671
Document: 1123
Page: 19
10/15/2014
1345294
98
confirmation to the merchant and the sale of the merchants goods or services is
completed. Id. The entire approval process takes a few seconds.
That instantaneity has made the card system enormously successful with
consumers and merchants alike. Consumers benefit from quick payment times and
the ease of making purchases freed from the limitations of available cash-on-hand.
See, e.g., K. Murphy Rep. 254, 113 (lodged with D.E.2088). Merchants benefit
from a consuming public with ready access to funds, promoting larger purchasing
volumevolume which might shrink if customers had to endure a substantial wait
for transactions to clear. See, e.g., id. 8086, 113 & n.126, 255, 273293;
D.E.1550 186187. The card system also enables on-line purchases, where use
of cash or checks is not possible. K. Murphy Rep. 254.
Merchants not only reap enhanced sales volume through the payment card
networks, they also receive payment without having to wait for cardholders to pay
their bills to banks, and without incurring risk of payment default if cardholders
fail to do so. That is because after each consumer transaction, the merchants
acquiring bank promptly pays the merchant, pursuant to an agreement between the
merchant and the acquiring bank. 1 Typically, the acquirer deducts a fee, known as
the merchant discount fee, from the face amount of the transactions as payment
for its services and expenses. SPA7SPA8.
1
The acquiring market is highly competitive and contractual agreements between

merchants and acquirers vary widely.
5
Case: 12-4671
Document: 1123
Page: 20
10/15/2014
1345294
98
In connection with the acquiring banks payment to the merchant, the

acquiring bank also receives a payment from the issuing bank (via the Visa or
MasterCard network) with respect to the transaction. SPA7; D.E.1478-4 11.
That leaves the issuing bank with the right to demand payment from the consumer.
T. Murphy Decl. 11 (lodged with D.E.2088). That right comes with a delay in
receiving any payment and a risk of default by the consumer. D.E.1550 168.
Generally, the issuing bank cannot return to the acquiring bank and demand
compensation if the cardholder fails to pay, nor may the acquiring bank turn to the
merchant. Indeed, the issuing bank must pledge to pay for charges incurred on its
cards as a condition of joining the Visa or MasterCard network. D.E.1478-4 12.
Just as the acquiring bank deducts a fee from its payment to the merchant,
each issuing bank deducts a fee from its payment to the acquiring bank (i.e., the
issuer pays the acquiring bank less than the full price the merchant charged the
consumer). That fee is known as the interchange fee, SPA7, and addresses, among
other things, the cost of services an issuer performs and risks it assumes. See
D.E.1478-4 12; D.E.1550 157. Any particular acquiring bank and particular
issuing bank are free to reach separately negotiated agreements governing
compensation, whether generally or with respect to a particular merchant client of
the acquiring bank. See SPA7SPA10; SPA16; D.E.1550 181187. To
promote efficiency, however, each network sets a schedule establishing the
Case: 12-4671
Document: 1123
Page: 21
10/15/2014
1345294
98
default interchange fees that will govern transactions processed over the network
in the absence of a separately negotiated agreement between the acquiring and
issuing banks. SPA7SPA10; SPA16; D.E.1550 181187.
Default interchange rules enable the networks to operate seamlessly by
eliminating the need for each of thousands of issuers and acquirers to negotiate
appropriate interchange fees for every possible combination of merchant,
transaction value, and specific card product (e.g., Chase Freedom, Chase Sapphire
Preferred, Chase Slate, and so on). D.E.1550 181; SPA29SPA30. Such
individualized negotiations would entail substantial transaction costs and would
likely inhibit merchants ability to accept all versions of the Visa- and MasterCardbranded cards. Moreover, by providing a default interchange fee, a network
prevents the possibility of negotiation impasse or hold-upe.g., a particular
issuer that declines to accept transactions forwarded by an acquiring bank from a
particular merchant absent receipt of a much higher fee.
Other network rules also ensure the widespread acceptance of cards. For
example, the Honor-all-Cards rule require[s] merchants to accept all the
networks credit cards . . . when proffered for payment, regardless of which bank
issued the card. SPA19. This rule ensures that the Visa network, for instance,
functions as a Visa network. Customers can purchase secure in the knowledge that
their particular Visa card will be accepted wherever any Visa card is welcome.
Case: 12-4671
Document: 1123
Page: 22
10/15/2014
1345294
98
SPA31SPA32. The rule also assures issuers that all cards carrying a particular
networks mark will be accepted on an equal basis. Honor-all-Cards thus parallels
the rule requiring issuing banks to accept payment obligations passed to them from
merchants through the acquiring bank. Just as an issuing bank must honor all
payment obligations for cardholder purchases made over a network, so, too, must a
merchant honor all network-branded cards presented to it. D.E.1478-4 12; see
D.E.5965 at 1415 (report of court-appointed expert Dr. Sykes). As the District
Court found, the Honor-all-Cards and default interchange rules are closely
interrelated, lay at the heart of Visas and MasterCards efforts to build the
successful networks they now have, and undeniably have significant
procompetitive effects. SPA16; SPA31.
Finally, as yet another means of assuring uniform acceptance and a reliable
customer experience, Visa and MasterCard each have maintained a no-surcharge
rule. See SPA9; SPA18; D.E.1478-4 2627, 3031. Absent restrictions on
surcharging, for example, an Honor-all-Cards rule could be undermined, as a
given merchant could impose an exorbitant surcharge on a given card, rather than
refusing it outright. K. Murphy Rep. 134; see id. 130134.
B.
Commencement And Litigation of Plaintiffs Claims
No American court has ever found these practices anticompetitive. SPA30.

In fact, a previous challenge to interchange was tried before a federal district judge
Case: 12-4671
Document: 1123
Page: 23
10/15/2014
1345294
98
and rejected, and the Eleventh Circuit affirmed the judgment. See Natl Bancard,
Corp. (NaBanco) v. VISA U.S.A., Inc., 596 F. Supp. 1231 (S.D. Fla. 1984), affd,
779 F.2d 592 (11th Cir. 1986). Nevertheless, merchants and affiliated trade
organizations sued to challenge Visas and MasterCards interchange-related rules
and structures. The actions relevant to this appeal were filed beginning in June
2005 and later amended after consolidation before Judge Gleeson. SPA18. 2 From
the outset, plaintiffs sought to present their antitrust claims via two discrete classes:
a Rule 23(b)(3) class seek[ing] damages only, and a Rule 23(b)(2) class
seek[ing] declaratory and injunctive relief only. D.E.317 97(a)(b)
(complaint). Both putative classes asserted that Defendants conspired to fix
interchange fees. SPA6. Plaintiffs allegations focused on the three sets of
network rules discussed above: (i) default interchange, (ii) Honor-all-Cards, and
(iii) no-surcharge and other alleged anti-steering rules. See SPA8SPA9;
SPA18SPA19. The putative (b)(3) class sought damages to compensate
2
The Judicial Panel on Multidistrict Litigation assigned these actions to Judge

Gleeson. In re Payment Card Interchange Fee & Merchant Discount Antitrust
Litig., 398 F. Supp. 2d 1356, 1358 (J.P.M.L. 2005) (MDL-1720). They followed
years of class and opt-out litigation before Judge Gleeson by retailers concerning
the networks Honor-all-Cards rules, particularly as applied to acceptance of
debit cards, and default interchange rules. See id. (discussing Judge Gleesons
familiar[ity] with the operation of the credit card networks); In re
Visa/MasterCard Antitrust Litig., 295 F. Supp. 2d 1379, 138081 (J.P.M.L. 2003)
(assigning MDL-1575 to Judge Gleeson who has become thoroughly familiar
with the allegations . . . as a result of his seven year involvement with the New
York class action litigation).
9
Case: 12-4671
Document: 1123
Page: 24
10/15/2014
1345294
98
merchants for allegedly inflated default interchange rates in the past, and the
putative (b)(2) class sought injunctive relief to revise the Visa and MasterCard
networks going forward.
During the pendency of these actions, there were significant changes to
Visas and MasterCards corporate structures and businesses. See SPA10; SPA17
SPA19. First, MasterCard and Visa completed IPOs in 2006 and 2008,
respectively, which fundamentally changed their organizational structures. At the
time of the initial complaints, Visa and MasterCard were bankcard associations
comprised of member banks, which plaintiffs claimed were therefore structural
conspiracies. SPA19; D.E.317 131135. Through its IPO, each network
became a standalone publicly traded compan[y] with no bank governance.
SPA10. Unable to rely on their former structural conspiracy allegations,
Plaintiffs filed amended complaints in January 2009 insisting that the networks still
functioned as conspiracies among the banks and Visa or MasterCard. 3
Second, the Durbin Amendment in the Dodd-Frank legislation, see 15
U.S.C. 1693o-2(b)(3)(A)(i), modified the networks no minimum purchase
rules and discounting rules to allow merchants greater ability to steer consumers
away from using credit cards. SPA10 & n.6; SPA17.
See D.E.1170-4 10, 147162; D.E.1170-2 8, 135150; D.E.1170-3 261

270, 429442, 444.
10
Case: 12-4671
Document: 1123
Page: 25
10/15/2014
1345294
98
Third, through consent decrees entered in 2011 to resolve targeted antitrust

suits brought by the U.S. Department of Justice (DOJ), Visa and MasterCard
agreed to modify various rules. See SPA 10, SPA17. 4 Visas and MasterCards
modifications of their no discounting and non-discrimination rules broadened
merchants discretion to offer discounts and other incentives for a range of
alternative types of payment, including for using other credit card brands or debit
cards, and to encourage customers to use other forms of payment. See SPA10.
Alongside these developments, the parties vigorously litigated this case.
They disputed, inter alia, whether the default interchange, Honor-all-Cards, and
no-surcharge rules were necessary to produce the procompetitive efficiencies that
each network indisputably generated. Eventually, Dr. Alan O. Sykes of the New
York University School of Lawwhom Judge Gleeson appointed pursuant to
Federal Rule of Evidence 706 to advise the court on economic issuesconcluded
that plaintiffs face a substantial probability of failure in efforts to establish that the
core practices that would remain in place after the proposed settlement violate the
antitrust laws. D.E.5965 at 2223. The parties further disputed whether there
was no antitrust conspiracy as a matter of law post-IPOs.
The government also brought suit against American Express. See United States
v. Am. Express Co., 10-cv-4496-NGG (E.D.N.Y. filed Oct. 4, 2010). Closing
arguments in a bench trial were held on October 9, 2014.
11
Case: 12-4671
Document: 1123
Page: 26
10/15/2014
1345294
98
The parties also engaged in wide-ranging expert discovery, and the experts
sparred over what the but-for world would have looked like without the
challenged practices, including absent any default interchange fee. As Dr. Sykes
later observed: To the best of my knowledge, no general purpose (non-debit) card
network of any consequence has ever operated without significant interchange fees
(or substantial merchant fees in a three-party network). D.E.5965 at 19.
Plaintiffs experts argued that the MasterCard and Visa networks could survive
with zero interchange and hypothesized a but-for world designed to produce
such a result. In that imagined (and improbable) world, issuers would be required,
by the networks rules, to accept all merchant transactions from acquirers and
required to pay acquirers the full amount of the merchant transaction without
receiving any compensation from acquiring banks or their merchants, not even for
the very real risk of non-payment by the consumer. But, as Professor Sykes later
advised the District Court, survival is not an antitrust standard, and plaintiffs
experts failed to show that a zero interchange competitive equilibrium would
ever realistically emerge, even with hypothesized changes in the networks rules.
See id. at 18, 2124.
By 2011, many issues were fully briefed and awaiting rulings, including
cross-motions for summary judgment, Defendants motions challenging the
admissibility of plaintiffs experts opinions, and Defendants opposition to
12
Case: 12-4671
Document: 1123
Page: 27
10/15/2014
1345294
98
plaintiffs motion for class certification. See, e.g., SPA10. Judge Gleeson held
these motions in abeyance pending the outcome of the parties settlement
discussions. SPA11 & n.9.
C.
Settlement
Even as they were aggressively litigating, starting in 2008 the parties made
effortsassisted by mediators, Magistrate Judge Orenstein and Judge Gleesonto
settle their disputes. See SPA11SPA12. Settlement efforts intensified in late
2011. Id. The parties sought a final resolution of their ongoing disputes regarding
the legitimacy of the networks respective interchange fee rules and other
challenged rules. After all, Visa and MasterCard had earlier settled the class action
in In re Visa Check/MasterMoney Antitrust Litigation, which had likewise
challenged, inter alia, Honor-all-Cards and default interchange, only to find
themselves embroiled in a new class action making the same claims almost
immediately thereafter. 5 Defendants had no desire to continue litigating these
same issues about rules that Visa and MasterCard each perceived as central to their
See, e.g., Wal-Mart Stores, Inc. v. Visa U.S.A., Inc., 396 F.3d 96 (2d Cir. 2005)
(Wal-Mart II) (class settlement); Wal-Mart Stores v. Visa U.S.A. Inc. (In re Visa
Check/Mastermoney Antitrust Litig.), 280 F.3d 124 (2d Cir. 2001) (Wal-Mart I)
(class certification), overruled on other grounds, Miles v. Merrill Lynch & Co. (In
re Initial Pub. Offerings Sec. Litig.), 471 F.3d 24 (2d Cir. 2006); Reyns Pasta
Bella, LLC v. Visa USA, Inc., 442 F.3d 741 (9th Cir. 2006) (effect of Wal-Mart II
settlement and release).
13
Case: 12-4671
Document: 1123
Page: 28
10/15/2014
1345294
98
operations. And plaintiffs viewed settlement as an opportunity to obtain otherwise

uncertain relief and change the complexion of network rules. See SPA11.
Despite the difficult and contentious issues involved and the evolving
litigation landscape, the parties reached a settlement in summer 2012. SPA12.
Judge Gleeson granted preliminary approval in November 2012 and provisionally
certified separate settlement classes for each of the two putative classes asserted by
Plaintiffs: one under Rule 23(b)(2) (asserting claims for injunctive relief) and one
under Rule 23(b)(3) (asserting claims for damages). After exhaustive approvalrelated proceedings, Judge Gleeson granted final approval to the settlement in
December 2013.
1.
The Settlements Terms
In the settlement with the (b)(3) class, Defendants agreed to make monetary
payments to class members valued at up to approximately $7.25 billion (before
reductions for opt-outs), primarily for releasing their claims for monetary damages
that had accrued up to the date of preliminary approval. SPA13. This fund
represents the largest-ever cash settlement in an antitrust class action. SPA35.
The (b)(3) class permits opt-outs because it sought only damages for past conduct.
In the settlement with the (b)(2) class, Defendants agreed to a package of
relief that modified the networks respective rules prospectively from the date of
preliminary approval and ensured the continuity of certain changes that occurred
14
Case: 12-4671
Document: 1123
Page: 29
10/15/2014
1345294
98
during this litigation. The (b)(2) class settlement makes material rules
modifications sought by plaintiffs, including significant changes to the alleged
anti-steering rules. This going-forward relief was provided in connection with a
non-opt out (b)(2) class because it involves uniform, across-the-board prescriptions
for each networks operation; the network rules themselves have always been and
must be consistent with respect to all merchants. For that reason as well, the (b)(2)
class was defined to include both existing and future merchants. SPA118 2(b).
First, Visa and MasterCard agreed to allow merchants to surcharge on Visaor MasterCard-branded credit card transactions at the brand level (i.e., all Visa or
all MasterCard transactions) and product level (e.g., all Visa Signature
transactions). SPA13. Merchants thus won the ability to pass their card
acceptance costs on directly to their customers.
Second, the settlement lock[s]-in the Durbin Amendments minimumpurchase and discounting provisions and those in the consent decree with the DOJ.
Id. Defendants agreed to continue to abide by those requirements, regardless of
legislative, judicial, or other developments that would otherwise dissolve them.
See id.
Third, the settlement makes clear that merchants who operate different
businesses under different banners or trade names can accept Visa- and
MasterCard-branded cards at some of those businesses but not others. Id.
15
Case: 12-4671
Document: 1123
Page: 30
10/15/2014
1345294
98
Fourth, the settlement obligates Visa and MasterCard to negotiate in good

faith with groups of merchants that wish to negotiate with the networks
collectively rather than one-on-one. Id.
The District Court concluded that the settlement
preserves the integrity of the rules that made (and continue to
make) the networks successful. At the same time, by further
relaxing merchant restraints regarding pricing, it provides for
transparency and competition at the point of sale. Merchants
who choose to use the power the proposed rules changes give
them will be able to exercise control over (and perhaps reduce)
their costs from accepting Visa and MasterCard credit cards.
SPA32.
In exchange for the agreement to modify or eliminate these existing
practices, the (b)(2) class agreed to permit on a going-forward basis certain other
conduct that had been the subject of the claims for injunctive relief. SPA118
2(b). Additionally, the (b)(3) and (b)(2) classes agreed to release claims that are
or could have been alleged in this case. SPA44; see SPA134SPA136 33
((b)(3) class release); SPA169172 68 ((b)(2) release). As the District Court
explained, [i]n exchange for a new, going-forward rules structure, Defendants
bargain[ed] for and receive[d] releases of claims that are or could have been
alleged based on the []identical factual predicate[] of the claims in this case.
SPA45SPA46. The provisions do not release the defendants from liability for
claims based on new rules or new conduct or a reversion to the pre-settlement
16
Case: 12-4671
Document: 1123
Page: 31
10/15/2014
1345294
98
rules. SPA46. Similarly, while Defendants must adhere to the agreed-upon rules
modifications until 2021, see, e.g., SPA84 & SPA87 13(a)(b) & (k)(m);
SPA151 45; SPA164 58, if they enact rules that are not substantially similar
to those agreed-upon provisions, the releases would not apply to those provisions,
see SPA82 12(c)(vii), SPA88 16(b)(vii). 6
2.
Bases for Approval
Only .05% of the approximately 12 million merchants compri[sing] the

class objected to the settlement. SPA23. As the District Court observed, 90% of
the objections were submitted on boilerplate forms downloaded from websites that
disseminated false and misleading information for the precise purpose of
drumming up objections and opt-outs. Id. A number of legitimate objectors,
however, argued that the settlement was procedurally and substantively
unreasonable. SPA13-SPA14; see SPA34SPA47 (discussing objections).
After extensive approval-related proceedings, see SPA14, SPA7, Judge
Gleeson issued a lengthy opinion approving the settlement. SPA1SPA55. The
court found that the settlement secures both a significant damage award and
meaningful injunctive relief for a class of merchants that would face a substantial
likelihood of securing no relief at all if this case were to proceed. SPA15; accord
Compare Merchant Appellants (hereinafter Merchants) Br. 8 (alleging that the

settlement prevents merchants . . . from challenging anticompetitive conduct
forever).
17
Case: 12-4671
Document: 1123
Page: 32
10/15/2014
1345294
98
D.E.5965 at 3 (Sykes). The court highlighted three significant sources of risk with
respect to liability and to the relief plaintiffs requested. SPA15SPA16; SPA25
SPA32.
First, Judge Gleeson concluded that the networks IPOs would undermine
plaintiffs ability to prove part of the core conduct [they] sought to address,
namely that Visa and MasterCard member banks [. . .] effectively control the
decisions of both Networks by setting rules and interchange fees for the networks
to serve their collective interest. SPA28 (quoting complaint) (second alteration in
original). The networks restructurings, the court explained, brought them out
from under the control of their member banks, which strengthened the
defendants argument that the setting of interchange fees was a unilateral network
activity, not the result of some structural or walking conspiracy. Id.
Second, the court was skeptical that plaintiffs could prove the unlawfulness
of the networks rules and practicesparticularly default interchange and the
Honor-all-Cards rule. SPA16; SPA28SPA32. Default interchange, Judge
Gleeson concluded, played an essential role in the construction of the networks at
issue here, and those networks provide substantial benefit to both merchants and
consumers. SPA30. Without default interchange, network participants would
need to execute countless bilateral agreements regarding unique interchange rates,
which would inflate costs and impair the now seamless system. See SPA29
18
Case: 12-4671
Document: 1123
Page: 33
10/15/2014
1345294
98
SPA30; D.E.1550 181. As the District Court noted, courts, economists and
practitioners have agreed that the Honor-all-Cards rule and similar rules [are]
procompetitive under the Rule of Reason. SPA31. Quoting the public remarks of
Objectors lead counsel below, the District Court explained that such rules
represent:
a classic example of a restraint that was actually
necessary for the functioning of the joint venture. . . .
We all take it for granted, but you needed to have a rule
that ensured to you, as a consumer, that when you proffer
the Visa card, the merchant is going to take it. Its not
going to say, Ill take a Chase Visa card, but I dont like
Citibank, so Im going to turn that one down.
SPA32 (quoting Panel Discussion II: Consumer Issues at 56 (Statement of Jeffrey
Shinder) (Fordham Univ. Sch. of Law 2008) (reproduced at D.E.5939-3)).
Those procompetitive effects, together with DOJs recent decision not to
challenge the default interchange rules despite the entreaties by Class Counsel that
it do so . . . further sugges[t] that the plaintiffs antitrust challenge to the rules
could easily fail. SPA30.
Third, even if plaintiffs were able to establish an unlawful agreement with
predominantly anticompetitive effects, the court concluded that plaintiffs faced
significant risks as to the relief they sought. For instance, the District Court
recognized that there was a compelling argument that Illinois Brick Co. v. Illinois,
431 U.S. 720 (1977), barred the (b)(3) classs ability to recover damages and
19
Case: 12-4671
Document: 1123
Page: 34
10/15/2014
1345294
98
would foreclose any future claims for monetary relief related to the interchange
system. See SPA27SPA28 & n.15.
Additionally, even if Illinois Brick did not extinguish damages entirely, the
court doubted that plaintiffs could fully overcome the problems and complexities
inherent in proving damages to the jury, SPA32, particularly becauseas the
court-appointed expert concludedplaintiffs lacked a model demonstrating what
the payment card market would look like in the absence of the challenged rules.
SPA33.
Furthermore, the court recognized that plaintiffs faced additional hurdles as
to the injunctive relief they pursued. It explained that many of plaintiffs
requestssuch as judicial regulation of interchange feeswere simply outside the
power of the federal judiciary. SPA14; SPA17.
After weighing all of the foregoing, the District Court approved the
settlement on December 13, 2013, and Objectors timely appealed.
SUMMARY OF ARGUMENT
The District Court acted well within its discretion in certifying two
settlement classes, one pursuant to Rule 23(b)(2) and one pursuant to Rule
23(b)(3), and in approving the settlement. Appellants challenge the judgment on
essentially two fronts. They assert that the (b)(2) class obtained relief that was
inadequate in light of the purported strength of the Classs claims, and that Judge
20
Case: 12-4671
Document: 1123
Page: 35
10/15/2014
1345294
98
Gleeson violated Rule 23 and absent class members due process rights by
approving both the non-opt-out (b)(2) settlement class and the release provided by
that class settlement. In addition, several of the networks competitors raise a
series of narrow objections. None of these arguments has merit or demonstrates an
abuse of discretion.
I. A class-action settlement must be fair, adequate, reasonable, and not the
product of collusion between the parties. Here, an irreproachable settlement
process yielded a settlement that was fair, adequate, and reasonable, in all respects.
Both settlement classes received substantial relief. The Rule 23(b)(3) class
obtained monetary relief valued at more than $7 billion (before reductions for optouts), the sufficiency of which Objectors do not challenge here. The modifications
to the networks rules secured by the (b)(2) class were likewise more than
adequate, especially in light of the serious litigation risks confronting plaintiffs.
Those risks were daunting. Any hope of the (b)(2) class to obtain goingforward relief was clouded by the IPOs, which eliminated the argument that the
networks were structural conspiracies, left the banks with no control over the
networks policies at issue, and undercut plaintiffs ability to demonstrate the
threshold antitrust requirement of an agreement. Moreover, as the District Court
recognized, the challenged network rules have significant procompetitive features,
and thus the class faced substantial obstacles in showing that they were unlawful.
21
Case: 12-4671
Document: 1123
Page: 36
10/15/2014
1345294
98
Default interchange streamlines card acceptance and eliminates the need for
separate, bilateral negotiations over the interchange fee schedule. Honor-all-Cards
is an indispensable element of the seamless payment experience that has driven the
expansion of cardholding, card usage, and merchant sales. Plaintiffs had not
shown that a credit-card system could survive, much less flourish, without the
challenged network rules. Yet, such a showing was essential to establishing that
the challenged rules were unlawful restraints.
Plus, if plaintiffs somehow prevailed on liability, they faced an equally
onerous task at the remedies phase. Because merchants do not directly pay
interchange feesacquiring banks doplaintiffs faced the significant risk that
Illinois Brick bars any past or future damages claims. And, if damages were not
entirely precluded, plaintiffs still faced a grave risk, as the District Court and the
court-appointed expert observed, that they would not recover substantial sums,
because plaintiffs damages model rested on an implausible but-for world. As to
injunctive relief, the District Court explained that plaintiffs demand for a
wholesale reshaping of the payment card networks was more than a federal court
could provide, especially given changes to the networks that occurred during the
litigation. Given the many weaknesses in plaintiffs case, the relief afforded was
more than fair.
22
Case: 12-4671
Document: 1123
Page: 37
10/15/2014
1345294
98
The District Court also did not abuse its discretion in certifying a mandatory,
non-opt-out (b)(2) settlement class here. Contrary to Objectors contentions, the
proper focus of the certification inquiry is on the claims presented and pursued by
the class, not on the relief ultimately obtained or the issues compromised by the
class in exchange for that relief. Here, a putative (b)(2) class had been, since the
filing of the first class complaint, challenging the lawfulness of core network rules.
Those rules applied to the class on the whole.
As a consequence, it was entirely proper under both Rule 23 and the Due
Process Clause that the (b)(2) settlement class agreed to release its ability to
challengewhether in a damages action or one for injunctive reliefthe
lawfulness of the post-settlement network rules. Such concessions must be within
the power of a (b)(2) settlement class, otherwise defendants could not enter
meaningful Rule 23(b)(2) settlements. Without such a release, whatever remedies
Defendants agreed to, and whatever changes they agreed to adopt, would be
subject to a new round of legal challenges by the same group of plaintiffs the
moment the settlement was approved. Settlements would settle nothing. There
is no reason to treat cases involving a non-opt-out settlement class as immune from
a negotiated conclusion.
II. The releases here were proper because they simply reflect the boundaries
of the identical factual predicate doctrine. The releases do nothing more than
23
Case: 12-4671
Document: 1123
Page: 38
10/15/2014
1345294
98
release the claims that were, or could have been, asserted on the factual predicate
underlying this action. That is what the settlement agreement says, what counsel
represented to the District Court, and the basis of the courts approval decision. Of
course, as Judge Gleeson recognized, there may be cases in which it is not
immediately clear whether particular claims fall within or without the scope of
these releases. But future courts can examine the actual claims raised and facts
alleged in such future cases to determine whether the release bars them.
III. Finally, the claims raised by competitors American Express, First Data,
and Discover are unavailing. Their principal concernthat the settlement releases
claims they may hold in their capacities as competitors to Visa and MasterCard
is belied by the text and context of the agreement. Discover also asserts that the
settlement enshrines an unlawful group boycott against it as a competing network,
but that novel claim cannot meet the standard for showing that a settlement violates
the antitrust laws.
ARGUMENT
This Court reviews the approval of a class settlementincluding the
decision to certify the settlement classesfor abuse of discretion. Wal-Mart I, 280
F.3d at 132; Joel A. v. Giuliani, 218 F.3d 132, 139 (2d Cir. 2000). The trial
judges views are accorded great weight . . . because he is exposed to the litigants,
and their strategies, positions and proofs. . . . Simply stated, he is on the firing line
24
Case: 12-4671
Document: 1123
Page: 39
10/15/2014
1345294
98
and can evaluate the action accordingly. Joel A., 218 F.3d at 139 (alterations in
original) (quoting City of Detroit v. Grinnell Corp., 495 F.2d 448, 454 (2d Cir.
1974)); accord Wal-Mart II, 396 F.3d at 117; Cnty. of Suffolk v. Long Island
Lighting Co., 907 F.2d 1295, 1323 (2d Cir. 1990). Moreover, [this] considerable
deference . . . is heightened where the trial judges experience has imparted to the
judge a particularly high degree of knowledge. Joel A., 218 F.3d at 139.
This case is the archetype for applying heightened deference. Judge
Gleeson has spent 16 years on the interchange firing lineeight during this
litigation and another eight courtesy of earlier actions between various merchantplaintiffs and the network defendants. See supra at 9 n.2, 1314 & n.5.
Regardless of the degree of deference, however, the result here would be the
same. A painstaking, arms-length settlement process negotiated while the parties
vigorously litigated the fundamental issues raised by plaintiffs claims produced a
deal that provides meaningful backward- and forward-looking relief for plaintiffs
and is a fair compromise in all respects. Despite Objectors mountain of briefing,
they fail to raise any serious question regarding Judge Gleesons evaluation of the
propriety of the settlement classes or the procedural or substantive fairness of the
settlement. The judgment should be affirmed.
25
Case: 12-4671
I.
Document: 1123
Page: 40
10/15/2014
1345294
98
JUDGE GLEESON PROPERLY APPROVED THE CLASS

SETTLEMENT.
Before a court may approve any class settlement, it must determine (1) that
the settlement is fair, adequate, reasonable, and not a product of collusion, per Rule
23(e); and (2) that the requirements of Rule 23(a) and (b) have been met. In re Am.
Intl Grp., Inc. Sec. Litig., 689 F.3d 229, 239 & n.8 (2d Cir. 2012) (AIG). These
requirements were satisfied here.
A.
The District Court Did Not Abuse Its Discretion In Approving

The Settlement.
The central question raised by the proposed settlement of a class action is

whether the compromise is fair, reasonable and adequate. Weinberger v.
Kendrick, 698 F.2d 61, 73 (2d Cir. 1982). That inquiry requires evaluating both
the settlements terms and the negotiating process leading to settlement. WalMart II, 396 F.3d at 116. A presumption of fairness, adequacy, and
reasonableness may attach to a class settlement reached in arms-length
negotiations between experienced, capable counsel after meaningful discovery.
Id.
The settlement process here was unimpeachable. As such, the agreement is
presumptively valid. But even absent that deference, the record confirms that the
settlement was more than appropriate in every respect, and the judgment should
therefore be affirmed.
26
Case: 12-4671
1.
Document: 1123
Page: 41
10/15/2014
1345294
98
The Settlement Was Procedurally Fair.
The parties were represented by a host of experienced and able classaction counsel, SPA21, who litigated the case full-throttle even during the
pendency of settlement talks, SPA11. Discovery was not merely meaningful,
Wal-Mart II, 396 F.3d at 116, but exhaustive and completed before settlement.
The parties produced tens of millions of documents and took hundreds of fact and
expert depositions. Summary judgment motions were awaiting decision and all
litigants were well-positioned to appreciate the strengths and weaknesses of their
(and their opponents) positions. SPA10SPA11; SPA16. The District Court
found that the parties talks were fair and conducted at arms length and rejected
Objectors suggestion that there was an indicat[ion of] collusion. SPA21.
Moreover, four third-party neutrals facilitated negotiationstwo highly respected
outside mediators (former U.S. Magistrate Judge Edward Infante and Professor
Eric Green) and, in the later stages, Magistrate Judge Orenstein and Judge Gleeson.
See id.; SPA11 & n.9; see also DAmato v. Deutsche Bank, 236 F.3d 78, 85 (2d
Cir. 2001) ([A] court-appointed mediators involvement in pre-certification
settlement negotiations helps to ensure that the proceedings were free of collusion
and undue pressure.).
27
Case: 12-4671
Document: 1123
Page: 42
10/15/2014
1345294
98
Because the process was fair and collusion-free, the resulting agreement is
presumed to be fair, adequate, and reasonable. Wal-Mart II, 396 F.3d at 116.
Nothing in the record or the briefing on appeal upsets that presumption.
2.
The Settlement Was Substantively Fair In Light Of The

Substantial Legal Defenses Plaintiffs Faced.
While Defendants agree with and therefore join the Class Plaintiffs
substantive-fairness arguments, see Class Br. II, we write to underscore the
reasonableness of the settlements terms in light of the significant litigation risks
the Classes would have faced if (as Objectors wish) they had spurned settlement.
See Grinnell, 495 F.2d at 463 (requiring court to consider litigation risk, among
other factors, in evaluating the fairness of a settlement); SPA20.
Everyone but Objectors recognizes that the Class Plaintiffs position at the
moment of settlement was precarious. The viability of their core challenges to
default interchange and Honor-all-Cardsalready highly doubtful under the Rule
of Reason, even for the period that the networks remained nonpublic bankcard
associationshad further weakened after the Visa and MasterCard IPOs. And
even assuming, arguendo, some antitrust violation could be shown, the merchants
had little chance of securing the relief they sought. Between Illinois Brick, a
serious Daubert challenge to their principal expert on antitrust injury, the lack of a
plausible vision for a functional payment-card market without the challenged rules,
and difficulties in persuading a jury to award billions in damages here, plaintiffs
28
Case: 12-4671
Document: 1123
Page: 43
10/15/2014
1345294
98
stood little chance of securing a substantial award. Plaintiffs also were unlikely to
obtain the sweeping injunctive relief they had demandedincluding, among
other things, invalidation of Honor-all-Cards and the resetting of default
interchange to some indeterminate level between zero and its current levelsdue
to the inherent limitations on the federal courts power to regulate the marketplace.
See SPA14; SPA17.
Plaintiffs faced serious odds of complete failure, and counsel who actively
litigate are in the best position to understand the risks. On appeal, however,
Objectors ignore those risks. The lead briefthe Merchants Briefbarely
addresses litigation risk. It first mentions the cases merits on page 78, but fails to
address the procompetitive effects of the challenged rules, the IPOs, or plaintiffs
risks at the remedies phase. Compare SPA20SPA36. The Merchant Trade
Groups pay marginally more attention to the merits, see Br. II, but never
reconcile their optimism about the likelihood of success with Judge Gleesons
findings or the record. See id. at 50 (asserting, without analysis, that the plaintiffs
best possible recovery was total elimination of Honor-all-Cards, default
interchange, and the no-surcharge rules) (quoting Grinnell, 495 F.2d 463). To the
extent Objectors are suggesting that a court reviewing a settlement should ignore
the substantial weaknesses in plaintiffs claimsthat a settlement should be
29
Case: 12-4671
Document: 1123
Page: 44
10/15/2014
1345294
98
viewed as if plaintiffs were highly likely to prevail in fullthey are wrong as a

matter of law. See Weinberger, 698 F.2d at 7374.
Objectors also erroneously suggest that the District Court had to fully
adjudicate plaintiffs claims, faulting Judge Gleeson for (among other things) not
resolving whether the networks possess market power. See, e.g., Merchant Trade
Groups Br. 38 (charging that the District Court skipp[ed] the first two parts in the
antitrust analysis); id. at 3941, 46. In fact, [t]he district court [need] not
determine the validity of the [plaintiffs] claim . . . . The very purpose of a
compromise is to avoid the determination of sharply contested and dubious issues
. . . . West Virginia v. Chas. Pfizer & Co., 440 F.2d 1079, 1086 (2d Cir. 1971)
(citing In re Prudence Co., 98 F.2d 559, 560 (2d Cir. 1938)); id. at 1085. As a
consequence, it is unremarkable that the District Court devoted its risk assessment
to the points discussed below and did not dwell on market power (though, it made
clear that it was aware of the issue, SPA26 n.14).
The District Court conducted the requisite realistic assessment of the
prospects of further litigation. See SPA14SPA19; SPA25SPA36. Having done
so, it fully understood the obstacles facing plaintiffs, including plaintiffs inability
to: (a) prove an unlawful agreement causing anticompetitive harms, and (b) obtain
30
Case: 12-4671
Document: 1123
Page: 45
10/15/2014
1345294
98
the damages or injunctive relief they sought, even if they prevailed on the merits. 7
In light of those risks, the settlement was more than fair.
a.
Plaintiffs Faced Substantial Hurdles In Proving A

Conspiracy And Anticompetitive Restraints.
To establish a right to any reliefmonetary or injunctiveplaintiffs would

have to prove (among other things) that the networks challenged rules (1) were
agreements in restraint of trade; and (2) were unlawfulthat is, anticompetitive
restraints. See, e.g., Bd. of Trade v. United States, 246 U.S. 231, 23839 (1918).
As Judge Gleeson found, plaintiffs face[d] a substantial likelihood of securing no
relief at all. SPA15.
First, as to the agreement, plaintiffs initially asserted that Visa and
MasterCard were structural conspiracies or walking conspiracies by virtue of
their organization as bankcard associations of member banks. SPA19; see SPA28.
That contention, even if arguable when this case was filed, crumbled when the
networks completed their IPOs. Post-IPO, member banks no longer retained their
ownership of Visa or MasterCard, lacked voting control over the networks
These were not the only hurdles plaintiffs had to surmount. They faced strong
opposition on their class-certification motion and would have confronted practical
problem[s] in convincing a jury to award them damages. See, e.g., SPA33
SPA34. And, even if plaintiffs had managed to eke out any sort of victory, an
appeal to this Court, potentially followed by further proceedings in the Supreme
Court or on remand to the District Court, would have increased their chances of
non-recovery, not to mention prolonged the already lengthy period that they could
expect to wait before obtaining any relief. See Grinnell, 495 F.2d at 457.
31
Case: 12-4671
Document: 1123
Page: 46
10/15/2014
1345294
98
respective Boards of Directors, and had no power to adopt, maintain, or modify

any of the networks payment-card rules or policies. See D.E.1477-7 181191.
As the District Court recognized, the IPOs brought the networks out from under
the control of their member banks and thereby strengthened the . . . argument
that . . . the setting of interchange fees cannot constitute horizontal price-fixing.
SPA28.
The Merchant Trade Groups argue that, post-IPOs, both Visa and
MasterCard maintained default interchange (like other rules plaintiffs challenged),
and that this somehow indicates a conspiracy among the Defendants to retain the
pre-existing anticompetitive rules. Br. 1516 & n.8, 4749. But the persistence of
the challenged rules simply shows that the rules are a procompetitive feature of a
well-functioning system. Plaintiffs faced serious obstacles to proving that, postIPOs, Visa and MasterCard failed to make independent decisions regarding the
challenged rules.8
Second, the District Courts lengthy discussion of the network rules that
plaintiffs challenged highlights plaintiffs likely inability to prove an unlawful
8
On July 18, 2014, Judge Gleeson denied Defendants motions to dismiss opt-out
complaints brought by certain merchants. In re Payment Card Interchange Fee &
Merchant Disc. Antitrust Litig.Opt Out Cases, No. 1:14-md-1720-JG (E.D.N.Y.
July 18, 2014) (minute order and transcript (D.E.104 & 105). The court did so on
the pleadings without passing on the underlying merits of the opt-out claims or
revisiting its earlier statements in the judgment on appeal about the litigation risks
to plaintiffs.
32
Case: 12-4671
Document: 1123
Page: 47
10/15/2014
1345294
98
restraint of trade. SPA16; SPA28SPA32. The District Court determined that the
default interchange and Honor-all-Cards rules undeniably have significant
procompetitive effects, SPA16, and the Eleventh Circuit has held default
interchange lawful, NaBanco, 779 F.2d at 602. Furthermore, other courts have
found lawful no-surcharge rules similar to those challenged by plaintiffs. 9
Here, the District Court observed that default interchange lay at the core of
the defendants successful business model, SPA29, and underscored that it
played an essential role in the construction of the networks at issue here, SPA30.
The Merchants Trade Groups nakedly assert that what has been true historically is
not necessarily true today given the alleged matur[ity] of the networks. E.g., Br.
38. But they ignore that the underlying rationales which courts have previously
found compelling remain just as vital today. Likewise, they ignore the abundant
evidence on this subject before the District Court, including expert analyses
provided not only by Defendants but also by the courts independent expert. See,
e.g., K. Murphy Rep. 3234, 98, 202, 209, 219224; D.E.5965 at 8, 1322.
See, e.g., Tennessean Truckstop, Inc. v. NTS, Inc., 875 F.2d 86 (6th Cir. 1989);
Kartell v. Blue Shield of Mass., Inc., 749 F.2d 922 (1st Cir. 1984) (Breyer, J.).
Defendants showed that without restrictions on surcharging, merchants could
attempt to nullify Honor-all-Cards by imposing an exorbitant surcharge on a
given card, D.E.1477-7 163, and harm consumers by offering low prices in
advertisements but effectively raising the price through surcharging at the point of
sale, id. 160, or opportunistically surcharging consumers lacking payment
alternatives, id. 161.
33
Case: 12-4671
Document: 1123
Page: 48
10/15/2014
1345294
98
Default interchange fees serve functions that have not changed with the
passage of time. SPA29. Default interchange still obviates the need for thousands
of issuers and acquirers to negotiate separately. See SPA29SPA30. Default
interchange thus eliminates a massive transaction cost on a system that has grown
tremendously in the last few decades, thanks in large part to the rules
minimization of such costs. D.E.1550 181; see SPA29; T. Murphy Decl. 24
28; Sheedy Decl. 2427 (lodged with D.E.2088). Courts have highlighted this
procompetitive efficiency, concluding that default interchange rules are of vital
import to the day-to-day functioning of the system because they eliminate the
costly uncertainty and prohibitive time and expense of price negotiations at the
time of the exchange between the thousands of [network] members. NaBanco,
596 F. Supp. at 125960; see In re ATM Fee Antitrust Litig., 554 F. Supp. 2d 1003,
1007 (N.D. Cal. 2008).
Moreover, as the Eleventh Circuit concluded in NaBanco, [f]or a payment
system like VISA to function, rules must govern the interchange of the
cardholders receivable, because, absent prearranged interchange rules,
universality of acceptancethe key element to a national payment systemcould
not be guaranteed. 779 F.2d at 602. Absent a system of mandatory acceptance of
all network-branded cards, a customer walking into a store that purports to accept
Visa would have no assurance that the Visa-branded card issued by his or her
34
Case: 12-4671
Document: 1123
Page: 49
10/15/2014
1345294
98
particular bank would function at that store. Acceptance would depend on whether
compensation arrangements had been reached with that particular customers
issuing bank covering transactions from this particular merchant. The fundamental
importance of universal acceptancei.e., that any card bearing a network brand
will work at any merchant that purports to accept the brandto network
functioning and image has not changed with time. Instead, universal acceptance
remains central to the Visa and MasterCard brands. See, e.g., Elzinga Rep. at 16,
7174 (lodged with D.E.2088); K. Murphy Rep. 111, 206224; T. Murphy
Decl. 2527; Sheedy Decl. 2428. 10
Thus, the District Court observed that those networks provide substantial
benefit to both merchants and consumers. SPA30. The court noted that
Defendants showings about procompetitive effects were essentially undisputed,
and rejected Objectors suggestion that these beneficial practices have become
antitrust violations by virtue of industry maturation. SPA29; see SPA30 (similar).
Instead, without making any merits determination, Judge Gleeson endorsed the
court-appointed economic experts finding that plaintiffs face considerable
difficulty in establishing [that] default interchange . . . cause[s] anticompetitive
10
The record showed many other procompetitive features, including enhancing

cardholder rewards programs, which increase card use and, unsurprisingly,
merchant sales, D.E.1550 186; Sheedy Decl. 31; guaranteeing payment to
acquirers and merchants, even in cases of fraud or nonpayment, D.E.1550 187; T.
Murphy Decl. 29; and funding innovations and enhancements to the networks,
D.E.1550 182183; see Sheedy Decl. 18; SPA29.
35
Case: 12-4671
Document: 1123
Page: 50
10/15/2014
1345294
98
harm that outweighs [its] pro-competitive benefits. SPA31 (emphasis added)

(crediting Professor Sykes view that plaintiffs face a substantial probability of
securing little or no relief at the conclusion of trial (alteration omitted)); accord
SPA15. Similarly, the court recognized that the prospect that [the default
interchange rules] anticompetitive effects remain outweighed by its
procompetitive ones is real. SPA30. These observations are consistent with
NaBanco, decided when Visa was a bankcard association, in which the Eleventh
Circuit affirmed that the interchange fee is more procompetitive than
anticompetitive. 779 F.2d at 605. Moreover, Judge Gleesons findings
demonstrate the incorrectness of the Merchant Trade Groups assertion that the
District Court failed to recognize the need to weigh procompetitive effects against
any anticompetitive ones. Br. 46.
Likewise, the Merchants Trade Groups are wrong that the allegedly
anticompetitive effects of default interchange were largely undisputed. Br. 40
41. To the contrary, Defendants produced evidence that network output increased,
which is the opposite of what one would expect from an anticompetitive system.
D.E.5965 at 1011. In any event, the Trade Groups argument misses the point.
The key question facing the District Court was whether there was a real risk that
plaintiffs would fail to carry their burden of proving that any anticompetitive
effects outweigh the rules procompetitive benefits. The many and substantial
36
Case: 12-4671
Document: 1123
Page: 51
10/15/2014
1345294
98
procompetitive features created a serious risk that plaintiffs would fail to make that
showing, and thus there was no reason for Judge Gleeson to focus on the relative
difficulty of proving whether default interchange had any anticompetitive effects.
Finally, the District Court recognized that [n]o American court has ever
held that Visa or MasterCards default interchange rules violate the antitrust laws.
SPA30. In addition to endorsing the NaBanco courts findings, id., Judge Gleeson
recognized (id.) that the Ninth Circuit more recently affirmed the dismissal of
claims that Banks conspired to fix the interchange fee, holding that merely
charging, adopting or following the fees set by a Consortium is insufficient as a
matter of law to constitute a violation of Section 1 of the Sherman Act. Kendall
v. Visa U.S.A., Inc., 518 F.3d 1042, 1048 (9th Cir. 2008). Given the evidence here,
plaintiffs faced a serious risk that the court would follow these decisions and find
default interchange lawful. SPA30SPA31.
The District Court concluded that plaintiffs Honor-all-Cards challenge was
similarly tenuous, given the reality that assurances that a networks cards will be
accepted wherever the networks logo is displayed [are] critical to customers
desire to carry such cards and to merchants willingness to accept them. SPA31.
If merchants could choose which Visa- or MasterCard-branded cards to accept, the
concept of a network and a network brand would lose all meaning. The court
recognized that courts, economists and practitioners have therefore found Honor-
37
Case: 12-4671
Document: 1123
Page: 52
10/15/2014
1345294
98
all-Cards or similar rules to be procompetitive. Id. (emphasis in original)

(analogizing Honor-all-Cards to the system upheld in Buffalo Broadcasting Co. v.
ASCAP, 744 F.2d 917 (2d Cir. 1984)); SPA32 (reiterating that procompetitive
features cast doubt on plaintiffs ability to succeed); Sheedy Decl. 3335;
T. Murphy Decl. 3740. As detailed, supra at 19, the District Court observed
that Objectors lead counsel had described the rule as a classic example of a
restraint that was actually necessary. SPA32.
In addition, as Professor Sykes noted, a showing that default interchange
and related network rules . . . are anticompetitive requires . . . a convincing
description of a counterfactual world in which the purportedly anticompetitive
practices of each network are eliminated, and in which the resulting market
equilibrium is demonstrably superior from an economic standpoint. D.E.5965 at
1920. Absent a plausible explanation of how a payment card market could be
viably maintained in the absence of the challenged rules, plaintiffs stood little
chance of proving that those rules violate the Sherman Act. SPA31 (endorsing
Professor Sykes conclusion that it would be difficult to establish that the alleged
anticompetitive harm of Honor-all-Cards outweighs its procompetitive effects).
Yet, as Defendants Daubert filings and own expert reports showed, Objectors
never presented a reliable expert or other analysis of what a realistic counterfactual
world would look like. See D.E.5965 at 2325 (Sykes criticisms of plaintiffs
38
Case: 12-4671
Document: 1123
Page: 53
10/15/2014
1345294
98
theory). Objectors fail to come to grips with the substantial risk faced by the Class
Plaintiffs on the merits.
If that were not enough, legal developments beyond this litigation further
eroded plaintiffs claims by removing some of the practices they challenged as
unlawful. See SPA10 & n.6; supra at 1011 (discussing rules changes triggered by
the Durbin Amendment and the DOJ consent decree). This increased the chances
that the going-forward interchange system would not, on balance, be deemed an
unlawful restraint of trade.
b.
Even If Plaintiffs Could Establish Liability, It Was

Doubtful That They Would Obtain Their Desired
Remedies.
Plaintiffs also faced long odds as to the relief they sought. See SPA32
SPA33.
i. The District Court rightly recognized that Illinois Brick raised serious
doubts that merchants could ever recover any damages. SPA27SPA28 & n.15.
Illinois Brick not only would foreclose the accrued damages that the (b)(3) class
sought, but also would extinguish plaintiffs ability to recover any future monetary
damages allegedly resulting from the interchange system that remained in place
following the (b)(2) settlement. Objectors contention that surrendering future
claims for damages was a substantial sacrifice is off-base, in large part because
they never confront Illinois Bricks potential impact on those hypothetical claims.
39
Case: 12-4671
Document: 1123
Page: 54
10/15/2014
1345294
98
In Illinois Brick, the Supreme Court held that the Clayton Act permits only
direct purchasers of a product to sue over alleged price-fixing. 431 U.S. at 735
37; see id. at 72629. Here, merchants allege that Defendants fixed the price of
interchange fees. SPA22. But the acquiring banks, not merchants, pay those fees
directly to the issuers. The challenged network interchange fee rules address only
the acquiring-bank-to-issuing-bank payment obligation. Merchants typically pay
their acquirers a merchant discount fee, which the evidence showed is not required
by anything in the network rules, not fixed by horizontal agreement among
acquirers, and not homogeneous in practice. See, e.g., D.E.1478-4 4344, 51
57; D.E.1550 131, 152; see SPA7SPA8. Plaintiffs contend that the discount
fee simply passes on the cost of the interchange fee jot-for-jot, but the Supreme
Court rejected that as a reason to depart from the direct-purchaser-only rule. Ill.
Brick, 431 U.S. at 74344 (Respondents here argue . . . that pass-on theories
should be permitted for middlemen that resell goods without altering them and for
contractors that add a fixed percentage markup to the cost of their materials in
submitting bids. . . . We reject these attempts to carve out exceptions . . . for
particular types of markets. (footnote omitted)); accord Kansas v. UtiliCorp
United, Inc., 497 U.S. 199, 216 (1990) (ample justification exists for [the Courts]
decision not to carve out exceptions to the [direct purchaser] rule for particular
types of markets.) (second alteration in original).
40
Case: 12-4671
Document: 1123
Page: 55
10/15/2014
1345294
98
This Court and the Ninth Circuit have ruled Illinois Brick bars damages
sought by downstream payors of payment-card fees and charges thatlike
interchange feesan intermediate party pays in the first instance. See Paycom
Billing Servs., Inc. v. MasterCard Intl, Inc., 467 F.3d 283, 29192 (2d Cir. 2006);
In re ATM Fee Antitrust Litig., 686 F.3d 741, 74445, 74950 (9th Cir. 2012), cert.
denied sub nom. Brennan v. Concord, EFS, Inc., 134 S. Ct. 257 (2013). Indeed, as
Judge Gleeson recognized, the Ninth Circuits ruling, which addressed another
system of interchange payments, rejected the same exceptions to the Illinois Brick
rule that plaintiffs here have relied on. SPA28 n.15 (citing ATM Fee, 686 F.3d at
75058). Thus, the District Court correctly concluded, the indirect purchaser
doctrine would be a source of significant uncertainty for the plaintiffs. SPA28.
Although Objectors fail to address Illinois Bricks application to plaintiffs
claims here, 11 the Merchant Trade Groups dismiss Judge Gleesons discussion of
the case, contending that the court committed a legal error in using Illinois Brick
to justif[y] the (b)(2) Settlement because Illinois Brick applies only to damages,
not injunctive relief. Br. 35. The argument is meritless.
To start, that is not what the District Court did. It acknowledged that Illinois
Brick holds only that indirect purchasers may not recover antitrust damages,
11
The Merchants, American Express, Blue Cross Blue Shield, Discover, First
Data, and the Retailers and Merchants Objectors do not cite it. U.S. PIRG appears
to recognize the potential merit of the Illinois Brick argument. Br. 26 n.6.
41
Case: 12-4671
Document: 1123
Page: 56
10/15/2014
1345294
98
SPA27, and repeatedly discussed the doctrine in that context, see, e.g., id. (trebledamages antitrust actions); id. (actions for antitrust damages); SPA28 (similar).
The court discussed Illinois Bricklike all of the litigation risks here, see
SPA25not as a risk to the (b)(2) class, but to plaintiffs generally. Moreover, the
Merchant Trade Groups claim that the (b)(2) settlement unjustifiably releases
merchants claims for future damages. Br. 51 (emphasis added, capitalization
omitted). As discussed below, there is nothing improper in the release. But to the
extent the Merchant Trade Groups are suggesting that plaintiffs gave up something
of value when releasing claims for future damages, the viability and value of those
claimsincluding the impact of Illinois Brickis highly relevant.
ii. Even if Illinois Brick was not an absolute bar, the history of antitrust
litigation is replete with cases in which antitrust plaintiffs succeeded at trial on
liability, but recovered no damages, or only negligible damages. Wal-Mart II,
396 F.3d at 118 (internal quotation marks omitted). Any past or future damages
claim here faced that risk. Given plaintiffs claim that the default interchange
system overcharges them, it was essential for plaintiffs to construct a rigorous
model for ascertaining what interchange rates would prevail in a payment card
market that lacked default interchange and Honor-all-Cards.
No such model exists. Although plaintiffs suggested a number of
possibilities, including an interchange fee of 0.0% and an interchange fee that was
42
Case: 12-4671
Document: 1123
Page: 57
10/15/2014
1345294
98
equivalent to the rate charged for debit transactions, D.E.5965 at 22 (Sykes Rep.),
those scenarios are unrealistic. Judge Gleeson found that [i]t is not likely that
credit card interchange fees would . . . become zero. SPA33. As Professor Sykes
explained, zero interchange is implausible because, at least to [his] knowledge[,]
no general purpose credit or charge card network of any consequence has ever
evolved with zero interchange. D.E.5965 at 2324. And debit-rate interchange
was unlikely given that credit cards are more costly to issue than debit cards due to
their many advantages to consumers (e.g., credit itself, float, purchase protection,
and reward programs). See SPA33; D.E.5965 at 16, 24.
iii. Finally, the Class Plaintiffs faced major obstacles as to the injunctive
relief sought, most notably the inherent limits on a federal courts remedial
authority and the many changes to the legal landscape since the case began.
Over the course of the litigation, there were numerous changes that
effectively eliminated many of the network structures that plaintiffs had
complained of: The IPOs terminated the banks control of Visa and MasterCard;
Dodd-Frank expanded merchants discounting authority; and the settlement with
DOJ enlarged the merchants discounting power still further. Additionally, the
settlement here, of course, permits surcharging on a going forward basis. 12 All of
12
Objectors decry the surcharging relief because some states laws prohibit
surcharging, which means that some merchants will not be able to surcharge Visa
and MasterCard transactions even in a post-settlement world. See, e.g., Merchants
43
Case: 12-4671
Document: 1123
Page: 58
10/15/2014
1345294
98
these developments both weakened plaintiffs claim that the going-forward system
was anticompetitive, see supra I.A.2.a, and narrowed the range of remedies
available to the District Court (beyond that achieved in the settlement) in the event
it concluded that interchange fees were in fact set at supracompetitive levels.
Put simply, by the time the District Court was evaluating the prospects of
further litigation in light of the settlement, the only other injunctive remedies that
remained possible concerned default interchange and Honor-all-Cards. But, as
discussed in detail above, those two features of the core network rules had been
repeatedly recognized by courts as, on balance, procompetitive, the record in this
case confirmed the continuing importance of those rules to the universal
acceptance that is central to the Visa and MasterCard brands, and both the courtappointed expert and Judge Gleeson expressed serious doubts that plaintiffs could
mount a successful challenge on the merits to those rules. Supra I.A.2. Any
request to enjoin default interchange and Honor-all-Cards out of existence entirely
would have required precisely the evidence plaintiffs had failed to bring forth: an
explanation of how a supposed-restraint-free world would have produced networks
without such rules at all.
Br. 15, 2223. The Classs brief discusses the going-forward importance of the
surcharging relief, Class Br. II.A.2, and we touch on the importance the class
placed on such relief throughout this litigation below, infra I.A.3.
44
Case: 12-4671
Document: 1123
Page: 59
10/15/2014
1345294
98
Plaintiffs might have asked the court to impose a different default

interchange rate than that imposed by the networks. But judicial regulation was
improbable, because courts enforcing the antitrust laws cannot set prices. See, e.g.,
Image Technical Servs., Inc. v. Eastman Kodak Co., 125 F.3d 1195, 1225 (9th Cir.
1997). But cf. Merchants Br. 14 (complaining that the settlement immunizes from
suit the default interchange schedules). Any effort to urge judicial regulation
also was compromised by the lack of a convincing counterfactual model showing
that a payment-card system could flourishor, for that matter, even function
with modified or judicially regulated interchange fees and card-acceptance
policies. As Judge Gleeson summarized the weaknesses in plaintiffs case for
injunctive relief:
Even if the plaintiffs spent several years pursuing this unwieldy
case to a successful conclusion (despite substantial odds against
such a result), this Court would be in no position to grant the
sweeping relief the objectors seek. It cannot regulate
interchange fees or enjoin nonparties or preempt state laws or
reform network rules that do not violate the antitrust laws. The
Sherman Act affords relief only from certain proven
anticompetitive business practices.
SPA17 (emphases added); see SPA14.
In sum, plaintiffs had little hope of obtaining any of the relief they sought.
The rules changes and billions of dollars they secured through this settlement
would have been in substantial doubt if this case went forward.
45
Case: 12-4671
3.
Document: 1123
Page: 60
10/15/2014
1345294
98
The Settlement Was Substantively Fair In Light Of The

Relief Provided To The Class.
Objectors offer no challenge to the more than $7 billion in monetary

compensation made available to merchants in the (b)(3) class. Instead, Objectors
complain that the changes to the network rules agreed to in the 23(b)(2) settlement
supposedly provide little more than a peppercorn, in exchange for which they were
forced to sacrifice valid and valuable claims. See Merchants Br. 3848; Merchant
Trade Groups Br. 3160; SPA36SPA37.
As discussed above, plaintiffs claims were seriously in doubt. Moreover,
Objectors are wrong in belatedly suggesting (post-settlement) that the ability to
surcharge is a recently invented (and insignificant) consideration. In reality,
obtaining the ability to surcharge was among the Classs primary goals from the
litigations outset. See, e.g., D.E.1170-2 49 (alleging that Defendants insulated
their anticompetitive practices from competitive pressures by adopting and
enforcing the No-Surcharge Rule and other Anti-Steering Restraints); id. 53. 13
As the JPML summarized in its initial transfer order assigning this MDL to Judge
Gleeson, [a]ll actions share factual questions arising out of allegations that the
13
Accord D.E.1170-3 190 (Visa and MasterCard impose the No-Surcharging

Rule . . . to prevent Merchants from incenting consumers to use less-expensive
payment methods.); id. 197 (no-surcharge rule insulat[es] Defendants from
competition and rais[es] prices for all consumers); id. 330 (The Anti-Steering
Restraints (and particularly the No-Surcharge Rule) are anticompetitive vertical
restraints.); id. 328329, 336338 (similar).
46
Case: 12-4671
Document: 1123
Page: 61
10/15/2014
1345294
98
imposition of a no-surcharge rule and/or the establishment of the interchange fee

causes the merchant discount fee to be set at supracompetitive levels. 398 F.
Supp. 2d at 1358 (emphasis added).
The Class Plaintiffs maintained that focus at summary judgment, long before
the settlement, describing surcharging as the most effective tool for merchants to
influence consumers payment choices. Mem. Supp. Summ. J. (D.E.1538) at 8
(citing expert reports); see, e.g., id. at 36, 54. As Judge Gleeson summarized, it
was a rule change that the Class and individual plaintiffs fought very hard to
obtain. SPA36. Accordingly, Objectors new antipathy toward that relief is
suspect.
Objectors also challenge the (b)(2) settlements value by claiming that the
settlement creates a worse result than if the Plaintiffs tried and lost the case.
Merchant Trade Groups Br. 52. This line of argument is doubly flawed. First, it
is principally a challenge to the scope of the releases, yetas discussed in depth
infra IIboth the language and effect of the releases are standard, and [t]he law
is well-established in this Circuit and others that class action releases may include
claims not presented and even those which could have not been presented. WalMart II, 396 F.3d at 107; accord Matsushita Elec. Indus. Co. v. Epstein, 516 U.S.
367, 37677 (1996). Second, Objectors rest their argument on the indefensible
notion that, if a certified (b)(2) class had lost at trial, the exact same plaintiffs
47
Case: 12-4671
Document: 1123
Page: 62
10/15/2014
1345294
98
could have filed an identical suit the following day becausein Objectors view
each card swipe made under the networks rules would be a wholly new antitrust
violation so different from those alleged in the earlier action that it would
somehow evade traditional rules of preclusion. Merchant Trade Groups Br. 52.
Objectors are off the mark. A final judgment on the merits of an action
precludes the parties or their privies from relitigating issues that were or could
have been raised in that action. SEC v. First Jersey Sec., Inc., 101 F.3d 1450,
1463 (2d Cir. 1996) (emphases added, alteration omitted) (quoting Federated
Dept Stores, Inc. v. Moitie, 452 U.S. 394, 398 (1981)). Res judicata represents
finality as to the claim or demand in controversy, concluding parties and those in
privity with them, not only as to every matter which was offered and received to
sustain or defeat the claim or demand, but as to any other admissible matter which
might have been offered for that purpose. Id. (quoting Nevada v. United States,
463 U.S. 110, 12930 (1983)).
In other words, if a (b)(2) litigation class had been certified and had
judgment entered against it, absent class members could not simply have relitigated the lawfulness of default interchange, Honor-all-Cards, no-surcharge, and
the other rules at issue here. See id. [U]nder elementary principles of prior
adjudication a judgment in a properly entertained class action is binding on class
members in any subsequent litigation. Cooper v. Fed. Reserve Bank of
48
Case: 12-4671
Document: 1123
Page: 63
10/15/2014
1345294
98
Richmond, 467 U.S. 867, 874 (1984). A judgment in favor of the defendant
extinguishes the claim, barring a subsequent action on that claim. Id. 14 Indeed, if
preclusion were not the result, trials of Rule 23(b)(2) claims would be little more
than an endless moot court for plaintiffs counsel, who could continue to press the
same claims over and over while hoping for a different result.
Thus, when plaintiffs settled, they did not put themselves in a position worse
than unsuccessful litigation would have produced. Instead, in exchange for
receiving going-forward settlement relief with respect to some of the rules, they
agreed to foreclose continued challenges to other rules. That type of compromise
is what settling parties always do, and the District Court properly found that
compromise fair and reasonable given the litigation risks.
14
Rule 23(e) itself, see Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623 (1997),
as well as the adequacy of representation doctrine under Rule 23(a)(4) (requiring
the representatives adequately protect the interests of the class), provide the
bulwark against unfairly binding an absent class member to a non-opt-out class
judgment. See, e.g., Hansberry v. Lee, 311 U.S. 32, 4243 (1940); Robinson v.
Metro-N. Commuter R.R., 267 F.3d 147, 165 (2d Cir. 2001), abrogated on other
grounds by Hecht v. United Collection Bureau, Inc., 691 F.3d 218 (2d Cir. 2012);
Marcera v. Chinlund, 595 F.2d 1231, 1240 n.13 (2d Cir.) ([D]ue process permits
binding absentees to a judgment with respect to common questions of law if they
have been adequately represented in the suit.), vacated on other grounds sub nom.
Lombard v. Marcera, 442 U.S. 915 (1979); see also Baby Neal ex rel. Kanter v.
Casey, 43 F.3d 48, 59 (3d Cir. 1994) (analysis of (b)(2) class certification
recognizes that absent plaintiffs will be[] bound by such judgment in the
subsequent application of principles of res judicata.). Defendants already have
shown why Rule 23(e) is satisfied here, supra I.A, and embrace the Classs
discussion of why its representation was adequate (and the Classs explication of
why the other Rule 23(a) factors are satisfied here). See Class Br. I.
49
Case: 12-4671
B.
Document: 1123
Page: 64
10/15/2014
1345294
98
The District Court Did Not Abuse Its Discretion In Certifying The
(b)(2) And (b)(3) Settlement Classes In This Case, Based On The
Specific Facts In This Record.
A district court that is presented with a proposed class settlement must,

separate and apart from the Rule 23(e) analysis, determine whether the
requirements for class certification in Rule 23(a) and (b) have been satisfied.
AIG, 689 F.3d at 238. Judge Gleeson did so here. SPA51SPA53 & n.20; see also
SPA36SPA43. As the Class Plaintiffs brief shows, Judge Gleeson did not abuse
his discretion in finding those prerequisites met in this case. Several points bear
additional mention.
Objectors assert that a (b)(2) class was improperly certified for settlement
because the case involved claims for money damages and injunctive relief, and the
settlement extinguishes claims for money damages. See Merchants Br. at 3266.
Objectors arguments blend a certification objection with an objection to the
release of speculative damages claims that might arise in the future, and assert
that reversal is proper because they were not permitted to opt out of the (b)(2)
class. See, e.g., id. at 3236, 4146, 52, 60. Objectors are mistaken about the facts
and the law.
50
Case: 12-4671
1.
Document: 1123
Page: 65
10/15/2014
1345294
98
The Non-Opt-Out (b)(2) Class Settlement Properly

Resolved Existing Claims For Injunctive Relief, Not Those
For Existing Monetary Damages.
Objectors challenge rests on the oft-repeated premise that the (b)(2) class
sought to resolve individualized claims for money damages. Merchants Br. 32
(capitalization omitted).15 The premise is false. From the outset of this case, the
putative (b)(2) class sought injunctive relief only. See D.E.317 97(b). The (b)(2)
class did not pursue any present, already-accrued claim for damages, and did not
receive any monetary payments through the settlement agreement. SPA84SPA87
13; SPA139 39 ([m]embers of the Rule 23(b)(2) Settlement Class shall receive
no money payments but shall receive only the rules modifications detailed in the
agreement). The (b)(2) release also does not require plaintiffs to forgo any
already-accrued claim for money damages. See SPA90 16.c; SPA92 16.c.ix
(releasing, inter alia, claims for damages or other monetary relief relating to the
period after the date of the Courts entry of the Class Settlement Preliminary
Approval Order). Any plaintiff who wishes to continue to litigate a claim for
money damages brought in this case may opt out of the (b)(3) class and face the
going-forward litigation risks, as various Objectors have done. The (b)(2) class
15
See also, e.g., Merchants Br. 32 (same); id. at 33 (individualized legal claims,
individualized awards of monetary damages, individualized monetary claims);
id. at 3334 (individual monetary claims); id. at 34 (individualized monetary
claims, individualized legal claims, individualized claim for money); id. at 35
(similar). (Some of these quotations have omitted alterations or internal quotation
marks for ease of reference.)
51
Case: 12-4671
Document: 1123
Page: 66
10/15/2014
1345294
98
compromises the originally asserted forward-looking claims for injunctive relief,

one of the most traditional remedies sought and received by (b)(2) classes, and any
liability that allegedly would flow from the post-settlement network rules. See,
e.g., SPA46; SPA18; SPA139SPA140 40; SPA153 53.
Objectors urge this Court to treat the non-opt out (b)(2) class as a class about
damages claims because the release covers possible future damages claims against
the going-forward network system, and therefore urge that they were entitled to opt
out under Rule 23 and as a matter of due process. But the propriety of (b)(2)
certification does not turn on the nature of the claims released; rather it rests on the
relief demanded in the present litigation.
Objectors primarily rely on Wal-Mart Stores, Inc. v. Dukes in support of
their theory, see, e.g., Merchants Br. 3335, 3839, but that case makes clear that
the appropriate certification focus is on the claims asserted. See 131 S. Ct. 2541,
2551 (2011) (claims must depend upon a common contention); id. at 2552
(class determination generally involves considerations that are enmeshed in the
factual and legal issues comprising the plaintiffs cause of action) (internal
quotation marks omitted). This Court has made the point with respect to a Rule
23(b)(3) settlement class, explaining that certification is examined based on
questions that preexist any settlement, and not on whether all class members have
a common interest in a fair compromise of their claims. AIG, 689 F.3d at 240
52
Case: 12-4671
Document: 1123
Page: 67
10/15/2014
1345294
98
(quoting Amchem, 521 U.S. at 623). In the (b)(2) setting, although some courts
have suggested the common bond should be stronger because of the mandatory
nature of the class and possible prejudice to individual claims, see, e.g., Barnes v.
Am. Tobacco Co., 161 F.3d 127, 14243 (3d Cir. 1998), the basic focus remains
the same: a common interest in questions that preexist any settlement. See, e.g.,
Dukes, 131 S. Ct. at 2557.
Dukes further instructs that Rule 23(b)(2) applies when a single injunction
or declaratory judgment would provide relief to each member of the class. Id.
This Court has similarly held that a Rule 23(b)(2) class action is a proper way for
seeking systematic changes and resolving outstanding questions about the
lawfulness of defendants practices. Robinson, 267 F.3d at 165; Marisol A. v.
Giuliani, 126 F.3d 372, 378 (2d Cir. 1997) (suit for injunctive relief to address
central and systemic failures of child welfare system satisfied Rule 23(b)(2));
contra Merchants Br. 3443.
These principles apply here, whereyears before the parties commenced
settlement negotiations, D.E.317 97(b)the (b)(2) class challenged the
networks core rules governing all merchants, and sought injunctive relief that
would displace and re-write those rules. The Class sought relief based on
Defendants act[ions] or refus[als] to act on grounds that apply generally to the
class, so that final injunctive relief or corresponding declaratory relief is
53
Case: 12-4671
Document: 1123
Page: 68
10/15/2014
1345294
98
appropriate respecting the class as a whole, Fed. R. Civ. P. 23(b)(2), and that
injunctive relief was sought for an alleged group harm, Robinson, 267 F.3d at
165. As Appellants admit, defendants practices affect all class members.
Merchants Br. 50. All class members participated in networks governed by the
same allegedly unlawful core rules (no-surcharge, no-discounting, default
interchange, Honor-all-Cards, etc.), and challenged those rules. See SPA52;
Marisol A., 126 F.3d at 378; cf. In re Nassau Cnty. Strip Search Cases, 461 F.3d
219, 22728 (2d Cir. 2006) (finding cohesion where a (b)(3) class sought to
impose liability based on defendants implementation of a blanket . . . policy).
And, regardless of how this case were to enddismissal on the merits, summary
judgment, jury verdict, or settlementall class members would continue to
participate in a network governed by whatever network rules emerged from that
judgment. In short, because plaintiffs claims concern the legality of the networks
governing rules and practices, the challenged conduct may be enjoined or
declared unlawful only as to all of the class members or as to none of them.
Dukes, 131 S. Ct. at 2557; see, e.g., Robinson, 267 F.3d at 165; SPA52 (the
structural relief is generally applicable to the class in the manner required by Rule
23(b)(2)); SPA46 (similar).
That is why the (b)(2) class was appropriately a mandatory, non-opt-out
class. Going forward, all Visa- and MasterCard-accepting merchants will
54
Case: 12-4671
Document: 1123
Page: 69
10/15/2014
1345294
98
necessarily operate within and be bound by the same post-settlement regime, just
as in the past, all Visa- and MasterCard-accepting merchants operated within and
were bound by the pre-settlement regime. The legal challenge plaintiffs raised is
not divisible as to individual merchants, and the relief sought [would] perforce
affect the entire class at once. Dukes, 131 S. Ct. at 2558.
Objectors urge that a non-opt-out class is improper because not all members
of the class will derive the same benefit from the relief provided by the settlement.
Merchants Br. 5052. But the law does not require all members of a (b)(2)
settlement class to benefit equally from the relief obtained in the settlement.
Indeed, if Objectors view were correct, not even the civil-rights cases at the core
of Rule 23(b)(2), Merchants Br. 37, would warrant certification of a (b)(2)
settlement class. After all, not every pupil would benefit equally from a schools
desegregation of its athletics program, see La. High Sch. Athletic Assn v. St.
Augustine High Sch. 396 F.2d 224 (5th Cir. 1968), nor will every employee derive
the same benefit from its employers adoption of more race- or gender-neutral
advancement policies, see Bishop v. Gainer, 272 F.3d 1009 (7th Cir. 2001), nor
every inmate from a change in prison medical treatment protocols, Parsons v.
Ryan, 754 F.3d 657 (9th Cir. 2014). But just as those classes are proper non-optout classes, so too is the (b)(2) settlement class here, because the different weight
that class members might place on various forms of relief does not alter the fact
55
Case: 12-4671
Document: 1123
Page: 70
10/15/2014
1345294
98
that the asserted injuries arise from a uniformly applied course of conduct, and that
their injuries can be remediated via a single, unitary injunction.
In sum, either default interchange, a surcharging ban, Honor-all-Cards, or
the networks other rules are unlawful restraints of trade that generate
supracompetitive interchange feesor they are not. If they are unlawful, they
impose anticompetitive restraints on all merchants within that network (including
any future merchants who participate in the network), and those restraints can be
enjoined only as to all merchants. Even as the networks allow for individualized
negotiation where the parties find it advantageous, there is no way, as a matter of
law, logic, or real-world commercial dealing, to offer a fully individualized
payment-card system for each of the millions of merchants populating the Visa and
MasterCard networks. The whole point of the networks, and an essential
contributor to their success, is that network rules are established without the need
for negotiation or fresh rule-making each time a new card is placed on the market
or a new merchant opens its doors (or comes on-line).
2.
The (b)(2) Class Was Not Improperly Certified For

Settlement Because It Released Claims For Future Liability
Stemming From The Post-Settlement Rules.
Although the Objectors are wrong that the (b)(2) class sought to resolve
individualized monetary damages claims, the (b)(2) class settlement does release
future claims for liability (whether seeking damages or injunctive relief) that
56
Case: 12-4671
Document: 1123
Page: 71
10/15/2014
1345294
98
purport to challenge the network rules deliberately left in place by the settlement.
That, however, does not demonstrate that the certification of the (b)(2) settlement
class was improper. Contra, e.g., Merchants Br. 3840. 16 It simply illustrates one
potential consequence of a proper (b)(2) class settlement, namely, the arrangement
implemented as a result of settlement can be insulated from the parties (and their
privies) future legal challenges. See In re Literary Works in Elec. Databases
Copyright Litig., 654 F.3d 242, 248 (2d Cir. 2011) (holding that the Settlements
release of claims regarding future infringements is not improper where the
complaint sought injunctive relief for future uses, and therefore contemplate[d]
these alleged future injuries). It is no more remarkable that agreements providing
that assurance would preclude plaintiffs from bringing any species of legal
challenge to the lawfulness of the post-settlement status quo, including claims for
money damages. See San Diego Police Officers Assn v. San Diego City Emps
Ret. Sys., 568 F.3d 725, 73436 & n.7 (9th Cir. 2009) (enforcing release in (b)(2)
class settlement to hold that monetary damages claims alleg[ing] the same injury
and the same wrong as in the released action were barred by the doctrine of
claim preclusion); Nottingham Partners v. Trans-Lux Corp., 925 F.2d 29, 3234
16
As a technical matter, Objectors arguments are little more than challenges to

releases scope. Nonetheless, because Objectors raise the release of unasserted
claims for future damages under their challenge to (b)(2) certification, we discuss
here briefly both why the release is lawful, and the lack of authority for Objectors
argument. For the fuller discussion of the lawful scope of the release, see II
below.
57
Case: 12-4671
Document: 1123
Page: 72
10/15/2014
1345294
98
(1st Cir. 1991) (enforcing release of claims in state court Rule 23(b)(2) action to
preclude damages claims by plaintiffs who fell within class definition, reasoning
[t]he two suits, notwithstanding any differences in remedies sought or theories of
recovery pleaded, shared a common gravamen. In sum, the instant case bore a
sufficiently close relation to the Dana complaint to come within the plain language
of the general release formulated as part of the Dana settlement.).
Objectors refuse to accept that plaintiffs settling claims challenging
generally applicable, uniform conduct (like the network rules here) must be able to
promise not to sue the defendants for following the settlements terms. This is
nothing less than a refusal to acknowledge that litigation has consequences, and
judgments have going-forward impacts. For example, in a litigated case, if a (b)(2)
plaintiff class loses on the merits of an injunctive claim because a challenged
practice is found to be lawful, there are consequences that go beyond the denial of
the injunctive relief requested. Since any future damages claim challenging the
same conduct depends on the ability to demonstrate liabilitythat is, some legal
violation by the defendantthe future damages claim will be doomed because any
effort to establish liability would be precluded by the prior judgment. See supra at
4749 (discussing res judicata). The legal system does not tolerate, much less
demand, endless litigation over the lawfulness of the same behavior. The release at
issue here has a similar effect.
58
Case: 12-4671
Document: 1123
Page: 73
10/15/2014
1345294
98
Were it otherwise, no (b)(2) settlement (or litigation on the merits) could

ever conclusively resolve the legality of a particular network, practice, or system of
rules. This Court has acknowledged that such resolutions are possible. E.g., TBK
Partners, Ltd. v. W. Union Corp., 675 F.2d 456, 460 (2d Cir. 1982) (recognizing
the essentiality of achiev[ing] a comprehensive settlement that would prevent
relitigation of settled questions at the core of a class action); see infra II
(discussing lawful scope of releases).
As Objectors would have itparticularly by framing their arguments in
terms of a purported due process right, see Merchants Br. 6, 3248; First Data Br.
924any plaintiff dissatisfied with the settlement must have an opt-out right to
preserve its ability to later claim that the settlement left an unlawful state of affairs
in place. Such plaintiffs, even if opting out, would get to enjoy the benefits of the
injunctive relief provided by the settlement even as they endlessly sue to force
further changes to the system at issue or recover going-forward damages based on
that system. That repeated cycle would destroy the stability on which networks
(and similar entities) rely for their efficient operation, if not their survival.
Objectors position means that each class member could hold a veto over the (b)(2)
settlement, which would make settlements that much harder to achieve in the first
59
Case: 12-4671
Document: 1123
Page: 74
10/15/2014
1345294
98
place. 17 This position is neither tenable, for the many reasons just discussed, nor
supported by the cases upon which Appellants rely.
a.
Dukes Does Not Preclude Certification Of The

23(b)(2) Settlement Class.
The Merchants brief invokes Dukes in support of its release-based

challenge to the certification of the (b)(2) settlement class. That effort fails.
To begin, Dukes is a case about certification of litigation classes, not about
the scope of a release. As noted above, supra I.B.1, Dukes strongly supports the
propriety of the Rule 23(b)(2) settlement class here based on the claims asserted by
plaintiffs.
Moreover, Dukes addressed only a single (b)(2) class that attempted to
include within its scope individual monetary claims for backpay based on alleged
previous discrimination. The Court said as much, concluding that the
combination of individualized and classwide relief in a (b)(2) class cannot be
squared with the history and structure of the rule. 131 S. Ct. at 255758 (emphasis
17
Particularly where, as here, defendants have already contested these issues across
multiple litigations and entered settlements that failed to bring repose, see, e.g.,
Wal-Mart II, 396 F.3d at 10103, 118 (settlement and releases concerning Honorall-Cards), a settlement would be impractical, if not impossible to reach, because a
release often is the primary benefit a defendant receives. See Literary Works, 654
F.3d at 24748; Wal-Mart II, 396 F.3d at 106; Sullivan v. DB Invs., Inc., 667 F.3d
273, 311 (3d Cir. 2011) (en banc) ([A]chieving global peace is a valid, and
valuable, incentive to class action settlements. . . . No defendants would consider
settling under [a] framework [where the release covered only certain qualifying
class members], for they could never be assured that they have extinguished every
claim from every potential plaintiff.).
60
Case: 12-4671
Document: 1123
Page: 75
10/15/2014
1345294
98
added). Therefore, the Court construed the rule to proscribe such combinations
where the monetary component of the relief sought was more than merely
incidental to the classs demand for injunctive or declaratory relief. Id. at 2557
60; see also id. at 254849 & n.2.
This case is completely different. Here, the (b)(2) class claims for injunctive
relief and the (b)(3) class claims for monetary damages arising from past conduct
have been separated, with different procedures adopted as to each settlement class
in light of their differing natures (including with respect to opt-out rights). The
(b)(2) class in Dukes sought to do precisely what the (b)(2) class here does not do:
bind class members to the resolution of non-incidental, individualized damages
claims actually asserted by the (b)(2) class.
Consequently, the quotations that Objectors lift from Dukes do not address
releases of future claims of liability, let alone claims of liability founded on
defendants adherence to the very regime achieved through the (b)(2) settlement.
See, e.g., Merchants Br. 3435, 45. 18 Rather, those quotations speak to the
compromise of already-accrued, presently available claims. See Dukes, 131 S. Ct.
at 2557 (concluding that claims for monetary relief may [not normally] be
certified under that provision) (emphasis added); id. at 2559 (observing that
18
In fact, the Dukes Court did not say a word about the relevance to (b)(2)
certification of either claims for future damages or claims released in class-action
settlements, let alone the relevance of a class settlements release of claims for
future damages.
61
Case: 12-4671
Document: 1123
Page: 76
10/15/2014
1345294
98
plaintiffs proposed test . . . creates perverse incentives for class representatives to

place at risk potentially valid claims for monetary relief, such as the plaintiffemployees claims for compensatory damages in Dukes itself, which the class
abandoned, instead pursuing only their more modest backpay claims). As the text
of the (b)(2) release makes plain, no such compromise took place here. SPA169
68.
And, far from being condemned by Dukes, the two-class approach has
flourished in its wake. See Gooch v. Life Investors Ins., 672 F.3d 402, 42728 (6th
Cir. 2012) (approving use of separate (b)(2) and (b)(3) classes in combination);
Johnson v. Meriter Health Servs. Emp. Ret. Plan, 702 F.3d 364, 371 (7th Cir.
2012) (suggesting that divided certification of a (b)(2) declaratory class and a
subsequent (b)(3) damages class would be consistent with Dukes); Huyer v. Wells
Fargo & Co., 295 F.R.D. 332, 34445 (S.D. Iowa 2013); Bristol Vill., Inc. v. La.Pac. Corp., 916 F. Supp. 2d 357, 36970 (W.D.N.Y. 2013); Sykes v. Mel Harris &
Assocs., LLC, 285 F.R.D. 279, 293 (S.D.N.Y. 2012) (Chin, J.) ([t]hat plaintiffs
are seeking substantial monetary damages is of no concern given the Courts
certification of separate Rule 23(b)(2) and Rule 23(b)(3) classes addressing
equitable relief and damages, respectively).
The simple truth is that Dukes does not discuss releases, compromises of
future claims, or even class settlements generally. Objectors repeated invocation
62
Case: 12-4671
Document: 1123
Page: 77
10/15/2014
1345294
98
of that decision does not demonstrate that the District Court abused its discretion
either in certifying the settlement classes or in approving the settlement.
b.
Shutts Also Does Not Support The Argument That

The Release Rendered The (b)(2) Class Improper.
Just as the (b)(2) settlement is proper under the Federal Rules, it does not
offend due process under Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 805
(1985), as the Merchants (at 3233) erroneously contend. Shutts involved alreadyexisting claimsspecifically, interest allegedly owed on already-paid natural gas
royalties. 472 U.S. at 800. Nothing in Shutts, however, suggested that a plaintiff
has a due process-protected property interest in an inchoate, unaccrued future
claim. See id. at 808. To the contrary, it is well-established that [n]o person has a
vested interest in any rule of law entitling him to insist that it shall remain
unchanged for his benefit. N.Y. Cent. R.R. v. White, 243 U.S. 188, 198 (1917)
(collecting cases).
c.
This Courts Cases Also Do Not Lead To A Different

Outcome.
This Courts cases cited by the Merchants (at 3637) do not show that the
releases here precluded certification of the (b)(2) settlement class.
By Objectors own admission, this Court in Wal-Mart I (a/k/a Visa Check),
280 F.3d 124, reserved decision on (b)(2) certification in that case, analyzing the
propriety of certification only under Rule 23(b)(3), see Merchants Br. 3637, and
63
Case: 12-4671
Document: 1123
Page: 78
10/15/2014
1345294
98
Hecht v. United Collection Bureau, Inc., 691 F.3d 218 (2d Cir. 2012), dealt with
the (here-irrelevant) issue of the adequacy of the notice of opt-out rights in a suit
adjudicating a backward-looking damages claim, see id. at 22223; Merchants Br.
36.
In Stephenson v. Dow Chemical Co., a group of Vietnam Veterans sued for
damages based on harms that had already been inflicted on them but whose Agent
Orange injuries did not manifest until after expiration of the settlement fund
established in an earlier lawsuit. 273 F.3d 249, 25758 (2d Cir. 2001), affd in
part by an equally divided Court and vacated in part on other grounds, 539 U.S.
111 (2003). Those claims were classic backward-looking damages claims, rather
than the claims Objectors are concerned with herefuture claims challenging the
legality of actions taken pursuant to a court-approved settlement. 19
Finally, the Merchants (at 37) are mistaken in attempting to dismiss Literary
Works, 654 F.3d 242, in which the parties settlement released the defendants from
future litigation over subsequent use of certain copyrighted works. The Merchants
contend that the Literary Works release is distinguishable from the one here
because it permitted class members to (1) opt out of the settlement entirely or (2)
19
Charron v. Wiener, 731 F.3d 241 (2d Cir. 2013), cert. denied sub nom. Suarez v.
Charron, 134 S. Ct. 1941 (2014), is similar. Although the court noted that the
settlement did not extinguish claims excluded from its scope, the claims at issue
were already-accrued, backward-looking claims. Id. at 244, 253. It says nothing
about the sort of going forward-based claims implicated here.
64
Case: 12-4671
Document: 1123
Page: 79
10/15/2014
1345294
98
opt out of the release for future use in particular. Merchants Br. 37 (quoting
654 F.3d at 24647). But the cited portion of the decision comes from this Courts
statement of the case, not its analysis. This Court never suggested that there was a
requirement that the settlement permit class members to opt out of the release for
future use in order for (b)(2) certification to be appropriate. Instead, consistent
with Defendants showings supra and infra, the court noted the breadth of typical
settlement releases, underscoring that [p]laintiffs in a class action may release
claims that were or could have been pled in exchange for settlement relief, and
that [p]arties often reach broad settlement agreements encompassing claims not
presented in the complaint in order to achieve comprehensive settlement of class
actions, particularly when a defendants ability to limit his future liability is an
important factor in his willingness to settle. 654 F.3d at 24748 (quoting WalMart II, 396 F.3d at 106). 20
In sum, none of the authorities cited by the Objectors supports their
argument that certification of the (b)(2) settlement class was improper simply
because the class released claims challenging the rules that exist post-settlement.
20
In addition to the cases discussed above, Objectors also contend that the
settlement classes certified in this case run afoul of the Supreme Courts decisions
in Amchem, 521 U.S. 591, and Ortiz v. Fibreboard Corp., 527 U.S. 815 (1999).
Merchants Br. 5266 (cohesion); id. at 6679 (adequacy). The Class Plaintiffs
aptly show why those opinions have no bearing here, see Class Br. I.D.2, and so
we incorporate that discussion here.
65
Case: 12-4671
II.
Document: 1123
Page: 80
10/15/2014
1345294
98
THE SCOPE OF THE RELEASES PROVIDED BY THE 23(b)(2)

CLASS IS LAWFUL.
In addition to Objectors argument that the release here was improperly
implemented through a mandatory 23(b)(2) settlement class, Objectors suggest

that, in all events, the releases were unlawful because they release claims beyond
those presented in this litigation. See, e.g., Merchants Br. 3132, 8090.
Objectors view is wrong. It contravenes decades of settled precedent,
erodes the courts longstanding policy of favoring the settlement of disputes
particularly class disputesand threatens to overwhelm defendants and courts with
a never-ending stream of litigation.
Contrary to Objectors rhetoric, [i]t is not at all uncommon for settlements
to include a global release of all claims past, present, and future, that the parties
might have brought against each other. Williams v. Gen. Elec. Capital Auto
Lease, Inc., 159 F.3d 266, 274 (7th Cir. 1998). That is because, [p]ractically
speaking, [c]lass action settlements simply will not occur if the parties cannot set
definitive limits on defendants liability. Wal-Mart II, 396 F.3d at 106 (second
alteration in original). Absent the ability to set such limits, defendants would . . .
face nearly limitless liability from related lawsuits in jurisdictions throughout the
country. Id.
Judge Gleeson did not abuse his discretion in approving the releases here.
66
Case: 12-4671
A.
Document: 1123
Page: 81
10/15/2014
1345294
98
The Identical Factual Predicate Doctrine Permits A Broad

Release Of Claims.
The releases are unremarkable and amply supported by this Courts

decisions. The law is well established in this Circuit and others that class action
releases may include claims not presented and even those which could not have
been presented as long as the released conduct arises out of the identical factual
predicate as the settled conduct. Wal-Mart II, 396 F.3d at 107 (quoting TBK
Partners, 675 F.2d at 460). Notwithstanding its name, the identical factual
predicate doctrine permits a class to release claims not presented in the
complaint and those involving events that have not yet occurred. Literary Works,
654 F.3d at 24748; see, e.g., Wal-Mart II, 396 F.3d at 107, 114 (claims not
presented and [that] might not have been presentable can nonetheless be released
(emphasis omitted)); TBK, 675 F.2d at 46061; Robertson v. NBA, 622 F.2d 34, 35
(2d Cir. 1980) (Robertson IV).
In Robertson IV, for example, this Court concluded that Wilt Chamberlains
suit against the NBA, which challenged the same rule that was the subject of a
prior class action and which had been modified in a prior class settlement, was
precluded by the release in that settlement. 622 F.2d at 35. The mere fact that the
later action challenged the rules application at a subsequent time was not
sufficient to remove the releases binding force under the identical factual
predicate doctrine.
67
Case: 12-4671
Document: 1123
Page: 82
10/15/2014
1345294
98
And in Wal-Mart II, this Court found the doctrine satisfied where in the
settled case (containing the release) the plaintiffs had pleaded a tying claim,
alleging that the exclusionary rules solidified Visa and MasterCards power in the
credit card market, enabling [them] to force plaintiffs to accept their debit cards.
396 F.3d at 107. The plaintiffs raising the arguably released claims, on the other
hand, had brought a Section 1 claim alleging an increase in credit-card transaction
costs. Id. Proving those disparate claims would have required proof of at least
some different facts (relating to the identity and contours of the relevant markets,
the existence and extent of damages, etc.), but the Court held that the doctrine was
satisfied because both cases involved the same central rules. See id. at 108.
Furthermore, the Court held that the doctrine permitted the release of claims
against non-parties to the action. See id. at 10809.
B.
The Releases Here Are Tailored To The Identical Factual

Predicate Doctrine.
Judge Gleeson properly recognized that the Releases here are lawful under
the identical factual predicate doctrine, because [t]hey do not release the
defendants from liability for claims based on new rules or new conduct or a
reversion to the pre-settlement rules. They appropriately limit future damages
claims based on the pre-settlement conduct of the networks. SPA45SPA46. As
the text of the Releases makes clear, they merely compromise claims:
68
Case: 12-4671
Document: 1123
Page: 83
10/15/2014
1345294
98
arising out of or relating in any way to any conduct, acts,

transactions, events, occurrences, statements, omissions, or
failures to act of any Rule 23(b)(2) Settlement Class Released
Party that are alleged or which could have been alleged from
the beginning of time to the date of the Courts entry of the
Class Settlement Preliminary Approval Order in any of the
Operative Class Complaints or Class Action complaints, or in
any amendments to the Operative Class Complaints or Class
Action complaints . . . .
SPA169-SPA170 68 (emphasis added). Included are claims related to the
lawfulness of default interchange, SPA170-SPA171 68(a), (g), and Honor-allCards, id. 68(c), (g). See supra I.A.2.a (discussing those claims and the factors
making it unlikely that plaintiffs would prevail on them). 21
What the releases do not affectcontrary to Objectors dark predictions
are claims based on new conduct. SPA46. If someday there are harmful rules,
practices, or actions that are not substantially similar to, SPA171 68(g), those
that were or could have been challenged in this case, the releases facially would
not apply. So, ifhypotheticallyVisa or MasterCard were to impose an entirely
21
The Merchants cite the Fixed Acquirer Network Fee (FANF) as an example of
a clai[m] beyond the scope of the case that was released by the (b)(2) settlement.
Merchants Br. 8788. But cf. SPA174 72(d) (release does not extend to FANFbased claims for injunctive relief). First, that contention is waived because, as
here, an argument made only in . . . footnote[s] [i]s inadequately raised for
appellate review. Norton v. Sams Club, 145 F.3d 114, 11718 (2d Cir. 1998);
see Home Depot Objection (D.E.2591) 8 n.8; Joint Objection (D.E.2670) 34 n.43.
Second, in any event, the point is meritless. Objectors concede that FANF was in
place prior to settlement. Merchants Br. 17, 4041. It thus could have been
litigated here andas suchis part of the factual predicate of this case,
notwithstanding that (as Objectors note) it was not specifically cited in the
complaint. See, e.g., Wal-Mart II, 396 F.3d at 107; TBK Partners, 675 F.2d at 460.
69
Case: 12-4671
Document: 1123
Page: 84
10/15/2014
1345294
98
new rule preventing merchants from steering customers away from paying with
credit cards, or to adopt entirely new, as opposed to substantially similar, rules
governing the use of mobile technologies (as Objectors claim to fear), the
lawfulness of those new rules would be fair game for a future antitrust suit.
For the same reason, the Releases do not, as the Merchants contend (Br. 80
82), effect an improper waiver of future liability under the federal antitrust
statutes. In re Am. Express Merchants Litig., 634 F.3d 187, 197 (2d Cir. 2011);
see Lawlor v. Natl Screen Serv. Corp., 349 U.S. 322, 329 (1955). The Merchants
assert: In Lawlor, 349 U.S. at 32829, the Supreme Court stated that
extinguishing claims which did not even then exist and which could not possibly
have been sued upon in the previous case . . . would in effect confer on
[defendants] a partial immunity from civil liability for future violations. Br. 81
(alterations in original) (emphasis added). The Merchants quotation of Lawlor is
disingenuous. The immunity which Lawlor prevents the parties from agreeing to
is immunity from antitrust liability for new allegedly anticompetitive conduct that
could not have been the subject of the previous suit. Lawlor, 349 U.S. at 328
(discussing post-settlement slow deliveries and allegedly illegal tie-ins). Here, the
release does not bar a future suit challenging any new, post-settlement allegedly
anticompetitive conduct that could not have been the subject of this suit. The
settlement here has precisely the effect that the Supreme Court gave to the
70
Case: 12-4671
Document: 1123
Page: 85
10/15/2014
1345294
98
settlement in Lawlor. The Court there held that the earlier judgment precludes
recovery on claims arising prior to its entry, [but] it cannot be given the effect of
extinguishing claims which did not even then exist and which could not possibly
have been sued upon in the previous case. Id. at 328. That is textbook res
judicata doctrine and says nothing about issue preclusion and does not undermine
the permissible scope of the class-settlement release at issue here.
Even Objectors skewed reading of Lawlor, however, does not help them
given the facts here. The releases were limited to claims based on facts that are
alleged or which could have been alleged here. SPA169SPA170 68 (emphasis
added). Having acknowledged this point and the fact that the releasesmuch like
long-arm statutes designed to be coterminous with the reach of the Due Process
Clausewere drafted only to releas[e] . . . claims that are or could have been
alleged based on the identical factual predicate of the claims in this case, SPA45
SPA46, Judge Gleeson properly recognized that the precise contours of the facially
valid releases is a subject for future cases. SPA47. For example, whether a
particular (presently hypothetical) claimbased on changed network rules, an
evolution in payment or processing technology, or other conductfalls within this
cases factual predicate is a determination to be made in that case, not this one. Id.
(substantial similar[ity] of rules will be decided in future litigation); see Reyns
71
Case: 12-4671
Document: 1123
Page: 86
10/15/2014
1345294
98
Pasta Bella, 442 F.3d at 74849 (barring suit based on conclusion that it was based
on identical factual predicate as Wal-Mart II); Robertson IV, 622 F.2d at 35.
Finally, the release of antitrust claims in this settlement can be overturned
only if Objectors could show, to a legal certainty, that the Releases (or some
other facet of the settlement) are unlawful. Robertson v. NBA, 556 F.2d 682, 686
(2d Cir. 1977) (Robertson II) (emphasis added); see id. (declaring that unless
the challenged practices have . . . been held to be illegal per se in any previously
decided case, the settlement may be approved); see also, e.g., Armstrong v. Bd. of
Sch. Dirs., 616 F.2d 305, 31920 (7th Cir. 1980) (same), overruled on other
grounds by Felzen v. Andreas, 134 F.3d 873 (7th Cir. 1998). Here, Objectors
obviously cannot make that showing, for all of the reasons discussed in Section
I.A.2, supra.
III.
THE COMPETITORS OBJECTIONS ARE EQUALLY MERITLESS.

American Express, First Data, and Discover also appeal the settlements
approval. Their challenges fare no better than those of other Objectors, as the
District Court recognized. SPA47. In addition to the Classs showings that the
Class Representatives adequately represent all absent class members, including
AmEx, First Data, and Discover in their limited capacity as merchant acceptors of
72
Case: 12-4671
Document: 1123
Page: 87
10/15/2014
1345294
98
Visa and MasterCard branded cards, see Class Br. I.D, 22 Defendants briefly set
forth additional infirmities in AmEx, First Data, and Discovers arguments.
A.
The Competitors Claims, As Competitors, Are Not Released.
AmEx and First Data principally claim that the settlement improperly and
unlawfully releases their claims as competitors of the network defendants. AmEx
Br. 2532; First Data Br. 924. 23 Those complaints are baseless. As Judge
Gleeson stated, they seek to make something of nothing, as the relevant release
language does not purport to release any claims that they may possess as
competitors. SPA47. Instead, as the District Court concluded, it is sufficiently
clear from both the text and context of the releases that these class members are
releasing only claims that merchants have alleged or could have alleged in this case
in their capacity as merchants. Id. More specifically, those releases provide that
claims that are alleged or which could have been alleged in this action (which
was brought by and on behalf of merchants that accept Visa and MasterCard), are
being released. SPA134 33; SPA169170 68; see also D.E.1740-2 at F212
(In general, the settlement will resolve and release all claims made by persons,
22
Competitors specific requestsAmEx primarily seeks to unwind the settlement,

see Br. 35; Discover seeks specific modifications to exclude it from portions of the
agreement, see Br. 6, 50; and First Data principally seeks an opt-out right, see, e.g.,
Br. 2do not materially change the response to their arguments.
23
Discover advanced a similar argument below, see Tr. (D.E.6094) 14445

(admitting that it objected to the release and that its objections overlap
substantially with First Data and AmExs), but shifts course here. See infra
III.B.
73
Case: 12-4671
Document: 1123
Page: 88
10/15/2014
1345294
98
businesses, and other entities that arise from or relate to their capacity as
merchants that accept Visa-Branded Cards and/or MasterCard-Branded Cards in
the United States . . . .) (emphasis added).24
American Express similarly argues that its inclusion in the class
demonstrates the lack the cohesion required by Rule 23, essentially for two
reasons: First, its interest in competing with Visa and MasterCard requires it to
oppose discriminatory treatment imposed by merchants (including discriminatory
surcharges)a position antithetical to the claims being settled by the Rule 23(b)(2)
class. AmEx Br. 1718; see also First Data Br. 3638 (similar). Second, many of
the merchants (and counsel) representing the Class are adverse to American
Express in a parallel antitrust suit, and the settlement here is contrary to AmExs
own litigation goals, interests, and strategies in that case. AmEx Br. 1819.
But those arguments, which merely rehash the competitors release-based
contentions, largely miss the point. Although First Data and AmEx do not
24
AmEx and First Data assert that the language of the settlement agreement calls
into question the breadth of the releases. See First Data Br. 1315, 2023; AmEx
Br. 2526. But if there were any question about the agreements breadth, it should
not be resolved by concluding that the agreement is unambiguously broad, contra
First Data Br. 2022, given that neither the parties to the agreement nor the District
Court have read the agreement in the manner AmEx and First Data propose.
Instead, the proper course would be to resort to parol evidence of the agreements
meaning and to defer to the parties representations belowconsistent with the
District Courts conclusion, SPA47that the releases do not bar claims based on
injuries as payment network competitors. Defs. Reply Supp. Final Approval 32
(D.E.5937) (emphasis and capitalization omitted); accord Pls. Reply Supp. Final
Approval 59 (D.E.5939).
74
Case: 12-4671
Document: 1123
Page: 89
10/15/2014
1345294
98
primarily do business in their merchant capacities, they nonetheless accept Visaand MasterCard-branded cards as modes of payment for services they offer.
AmEx Br. viii; First Data Br. 24. That brings them within this class, and they offer
no basis for concluding that the District Court erred in concluding that the class
representatives and counsel adequately represented them in that capacity, see
Literary Works, 654 F.3d at 249; Class Br. I.Dmuch less that it abused its
discretion, Joel A., 218 F.3d at 139.
B.
Discovers Group-Boycott Claim Lacks Merit.
Discover also attacks the settlements, claiming that the Level Playing
Field provisions are unlawful group boycotts and impose serious administrative
burdens on any merchant that wishes to accept both Discover and Visa and/or
MasterCard. Discover Br. 56, 2227, 3949. The challenged provisions ensure
that surcharging of Visa or MasterCard transactions will be permitted only under
the conditions pursuant to which the merchant is allowed to surcharge cards from
networks with more-restrictive surcharging rules. See SPA141 & SPA 148
42(a)(iv), (c); SPA154SPA155 &SPA161SPA162 55(a)(iv), (c). Discovers
arguments lack merit, and, in all events, are insufficient to disturb Judge Gleesons
approval decision.
Arguments, as here, that a class settlement enshrines an unlawful agreement
face a high bar at the settlement-approval phase, where the district courts analysis
75
Case: 12-4671
Document: 1123
Page: 90
10/15/2014
1345294
98
of both the pre- and post-settlement status quo isand must beless rigorous than
it would be in the context of a full trial on the merits. Robertson II, 556 F.2d at
686 (upholding a settlement approval because challengers failed to demonstrate, to
a legal certainty, that the settlement was unlawful); see supra at 3031 (court
need not fully adjudicate plaintiffs claims). Unless the challenged practices have
. . . been held to be illegal per se in a previously decided case, there is no error in
approving an otherwise-adequate settlement accord. Robertson II, 556 F.2d at 686
(rejecting claim that settlement agreement cannot be approved because it
perpetuates . . . classic group boycotts).
Here, Discovers group-boycott claim is doomed by its failure to identify
even a single case in which settlement provisos or contractual terms such as the
Level Playing Field provisions have been deemed a group boycott and held
unlawful per se.
Additionally, Discovers attack on the Level Playing Field provisions as
unfair and unreasonably harmful to it as a third-party, see Discover Br. 3942, is
unavailing, because those provisions are, at bottom, nothing more than most
favored nations clauses. Those clauses ensure merchants cannot use a surcharge
to make paying with a Visa- or MasterCard-branded card more expensive for
consumers than paying with a card from a higher-cost Competitive Card Brand.
Far from being unlawful per se, courts repeatedly have upheld most favored
76
Case: 12-4671
Document: 1123
Page: 91
10/15/2014
1345294
98
nations provisions given their legitimate, competitive features. See, e.g., Blue
Cross & Blue Shield United of Wis. v. Marshfield Clinic, 65 F.3d 1406, 1415 (7th
Cir. 1995); Ocean State Physicians Health Plan, Inc. v. Blue Cross & Blue Shield
of R.I., 883 F.2d 1101, 1102, 1110 (1st Cir. 1989).
Finally, Discovers protestations regarding the burdensome calculations
required by the Level Playing Field provisions ring hollow in light of the fact
that Discover itself maintains an Equal Treatment Rule that can operate to limit
surcharges against Discover where Discover matches or beats the pricing of a rival
that is not surcharged. Discover Br. 20 (emphasis added). In other words, it
appears that even Discovers own rules can necessitate an inter-network cost
comparison, which illustrates the point that such comparative exercises are merely
a cost of doing business in the payment card industry. They are hardly a basis for
denying settlement approval, let alone for overturning an already-approved
settlement on abuse-of-discretion review.
77
Case: 12-4671
Document: 1123
Page: 92
10/15/2014
1345294
98
CONCLUSION
For the foregoing reasons and those set forth by the Class Plaintiffs, the
judgment should be affirmed.
Dated: October 15, 2014.
SIDLEY AUSTIN LLP
By: /s/ Carter G. Phillips
Carter G. Phillips
SIDLEY AUSTIN LLP
1501 K Street, NW
(202) 736-8000
cphillips@sidley.com
David F. Graham
Robert N. Hochman
One South Dearborn Street
Chicago, IL 60603
Benjamin R. Nagin
Eamon P. Joyce
Mark D. Taticchi
787 Seventh Avenue
New York, NY 10019
Citigroup Inc., Citibank, N.A., and
Citicorp
78
Case: 12-4671
Document: 1123
Page: 93
10/15/2014
1345294
98
ARNOLD & PORTER LLP
By: /s/ Robert C. Mason

Robert C. Mason
399 Park Avenue
New York, NY 10022-4690
(212) 715-1000
robert.mason@aporter.com
Keila D. Ravelo
Wesley R. Powell
Matthew Freimuth
787 Seventh Avenue
New York, NY 10019-6099
PAUL, WEISS, RIFKIND, WHARTON &
GARRISON LLP
Robert J. Vizas
Three Embarcadero Center, 10th Floor
San Francisco, CA 94111-4024
By: /s/ Kenneth A. Gallo

Kenneth A. Gallo
2001 K Street, NW
Washington, DC 20006-1047
(202) 223-7300
kgallo@paulweiss.com
Mark R. Merley
Matthew A. Eisenstein
555 12th Street, NW
Washington, DC 20004-1206
Gary R. Carney
New York, NY 10019-6064

By: /s/ Michael S. Shuster
Richard J. Holwell
Michael S. Shuster
Demian A. Ordway
New York, NY 10004
(646) 837-5152
mshuster@hsgllp.com

MasterCard Incorporated and
MasterCard International Incorporated
Attorneys for Defendants-Appellees Visa

Inc., Visa U.S.A. Inc., and Visa
International Service Association *
Arnold & Porter LLP is counsel to the Visa Defendants-Appellees except as to ObjectorsAppellants Barnes & Noble, Inc., Barnes & Noble College Booksellers LLC, J.C. Penney
Corporation, and The TJX Companies, Inc. and related entities.
79
Case: 12-4671
Document: 1123
Page: 94
10/15/2014
1345294
By: /s/ Mark P. Ladner

Mark P. Ladner
Michael B. Miller
250 West 55th Street
New York, NY 10019
(212) 468-8000
mladner@mofo.com
By: /s/ James P. Tallon

James P. Tallon
New York, NY 10022-6069
(212) 848-4000
jtallon@shearman.com

Bank of America, N.A., BA Merchant
Services LLC (f/k/a Defendant National
Processing, Inc.), Bank of America
Corporation, and MBNA America Bank,
N.A.
98

Barclays Bank plc (in its individual
capacity and as successor in interest to
Barclays Financial Corp.) and Barclays
Bank Delaware
By: /s/ Andrew J. Frackman

Andrew J. Frackman
Abby F. Rudzin
Times Square Tower
7 Times Square
New York, NY 10036
(212) 326-2000
afrackman@omm.com
By: /s/ Richard L. Creighton, Jr.

Richard L. Creighton, Jr.
Drew M. Hicks
One East Fourth Street, Suite 1400
(513) 579-6400
rcreighton@kmklaw.com

Capital One Bank(USA), N.A., Capital
One F.S.B., and Capital One Financial
Corp.
Attorneys for Defendant-Appellee Fifth

Third Bancorp
80
Case: 12-4671
Document: 1123
KUTAK ROCK LLP

By: /s/ John P. Passarelli
John P. Passarelli
James M. Sulentic
The Omaha Building
1650 Farnam Street
Omaha, NE 68102-2186
(402) 346-6000
John.Passarelli@KutakRock.com
Page: 95
10/15/2014
1345294
98
WILMER CUTLER PICKERING HALE

AND DORR LLP
By: /s/ Ali M. Stoeppelwerth
Ali M. Stoeppelwerth
1875 Pennsylvania Avenue, NW
(202) 663-6589
ali.stoeppelwerth@wilmerhale.com
Attorneys for Defendant-Appellee First

National Bank of Omaha

HSBC Finance Corporation, HSBC
North America Holdings Inc., and
HSBC Bank USA, N.A.
SKADDEN, ARPS, SLATE, MEAGHER

& FLOM LLP
MASSEY & GAIL, LLP
By: /s/ Peter E. Greene

Peter E. Greene
Boris Bershteyn
Peter S. Julian
Four Times Square
New York, NY 10036
(212) 735-3000
peter.greene@skadden.com
By: /s/ Jonathan S. Massey

Jonathan S. Massey
Leonard A. Gail
1325 G Street NW
Suite 500
(202) 652-4511
jmassey@masseygail.com
Attorneys for Defendants-Appellees JPMorgan Chase & Co., JPMorgan Chase

Bank, N.A., Chase Bank USA, N.A., Chase Manhattan Bank USA, N.A., Chase
Paymentech Solutions, LLC, Bank One Corporation, Bank One, Delaware, N.A.,
and J.P. Morgan Chase Bank, N.A. as acquirer of certain assets and liabilities of
Washington Mutual Bank
Skadden, Arps, Slate, Meagher & Flom LLP is counsel to the Chase Defendants-Appellees
except as to Objectors-Appellants American Express Co., American Express Travel Related
Services Company, Inc., American Express Publishing Corp., Serve Virtual Enterprises, Inc.,
ANCA 7 LLC d/b/a Vente Privee, USA, AMEX Assurance Company, Accertify, Inc., Wal-Mart,
Inc., Alon USA, LP, Amazon.com, Zappos.com, Foot Locker, Inc., and J.C. Penney Corporation,
Inc. and related entities.
81
Case: 12-4671
Document: 1123
Page: 96
10/15/2014
1345294
JONES DAY
ALSTON & BIRD LLP
By: /s/ Joseph W. Clark

John M. Majoras
Joseph W. Clark
(202) 879-3939
jwclark@jonesday.com
By: /s/ Teresa T. Bonder

Teresa T. Bonder
Valarie C. Williams
Kara F. Kennedy
One Atlantic Center
1201 W. Peachtree Street, NW
Atlanta, GA 30309
(404) 881-7000
teresa.bonder@alston.com

National City Corporation, National
City Bank of Kentucky
98

SunTrust Banks, Inc. and SunTrust Bank
PATTERSON BELKNAP WEBB & TYLER
LLP
By: /s/ Jonathan B. Orleans

Jonathan B. Orleans
Adam S. Mocciolo
850 Main Street
Bridgeport, CT 06601-7006
(203) 330-2000
jborleans@pullcom.com
Attorneys for Defendant-Appellee Texas
Independent Bancshares, Inc.
By: /s/ Robert P. LoBue

Robert P. LoBue
William F. Cavanaugh
New York, NY 10036
(212) 336-2000
rplobue@pbwt.com
Wachovia Bank, NA., Wachovia
Corporation, and Wells Fargo
& Company
82
Case: 12-4671
Document: 1123
Page: 97
10/15/2014
1345294
98
CERTIFICATE OF COMPLIANCE WITH FEDERAL RULE OF

APPELLATE PROCEDURE 32(a)
This brief complies with the type-volume limitations of Fed. R. App. P.
32(a)(7)(B) and this Courts October 12, 2014 order (Docket Entry 1119) granting
Defendants-Appellees motion for leave to file an oversized brief of up to 19,000
words because this brief contains 18,351 words (as determined by the Microsoft
Word 2007 word-processing program used to prepare this brief), excluding those
parts of the brief exempted by Fed. R. App. P. 32(a)(7)(B)(iii).
This brief complies with the typeface requirements of Fed. R. of App.
P. 32(a)(5) and the type style requirements of Fed. R. App. P. 32(a)(6) because this
brief has been prepared in a proportionally spaced typeface using the Microsoft
Word 2007 word-processing program in 14-point Times New Roman font.
/s/ Carter G. Phillips
83
Case: 12-4671
Document: 1123
Page: 98
10/15/2014
1345294
98
I hereby certify that on this 15th day of October 2014, I electronically filed
and served the foregoing brief using the CM/ECF System.
/s/ Carter G. Phillips
Carter G. Phillips
84
Case 1:11-md-02221-NGG-RER Document 513 Filed 07/14/14 Page 1 of 53 PageID #: 24336

HIGHLY CONFIDENTIAL SUBJECT TO PROTECTIVE ORDER
UNITED STATES DISTRICT COURT

EASTERN DISTRICT OF NEW YORK
-----------------------------------------------------------------------x
IN RE: AMERICAN EXPRESS ANTI-STEERING
RULES ANTITRUST LITIGATION
This Document Relates To:
CONSOLIDATED CLASS ACTION
-----------------------------------------------------------------------x
THE MARCUS CORPORATION,
on behalf of itself and all similarly situated persons,
11-MD-02221 (NGG) (RER)
13-CV-07355 (NGG) (RER)

Plaintiff,
- against AMERICAN EXPRESS COMPANY et al.,
Defendants.
-----------------------------------------------------------------------x
REPLY DECLARATION OF ALAN S. FRANKEL, PH.D.

1.
2.
3.
4.
INTRODUCTION AND OVERVIEW ......................................................................................................... 1

1.1.
My Analysis of American Expresss NDP and the Agreement ....................................................... 1
1.2.
Professor Stiglitzs Declaration ...................................................................................................... 3
1.3.
Professor Hausmans Report ......................................................................................................... 4
ECONOMIC CONTEXT OF THE AGREEMENT......................................................................................... 7

2.1.
The Incremental Reform of Anti-Steering Rules ........................................................................... 7
2.2.
The Agreement Liberalizes Competition Relative to the Status Quo.......................................... 10
2.3.
The Prospects for Further Competitive Reform and the Reasonableness of the Settlement .... 14
MARKET DEFINITION, MARKET POWER, AND STEERING IN COMPETITIVE MARKETS .................... 15

3.1.
Market Definition ........................................................................................................................ 15
3.2.
Market Power.............................................................................................................................. 19
3.3.
Steering in Competitive Markets ................................................................................................ 19
PROFESSORS STIGLITZ AGREES WITH EACH ELEMENT OF MY ECONOMIC ANALYSIS AND

PROFESSOR HAUSMANS CRITICISMS HERE ARE UNPERSUASIVE AND INCONSISTENT WITH
HIS EARLIER POSITIONS...................................................................................................................... 19
4.1.
Credit Cards are Substantially More Costly to Merchants Than Other Payment Methods,
Including Debit Cards .................................................................................................................. 20
4.2.
At the Level of Credit Card Fees Prevailing in the United States, Many Merchants are
Likely to Surcharge Credit Card Transactions.............................................................................. 21
4.2.1.
My Views About the Likelihood that Merchants Will Surcharge Credit Card
Transactions Have Not Changed ......................................................................................... 21
4.2.2.
The Evidence From Australia Supports a Conclusion that Many Merchants Will
Surcharge Credit Card Transactions in the United States ................................................... 23
4.2.3.
Surcharging is More Likely in the United States than in Australia or the U.K. at Current
Levels of Merchant Fees...................................................................................................... 27
4.3.
Surcharging Permits Merchants to Recoup the Cost of Credit Card Payments Directly From
Customers Who Use Credit Cards and Post Lower Shelf Prices .................................................. 28
4.4.
Many Customers Will React to Credit Card Surcharges by Using Lower Cost Alternatives at
Merchants that Surcharge, Reducing Costs and Prices ............................................................... 29
4.4.1.
Narrow Relevant Markets Do Not Imply That Credit Card Surcharges Cannot Shift
Transaction Volume to Debit Cards .................................................................................... 29

4.4.2.
Professors Stiglitz and Hausman Have Agreed that Credit Card Surcharges Shift Some
Credit Card Transactions to Debit Cards ............................................................................. 32
4.4.3.
A Recent One-Year Change in Credit and Debit Card Volume in Australia Does Not
Disprove The Usefulness of Surcharges at Steering to Debit.............................................. 36
4.5.
The Threat to Networks of Lost Transaction Volume From Credit Card Surcharges Will
Generate a Previously Suppressed Competitive Constraint on the Level of Credit Card
Merchant Fees............................................................................................................................. 39
4.5.1.
Networks Lose More Transactions as Merchant Fees Increase if Merchants Can

Surcharge Than If Merchants Cannot Surcharge ................................................................ 39
4.5.2.
Surcharging Intensifies Constraints on Merchant Fees Relative to the Status Quo Even
if Uniform Across Brands..................................................................................................... 41
4.5.3.
Professor Hausman is Incorrect in His Claim that Surcharging Has Not Caused a
Reduction in American Expresss Merchant Fees in Australia ............................................ 43
4.6.
The Agreement Benefits Merchants With Respect to All Current Credit Card Usage, Not
Only American Express Card Usage............................................................................................. 47
5.
STATE RESTRICTIONS.......................................................................................................................... 48
6.
CONCLUSION ...................................................................................................................................... 49

1. INTRODUCTION AND OVERVIEW

1.1. My Analysis of American Expresss NDP and the Agreement
1.
On April 10, 2014, I submitted a declaration (Frankel Declaration) concerning
the proposed class settlement (Agreement) in this litigation between American Express and a
settlement class consisting of the millions of U.S. merchants that accept American Express
cards, under which American Express would partially relax its non-discrimination policy
(NDP). 1 The Agreement follows settlements reached between a similar class of merchants
and MasterCard and Visa in MDL-1720 under which merchants obtained the ability to surcharge
MasterCard and Visa credit card transactions, with certain restrictions including linkage to the
merchants ability to surcharge American Express card transactions (if the merchant accepts
American Express cards and they are more costly than MasterCard or Visa credit cards).
2.
American Expresss NDP currently prohibits a merchant from surcharging
American Express card transactions unless the merchant surcharges all other card transactions,
including those using debit cards, by at least as much as the merchant surcharges American
Express card transactions. But MasterCard and Visa prohibit surcharging of their debit cards
(and merchants generally would not want to surcharge debit cards, which are lower cost and
the use of which merchants wish to encourage), so merchants that accept American Express
cards generally cannot surcharge any credit cards. The Agreement would permit merchants to
surcharge American Express card transactions so long as the merchant surcharges all credit
cards by at least as much i.e., the requirement that debit cards also be surcharged in order to
Declaration of Alan S. Frankel, Ph.D., April 10, 2014 (Frankel Declaration). My qualifications are described in
the Frankel Declaration. My updated CV is attached in Appendix A.

surcharge American Express cards will be removed. 2 Merchants still will be prohibited from
surcharging American Express but not other credit cards or setting a higher surcharge for
American Express cards than for other credit cards. I refer to this continued American Express
restriction as its no-differential surcharge or NDS policy.
3.
I concluded that [a]lthough further liberalization [of American Expresss NDP]
would benefit merchants, the relief obtained in the Agreement will significantly benefit
merchants throughout the United States. 3 I based my conclusion on a straightforward
economic analysis built on the following logical steps and supported by the available evidence.
Differential pricing or promotion at retail is a principal mechanism by which
competition between merchants suppliers occurs. 4 Payment card networks like
American Express have used anti-steering rules to suppress this basic
competitive process with respect to the networks card acceptance services
provided to and paid for by merchants.5 Anti-steering rules are used by card
payment networks to restrict competition, enhance market power, and maintain
anticompetitively high merchant fees. 6
At the level of credit card fees prevailing in the United States, many merchants
are likely to surcharge credit card transactions. 7
Surcharging permits merchants to recoup the cost of credit card payments
directly from customers who use credit cards and set lower posted prices. 8
Surcharging credit card transactions will induce many customers to use
alternatives, especially debit cards, which directly reduces merchant costs.9
3
4
5
6
7
8
9
Under the terms of the Agreement, American Express also may limit the amount of a merchant surcharge to
the amount that the merchant pays to American Express to accept the card transaction. Class Settlement
Agreement, January 7, 2014 (Agreement), p. 21.
Frankel Declaration, 74.
Frankel Declaration, Part 3.2.
Frankel Declaration, 24 and Part 3.4.

The threat to networks of lost transaction volume from credit card surcharges
will generate a previously suppressed competitive constraint on the level of
credit card merchant fees. 10
The increased prevalence of credit card surcharging will tend to shift general
payment preferences and patterns towards debit card use at all merchants,
whether they surcharge or not.11
Statutes that may restrict the ability to surcharge in some states may reduce, but
not eliminate, the value of the right to surcharge all credit cards under network
rules. 12
American Expresss NDP significantly reduces the number of merchants that can
economically surcharge any credit card transactions even after the MDL-1720
Settlements, so that the economically relevant benefits from the Agreement
extend beyond reductions in costs currently incurred to process American
Express transactions, by permitting merchants more effectively to reduce all of
their credit card transactions and fees.13
1.2. Professor Stiglitzs Declaration
4.
Professor Joseph Stiglitz has submitted a declaration on behalf of a group of
major supermarket and drug store merchants (Individual Merchant Plaintiffs) that object to
the Agreement and which are currently engaged in litigation with American Express over its
NDP (and which previously litigated similar issues against MasterCard and Visa in MDL-1720).
5.
Professor Stiglitz neither addresses nor disputes any of the elements of my
analysis that show that merchants will benefit from the ability to surcharge all credit card
transactions without surcharging debit card transactions. Professor Stiglitzs opinions about
10
11
12
13
Frankel Declaration, Part 3.5. Professor Hausman oddly states, Dr. Frankel makes no claims that the
settlement will lead to reduced AMEX merchant fees. Report of Professor Jerry Hausman, June 6, 2014
(Hausman Report), 9 (at p. 7). But I do believe that the settlement will lead to reduced fees by all of the
credit card networks, including American Express, as I stated in my initial declaration. Frankel Declaration,
15, 24, 58, 65, and Professor Hausman elsewhere cites my conclusion to that effect (Hausman Report, 9
(at p. 5), 44, 57 (and disagrees with my conclusion)).
Frankel Declaration, 15, 24.

competition in card payment systems and the anticompetitive effects of anti-steering rules,
which he has described in his previous reports, are very similar to my own, and should lead him
to agree with me that the ability to surcharge credit card transactions is valuable to merchants
and to the public.
6.
In his declaration, Professor Stiglitz focuses instead on some elements of the
Agreement such as American Expresss ability to continue enforcing its NDS policy which he
contends cause the Agreement to harm merchants. As I explain, however, Professor Stiglitz is
only correct if the alternative to the Agreement is assumed to be better, more valuable relief,
not the status quo. A settlement is invariably a compromise, however, and I do not evaluate
the Agreement relative to relief that I believe would most fully make the market competitive,
rather I evaluate it relative to the status quo in determining that the Agreement benefits U.S.
merchants.
1.3. Professor Hausmans Report
7.
Professor Jerry Hausman has submitted a report on behalf of a group of 41 other
merchants and merchant groups (Constantine Cannon Objectors) that object to the
Agreement.
8.
Professor Hausman and I agree about many of the competitive economic
features of the payment card marketplace, but he disagrees with my conclusion that the ability
to surcharge credit cards without surcharging debit cards under the terms of the Agreement
will benefit U.S. merchants. Professor Hausman explains why American Express possesses
significant market power, but he neither offers a diagnosis of any conduct he believes American
Express has engaged in that has had significant anticompetitive effects in the United States nor

prescribes any relief which might reduce or eliminate American Expresss market power. In
fact, it is unclear whether Professor Hausmans opinions, if accepted, could even be consistent
with merchants succeeding in an antitrust case against American Express.
9.
To explain, in MDL-1720, Professor Hausman contended that [s]urcharging will
not eliminate or discipline the market power exercised by Visa and MasterCard. 14 In that case
and in his report here he contends that surcharging even differential surcharging of American
Express cards has had no effect on the level of American Express merchant fees in Australia,
where surcharging and differential surcharging have been common, and he claims that a far
lower share of U.S. merchants are likely even to try surcharging card transactions in the United
States. Thus, logically, Professor Hausman cannot argue that the continued American Express
NDS policy is conduct that has a significant anticompetitive effect by preventing competition
from reducing American Expresss market power.
10.
Professor Hausman credits the decline in American Express fees in Australia
exclusively to merchants ability there to discontinue acceptance of American Express cards

(which he says merchants are less able to do in the United States) coupled with regulatorymandated reductions in MasterCard and Visa interchange fees not to surcharging of American
Express cards.15 But relief that would force MasterCard and Visa to reduce their credit card
interchange fees is unavailable in a case being litigated against American Express (and American
Express itself is a single corporate entity, and not one created by an association of banks that
otherwise compete). American Express does not permit open competition among acquiring
14
15
Report of Professor Jerry Hausman, In re Payment Card Interchange Fee and Merchant Discount Antitrust
Litigation, May 28, 2013 (Hausman MDL-1720 Report), 10.
Hausman MDL-1720 Report, 24, 65-71.

banks to set the level of American Express merchant fees (and unlike MasterCard and Visa,
American Express can control the pricing of all of its merchant agreements without any explicit
interchange fee system).
11.
Professor Hausman argued in MDL-1720 that [m]ore extensive reform which
eliminates the Honor All Cards (HAC), Honor All Issuers (HAI), and no bypass rules is necessary
to constrain the supra-competitive interchange fees of MasterCard and Visa. 16 He also
identified default interchange fee schedules fees MasterCard and Visa mandate flow from
the merchants bank to the card issuing bank as a restrictive practice of those networks.
Many individual banks issue MasterCard and Visa credit cards, and Professor Hausman argued
that if merchants could negotiate with individual member banks of MasterCard or Visa over the
terms of acceptance of credit cards issued by those individual banks, then the merchants would
be able to counter the market power otherwise exercised by MasterCard and Visa on behalf of
themselves and their member banks collectively. But, although American Express has enlisted
some U.S. banks to use American Express branded cards, it continues to control the merchant
side of its network and the majority of American Express card issuing volume internally. 17
12.
In short, while Professor Hausman casts doubt on the value to merchants of the
relief provided by the Agreement, he does not offer the prospect that antitrust litigation against
American Express should produce a better result (or any positive result) for merchants.
However, he is incorrect about the value of the ability for merchants to surcharge as provided
for by the Agreement, as I will discuss further below.
16
17
Hausman MDL-1720 Report, 10 (emphasis added).

No bypass rules refer to Visa restrictions on merchants and banks using an alternative to Visa to provide
authorization, clearing, and settlement services in interbank Visa card transactions. There is no analogue in
the American Express network because American Express is predominantly a unitary organization.

2. ECONOMIC CONTEXT OF THE AGREEMENT

2.1. The Incremental Reform of Anti-Steering Rules
13.
The NDP is American Expresss version of anti-steering rules requirements
which impede or prohibit merchants from discouraging their customers use of a more costly
payment method (short of discontinuing acceptance of that method) or encouraging the use of
less costly payment methods. In my initial declaration (as in my published research), I
explained that anti-steering rules have had anticompetitive effects.
[C]ard networks use anti-steering rules (especially no-surcharge rules) to
restrict competition, enhance market power, and maintain
anticompetitively high merchant fees. In my view, merchants should be
free to compete with other merchants over the terms of sale, including
any surcharges, discounts, or other incentives with respect to payment
methods. 18
14.
Anti-steering rules have taken a variety of forms. Card networks originally
prohibited merchants accepting credit cards from treating customers using those cards less
advantageously than the merchant treated any other customers including cash customers.
Legislation has relaxed some aspects of American Expresss NDP and similar rules of other card
networks. In the early 1980s, the Cash Discount Act permitted merchants to offer discounts to
customers who use cash.19 In 2010, the Dodd-Frank Act expanded that to permit merchants to
offer discounts to customers who use cash, checks, or debit cards overriding network rules to
the contrary and permitted merchants to require a minimum purchase amount (up to $10) for
use of credit cards. 20
18
19
20

Public Law 97-25, July 27, 1981, Sec. 101.
Public Law 111-203, July 21, 2010, Sec. 1075.

15.
Under the card networks rules, merchants were still effectively prohibited from
offering discounts (or similar benefits) to users of one brand of credit cards but not others, or
adding a surcharge to any credit card transactions. With respect to brand-specific credit card
discounts, the United States Department of Justice brought an action against MasterCard, Visa,
and American Express, and reached a settlement with MasterCard and Visa under which those
networks permit merchants to offer discounts to customers who use alternative brands of
credit cards. American Express maintains its prohibition, however, and the Department of
Justice and American Express are currently litigating the issue. 21
16.
Until recently, and with only very limited exceptions, MasterCard and Visa
prohibited surcharges on the use of their respective credit card brands. Discover Card requires
that a merchant can only surcharge Discover Card transactions if the merchant also surcharges
other credit card transactions by at least as much as the merchant surcharges Discover Card
transactions.22 American Express prohibits a merchant from surcharging American Express card
transactions unless the merchant also surcharges all other credit card and debit card
transactions by at least as much.
17.
In 2012, MasterCard and Visa agreed to relax their no-surcharge rules as part of
their respective settlements with a merchant class (MDL-1720 Settlements). In particular,

those networks agreed to permit a merchant to surcharge MasterCard or Visa card
transactions, but only if the merchant also surcharges more costly credit cards (i.e., typically
21
22
Discover, Merchant Operating Regulations, October 14, 2011, DFS0809209 at 226.

American Express) by at least as much. The MDL-1720 Settlements permits MasterCard and
Visa to continue their prohibitions of surcharges on MasterCard or Visa debit card transactions
(and merchants generally would be unlikely to want to surcharge debit cards which are much
less costly than credit cards). But American Express forbids merchants from surcharging
American Express card transactions unless they also surcharge debit cards. Thus, if a merchant
accepts American Express cards, it typically cannot surcharge any credit cards.
18.
As I explained in my initial declaration, the absence of relief from American
Expresss linkage of credit card surcharges to debit card surcharges significantly attenuated (but
did not eliminate) the value of the relief in MDL-1720. 23 Relief from the American Express
linkage of surcharges of its cards to surcharges on debit cards would make credit card
surcharges a more practical option for many merchants. Indeed, objectors to the MDL-1720
Settlements cited the American Express linkage between credit and debit card surcharges as a
reason why they claimed that those settlements would provide no benefit to merchants. 24
Professor Hausman likewise cited that linkage as a reason why most U.S. merchants would not
surcharge credit cards and so would not, in his view, benefit from the MDL-1720 Settlements. 25
23
24
25
Frankel Declaration, 63, 74; see also See, Declaration of Alan S. Frankel, Relating to the Proposed Class
Settlement, April 11, 2013 (Frankel MDL-1720 Declaration), Part 4.6.2 and Reply Declaration of Alan S.
Frankel, Ph.D., Relating to the Proposed Class Settlement, August 16, 2013 (Frankel MDL-1720 Reply
Declaration), Part 4.4.2.
As National Retail Federation argued in objecting to the MDL-1720 Settlements, In order to surcharge,
retailers must be willing to sacrifice their business relationship with American Express which few if any can
afford to do. National Retail Federation Statement of Objection to Final Approval of the Proposed Rule
23(B)(2) Agreement, In Re Payment Card Interchange Fee and Merchant Discount Antitrust Litigation, May 28,
2013 (NRF MDL-1720 Objection), pp. 5, 18-19
Hausman MDL-1720 Report, 64-67. Unlike the Constantine Cannon Objectors, the National Retail
Federation and Professor Hausman, the Independent Merchant Plaintiffs supported the MDL-1720
Settlements, arguing that while additional relief against American Express was important to obtain, the relief
with respect to MasterCard and Visa no-surcharge rules was nevertheless valuable. Richard Arnold in MDL1720 Fairness Hearing, September 12, 2013, pp. 44-48.

The Agreement will now eliminate American Expresss linkage of credit card and debit card
surcharging, so merchants that accept American Express cards (which collectively account for a
large share of retail sales) 26 can surcharge credit card transactions without dropping acceptance
of American Express.
2.2. The Agreement Liberalizes Competition Relative to the Status Quo
19.
Professors Stiglitz and Hausman (and the objectors) misleadingly suggest that
the Agreement creates new anticompetitive restrictions that will harm merchants. Professor
Stiglitz claims that the Agreement will
27
He
cites, for example, the ability under the MDL-1720 Settlements for merchants to differentially
surcharge at the product level i.e., a different surcharge for premium MasterCard or Visa
credit cards that cost the merchant more to accept than non-premium credit cards. 28 Professor
Stiglitz contends that the Agreement
29 He explains:
26
27
28
29
Declaration of Prof. Joseph Stiglitz, May 27, 2014 (Stiglitz Declaration), 26 (emphasis added).
Stiglitz Declaration, 28.
Stiglitz Declaration, 30. Although product-level surcharging is an option that I believe merchants should
have, there are practical impediments with implementing such a strategy. In Australia, the Reserve Bank of
Australia has encouraged merchants to engage in product-level surcharging but has found that few, if any,
merchants use this strategy. See, Reserve Bank of Australia, A Variation to the Surcharging Standards: Final
Reforms and Regulation Impact Statement, http://www.rba.gov.au/paymentssystem/reforms/cards/201206-var-surcharging-stnds-fin-ref-ris/pdf/201206-var-surcharging-stnds-fin-refris.pdf, June 2012, p. 6.
10

. 30
20.
But American Express currently forbids any surcharging unless the merchant also
surcharges all other card transactions, including debit card transactions, by at least the same
amount. And the MDL-1720 Settlements did not eliminate the pre-existing MasterCard and
Visa no-surcharge rules with respect to debit card transactions. Thus, under the status quo
rules of these networks, only if a merchant drops acceptance of American Express cards can it
typically surcharge any MasterCard or Visa credit cards by any amount.
21.
Professor Stiglitz thus is not comparing the competitive strategies available to a
merchant under the terms of the Agreement to the strategies available to them today. Instead,
he is comparing their available options under the Agreement to a hypothetical set of even
better options that would permit a merchant to engage in the more refined pricing strategies
he describes, and which he says are foreclosed by the Agreement. Professor Stiglitz states that
31 But that is not true. The Agreement eliminates a

key American Express competitive restriction while not eliminating other restrictions. Its net
effect cannot be to make the situation worse than it already is. It is only worse than it would be
compared to a but-for world in which some or all of the remaining restrictions were also
eliminated.
22.
Professor Stiglitz is not alone in attributing anticompetitive effects to the
Agreement rather than to the status quo. Professor Hausman states:

30
31
Stiglitz Declaration, 31.

Stiglitz Declaration, 26 (emphasis added).
11

The proposed settlement will eliminate any economic incentive for a

merchant to attempt to steer business away from AMEX [and towards
Visa, MasterCard, or Discover credit cards] using surcharges, since the
permitted surcharge for AMEX can be no larger than the surcharges for
VMC or Discover cards, even though the costs of those cards are lower
than AMEX cards. 32
23.
Professor Hausman misleadingly claims that the Agreement will create oligopoly
behavior 33 and provides American Express with more market power than it has today:
[S]ince merchants will not be permitted to surcharge AMEX credit-card
transactions unless they also surcharge VMC credit-card transactions
equally, I find it likely that AMEX will increase its merchant fees relative to
VMC or Discover because AMEX knows merchants cannot steer
consumers to VMC or Discover credit cards by not surcharging them or
surcharging them less than AMEX credit cards. 34
24.
He suggests that the Agreement will lead to an anticompetitive outcome by
permitting American Express to enforce its no-differential surcharging restriction 35 and states
that [a]n even worse possible result of the Agreement is the price fix outcome which he
explains, means that [s]ince AMEX and VMC will all realize that the allowed surcharge will be
identical among the three cards, they may all increase their merchant fees. 36
25.
The Constantine Cannon Objectors similarly state:

[T]he Proposed Settlement not only eliminates the one benefit of
surchargingthe ability of at least some merchants to benefit from
interbrand competition through differential surchargingwhile [sic]
32
33
34
35
36
Hausman Report, 42 (emphasis added).

Hausman Report, 61.
Hausman Report, Part VII section heading (emphasis added).
Hausman Report, 62.
12

empowering American Express to raise its rates to further entrench its

power. 37
26.
Notwithstanding its earlier argument that the ability to surcharge MasterCard
and Visa transactions provided by the MDL-1720 Settlements was not valuable to merchants
due to a merchants need to drop American Express cards to take advantage of it, the National
Retail Federation similarly states that the Agreement affirmatively undercuts merchants
surcharging abilities, 38 because merchants that wish to surcharge Amex must surrender their
right to engage in the differential, interbrand surcharging that the MDL 1720 settlement
permits. 39 But the Agreement does not permit American Express to begin enforcing a nodifferential surcharge provision with respect to other credit card brands American Express
already enforces that policy without the Agreement. The NRF criticized the MDL-1720
Settlements because, with American Express NDP in place, they said few merchants would
surcharge at all. Now they suggest elimination of the American Express restriction is taking
away a valuable option from merchants.
27.
In fact, the Agreement provides a new option surcharge all brands of credit
cards but not debit cards that merchants did not previously have. That cannot lead to an
anticompetitive outcome or
or create a price fix
outcome that lets American Express increase its merchant fees relative to what it would
charge absent the Agreement. Relative to the status quo, the Agreement clearly does none of
these things.
37
38
39
Memorandum of Law in Opposition to Motion for Final Approval of Settlement on behalf of: 7-Eleven, Inc., et
al., June 6, 2014 (CC Brief), p 35 (emphasis added).
Statement of Objections to the American Express Class Action Settlement of Absent Putative Rule 23(B)(2)
Class Member National Retail Federation, May 21, 2014 (NRF Statement of Objections), p. 7.
Id., p. 8.
13

2.3. The Prospects for Further Competitive Reform and the Reasonableness of the
Settlement
28.
I explained in my initial declaration and revisit below why the ability to surcharge
all credit cards will cause a relative shift of transaction volume to debit cards (and cash) and
why that will benefit merchants. I agree with Professor Stiglitz that no-differential surcharge
rules are anticompetitive and merchants should have the ability to establish whatever
surcharges or discounts they wish in competition with other merchants.
29.
American Express, MasterCard, and Visa all disagree, and they all have retained
economic experts in this case or MDL-1720 previously who argue that the networks lack market
power and their anti-steering rules are procompetitive and benefit the public. I believe that
those arguments are incorrect. But any claim that the Agreement has anticompetitive effects
by enabling American Express to enforce a no-differential surcharge policy implicitly is basing
that claim on the assumption that the relevant status quo is not the actual world but rather a
but-for world in which all aspects of American Expresss NDP have been eliminated. In other
words, absent the Agreement some plaintiff will litigate or arbitrate to a successful verdict and
obtain injunctive relief on behalf of all merchants in the United States (rather than a narrower
injunction or settlement or one applicable only to a particular plaintiff) that permits all
merchants to engage in differential surcharging, and that outcome is upheld on appeal. That
requires a litigation risk analysis in addition to an economic analysis. 40 If it were certain that
rejection of the Agreement would be followed quickly by even more effective competitive
40
Both Professors Stiglitz and Hausman avoid the litigation risk issue by making the extreme claim that the
Agreement provides no benefits (so that any possibility of more complete relief is more valuable than the
Agreement). Although I am confident that American Expresss NDP is anticompetitive and should be
abolished, I have not undertaken a litigation risk analysis and have no opinion on the likelihood that a better
outcome will materialize for the class as a whole if the Agreement is rejected.
14

reforms applicable to the millions of merchants in the class, then it would make sense to reject
the Agreement. But neither Professor Stiglitz nor Professor Hausman can provide any such
certainty (and Professor Hausman does not even identify any anticompetitive American Express
conduct).
30.
Professor Stiglitz criticizes the Agreement on the grounds that the Agreement
does not eliminate American Expresss restrictions on merchants discounting other credit card
brands or engaging in soft steering. 41 But the United States Department of Justice is trying
those issues in a separate case. Even in a but-for world setting (i.e., comparing the Agreement
to an even better outcome rather than the status quo), to judge the Agreement as deficient on
this basis is equivalent to assuming that if the Department of Justice loses its case, one or more
merchants nevertheless would have subsequently prevailed and obtained market-wide relief on
the same issues. 42 Moreover, I understand that the Department of Justice can still bring an
action challenging remaining aspects of American Expresss NDP.
3. MARKET DEFINITION, MARKET POWER, AND STEERING IN COMPETITIVE MARKETS
3.1. Market Definition
31.
Professor Hausman devotes considerable attention to the issue of market
definition. Market definition is typically performed as a part of an indirect analysis of market

power. From an economic perspective, market definition may not be particularly important
when market power or anticompetitive effects of conduct (or relief from conduct) can be
directly evaluated, as they can in this case. The Constantine Cannon Objectors and the National
41
42
Stiglitz Declaration, 27, 45.

Moreover, if the Department of Justice fails to prevail in its case against American Express with respect to
discounting, that presumably could affect the likelihood that a merchant would prevail with respect to
surcharging.
15

Retail Federation, however, use (or rather misuse) market definition in a purely semantic way
to argue that merchants cannot benefit by steering customers from credit card transactions to
debit cards an issue I discuss in Part 4.4.
32.
Although Professor Hausman and Professor Stiglitz (and Individual Merchant
Plaintiff expert Dr. Christopher Vellturo) have used somewhat different language to describe
the relevant markets in the payment card industry, our respective analyses have been similar.
In MDL-1720, I concluded that relevant antitrust markets exist for services provided to
merchants separate from services provided to cardholders, for credit cards separate from debit
cards, and in part because of the effects of network anti-steering rules for different brands
of credit card or debit card acceptance services sold to merchants. 43 As I explained, [a]ntisteering rules by their nature tend to create separate relevant markets since they reduce the
substitutability of other payment methods. 44 Even in a market defined to include all credit
card acceptance services (or network services) sold to merchants, I explained that the loss of
profits to most merchants from dropping any one brand, the inability to process one brand of
card transactions using another network (and its fees), and the anti-steering rules, meant that
market shares in the broader general purpose credit card network services market, standing
alone, would tend to understate the degree of market power that each network could
exercise. 45
43
44
45
See, Report of Alan S. Frankel, MDL Docket No. 1720, July 2, 2009 (Frankel MDL-1720 Report), 11.
Id., 50.
Id., 15 (Anti-steering rules have effectively divided the market so that each brand MasterCard and Visa
of a particular type of card network service is supplied in a relevant market separate from the other brands.
Equivalently shares in the broader market for general purpose card, offline debit card, and PIN debit card
network services can tend to underemphasize each networks market power, due to their anti-steering
rules.).
16

33.
Professor Stiglitz reaches the same conclusion as I did. In MDL-1720, he
explained:
46
34.
With respect to American Express, he similarly explains that
47
35.
Professor Hausman concludes that there is a relevant market for general
purpose credit card network services (i.e., services to merchants). 48 His focus regarding market
definition in this case is in distinguishing credit cards from debit cards, and he does not (in this
46
Expert Report of Joseph Stiglitz, Ph.D., In Re: Payment Card Interchange Fee and Merchant Antitrust Litigation,
June 25, 2009, (Stiglitz MDL-1720 Report), 10(I). See also Expert Report of Dr. Christopher A. Vellturo, In
Re: Payment Card Interchange Fee and Merchant Antitrust Litigation, July 2, 2009 (Vellturo MDL-1720
Report), 144
47
Expert Report of Professor Joseph Stiglitz, In Re American Express Anti-Steering Rules Antitrust Litigation, April
2, 2013, (Stiglitz Report), 36. Dr. Vellturo also reaches the same conclusions that he reached in MDL-1720.
Expert Report of Dr. Christopher A. Vellturo, In Re American Express Anti-Steering Rules Antitrust Litigation,
April 3, 2013 (Vellturo Report), 278.
Hausman Report, 20; Hausman MDL-1720 Report, 27.
48
17

case) explore whether the market can be defined even more narrowly (to single brands) due,
for example, to the card networks anti-steering rules. 49 In a case brought by the New Zealand
Commerce Commission against MasterCard and Visa in which both Professor Hausman and I
served as experts on behalf of the Commission, Professor Hausman defined a relevant market
consisting of MasterCard and Visa services provided to merchants. But he explained like I have
that [o]ne could define a more narrow market of Visa acquiring [i.e. merchant services] and
similarly a market for MasterCard acquiring and that none of the subsequent economic
analyses would change. 50 Professor Hausman also explained in New Zealand that instead of a
single-brand market or a MasterCard-Visa credit card acquiring market, one could define a
broader market for all brands of merchant credit card acquiring services (e.g., in New Zealand
including American Express and Diners Club) without altering the economic analysis. 51
36.
Professor Stiglitz does not address the geographic extent of the market, but
Professor Hausman and I agree that the relevant geographic market is the United States. 52
37.
In short, there is no substantive difference between my opinions and those of
Professors Hausman and Stiglitz concerning market definition.

38.
I will return to the implications of market definition (or lack thereof) for analysis
of the effects of surcharging in Part 4.4.1.

49
50
51
52
His analysis of each networks market power and the need of most merchants to accept each of the major
brands, however, is the same as mine, Professor Stiglitzs, and Dr. Vellturos. For example, he explains that
MasterCard, Visa, and American Express potentially could compete against each other regarding merchant
fees but that this competition has not occurred in practice. Hausman Report, 40.
Brief of evidence of Professor Jerry Hausman (redacted open version) in Commerce Commission v. Cards NZ
Limited, et al. and DSE (NZ) Limited, et al. v. Cards NZ Limited et al., 4 May 2009 (Hausman NewZealand Brief
of Evidence), 4.19.
Id., 4.20.
Hausman Report, 38; Frankel Declaration, 73. Dr. Vellturo states his understanding that
.
Vellturo Report, FN140.
18

3.2. Market Power

39.
I agree with Professor Hausman and Professor Stiglitz that MasterCard, Visa, and
American Express all possess market power. 53

3.3. Steering in Competitive Markets
40.
Professor Hausman, Professor Stiglitz, and I are in agreement that, as Professor
Hausman puts it, price differences through surcharges or discounts are common in the U.S.
retail economy. 54 However, Professor Hausman misleadingly asserts that the proposed
settlement prohibits just this type of competitive behavior (referencing American Expresss nodifferential surcharge policy). 55 As I explained in Part 2, however, it is American Expresss NDP
that prohibits differential surcharges not merely between American Express cards and other
brands of credit cards, but also between American Express cards and debit cards. The
Agreement liberalizes the American Express policy by removing this link to debit cards. The
MDL-1720 Settlement permits merchants to surcharge MasterCard and Visa credit card
transactions if the merchant surcharges American Express transactions, and with the proposed
change to American Expresss NDP under the terms of the Agreement, merchants will be able
to surcharge American Express, MasterCard, and Visa credit card transactions.
4. PROFESSORS STIGLITZ AGREES WITH EACH ELEMENT OF MY ECONOMIC ANALYSIS AND
PROFESSOR HAUSMANS CRITICISMS HERE ARE UNPERSUASIVE AND INCONSISTENT WITH
HIS EARLIER POSITIONS
41.
In the remainder of this Reply, I explain why I believe that economic evidence
and analysis shows that the ability to surcharge credit card transactions (equally) without a
53
54
55
Hausman Report, Part V; Stiglitz Declaration, 12 (

; Stiglitz MDL1720 Report, 66 (
); Frankel MDL-1720 Report, Part 4.
Hausman Report, 60; Stiglitz Declaration, 17; Frankel Declaration, Part 2.
Hausman Report, 60.
19

requirement to also surcharge debit card transactions will benefit U.S. merchants. Professor
Stiglitz criticizes the Agreement but never addresses this central question. I show that he
agrees with each element of my economic analysis that leads to my conclusion. Professor
Hausman claims here that U.S. merchants wont surcharge credit cards, and surcharging credit
cards would neither induce a relative shift of transaction volume to debit cards nor create any
pressure on credit card networks to reduce their merchant fees. But he has previously agreed
with me on these points. He distinguishes the usefulness of surcharging to U.S. merchants from
its usefulness to merchants in Australia because the U.S. merchant sectors are highly
competitive, but he has previously agreed that surcharging is beneficial even in competitive
markets.
4.1. Credit Cards are Substantially More Costly to Merchants Than Other Payment
Methods, Including Debit Cards
42.
In my initial declaration, I showed that it costs U.S. merchants substantially more
when a customer pays for a purchase with a credit card compared to the same transaction paid
for using other payment methods, in particular, a debit card. American Expresss merchant
fees, on average, are higher than those for Discover Card, MasterCard, and Visa credit cards,
but each of those (and the weighted average of them together) is substantially higher than the
cost of accepting debit cards. 56 Neither Professor Stiglitz nor Professor Hausman disagree that
credit cards cost merchants far more than debit cards.
56
Frankel Declaration, Part 3.1 and Figure 3.
20

4.2. At the Level of Credit Card Fees Prevailing in the United States, Many Merchants are
Likely to Surcharge Credit Card Transactions
43.
In my initial declaration, I explained that many merchants are likely to adopt
surcharges on credit card transactions as a strategy for coping with high credit card fees in this
country, particularly given the significantly lower fees incurred to accept debit card
transactions.57 Professor Stiglitz agrees with me that the incentive for a merchant to surcharge
increases with the level of merchant fees.58
4.2.1. My Views About the Likelihood that Merchants Will Surcharge Credit Card
Transactions Have Not Changed
44.
Professor Hausman contends that few U.S. merchants are likely to surcharge
credit card transactions due to a business stealing effect in which merchants that do not
surcharge credit card transactions gain customers and profits from those that do, which
alienates credit card customers. 59 He states that I previously agreed with his analysis. 60 That is
not correct. Professor Hausman suggests that the business-stealing impediment to surcharging
is permanent, deters almost all merchants from surcharging (even at the level of credit card
merchant fees prevailing in the United States), and essentially eliminates any benefits from the
ability to surcharge. I have never agreed with that analysis. He cites part of a passage from a
paper of mine published by the Reserve Bank of Australia (RBA) in suggesting that I agree that
merchants are unlikely to surcharge, but in that passage I say merely that surcharging is not a
complete solution to the legacy of anticompetitive effects of the Visa and MasterCard
57
58

Stiglitz Declaration, FN27 (in the context of American Express:
.
59
60
Hausman Report, 47.

Hausman Report, 46.
21

monopolies and their conduct over the decades, and Professor Hausman omits half of a
sentence in which I explain that it is particularly difficult for merchants to communicate the
availability of lower prices using payment methods other than credit cards if the credit card
networks have other rules which they did in the United States that restricted merchants
ability to promote their non-credit card prices.61 In the paper cited by Professor Hausman, I
explain at length why the ability to surcharge credit card transactions is an important
competitive reform that benefits merchants and the public.
[C]ompetitive pressure on networks to constrain the amount of the interchange
fee is more effective if a merchant can choose the network, reflect its relative
costs in point-of-sale surcharges and discounts, or otherwise effectively
influence consumers to choose the merchants preferred network. This is likely
why the networks often deter or prohibit merchants from influencing payment
choices. 62
Even merchants mere ability to impose surcharges on credit card transactions
can have procompetitive effects. 63
Prohibiting surcharges therefore has anticompetitive effects. 64
Although merchants ability to surcharge will not prevent networks entirely
from using interchange fees to artificially increase merchant fees, it will
constrain the amount of overcharges imposed through interchange fees. 65
The ability to surcharge can increase the number of merchants accepting cards,
pressure networks to reduce merchant fees, and induce consumers to make
more efficient payment choices. 66
61
62
63
64
65
66
Alan S. Frankel, Towards a Competitive Card Payments Marketplace, Reserve Bank of Australia and
Melbourne Business School, Payment System Review Conference, April 2008
(http://www.rba.gov.au/payments-system/resources/publications/payments-au/paymts-sys-rev-conf/2007/5compet-card-payment.pdf) (Frankel RBA Paper), p. 53.
Frankel RBA Paper, p. 52.
Id., p. 53.
Id.
Id.
Id., p. 54.
22

Incremental reform of contractual restrictions, such as elimination of the nosurcharge and no-discrimination rules, is helpful and beneficial to the public, but
may be insufficient to erode interchange fees to restore fully competitive pricing
in the marketplace, given the four-party credit card duopoly and vertically
integrated three-party card networks. 67
45.
It is true, as I acknowledged in Australia, that probably due to transaction costs
- with low enough fees, most merchants will not surcharge, but even in Australia, where
merchant fees are significantly lower than in the United States, the ability to surcharge and
actual use by some merchants of surcharges can significantly constrain merchant fees or reduce
consumer usage of the most expensive cards. 68
4.2.2. The Evidence From Australia Supports a Conclusion that Many Merchants Will
Surcharge Credit Card Transactions in the United States
46.
Professor Hausman agrees that the removal of no-surcharge rules in Australia is
a natural experiment about the effects of such reforms.69 In my initial declaration I presented
information about the trend in merchant credit card surcharging in Australia. For this purpose,
I relied on periodic surveys published by the Australian consulting firm East and Partners, the
same source cited routinely by the Reserve Bank of Australia (RBA). 70 The East and Partners
surveys do not provide comprehensive data on surcharging in Australia, but they show an
unmistakable trend towards increased adoption of surcharging by Australian merchants.
67
68
69
70
Id., p. 58.
Id., p. 35 (emphasis added). I have concluded from my research that in a fully competitive credit card industry
there would not have been interchange fees in the MasterCard and Visa networks i.e., transactions would
have been settled at par between banks as are check transactions, and merchants would probably pay fees of
less than half of one percent, on average, to accept credit cards. The Reserve Bank of Australia asked me to
present a paper addressing whether reforms to the rules and structure of credit card networks now could be
sufficient to attain a fully competitive market. It is in that context that I explained that reforms now not
retrospectively at the inception of the anticompetitive practices would be beneficial but not likely a
complete solution to the legacy of anticompetitive effects.
Hausman Report, 52.
Frankel Declaration, 37. See also, e.g., Reserve Bank of Australia, Review of Card Surcharging: A
Consultation Document, June 2011 (RBA June 2011 Consultation Document), p. 2.
23

47.
Professor Hausman discounts the importance of the growing fraction of
merchants that have adopted surcharging in Australia. He notes that in a payment panel study
of Australian households undertaken by the RBA, respondents reported that they paid
surcharges on only 5% of their credit card transactions in both 2007 and 2010. 71 He claims that
it is the volume of transactions that are surcharged which provides the important economic
factor. 72 I disagree. The percentage of merchants that actually surcharge credit cards is
clearly relevant to a claim that merchants will not surcharge credit cards. Moreover, in
Australia, the largest merchants obtained very substantial reductions in the level of their credit
card fees so that they would not surcharge credit card transactions.73 In addition, many
Australian merchants differentially surcharged i.e., applied surcharges to American Express
(and Diners Club if they accept that less commonly accepted brand) transactions but not to
MasterCard or Visa credit card transactions (and not to debit cards). Merchants did this
because of the higher merchant fees charged by American Express and Diners Club, and many
customers use non-surcharged cards at merchants that surcharge.74
48.
Professor Hausman states that I neglect to explain that under the terms of the
Agreement, the amount of any surcharge on American Express transactions will have to equal
the amount that the merchant surcharges MasterCard and Visa credit card transactions. 75 To
the contrary, I devote Part 4.1 of my initial declaration to that issue. It is Professor Hausman
who neglects to address the fact that the weighted average merchant fee for all credit card
71
72
73
74
75
Hausman Report, 72.

Id.
Merchants are also more likely to surcharge or credibly threaten to surcharge the more common surcharging
becomes generally in the retail economy.
Hausman Report, 74. American Express undertook efforts to curb differential surcharging and
Hausman Report, 74.
24

transactions in the United States exceeds the American Express fee in Australia, and the
average difference between credit card and debit card fees in the United States substantially
exceeds the difference between American Express and MasterCard/Visa fees in Australia.
Moreover, most merchants in Australia that surcharge credit card transactions appear to apply
the same surcharge to all brands. 76
49.
Professor Hausman contends that the willingness of merchants to surcharge
credit card transactions in Australia is not representative of their likely willingness to surcharge
credit card transactions in the United States because there is greater competition in most
retail sectors in the U.S. than in Australia. 77 He concedes that a limited number of
merchants will surcharge and receive an economic benefit from the settlement, but he assumes
that these will be large-size merchants, and that [m]erchants in more competitive sectors and
especially small- and medium-size merchants are unlikely to surcharge. 78
50.
I disagree that the competitiveness of merchant sectors makes it unlikely that a
significant number of merchants will surcharge. In fact, in the New Zealand litigation, Professor
Hausman agreed that there is intense competition between merchants (i.e., in New
Zealand), 79 yet he argued that the evidence from Australia was contrary to an opposing
76
This can be seen from the 75% of American Express merchants that East & Partners reported applied
surcharges in 2010 together with
77
Hausman Report, 76.

Hausman Report, 9, 89. In citing Australian data, Professor Hausman focuses on the smaller share of
transactions surcharged than merchants that surcharge, caused in part by the large merchants obtaining lower
fees not so that they would not surcharge. When predicting the effects of elimination of no-surcharge rules in
the United States, he inconsistently focuses on his prediction that a small number of merchants will surcharge,
despite his contention that these will tend to be large merchants.
Evidence in reply of Professor Jerry Hausman (redacted open version) in Commerce Commission v. Cards NZ
Limited, et al. and DSE (NZ) Limited, et al. v. Cards NZ Limited et al., 4 September 2009 (Hausman New
Zealand Brief of Evidence in Reply), 7.6.
78
79
25

experts claim that merchants would be unlikely to surcharge in New Zealand. 80 In Australia,
some of the largest merchants were able to obtain significantly lower credit card merchant fees
to keep them from surcharging credit cards.
81
In any
event, as I showed in my initial declaration, small- and medium-sized merchants, not just large
merchants, have adopted surcharging in large numbers. In New Zealand, Professor Hausman
noted evidence that in New Zealand both travel agents and computer companies surcharged
credit card transactions, even though it violated their agreements with acquirers and that
[t]hese market actions refute the claim that only large merchants, with market power, will
surcharge credit card transactions if permitted to do so. 82
51.
In his report in MDL-1720, i.e., litigation against MasterCard and Visa in the
United States, Individual Merchant Plaintiff expert Dr. Vellturo states that
. 83 He relies on the experience in Australia in
84
which [
Like me, Dr. Vellturo
relies on the level of American Express and Diners club fees in Australia and the fact that those
high fees induce surcharging to support his opinions about MasterCard and Visa surcharge rules
80
Id., 6.22
81
82
83
84
Hausman New Zealand Brief of Evidence, pp. 15-16.

Vellturo MDL-1720 Report, 201.
26

in the United States context. 85 Dr. Vellturo addressed and rejected claims by one of the experts
for the defendants in that case that few merchants would surcharge because of the fear that
they would lose customers to non-surcharging merchants i.e., the same claim made here by
Professor Hausman. 86
4.2.3. Surcharging is More Likely in the United States than in Australia or the U.K. at
Current Levels of Merchant Fees
52.
Many merchants chose to surcharge American Express card transactions in
Australia because the average American Express merchant fee was 2.51% in 2003 (even after
pre-emptive reductions in 2002, discussed below). By 2005, the average American Express
fee was still over 2.3% and surcharging accelerated. The average MasterCard and Visa
merchant fee in Australia was less than 1.0% by 2005 (and trending lower), and merchants were
less inclined to apply surcharges to those cards. In the United States, by contrast, merchant
fees for all of the major brands on average exceed 2.3%. Thus, merchants here are more likely
than in Australia to be interested in surcharging all of those brands.
53.
Professor Hausman points to the United Kingdom where he says that surcharging
is infrequent. 87 He fails to note that MasterCard and Visa interchange fees are far lower in the
United Kingdom than in the United States, and the average merchant fee for those brands is
about 1.4%. 88 Moreover, the U.K. Office of Fair Trading found that despite those lower fees,
85
86
87
88

Reply Report of Dr. Christopher A. Vellturo, In Re: Payment Card Interchange Fee and Merchant Antitrust
Litigation, June 22, 2010, 110.
Hausman Report, 81-82.
See, Frankel MDL-1720 Reply Declaration, 32.
27

14% of businesses in the U.K. applied surcharges to credit cards, and 9% applied surcharges to
charge cards.89
4.3. Surcharging Permits Merchants to Recoup the Cost of Credit Card Payments Directly
From Customers Who Use Credit Cards and Post Lower Shelf Prices
54.
In my initial declaration, I stated that a direct effect of surcharging (or, as
Professor Stiglitz puts it, pricing of credit card transactions) is that merchants that surcharge
credit card transactions recoup the cost of those transactions from customers who use credit
cards rather than from all customers generally. 90 I do not believe that this issue is in dispute.
For example, Professor Hausman has explained that [e]vidence from Australia demonstrates
that some merchants will levy surcharges for credit card use to recover the [merchant fees]
when the no surcharge rule is eliminated. 91 By contrast, without surcharging, merchants
recover [the merchant fee] by increasing their prices to all consumers, whether or not the
consumers are using credit cards. 92 This is a central theme of Professor Stiglitz.93
55.
In his discussion of business stealing, however, Professor Hausman ignores the
fact that merchants surcharging credit card transactions will tend to have lower posted prices
89
90
91
92
93
United Kingdom Office of Fair Trading, Payment Surcharges: Response to the Which? Super-Complaint, July
2012, 5.5. Charge cards consist of American Express and Diners Club, and East & Partners reports that only
10.1% of 505 surveyed merchants in the United Kingdom accepted American Express or Diners Club cards in
2009. East & Partners, Australian and UK Credit Card Surcharging Perspectives: Custom Analysis for NERA
Australia, November 2009 (HOUSTON001120), Table 17. The East & Partners survey suggests that a
significant percentage of merchants that accept American Express or Diners Club cards in the U.K. apply
surcharges to those cards, but the percentage of merchants that accept charge cards is only roughly suggested
by the East & Partners report, which found only 51 U.K. merchants that accepted American Express or Diners
Club in addition to MasterCard and Visa, 19 of which (37.3%) applied surcharges. Id., Table 20. (East &
Partners states that this reported percentage is for indicative purposes only and not likely reliable as a
specific percentage.)
Frankel Declaration, 45-46.
Hausman New Zealand Brief of Evidence, 5.2.
See, e.g., Stiglitz Declaration, 23 (with the NDP in place
).
28

than merchants that do not surcharge, thereby making additional sales and profits to
customers who do not use credit cards (i.e., business stealing in the other direction). 94
4.4. Many Customers Will React to Credit Card Surcharges by Using Lower Cost
Alternatives at Merchants that Surcharge, Reducing Costs and Prices
4.4.1. Narrow Relevant Markets Do Not Imply That Credit Card Surcharges Cannot
Shift Transaction Volume to Debit Cards
56.
American Express, MasterCard, and Visa have all enforced anti-steering rules,
including importantly no-surcharge rules, which have impeded competition between them at
the point of sale. Professor Stiglitz, Dr. Vellturo, Professor Hausman, and I all agree that most
merchants that accept MasterCard, Visa, and American Express cards find it necessary to accept
those cards despite supracompetitive levels of fees for all of those networks because of the loss
of transaction volume and thus profits which would be incurred if the merchants stopped
accepting any of those brands. Moreover, as I explained in Part 3.1, we are in agreement that
from an economic perspective (because merchants have not been able to shift transactions
even from one of those credit card networks to another), one can define separate relevant
markets for each of those brands of credit cards.
57.
The economic support for the existence of brand-specific relevant merchant
credit card markets does not mean that differential surcharges, for example, on more costly
American Express cards would not induce a significant number of consumers to choose instead
to use a lower cost brand of credit card. The objectors, Professor Stiglitz, and Professor
Hausman, in fact, all point to the continued prohibition of differential surcharges under the
terms of the Agreement as having precisely that effect. The cumulative effects of the card
94
29

networks prohibitions on price differentials (and other forms of interbrand steering) are the
main reasons why one can define brand-specific relevant markets. Professor Stiglitz and Dr.
Vellturo emphasize that the relevant market would have been broader in a but-for world in
which the anti-steering rules had not been in effect. 95
58.
Yet the National Retail Federation relies on the existence of separate relevant
credit card and debit card antitrust markets (in the presence of anti-steering rules) to claim it is
a fallacy to conclude that surcharges on credit cards will cause a relative shift of transaction
volume to debit cards. 96 The Constantine Cannon Objectors likewise cite the existence of
separate relevant markets for credit cards and debit cards to claim it is a faulty premise that
the Agreement will allow merchants to steer transactions from credit cards to debit cards. 97
59.
These inferences cannot be drawn from the existence of separate relevant credit
card and debit card markets.

60.
Like Professor Stiglitz and Dr. Vellturo, my analysis of the history of payment
systems in the United States has led me to conclude that in a but-for world in which card
payment networks had not operated with anticompetitive restraints, the relevant market
would likely be broader than it is in the actual world today. That is not to say that removing the
prohibition on credit card surcharging today will immediately create a broader relevant market.
Indeed, even if widespread credit card surcharging under the settlement were to cause great
95
For example, in MDL-1720, Professor Stiglitz explained that
96
Stiglitz MDL-1720 Report, 69.

NRF Statement of Objections, p. 8.
CC Brief, p. 31.
97
30

numbers of consumers to switch to debit, that still would not necessarily imply a broader
relevant market. 98
61.
Today, merchants use blended or bundled pricing in which customers using
high cost credit cards and low cost debit cards all pay the same retail price. Moreover, as
Professor Stiglitz and Professor Hausman note, credit card issuers use some of their merchant
fee revenue to offer rebates and other rewards (i.e., negative prices) to credit card users, so
that consumers are provided with an incentive to use the more costly payment method. If a
merchant unbundles these prices, the net price for credit card customers will increase and
the net price for debit card customers (and cash customers) will decrease. Even in the unlikely
scenario that no customer switched from using a credit card to using a debit card, basic
economic principles imply that merchants will increase their sales to debit card and cash
customers and reduce their sales to credit card customers.
62.
Of course, almost all consumers who possess a credit card also possess a debit
card and most use a debit card for at least some transactions, and, as Professor Stiglitz has
explained, the proportion of transactions for which it will be optimal for cardholders to use
their credit cards will tend to decline and the proportion for which it will be optimal to use their
98
Conventional relevant market definition methods can be confusing in this context because consumers often
face no price or a negative price (due to rewards) and surcharges are often expressed in percentage terms.
Converting to dollars simplifies the point. Suppose a $100 purchase generates a reward worth $1 to a
consumer, so that the price to use a credit card is -$1. If a merchant charges a $2 surcharge for use of a credit
card, the new price to the cardholder is +$1, a change of +200% in the price to the cardholder to use a credit
card even though that surcharge is only 2% of the price of the items purchased from the merchant. The net
price to the merchant has decreased by roughly 100% because the merchant fee is offset by the surcharge
revenue. A 5% or 10% increase in the price of a credit card transaction to a cardholder (i.e., a conventional
market definition threshold increase) would be generated by a surcharge of only 0.05% to 0.10% on the
overall purchase amount. In other words, only if very small surcharges were sufficient to cause credit card
networks to reduce their fees to the level of debit card fees would the two likely be in the same relevant
market from the perspective of merchants.
31

debit cards will tend to increase. This, too, is basic economics. This substitution in the face of
surcharges does not have to be so great as to imply a single relevant antitrust market consisting
of credit card and debit card network services together in order for surcharging to benefit
merchants. With a credit card surcharge, each transaction that migrates to a debit card saves
the merchant costs, and each transaction made using a credit card generates additional
revenue to offset at least some of the extra cost associated with credit cards.
4.4.2. Professors Stiglitz and Hausman Have Agreed that Credit Card Surcharges Shift
Some Credit Card Transactions to Debit Cards
63.
Professor Stiglitz and Professor Hausman emphasize how differential surcharges
on more costly credit cards (such as American Express cards in Australia) will tend to cause a
shift in use towards lower cost credit cards. I agree. But by the same token, surcharges on all
credit cards will cause some customers to switch to debit cards for some transactions. In fact,
Professor Stiglitz and Professor Hausman have previously agreed with me that credit card
surcharges drive transaction volume to non-surcharged cards, including debit cards. In his
MDL-1720 reports concerning MasterCard and Visa anti-steering rules, Professor Stiglitz
repeatedly explained that merchant restraints (including no-surcharge rules) prevented an
efficient migration of some credit card transactions to debit cards.
99
99
32

100
101
102
103
104
100
101
102
103
104

Stiglitz MDL-1720 Report, 81 (citing Reserve Bank of Australia, Reform of Australias Payments System:
Preliminary Conclusions of the 2007/08 Review 1-2 (Apr. 2008).
Stiglitz MDL-1720 Report, 82 (citing Neelie Kroes, Eur. Commr for Competition Policy, Speech at the
European Retail Round Table Conference: Europes Payment Systems after the MasterCard Decision (Jan 14,
2008)).
Reply Report of Joseph Stiglitz, Ph.D., In Re: Payment Card Interchange Fee and Merchant Antitrust Litigation,
June 22, 2010, p. 63.
33

64.
In his report in the American Express litigation, Professor Stiglitz writes:
105
106
65.
In New Zealand, like here, Professor Hausman defined separate credit card and
debit card markets. 107 Yet he also explained that credit card surcharges in various possible
scenarios would cause some credit card transaction volume to migrate to debit cards (called
EFTPOS cards in New Zealand).
The least cost acceptance vehicle for many merchants is EFTPOS. The use of EFTPOS
will increase if surcharges are levied on credit card transactions. Also, I would expect
the usage of credit cards with lower [merchant fees] to increase because surcharges on
those cards would be lower than the surcharges on cards with higher [merchant
fees]. 108
[E]conomic analysis would typically find that merchants with market power would
increase prices above the competitive level to all customers, not only credit card
customers who could easily shift to a competing payment form, e.g., EFTPOS. 109
I consider the likely outcome if the [honor all cards] rule remains but the no surcharge
and no discrimination rules are eliminated along with the [multilateral interchange fee].
I expect an increase in competition compared to the current situation because
merchants will levy (higher) surcharges on credit cards with high [merchant fees] so long
as merchant acquirers unbundle their [merchant fees] Merchants will also attempt to
steer consumers to the use of lower cost payment options such as EFTPOS or credit
cards with lower [merchant fees]. 110
105
106
107
108
109
110
Stiglitz Report, p. 19.

Stiglitz Report, 29.
Hausman New Zealand Brief of Evidence, 4.4-4.6, 4.12, 4.14.
Hausman New Zealand Brief of Evidence, 5.2 (p. 17).
Hausman New Zealand Brief of Evidence, pp. 36-37.
34

[I] have discussed how the agreement among the banks to enforce the Visa and
MasterCard rules leads to anti-competitive restrictions, e.g. the no surcharge rule, on
possible merchant strategies to cause consumers to use lower cost payment options,
e.g. EFTPOS cards. These Visa and MasterCard rules lead to higher effective MSFs paid
by merchants... I find that in the counterfactuals if the agreement to enforce some or all
of the rules is eliminated, the expected outcome is a reduction in the MSFs paid by
merchants in the acquiring market, which is a pro-competitive outcome. 111
With lower [interchange fees], more consumers would instead use EFTPOS or cash so
merchants costs would be lower. 112
In his further discussion of surcharging as free-riding Prof. Wright neglects to take
into account that many consumers carry more than one credit card which they can
switch to or use their EFTPOS card. 113
Prof. Wright states that without the no surcharge rule and HAC rule interchange fees
might become negative This outcome has not been observed in countries where the
no surcharge rule has been eliminated. Further, the cost of using credit cards would
increase, especially for transactors, i.e. credit card users who do not have a credit
balance. Many of these credit card users would switch to EFTPOS. 114
If merchants have the ability to steer or surcharge individual issuers, they will establish
steering policy or set surcharges to offset an attempt by an issuer to charge higher
interchange. As elsewhere in his brief, Prof. Wright fails to take into account that many
people carry more than one credit card and almost all credit card users also have an
EFTPOS card. Customers who do not want to pay the surcharge will switch to another
credit card or use their EFTPOS card. 115
Merchants would establish steering policies and/or set a high surcharge fee on the
given issuers' transactions which would cause credit card users to switch to another
credit card or to use EFTPOS. 116
66.
In short, these experts and I agree both that there are narrow relevant markets
in the actual world in which surcharging has been prohibited, and that when merchants are
able to surcharge credit cards, at least some credit card transaction volume will migrate to
debit cards, benefitting merchants. Posted (debit card and cash) prices will tend to be lower
111
112
113
114
115
116

Hausman New Zealand Brief of Evidence in Reply, 4.3.
35

and credit card prices (with the surcharge) higher, but the relative shift of some transaction
volume from credit to debit will mean lower average costs overall (and thus greater sales), even
apart from any reduction in the level of credit card merchant fee rates.
4.4.3. A Recent One-Year Change in Credit and Debit Card Volume in Australia Does
Not Disprove The Usefulness of Surcharges at Steering to Debit
67.
Professor Hausman claims here that surcharging of credit cards has not led to a
significant shift to debit cards in Australia. 117 His view is not shared by the Reserve Bank of
Australia or by the Australia Payment Clearing Association (APCA), as I cited in my initial
declaration.118 Professor Hausman discounts my citation to these observers as selective
quotes but he presents no contradictory views by other Australian observers. 119 He suggests
that an empirical test of whether credit card surcharging has led to a significant shift to debit
card usage is a comparison of the growth rate of the dollar amount of credit-card usage with
that of debit-card usage in Australia over the same period. 120 He reports that credit card
dollar charge volume increased by 10.2% between December 2012 and December 2013, and
that growth rate was the highest in the past six years. Further, he reports that the growth rate
of credit card charge volume exceeded the increase in debit card charge volume during the
same twelve month period. 121
117
118
119
120
121
Hausman Report, 78.

Frankel Declaration, 56. The RBA is the countrys central bank. The APCA describes itself as the selfregulatory body for Australias payments industry. APCA has 90 members including Australias leading
financial institutions, major retailers and other principal payments service providers.
http://www.apca.com.au/about-apca
Hausman Report, 78.
Id.
Id.
36

68.
I am familiar with the data relied on by Professor Hausman for this analysis,
which are published by the RBA. I have extended his analysis in two ways. First, rather than
focus on a single 12-month period ending in December 2013 a decade after the elimination of
no-surcharge rules in Australia I have replicated his computation for each period ending in
December of each of the available previous years as well. I show the result in Figure 1. Second,
instead of focusing only on year-over-year changes for each December, I have computed the
aggregate dollar charge volume on debit cards in Australia as a percentage of the charge
volume on credit cards in each month since both series have been available beginning in August
2002. I show the result of this analysis in Figure 2.
Figure 1
12-Month Growth Rate: Credit vs. Debit Volume in Australia
25.0%
21.4%
20.0%
Credit
Debit
10.14%
16.2%
15.4%
15.1%
10.15%
15.0%
13.3%
13.2%
11.5%
11.3%11.4%
10.7%
9.7%
10.0%
9.5%
9.5%
7.0%
6.0%
5.6%
4.4%
5.0%
3.4%
0.2%
0.0%
-0.9%
-5.0%
2003
2004
2005
2006
2007
2008
2009
12 Months Ending December:
2010
Sources: http://www.rba.gov.au/statistics/tables/xls/c01hist.xls?accessed=2014-06-23-15-38-54 and

http://www.rba.gov.au/statistics/tables/xls/c05hist.xls?accessed=2014-06-23-15-38-54.
37
2011
2012
2013

Figure 2
Ratio of Debit to Credit Card Charge Volume in Australia 2002-2014
90.0%
80.0%
70.0%
60.0%
50.0%
40.0%
Professor
Hausman's
Comparison
30.0%
20.0%
10.0%
69.
Apr-2014
Dec-2013
Apr-2013
Aug-2013
Dec-2012
Apr-2012
Aug-2012
Dec-2011
Apr-2011
Aug-2011
Dec-2010
Apr-2010
Aug-2010
Dec-2009
Apr-2009
Aug-2009
Dec-2008
Apr-2008
Aug-2008
Dec-2007
Apr-2007
Aug-2007
Dec-2006
Apr-2006
Aug-2006
Dec-2005
Apr-2005
Aug-2005
Dec-2004
Apr-2004
Aug-2004
Dec-2003
Apr-2003
Aug-2003
Dec-2002
Aug-2002
0.0%
Figures 1 and 2 paint a picture quite different from that suggested by Professor
Hausmans cherry-picking of December 2012 and December 2013 as his dates for comparing
the growth of debit card and credit card usage. It is true, as Professor Hausman reports, that
credit card usage grew by 10.2% in the year ending December 2013, and that this exceeded the
growth rate of debit card usage during that same period. Closer examination, however, shows
that the difference in these growth rates was only 0.018% during 2013, and that this 12-month
period was aberrational. As Professor Hausman notes, credit card usage grew at the fastest
rate in six years in 2013. Figure 1 shows that, aside from the one-fiftieth of one percent faster
credit card growth than debit card growth in 2013, in every other year ending each previous
December since 2003 (the year credit card surcharging began), debit card growth exceeded
credit card growth in Australia. Figure 2 shows the cumulative effect of this differential
38

growth. 122 Aggregate debit card spending was about 40-45% of credit card spending
throughout most of the period August 2002-2005, but since then this ratio has increased
sharply. The very slightly faster growth rate of credit card usage in the brief, recent period
examined by Professor Hausman saw this ratio decline by 0.01%, but that left the ratio at 76.3%
far above the level prevailing in 2003. Moreover, there is no sign that the upward trend has
ceased.
4.5. The Threat to Networks of Lost Transaction Volume From Credit Card Surcharges Will
Generate a Previously Suppressed Competitive Constraint on the Level of Credit Card
Merchant Fees
4.5.1. Networks Lose More Transactions as Merchant Fees Increase if Merchants Can
Surcharge Than If Merchants Cannot Surcharge
70.
As I have already explained, at the current level of credit card fees merchants
benefit from surcharging credit cards even without consideration of any change to the level of
merchant fees resulting from surcharging. Merchants that surcharge recoup credit card costs
specifically from customers who use credit cards and so can charge lower prices to cash and
debit card customers, which increases sales to those customers. Moreover, some consumers
will alter their payment choices for at least some of their transactions. I showed in my initial
declaration (and neither Professor Stiglitz nor Professor Hausman has disputed) that
transactions migrating from credit cards to debit cards will save merchants approximately
1.57% of the transaction amount. Cost reductions also lead to price reductions, leading to
increased overall sales.
122
Figure 2 uses the same sources as Figure 1 and the same as those used by Professor Hausman. Gaps in the
trend line in Figure 2 indicate months in which the RBA reports series breaks due to changes in reporting.
39

71.
Credit card networks have long imposed and fought to preserve no-surcharge
rules for obvious reasons. If card networks are more costly to merchants than the alternatives,
and merchants can price the more costly networks at a higher level, then customers will have
an increased economic incentive to choose an alternative, non-surcharged payment method
than if merchants cannot surcharge the more costly cards. This increases the elasticity of
demand faced by a card network with respect to merchant fees, which as a result will have an
economic incentive to set lower merchant fees than if merchants cannot surcharge. Nosurcharge rules, on the other hand, leave merchants with an all-or-nothing choice whether to
accept credit card brands, reduces the elasticity of demand faced by each network, and results
in higher merchant fees.
72.
I first described this analysis of the effects of no-surcharge rules on the level of
merchant fees in the mid-1990s. 123 Since then, many others have reached the same conclusion.
Professor Stiglitz agrees that anti-steering rules (including no-surcharge rules) reduce the
elasticity of demand facing card networks and results in higher merchant fees. 124 He agrees
that when merchants surcharge, the
125
73.
In New Zealand, Professor Hausman agreed that no-surcharge rules lead to
higher merchant fees:

I have discussed how the agreement among the banks to enforce the Visa
and MasterCard rules leads to anti-competitive restrictions, e.g. the no
surcharge rule, on possible merchant strategies to cause consumers to
123
124
125
See, e.g., Alan S. Frankel, Monopoly and Competition in the Supply and Exchange of Money, 66 Antitrust Law
Journal 313 (1998), at pp. 343-47.
Stiglitz Report, 34; Stiglitz MDL-1720 Report, 10(I), 41.
40

use lower cost payment options, e.g. EFTPOS cards. These Visa and
MasterCard rules lead to higher effective MSFs [merchant fees] paid by
merchants. The MSFs are a variable cost for merchants and are passed on
in higher prices to all consumers, whether they use credit cards or
another form of payment. I find that in the counterfactuals if the
agreement to enforce some or all of the rules is eliminated, the expected
outcome is a reduction in the MSFs paid by merchants in the acquiring
market, which is a procompetitive outcome. 126
74.
The elasticity of demand with respect to merchant fees would be enhanced the
most, putting the most downward pressure on credit card merchant fees, if merchants had
complete flexibility to accept or reject, encourage or discourage, and surcharge or discount
whichever cards they wished. But the effect even of uniform credit card surcharges is the same
directionally, if not in magnitude.
4.5.2. Surcharging Intensifies Constraints on Merchant Fees Relative to the Status
Quo Even if Uniform Across Brands
75.
If merchants could differentially surcharge in the United States, this would tend
to generate a relatively greater competitive constraint on American Express than on

MasterCard or Visa (or Discover Card) because of American Expresss relatively higher merchant
fee. But the use of differential surcharges would unlikely be as widespread as in Australia.
While there is variation across merchants, in the United States I estimate that the average
American Express transaction migrating to MasterCard or Visa today would save a merchant
126
41

approximately 17 basis points.127 In Australia, in 2004-05, the difference was approximately

140 basis points. 128
76.
When merchant fees for different brands of credit cards are relatively close to
one another, fewer merchants will differentially surcharge even if they have the contractual
ability to do so. Even with merchants that differentially surcharge in Australia, it is common to
see merchants charge the same surcharge for American Express and Diners Club transactions,
despite likely differences in the cost of each brand. It is also common for merchants in Australia
to parity surcharge all credit cards at the same level. As the National Retail Association of
Australia has observed, in filings with the RBA:
For retailers who elect to surcharge credit card purchases, simplicity and
efficiency are primary considerations in setting such fees. . . . Surcharges
must be easy for consumers to identify and calculate and easy for retail
sales staff to administer. Often a single surcharge for all credit cards will
be preferred by retailers due to its simplicity of operation and the ability
of the retailer to look at their total costs related to such transactions in
establishing an appropriate fee.129
77.
Even if differential surcharging in the United States hypothetically eliminated
differences between merchant fees for different credit card brands, surcharging would
continue to exert a constraint on network merchant fees, even if merchants thereafter used
only uniform surcharges. The credit card network industry is highly concentrated, and each
network would bear a significant share of the lost transaction volume resulting from merchant
127
Frankel Declaration, Figure 3.
128
http://www.rba.gov.au/statistics/tables/xls/c03hist.xls
National Retail Association Submission to Reserve Bank of Australia: Review of Card Surcharging, July 25, 2011.
http://www.rba.gov.au/payments-system/reforms/submissions-card-surcharging/nra-card-surcharging.pdf
129
42

surcharging.130 The competitive pressure on a network increasing its merchant fees would not
be as great as would exist if some merchants singled out that network for differential
surcharging, but there will be a more effective competitive constraint than exists in the current
marketplace in which networks face no effective constraint when they raise fees and
consumers are provided no incentive to avoid using credit cards in response to those higher
fees.
4.5.3. Professor Hausman is Incorrect in His Claim that Surcharging Has Not Caused a
Reduction in American Expresss Merchant Fees in Australia
78.
Professor Hausman never identifies any anticompetitive conduct that he believes
American Express has engaged in or any remedies that would erode its significant market
power. He criticizes the Agreement for leaving intact American Expresss no-differential
surcharge policy with respect to alternative credit cards, but that is puzzling, because Professor
Hausmans analysis (incorrectly) suggests that that American Express policy has no significant
anticompetitive effects. He contends that even in Australia, where American Expresss
merchant fee was 140 basis points above those of MasterCard and Visa, differential surcharging
had no effect on American Expresss merchant fees. But if the ability to differentially surcharge
had no effect at reducing even American Expresss fees when they were 140 basis points above
those of other credit card networks, that ability cannot logically have a significant competitive
effect in the United States where the major card networks fees are much closer and he
contends merchants are less likely to surcharge.
130
That is not to say that some merchants will not finely distinguish different surcharges for different brands or
that some major merchants would not negotiate favored no-surcharge agreements with particular brands in
exchange for lower fees if differential surcharging were an available option. Differential surcharging would
continue to be a valuable competitive merchant tool even if all brands merchant fees started out the same.
43

79.
Professor Hausman goes further, suggesting that the relevant measure of
American Expresss merchant fees has actually increased in Australia despite the ability to
surcharge and differentially surcharge. But American Expresss average merchant fee has
decreased by 78 basis points since March 2003, compared to 61 basis points for the
interchange-regulated MasterCard and Visa.
31
80.
In other words, the effective American Express merchant fee in Australia has
declined by between 83-113 basis points, compared to a decline of 61 basis points for
MasterCard and Visa, notwithstanding the lack of direct regulatory intervention with respect to
American Expresss rates.
131
Vellturo Report, 204, citing Malhotra Deposition Exhibit 352.
44

, and there is no need for me to repeat all of that evidence here. 132
81.
Professor Hausmans claim that surcharging had no effect on American Expresss
rates, and in fact that AMEX has grown relatively more expensive than VMC since Australia
began allowing surcharging, is based on his computation of the ratio of American Expresss
average merchant fee to that of MasterCard and Visa. 133 He contends that is the economically
relevant measure of price for purposes of evaluating the effects of surcharging. But it is not.
82.
Professor Hausman asserts that the decreases in AMEXs fees have been due to
the lower regulated interchange rates of VMC and merchants ability to refuse to accept
AMEX. 134 But his use of the declining level of American Expresss fees to evaluate the claimed
effectiveness of merchants ability to refuse to accept American Express cards and the ratio to
MasterCard and Visa fees to evaluate the effectiveness of surcharging is arbitrary.
83.
I agree that reductions in MasterCard and Visa merchant fees contributed to the
decline in American Express merchant fees, but so did the elimination of no-surcharge rules.
132
133
134
Vellturo Report, 205-228, 235-249.

Id. (emphasis added).
45

135
84.
If American Expresss no-surcharge policy had been left in place in 2003, it would
have faced less pressure to reduce its merchant fees. Professor Hausman, however, suggests
that the RBA agrees with his conclusion that surcharging has played no role at constraining
merchant fees. 136 In 2009, the then-Head of Payment Policy at the RBA agreed with me.
I think there are two things driving this [continued reduction in American
Express merchant fees] (1) the fact that the spread widened after
interchange fees fell meant that Amex had to lower its fees over time to
remain competitive and to increase acceptance and (2) surcharging has
hit Amex much harder that [sic] Visa and MasterCard which has in some
cases resulting [sic] in it having to lower its [merchant service fees] to
avoid surcharging.137
85.
In its 2011 review of card surcharging, the RBA affirmed its view that no-
surcharge rules limited the ability of merchants to put downward pressure on their merchant
service fees and interchange fees by threatening to charge the customer for using a credit or
scheme debit card. 138
86.
Professor Hausman cites data concerning the (minor) growth in the combined
market share of American Express and Diners Club (relative to MasterCard and Visa) based on
transaction dollar volume as evidence that surcharging has not constrained AMEX and Diners
merchant fees. 139 That is a non sequitur. In fact, despite the issuance of companion
135
136
137
138
139
Hausman Report, 68.

Rebuttal Report of Alan S. Frankel, Ph.D., In Re Payment Card Interchange Fee and Merchant Discount
Antitrust Litigation, June 22, 2010, citing email from Michelle Bullock to Alan Frankel, May 31, 2009.
RBA June 2011 Consultation Document, p. 2.
Hausman Report, 71. At least part of that reported increase results from a change in the RBAs reporting
methodology, as Visa debit transactions are now being excluded from that networks reported market share.
http://www.rba.gov.au/statistics/tables/exp-note/c2-note-120510.html
46

American Express cards by Australian banks and despite American Expresss much higher
merchant fees and the ability to use those fees to continue funding reward programs, the
death spiral that MasterCard and Visa warned about in Australia did not materialize.
Professor Hausman contends that a slight uptick in the combined share of American Express
and Diners Club shows that surcharges have failed to constrain those networks. In fact, it
shows the opposite. Banks that issue MasterCard and Visa credit cards lost about 45 basis
points of merchant (interchange) fee revenue due to RBA regulation. The major banks began
issuing companion American Express cards and encouraged the use of those cards with
enhanced rewards for use of American Express instead of the MasterCard or Visa credit card
linked to the same account. Yet American Express has lost over 83-113 basis points of
merchant fee revenue and has been unable to induce a significant migration to usage of its
cards.
4.6. The Agreement Benefits Merchants With Respect to All Current Credit Card Usage, Not
Only American Express Card Usage
87.
In my initial declaration, I estimated roughly the benefits to merchants for each
one percent of charge volume that becomes subject to credit card surcharges in the United
States. The Target Objectors state that my computation should be disregarded because I do
not separate the value from surcharging American Express from the value of surcharging
MasterCard and Visa. 140 But I have explained in my declarations in MDL-1720, 141 and again in
this case, 142 that American Expresss NDP prevented most merchants from applying surcharges
140
141
142
Omnibus Objections to the American Express Class Action Settlement of Absent Putative Rule 23(B)(2) Class
Members Target Corporation, et al., June 6, 2014, p. 26.
Frankel MDL-1720 Declaration, Part 4.6.2 and Frankel MDL-1720 Reply Declaration, Part 4.4.2.
Frankel Declaration, Part 3.6 and Part 2 above.
47

to MasterCard and Visa credit card transactions permitted under the relief obtained in MDL1720. As I explained in MDL-1720, American Expresss policy significantly reduced the value
that merchants otherwise would have obtained from the relief in that case. The relief proposed
in the Agreement will free merchants now to surcharge all credit card brands. As an economic
matter, the benefits to merchants from the settlement thus include the benefits not only from
the ability to surcharge American Express card transactions, but also the ability to surcharge all
brands by merchants that accept American Express cards.
5. STATE RESTRICTIONS
88.
In my initial declaration I noted arguments that state statutes restricting credit
card surcharges eliminate any value from elimination of network no-surcharge rules. 143
Professor Hausman repeats this argument and computes the percentage of U.S. commerce in
ten states he says prohibit surcharging.144 He notes as I did that a federal court has determined
that New Yorks statute is unconstitutional, and he states how his computed percentage of
commerce in no-surcharge states would change if that courts judgment is reversed, but he
does not compute how it will change if instead other courts decide similarly to that in the New
York case. In addition to legal challenges, I also explained that state restrictions may be
subjected to competitive challenges as merchants that can choose their location may consider
surcharge rights in addition to other factors.
143
144
Frankel Declaration, 4.2.

Hausman Report, 52. Professor Hausman notes that some merchants have significantly more than 35% of
their business in the states which prohibit surcharging. Hausman Report, 53. If any state prohibits
surcharging that will be true, because there are single-location merchants in every state. But that also means
that there are many merchants with none of their business in no-surcharge states.
48

89.
There is no dispute in this case that the relevant geographic market is the United
States, and I noted that Visa and MasterCard, for example, do not set separate interchange
fees in separate states so that competitive pressure resulting from surcharging will have a
nationwide effect.145 Professor Hausman criticizes this on the grounds that how Visa and
MasterCard price says nothing about the effect of the settlements limited surcharging on
AMEXs merchant fees. 146 But as I have explained, the settlement frees merchants that accept
American Express to apply surcharges also to MasterCard and Visa credit card transactions.147
6. CONCLUSION
90.
One of the factors that has made surcharging a relatively effective strategy for
merchants in Australia was the endorsement of surcharging as a legitimate and pro-consumer

business strategy by regulators and merchants alike. In the United States, the National Retail
Federation has stated that surcharging runs 180 degrees from the goal of litigation to reduce
credit card merchant fees and prices paid by consumers. 148 That is exactly wrong, and it is
unfortunate. The ability to surcharge is a tool that merchants can use to reduce the very
substantial market power exercised by card networks in the United States. Demonizing
surcharging as anti-consumer and ineffective will not help merchants to use surcharging
successfully to generate its potential procompetitive effects.
91.
Professor Stiglitz, Dr. Vellturo, and I agree about the fundamental economic
issues and forces that lead me to conclude that the ability to surcharge, even if restricted to
145
146
147

Hausman Report, 57.
Professor Hausman states that AMEX does not have nationwide merchant fee tables as do Visa and
MasterCard. Id.
.
148
https://nrf.com/media/press-releases/nrf-says-merchants-unlikely-surcharge-credit-card-use
49

HIGHLY CONFIDENTIAL -SUBJECT TO PROTECTIVE ORDER
uniform surcharges, will have beneficial effects for merchants and their customers. I agree that
more complete reforms of the card networks would be even more beneficial, but that is not an
option currently available, and I am not able to opine on the likelihood that rejection ofthis
Agreement would be followed by more complete, market-wide relief. Notwithstanding
Professor Hausman's criticisms of my analysis, he and I also agree on most ofthe fundamental
economic issues, including the substantial market power exercised by all of the major U.S.
credit card networks. Where I differ from both him and Professor Stiglitz is my conclusion that
the ability to apply uniform credit card surcharges is beneficial (even though it is easy to
imagine even more beneficial market arrangements). Professors Stiglitz and Hausman, on the
other hand, take an extreme position -that the ability to surcharge all credit card transaction
has no value to U.S. merchants. I do not believe that position is supported by economic
analysis or the evidence. Moreover, Professor Hausman does not even offer a coherent
prescription for relief that American Express on its own could provide that would significantly
reduce its market power.
July 4, 2014
so
HOW THE PROPOSED PAYMENTS LEGISLATION WILL RESTRAIN

COMPETITION AMONG PAYMENT CARD SCHEMES AND HARM
CONSUMERS IN THE EUROPEAN UNION
!
David S. Evans*
1 October 2014
The European Commissions proposed payments card legislation and the common position reached by the
European Parliament in April 2014 will harm competition, innovation, and consumers if broadly endorsed by the
European Council in the coming months. The interchange fee price caps will soften competition between MasterCard
and Visa, the global four-party bankcard systems, and disadvantage domestic card systems. The limits on what, in effect,
merchants pay for cards will shift billions of euros of costs to European consumers. The infirmity of the legislation is
particularly apparent from its treatment of the three-party card schemes. These three-party schemes, which have small
shares of payment cards in European countries, provide an important source of competition. The proposed legislation
impairs the ability of these smaller three-party systems to compete by permitting merchants to surcharge cards from the
smaller three-party systems but not the larger four-party ones, and by potentially prohibiting three-party systems from
only entering into select partnerships. They may also face arbitrary price caps. These anti-competitive restrictions on
small rivals, advanced in the name of competition, demonstrate the lack of serious analysis behind the proposed
legislation. For European consumers the proposed payments legislation will lead to a hefty price tag, diminished choice,
and depressed innovation.
__________________________________________________________
Chairman, Global Economics Group; Executive Director, Jevons Institute for Competition Law and Economics and
Visiting Professor, Faculty of Laws, University College London; and Lecturer, University of Chicago Law School. I
would like to thank Steven Joyce and Alexis Pirchio for excellent research help and American Express for financial
support. The views in this paper are my own and do not necessarily reflect the view of any of the people or institutions
mentioned above.
*
Electronic copy available at: http://ssrn.com/abstract=2503696
I.
INTRODUCTION AND SUMMARY
The European Parliament, in early April 2014, endorsed and further extended draft
legislation,1 originally proposed by the European Commission, which will impose sweeping
regulation on payment card businesses.2 The backers of the legislation claim that it will nurture
competition, innovation, and consumer choice in the European Union.3 In fact, if adopted in the
current form, it will reduce competition among payment systems in the EU, impede the entry of
new schemes, weaken innovation, and decrease consumer choice. European consumers will end up
paying billions of euros more in fees. The legislation will squelch virtually all challengers to
MasterCard and Visa.
One doesnt have to speculate about these effects. There are already dead and wounded
victims in plain sight. The European Commissions recent policies have eliminated the most serious
emerging pan-European challenger to the global card networks. A group of 24 banks drawn from
across major countries in Europe tried to start a competing pan-European card system a few years
ago. After being rebuffed by an intransigent Commission, set on shifting the cost of payments from
merchants to consumers, the Monnet Project folded in April 2012. Several other attempts are all but
shuttered. European consumers have already lost competition, choice, and innovation as a result.
The cornerstone of the draft legislation involves caps on the multilateral interchange fees
(MIF) that banks that service merchants pay to banks that service consumers when consumers use
their cards to pay at merchants. These caps apply to banks that are members of the four-party bankcard networks.4 The fee caps will reduce the revenue that cardholders banks receive from
European Parliament, Amendments adopted by the European Parliament on 3 April 2014 on the proposal for a
directive of the European Parliament and of the Council on payment services in the internal market and amending
2 The Commissions initial proposals are currently being considered by the European Council, before Trialogue
discussions commence at the end of 2014, or beginning of 2015, with a view to finalizing the legislation for adoption by
the European Council and European Parliament.
3 European Commission, Proposal for a Regulation of the European Parliament and of the Council on Interchange
Fees for Card-Based Payment Transactions. COM(2013) 550 final. July 24, 2013. Available at:
http://www.ipex.eu/IPEXL-WEB/dossier/document/COM20130550.do; European Commission, Proposal for a
Directive of the European Parliament and of the Council on Payment Services in the Internal Market and Amending
Directive 2002/65/EC, 2013/36/EU and 2009/110/EC and repealing Directive 2007/64/EC. COM(2013) 547 final.
July 24, 2013. Available at: http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2013:0547:FIN:EN:PDF
4 The three in three-party systems refers to the cardholder, the card company, and the merchant. The four in fourparty systems refers to the cardholder; the cardholders bank; the merchants bank, and the merchant. Of course, that
list doesnt include the network operator, which would make five, but since this is the normal nomenclature I will use it.
The network operator for the four-party system and the card company for the three-party system serve very different
roles. The card company for the three-party system has direct relationships with cardholders and merchants. The
network operator for the four-party system does not have direct relationships with cardholders and merchants.
1
merchants banks by as much as 84 percent for debit and 73 percent for credit in some European
countries. That is an experiment that several countries around the world have already performed
with widely reported disastrous results for cardholders. When the merchant-side pays less the
consumer-side pays more. Consumer fees go up, or services go down, and by far more than
consumers may ever see back in lower prices from merchants.
It is even worse than that. By making every four-party bank network, in every country in
Europe, have exactly the same MIF for every transaction regardless of the amount or type of
merchant or any other factor, the legislation limits the ability of these four-party networks to
compete through price and product differentiation. Perversely, given the claimed purpose of the
legislation, this approach will soften competition between MasterCard and Visa. That will exacerbate
the harm to European consumers through less choice, higher prices, and less innovation.
The defects of the legislation are most apparent in the treatment of the smaller card systems
that operate primarily as standalone companies and do not involve large networks of banks. These
companies, which are called three-party systems, account for less than 5 percent of debit, credit
and charge card volume in the European Union. Their presence is known to be modestless than
10 percent in virtually all EU Member States. Yet the legislation sweeps them into regulations that
were originally motivated by competition concerns about the large four-party bank networks.
According to the legislation, if one of these standalone card companies decides to
collaborate with even one bank to issue or acquire cards, the company may have to make its card
brand available for all banks, in every country in the EU, to issue and acquire as well.5 That
requirement may lead these three-party systems to withdraw from a number of the smaller European
markets where they have entered and extended their reach and coverage through perfectly legitimate
individually and confidentially negotiated vertical agreements with a bank or payment institution
partner. It may therefore perversely reverse the competitive entry that has taken place over the last
two decadesentry that was enabled by a competition law-based intervention brought by the
European Commission against Visa in the mid-1990s. It also deters three-party systems, such as
Cetelem in France, from considering a business model that involves partnerships and thereby
arbitrarily limits the ability of these domestic three-party systems from expanding beyond their
borders.
The legislation proposed by the European Parliament6 also prohibits merchants from
5
Article 29(1) of the proposed revised Payment Service Directive purports to extend the open access obligation for
four-systems to three-party systems, requiring them to establish criteria for participation in the system by unrelated
institutions which are objective, non-discriminatory and proportionate and do not inhibit access more than is necessary
to safeguard against specified risks. The Commissions proposal on this issue has been endorsed by the European
Parliament.
6 See Article 55, paragraphs 3-4 European Commission, Proposal for a Directive of the European Parliament and of
imposing added fees (surcharges) on consumers who use cards from the four-party bank networks
that account the preponderance of card use. It then, in a peculiar twist, specifically permits
merchants to impose added fees on consumers who use cards issued by the smaller companies that
compete with these large bank networks. This makes no sense at all. The ostensible rationale for the
legislation is to provide tools to merchants in circumstances where they have to accept the cards
issued by the must have four-party bank networks. That reasoning doesnt extend to smaller
systems whose cards are not must have but, in fact, are dont have at many merchants.
The legislation proposed by the European Commission and European Parliament provides
for extending price regulation to the smaller three-party systems. The price caps would appear to
apply whenever these standalone card companies enter into an individually-negotiated vertical
agreement with an arms length partner. In that case the proposals suggest that the three-party
systems should be treated as if they are a four party scheme.7 It is unclear what these provisions
mean in practice since these systems do not have a MIF that could be subject to a cap.
The proposed regulations on smaller players are inconsistent with sound competition policy,
which imposes special obligations only on firms that are dominant in a market, and demands open
access only in the extreme case of essential facilities such as telecom monopolies. Indeed, one sees
how absurd the proposed legislation is from the effect of the combination of the proposed
regulations on the smaller card companies. Several of the regulations make it harder for these
companies to compete against the likes of MasterCard and Visa, and risk undermining the
competitive entry that the European Commission has previously sought to enable.8
The proposed payments legislation left behind by the outgoing European Commission and
European Parliament is anti-consumer and anti-competition. The European Council should not
the Council on Payment Services in the Internal Market and Amending Directive 2002/65/EC, 2013/36/EU and
2009/110/EC and repealing Directive 2007/64/EC. COM(2013) 547 final. July 24, 2013. Available at: http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2013:0547:FIN:EN:PDF. See also Amendments 111 and 112,
European Parliament, Amendments adopted by the European Parliament on 3 April 2014 on the proposal for a
directive of the European Parliament and of the Council on payment services in the internal market and amending
Directives 2002/65/EC, 2013/36/EU and 2009/110/EC and repealing Directive 2007/64/EC (COM(2013)0547 C70230/2013
2013/0264(COD)).
April
3,
2014.
Available
at:
http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//TEXT+TA+P7-TA-20140280+0+DOC+XML+V0//EN.
7 See Article 1, paragraph 3 (c), Article 2 (15) and Articles 3-5, European Commission, Proposal for a Regulation of the
European Parliament and of the Council on Interchange Fees for Card-Based Payment Transactions. COM(2013) 550
final. July 24, 2013. Available at: http://www.ipex.eu/IPEXL-WEB/dossier/document/COM20130550.do. See also
Amendment 21, Amendment 28, and Amendments 29-34, European Parliament, Amendments adopted by the
European Parliament on 3 April 2014 on the proposal for a regulation of the European Parliament and of the Council
on interchange fees for card-based payment transactions (COM(2013)0550 C7-0241/2013 2013/0265(COD)).
April 3, 2014. Available at: http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//TEXT+TA+P7-TA2014-0279+0+DOC+XML+V0//EN.
8 See discussion below and footnote 29.
approve it in its current form.
II. THE COMPETITIVE LANDSCAPE FOR PAYMENT CARDS IN THE

EUROPEAN UNION
Networks of banks issue most debit, credit, and charge cards in the European Union. They
are part of four-party systems in which one bank handles card payments for merchants, another
bank handles card payments for cardholders, and the network company that operates the system has
no direct relationship with either the merchants or the cardholders, but takes care of authorizing and
clearing transactions between these banks, who are paid a common fee, the MIF, set by the network.
In some countries such as the United Kingdom, most banks belong to networks operated by one of
the global card networksMasterCard and Visa.9 In other countries, many banks belong to
independent domestic networkssuch as ServiRed in Spain and Cartes Bancaires in France. The
domestic networks are often affiliated with MasterCard and Visa so their cardholders can use their
cards in other countries using the global networks. Four-party systems almost always have a MIF
that is adhered to by default and which the bank that issues the card receives from the bank that
services the merchant.
Three-party systems also issue debit, credit, and charge cards in the EU. They typically sign
up merchants directly and take care of reimbursing them for payments made on cards they issue;
they also sign up and service consumers directly. Because they are single integrated enterprises they
do not have interchange fees. In some countries the multinational three-party systems, American
Express and Diners Club, work with a local partner that issues cards and in some cases may also
work with merchants. Nonetheless, even in such circumstances, they continue not to have
multilateral or bilateral interchange fees. As with any freely and bilaterally negotiated agreement, the
two parties agree on how to allocate the revenues from their joint activities.
To understand the competitive landscape for payment card systems I have examined
industry data for EU Member States that account for approximately 92 percent of EU GDP and 91
percent of EU population. The multi-national four-party networks account for more than 80 percent
of debit, credit and charge card spend in most these countries and their average share, weighted by
GDP, is almost 60 percent.10 By contrast the multi-national three-party systems, in total, account for
Visa International is a publicly traded global card system. Visa Europe is an association of European banks which are
affiliated with Visa International and have entered into a deal in which they have the option of selling Visa Europe in
return for equity in Visa International.
10 The European Commission has claimed that credit, charge and debit cards are a separate relevant antitrust market and
9
less than 10 percent of card spend in virtually every country and their average share, weighted by
GDP, is about 3 percent. Visa and MasterCard dominate the payment card landscape in the EU.
As mentioned earlier, the multi-national three-party systems sometimes partner with banks
or payment institutions in particular countries. Table 1 shows the extent of these partnership
relationships for American Express for all EU countries. American Express operates directly in 11
EU Member States. It has partners in 17 Member States. It is apparent that American Express
typically operates on its own in the larger economies but chooses to partner with banks in the
smaller economies. The average GDP per capita of countries in which it operates on its own is
37,359 while the average GDP per capita in countries for which it has a partner is, at 15,889, more
than 57 percent lower.
The main increase in competition occurred after 1996 when American Express began
entering into bank partnerships to issue cards for use in various countries. That happened following
a competition law-based intervention by the European Commission that challenged the introduction
of a Visa rule that prohibited its member banks from issuing cards for a competitor other than
MasterCard. These partnership relationships were mainly entered into over the course of the first
decade of the century and reflect entry into these countries over that time period.
III. THE FAILED QUEST FOR A NEW PAN-EUROPEAN CARD SYSTEM

The European Commission has encouraged the creation of a pan-European system that
could obtain a global presence. The Commission saw Chinas UnionPay as an example. The Chinese
government established UnionPay as the card network for banks in China in March 2002.11 China
UnionPay cards accounted for over 9 billion card transactions in 2012.12 Although the UnionPay
that they do not compete with cash, checks, or other means of payments. EC. Mastercard. COMP/34.579. December
19, 2007. Available at: http://ec.europa.eu/competition/antitrust/cases/dec_docs/34579/34579_1889_2.pdf. That
definition is not consistent with the fact that payment card systems compete aggressively with these other forms of
payments, such as cash, and that consumers and merchants can and do substitute readily between different forms of
payment. Nevertheless, for the purpose of this paper I use the Commissions view as a reference point, as this underlies
the proposed legislation, that credit, charge and debit cards are together the relevant market.
11 China UnionPay, Overview. http://en.unionpay.com/comInstr/aboutUs/file_4912292.html.
12 The Nilson Report, #1043.
Table 1: American Express Entities and Partners in Europe

Country
Entities
Austria
American Express
Belgium
Alpha Card (American Express / BNP Paribas Fortis JV)
Bulgaria
Eurobank EFG Bulgaria
Croatia
PBZ Card
Cyprus
Bank of Cyprus
Czech Republic
Global Payments Europe
Denmark
Teller
Estonia
Swedbank
Finland
American Express
France
American Express, Credipar, Credit Mutuel
Germany
American Express
Greece
Alpha Bank
Hungary
OTP Bank
Ireland
Elavon Merchant Services
Italy
American Express
Latvia
Citadele Banka
Lithuania
Citadele Bankas
Luxembourg
Alpha Card (American Express / BNP Paribas Fortis JV)
Malta
Bank of Valletta
Netherlands
American Express
Poland
Bank Millennium, First Data
Portugal
Millennium bcp, Banco Espirito Santo
Romania
Bancpost / EFG Retail Services
Slovakia
VUB Bank
Slovenia
Banka Koper
Spain
American Express, Bansamex (American Express / Santander JV), La Caixa, Iberia Card, Banco
Popular Espanol
American Express, Entercard
Sweden
United Kingdom
American Express, BarclayCard, Lloyds Banking Group, TSB Bank plc, MBNA (Bank of
America Europe Card Services)
cards are primarily issued to Chinese nationals, the cards are accepted in 141 countries and regions
outside of China.13 Another way to look at the situation in Europe is to consider several other large
countries. The United States, China, Japan, and Russia all have large payment systems with roots in
those countries.14 These US, China and Japan systems have secured worldwide distributions for their
domestic card systems.
Several bank groups considered starting pan-European systems in the late 2000s. These
included the European Alliance of Payment Schemes (EAPS), the Monnet Project, and payFair. The
experience of the Monnet Project is instructive. The idea for starting a new pan-EU card system
came about around 2008. A number of banks met in Madrid in 2010 to discuss the initiative and
made plans for moving it forward. By 2011 the Monnet Project had developed detailed technical and
business plans for starting a pan-European system. By then it included 24 banks drawn from seven
countries including the EU-5 as well as Belgium and Portugal.15 One of their key plans was to
develop a mobile payments system for Europe.
The proponents of the new system, however, did not believe they could develop a viable
business model that did not include economically meaningful interchange fees for the participating
banks.16 They took their concerns to the European Commission. The Commission, however,
apparently would not entertain any system, including a new entrant, having interchange fees in
excess of the low levels that the Commission was pursuing. Absent a clear revenue stream for
issuing banks, the Monnet Project believed it could not move forward. It disbanded in April 2012
owing, as the European Central Bank put it, to the perceived absence of a viable business
model.17
Meanwhile EAPS and payFair have not obtained much traction in Europe. EAPS is a
coalition of the domestic independent card systems in Europe. According to the European Central
Bank the number of participating systems has declined from six to three.18 EAPS webpage lists
13
China UnionPay, Overview. http://en.unionpay.com/comInstr/aboutUs/file_4912292.html.

Visa and Mastercard in the United States, UnionPay in China, JCB in Japan, and several domestic systems in Russia.
BIS (2011) Payment, Clearing and Settlement Systems in Russia. CPSS Red Book 2011. Available at:
http://www.bis.org/publ/cpss97_ru.pdf; BIS (2003) Payment Systems in the United States. CPSS Red Book
2003. Available at: http://www.bis.org/cpss/paysys/UnitedStatesComp.pdf; BIS (2012) Payment, Clearing and
Settlement Systems in China- CPSS Red Book 2012. Available at: http://www.bis.org/publ/cpss105_cn.pdf; Bank
of Japan (2003) Payment System in Japan. CPSS Red Book 2003. Available at:
https://www.boj.or.jp/en/paym/outline/pay_boj/pss0305a.pdf.
15
EU
Banks
Ready
to
Break
Visa/MasterCard
Duopoly,
FinExtra.
June
15,
2011
http://www.finextra.com/news/fullstory.aspx?NewsItemID=22662.
16
EU
Banks
Ready
to
Break
Visa/MasterCard
Duopoly,
FinExtra.
June
15,
2011
http://www.finextra.com/news/fullstory.aspx?NewsItemID=22662.
17 Thumbs Down for Monnet, PaySys SEPA Newsletter, May 2012. http://www.paysys.de/download/SepaMay12.pdf.
18 European Central Bank, Card Payments in Europe A Renewed Focus on SEPA for Cards, at p. 32. Available at
http://www.ecb.europa.eu/pub/pdf/other/cardpaymineu_renfoconsepaforcards201404en.pdf.
14
Consorzio BANCOMAT, EUFISERV, and the German Banking Industry Committee. However,
their webpage provides no information on commercial activity after 2012. The latest news section
on the site has only one item from April 2012. PayFair was started in 2007 by industry professionals
and has attempted to develop a pan-European payment system. It highlights on its web site that it
did its one-millionth transaction in 2013. Unfortunately, one million transactions, in total, over six
years, is not an impressive number.19 There were, for example, more than 1.2 billion transactions in
Belgium in just one year, 2013.
Faced with the obstacles set in place by the European Commission there is, at this point, no
significant effort underway, to my knowledge, to create a pan-European system.20 Despite the
prospect of legislation that claims to nurture competition in Europe it does not appear that
anyone is waiting in the wings anxious to make another attempt to start a pan-European system.
These facts strongly suggest that the legislation is not the solution but rather the problem along with
the regulatory barriers to entry erected by the European Commission.
IV. OVERVIEW OF THE PAYMENTS LEGISLATION ENDORSED BY THE

EUROPEAN PARLIAMENT
The legislation endorsed by the European Parliament in April 2014 shifts most of the cost of
running domestic payment systems in Europe from merchants to consumers and favors MasterCard
and Visa at the expense of domestic systems and smaller multinational competitors.
The European Central Bank also mentions EUFISERV as one of the entities trying to establish a pan-European
system. I note that their webpage has a 2012 date on it. Suffice it to say that it does not have much presence in Europe.
http://www.eufiserv.com/home.aspx. Note that EUFISERV is also a member of PayFair.
20 One possibility concerns the banks that belong to Visa Europe. Visa Europe is not owned by Visa. However, under
the terms of an agreement Visa Europe has an option to sell itself to Visa International. There have been some
discussions that Visa Europe would exercise that option after which the banks that belong to Visa Europe would
establish their own pan-European debit card system. Visa Likely to Purchase Europe Payments System, Banking Services
Payments, March 20, 2013. http://payments.banking-business-review.com/news/visa-likely-to-purchase-europepayments-system-200313.
19
10
A.
THE PROPOSED INTERCHANGE FEE REGULATIONS
The interchange fee is paid by the merchants bank to the cardholders bank in the four-party
model. Typically the merchants bank passes on most of the cost of the interchange fee payments to
the merchant and the cardholders bank passes on most of the benefit of the interchange fee
revenue to the consumer in the form of lower fees and product enhancements. As a result the MIF
balances how much one group of customers (merchants that accept cards) pays relative to another
group of customers (cardholders). Increasing the interchange fee usually lowers what consumers pay
for using cards and increases what merchants pay for using cards. In some cases, the two individual
banks, one an acquirer for merchants and the other an issuer to cardholders, negotiate a bilateral
interchange fee. Such bilateral negotiations are seldom practical for four-party bank networks with
many participants that have to deal with each other. As a result, four-party bank networks typically
set a default interchange fee that applies whenever there is no alternative bilaterally agreed fee.
Notably, European competition authorities have never questioned interchange fees that are
negotiated bilaterally between banks. Three-party systems, as noted earlier, do not have interchange
fees.
Four-party systems use the interchange fee to compete with each other and with the
companies that operate the so-called three-party system. A higher interchange fee helps attract banks
to the system. And since banks pass savings on to cardholders the higher interchange fee also
attracts cardholders, which in turn is critical to ensuring merchants are interested in accepting the
networks cards. Card systems also have to consider the impact on merchant acceptance. Acquiring
banks may pass on some, or all, of the interchange fee to merchants, so a higher interchange fee
results in a higher merchant fee. The companies that operate three-party systems also charge
merchant fees, but do not have interchange fees. Four and three-party systems strike different
balances between the prices to merchants and consumers. That is consistent with normal
competition where businesses differentiate themselves based on price and many other features.
Four-party bank card systems also reach different judgments on the interchange fee, along
with other prices, across EU Member States. Thats not surprising. As much as Europeans might
aspire for more similarity across countries, the countries differ enormously from one another in so
many waysfrom income levels, to the role of large merchants, to cultural preferences concerning
credit. In fact, given the obvious differences it would be astonishing if the rate structures for cards
were the same across Europe. Figure 1 shows the median interchange fees for credit and debit
taken at the EMV rate when availablefor most of the EU countries.21
21
See the appendix for details. To show interchange fees on a comparable basis we used the interchange fees for non-
11
Figure 1: Interchange Fees in EU Countries

1.40%
1.20%
1.00%
0.80%
0.60%
Debit Interchange Fee

Credit Interchange Fee
0.40%
Average Interchange Fee
0.20%
Au
s
Be tria
lgi
Bu um
lga
r
Cr ia
oa
ti
Cz
ec Cyp a
h
Re rus
pu
D blic
en
m
a
Es r k
to
n
Fi ia
nl
an
Fr d
G anc
er e
m
an
G y
re
H ece
un
ga
Ire ry
lan
d
Ita
ly
La
L tvi
Lu ithu a
xe ani
m a
bo
ur
g
N M
et al
he ta
rla
nd
Po s
la
Po nd
rtu
Ro gal
m
a
Sl nia
ov
a
Sl kia
ov
en
ia
Sp
U
ai
ni
te Sw n
d ed
K en
in
gd
om
0.00%
Source: See Appendix.
There is even more variability than shown in the figure. Four-party bank card schemes
typically set different interchange fees for different kinds of payment cards. There are other
differences as well. Rates for chip-and-pin cards used at the brick-and-mortar locations where the
consumer is present when they are paying with the card are, for example, lower than the rates for
online transactions. Rates can also vary by industry so in some cases rates for petrol are lower than
for retail. These variations in the interchange fees are another source of competitive differentiation
among the systems.
The Commissions proposals, broadly endorsed by the European Parliament, impose caps
on the MIF adopted by four-party networks of 0.20 percent of the transaction amount for debit
cards and 0.30 percent of the transaction amount for credit cards. For a 50 payment the issuing
bank would receive 10 eurocents for debit and 15 eurocents for credit. The same interchange fee
premium consumer cards used in face-to-face transactions.
12
caps would apply in every country, to every industry, for every merchant, online and offline, and for
every size and type of transaction. The Commission has not, to my knowledge, provided any serious
economic support for the level of these proposed caps, which are apparently wholly arbitrary. The
interchange fee is the only method available for four-party systems to balance their relative prices to
merchants and consumers since different banks serve these two sets of customers. Therefore the
interchange fee caps prevent the card systems from differentiating themselves based on their relative
prices to merchants and cardholders. Under the cap, none of the systems would be able to use
interchange fees to attract banks from other systems.
The interchange fee caps would lead to a dramatic reduction in the fees collected by issuing
banks in most EU countries. Table 2 shows the impact of the legislation by country. It shows the
percent reduction in interchange fee revenue received by issuing banks. The figures are based on the
average interchange fee for debit and credit cards for each country weighted by the volume of
transactions for debit and cards.22 The median reduction in fees is 66 percent. The reductions range
from a low of 0 percent in Hungary to a high of 82 percent in Romania. They exceed 65 percent in
15 of the 28 Member States. In a few pages Ill show what these reductions mean for European
consumers.
B.
THE PROPOSED REGULATION OF THREE-PARTY SYSTEMS
The European Parliaments proposals prohibit merchants from imposing surcharges when
consumers present a card from a four-party card system. At the same time it specifically allows
merchants to impose surcharges when consumers present a card from a company that operates a
three-party system. The law today, as set out in the Payment Services Directive, allows merchants to
surcharge but gives Member States the option of banning merchant surcharging. As of February
2013, 14 EU countries had done that.23 The new proposals do not allow EU Member States to opt
out. Therefore under the European Parliaments proposals, merchants would be able to surcharge
three-party systems throughout the EU.
See Appendix.
They are Austria, Bulgaria, Cyprus, Czech Republic, France, Greece, Hungary, Italy, Latvia, Lithuania, Luxembourg,
Portugal, Romania, and Sweden. See: London Economics, iff, and PaySys (2013), Study on the Impact of Directive
2007/64/EC on Payment Services in the Internal Market and on the Application of Regulation (EC) No 924/2009 on
Cross-Border
Payments
in
the
Community
(Table
17,
page
70).
Available
at
http://ec.europa.eu/internal_market/payments/docs/framework/130724_study-impact-psd_en.pdf.
22
23
13
Table 2: Reduction in Interchange Fees by Country

Country
Reduction in Debit Card

Reduction in Credit Card Reduction in Overall
Interchange
Interchange
Interchange
Austria
69%
70%
69%
Belgium
0%
56%
11%
Bulgaria
73%
63%
70%
Croatia
83%
76%
80%
Cyprus
78%
67%
73%
Czech Republic
80%
71%
79%
Denmark
33%
59%
36%
Estonia
75%
63%
73%
Finland
0%
45%
7%
France
21%
0%
10%
Germany
67%
67%
67%
Greece
69%
71%
70%
Hungary
0%
0%
0%
Ireland
4%
56%
26%
Italy
56%
52%
55%
Latvia
48%
39%
45%
Lithuania
79%
68%
77%
Luxembourg
0%
56%
32%
Malta
0%
56%
26%
Netherlands
0%
56%
6%
Poland
84%
76%
82%
Portugal
65%
69%
66%
Romania
84%
73%
82%
Slovakia
70%
57%
69%
Slovenia
78%
71%
76%
Spain
70%
62%
67%
Sweden
0%
56%
15%
United Kingdom
17%
62%
31%
Median
66%
62%
66%
Source: Based on our calculations; see appendix for details on the calculation of rates.
14
There is now extensive data and research on the results of surcharging.24 We know from the
experience of countries in Europe, and elsewhere, that most merchants do not surcharge when given
the opportunity to do so. Some, however, use the ability to surcharge to act opportunistically
towards consumers and exploit them. Depending on the interpretation, the proposed legislation also
allows merchants that have agreed to accept cards from a three-party system to selectively refuse to
take some of the systems cards for payment.25 For example, a merchant could potentially choose to
accept an American Express corporate card but not take the classic green consumer card product.
When any merchant does that it creates uncertainty for consumers on whether other merchants will
do the same.
In the eyes of consumers the regulations make the cards of three-party systems less
desirable. Consumers may see signs at merchants alerting customers that they will surcharge certain
three-party system cards. And consumers that have these cards will occasionally face opportunistic
surcharging. The proposed legislation therefore seeks to incite the merchant community to
participate in what would become a massive advertising campaign against the three-party card
companies. Consumers will learn in no uncertain terms that if they want to be confident that they
can use their cards to pay and be safe from opportunistic merchant behavior they should stick to
MasterCard and Visa. The proposed legislation will taint the smaller three-party systems, which have
been the main source of new competition in many countries, with a badge of inferiority and will
create a two-tier structure of card products in which the three-party cards are inherently open to and
most likely to be subject to unfavorable treatment.
The legislation also appears to impose open-access regulation on these three-party systems.
To understand the implications of this requirement a short digression into the modern business
See, for example, European Commission (2001), Commission Decision of 9 August 2001 Relating to a Proceeding
Under Article 81 of the EC Treaty and Article 53 of the EEA Agreement, Case No. COMP/29.373 (Visa
International),
2001
O.J.
(L
293)
24.
Available
at
http://eur-lex.europa.eu/legalcontent/EN/TXT/PDF/?uri=CELEX:32001D0782&from=EN. European Commission (2006), Interim Report I
Payment Cards: Sector Inquiry Under Article 17 Regulation 1/2003 on Retail Banking. Available at
http://ec.europa.eu/competition/sectors/financial_services/inquiries/interim_report_1.pdf. London Economics, iff,
and PaySys (2013), Study on the Impact of Directive 2007/64/EC on Payment Services in the Internal Market and on
the Application of Regulation (EC) No 924/2009 on Cross-Border Payments in the Community. Available at
http://ec.europa.eu/internal_market/payments/docs/framework/130724_study-impact-psd_en.pdf. Reserve Bank of
Australia (2007), Reform of Australias Payments System: Issues for the 2007/08 Review. Available at
http://www.rba.gov.au/payments-system/reforms/review-card-reforms/pdf/review-0708-issues.pdf. Reserve Bank of
Australia
(2011),
Review
of
Card
Surcharging:
A
Consultation
Document.
Available
at
http://www.rba.gov.au/publications/consultations/201106-review-card-surcharging/pdf/201106-review-cardsurcharging.pdf. Reserve Bank of Australia (2012), A Variation to the Surcharging Standards: Final Reforms and
Regulation Impact Statement. Available at http://www.rba.gov.au/payments-system/reforms/cards/201206-varsurcharging-stnds-fin-ref-ris/pdf/201206-var-surcharging-stnds-fin-ref-ris.pdf.
25 See Article 10, proposed MIF Regulation.
24
15
model of these three-party systems is helpful.

Some of the standalone card companies have decided to enter into selective partnerships
with banks to help expand their businesses.26 This strategy helps them secure scale economies and
network effects by issuing more cards and acquiring transactions in countries where they might
otherwise have no presence, and thereby makes them stronger competitors to the large four-party
systems in those countries and globally. Partnerships also enable three-party systems to enter
countries without making significant investments and thereby reduce barriers to entry into the
payments sector of these countries. American Express used this strategy starting in the late 1990s to
enter 17 EU countries, mainly less wealthy ones as mentioned above, through partnerships with
financial institutions. Diners Club also has bank partners in 8 countries that it has developed over
the last several decades.
The European Commission and European Parliament proposals apparently require threeparty systems to enter into partnerships on the basis of objective, proportionate and nondiscriminatory criteria with any and all banks or payment institutions that want to issue or acquire
cards for three-party systems if those systems make a deal with a single bank or payment institution
in the EU. That requirement increases entry barriers for three-party systems into domestic payment
markets because a decision to enter into a bank partnership triggers a requirement to provide access
to banks throughout the EU as a result. As a practical matter the proposed legislation appears to
subject the small three-party systems to essential-facility regulation that is commonly applied to
domestic monopolies in energy, ports, and telecom. Such an approach imposes entirely
disproportionate burdens and has no basis in competition or regulatory policy.
Finally, the proposed legislation appears to subject the three-party systems to price caps as
well. Whenever the three-party system enters into a licensing deal to issue a card, it appears the
system would be subject to the MIF price caps developed for and applied to four-party systems. It is
unclear how the proposed legislation envisions this provision would apply in practice. The threeparty systems do not establish an interchange fee that flows from an acquiring bank to an issuing
bank. They do enter into a bilateral negotiation with a potential partnerwhich may have its own
valuable brand and other assets in a particular countryover the allocation of revenues resulting
from their joint activities. The viability of these confidential and bilateral negotiations appears under
threat.
Beyond the issue of practicality is the question of what the possible justification for the price
caps could be. The European Commission and other competition authorities that have challenged
interchange fees for four-party systems have claimed that these fees violate the competition laws
Visa used to prohibit its member banks from entering into these partnerships. Antitrust action taken by the European
Commission forced it to allow these partnerships.
26
16
because they are set collectively and have also pointed to the dominant position of these four-party
systems in a claimed market for debit, credit, and charge cards. To my knowledge, no competition
authority, regulatory authority, or court has complained about commercial terms that are agreed
bilaterally and it is hard to see under what basis these authorities could do so. Likewise, no
competition authority, regulatory authority, or court, at least to my knowledge, has found that the
merchant fees agreed between merchants and three-party systems are anticompetitive. Given the
small European wide share that three-party systems have, in total, of debit, credit and charge card
volume (the market identified by the European Commission) none of these systems is even remotely
dominant.27
The rationale for capping interchange fees does not apply to the financial terms that threeparty systems negotiate with a licensee that issues cards. Interchange fees are direct payments from
the merchants acquirer to the cardholders issuer and are typically passed on by the acquirer to the
merchant. Competition and regulatory authorities have sought to reduce the impact of interchange
fees on merchants and have done so by capping those fees. There is no pass through, however,
between the fees that a three-party system negotiates individually with a licensee and the fees that
the three-party system, or an acquiring partner, negotiates with merchants.
These proposed regulations of smaller three-party payment card systems are unprecedented
outside the EU.28 What is remarkable is the length to which the proposed legislation has gone to
squelch competition by three-party card systems. The proposed legislation impairs the ability of
these smaller systems to compete by permitting merchants to surcharge the smaller three-party
systems but not the larger four-party ones and by making it costly for three-party systems to enter
into select partnerships. But, then, just in case (against the odds) the three-party systems are
successful, the proposed legislation empowers the European Commission to impose even more
restraints if they surpass some undefined threshold.
27
The proposed legislation involving MIFs exposes three-party systems to other risks. In the text proposed by the
European Parliament, the entire set of interchange fees regulations for four-party systems can be applied to three-party
systems that exceed a threshold set by the European Commission. In practice, it is difficult to envision what this means.
See Amendment 21, European Parliament, Amendments adopted by the European Parliament on 3 April 2014 on the
proposal for a regulation of the European Parliament and of the Council on interchange fees for card-based payment
transactions (COM(2013)0550 C7-0241/2013 2013/0265(COD)). April 3, 2014. Available at:
http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//TEXT+TA+P7-TA-20140279+0+DOC+XML+V0//EN.
28
Spain recently adopted the Commissions proposals on this issue word-for-word, pending the adoption of final
legislation (at which point Spain indicated it would be prepared to reverse course).
17
V. IMPACT OF PROPOSED LEGISLATION ON COMPETITION IN THE

EUROPEAN UNION
The proposed payments legislation has an Alice-in-Wonderland up is down, left is right

flavor to it. In the name of nurturing competition the European Commission and European
Parliament have come up with an approach that places an oppressive thumb on the smallest
competitors, discourages the challengers, and weakens competition between the two giant systems
left standing. Despite a vision of creating a European born-and-bred system the legislation pushes
and shoves consumers to the dominant global brands. Then, in a final flourish, the legislation
threatens the smaller systems with even more regulation if they are nonetheless able to continue
providing a degree of increased competition. This is legislation that only Lewis Carroll could have
written.
A.
RESTRAINING THREE-PARTY SYSTEM COMPETITION
The proposed payments legislation restrains the ability of three-party systems to compete
with the four-party card schemes, which are based on networks of banks, in at least three ways.
First, it imposes rules that make cards from the dominant four-party bank card systems
preferablein the sense of having fewer regulatory-imposed annoyancesfor cardholders and
merchants than cards from the smaller three-party systems. Consumers will learn that the three-party
system cards are the ones that merchants can surcharge, possibly opportunistically, and reject
altogether even though the merchant has a sign at their store claiming they accept the card.
Consumers may find this out directly, from the media, or from friends, family and colleagues. It is
easy to imagine the media reports advising consumers to stick with the major brandsMasterCard
and Visato avoid having merchants subject them to a surcharge.
Second, the legislation could result in the three-party systems simply withdrawing as
competitors in countries where they operate with a bank or payment institution partner. A single
partnership agreement, anywhere in the European Union, exposes a three-party system to the risk
that banks, including members of MasterCard and Visa, will insist on being able to issue or acquire
the three-party systems cards as well. Under the legislation a three-party system that has entered a
partnership with a single bank or payment institution, anywhere in the EU, may be required to offer
the same terms to every other bank or payment institution that approaches the system. No longer
could a three-party system decide to partner selectively and to do so in countries of its own
18
choosing. It must either operate with no partners, or it must mimic the business model of the fourparty bank associations and open itself up to all. The European Commission and the European
Parliament have offered no credible explanation why three-party payment card schemes, that have
miniscule shares in the market the European Commission has defined, should be subject to these
requirements, nor have they considered apparently the consequence on competition of hobbling the
smaller card rivals.
The legislation will make it harder for three-party systems to negotiate mutually profitable
agreements with potential partners. A three-party system that enters into a deal with a bank or
payment institution in a country may have to extend that deal to other banks or payment institutions
in the country. Its chosen partner would find this unattractive for a variety of reasons. The
prospective partner would effectively have to share the business opportunity of a smaller system
with other banks and payment institutions, while still having to invest into the establishment and
growth of that system, for the benefit of all comers. For example, any funds that the chosen bank or
payment institution partner spends on marketing and advertising the brand would benefit other
banks that insist on issuing the three-party system card as well. The legislation therefore undermines
the investments of a three-party systems partner and thus makes it harder for three-party systems to
reach acceptable partnership deals.
The legislation also makes it less profitable for a three-party system to enter into any
partnership deal and thereby eliminates all the competition that results from these relationships. A
three-party system that negotiates a partnership arrangement in a single country loses its ability to
manage its system throughout the EU. It could be forced to partner unwillingly with banks or
payment institutions in that country as noted above. In addition, however, it could be forced to
partner with banks and payment institutions in countries in which it would prefer to operate alone
and with banks that it would prefer not to work with at all.
The legislation therefore poses a serious risk that three-party systems that rely on partnership
deals to extend their reach across the EU will either end these partnership deals because of the risks
and costs they pose or that their partners will end these deals if the system cannot assure them of an
exclusive deal in that country. That is an odd coda to the competition policy decision the European
Commission took in 1996 when it forced Visa to allow its member banks to enter into selective
partnerships with three-party systems.
Third, the payments legislation will retard future entry of three-party systems throughout the
EU in addition to unraveling past entry. Existing three-party systems will likely refrain from
considering entering countries in the EU through partnership deals. Consider Cetelem, which
operates a standalone card system in France. Suppose that, as the single EU market evolves, Cetelem
would like to enter some countries by forming a partnership with a bank or payment institution. As
soon as it enters into one partnership in one country, it opens itself up to demands that it provide
similar deals to banks and payment institutions in that country and every other EU country. That
19
prospect would likely deter Cetelem from ever considering partnerships and therefore limit its ability
to compete on a pan-European basis.
More importantly, the legislation makes the EU a very unwelcome area for entry by any new
three-party system. Suppose, for example, that a three-party global mobile payments system emerges
and that system needs to consider where to enter around the world. The EU will be the only place in
the world in which merchants are allowed to surcharge, and reject, the cards of three-party systems
but not of MasterCard or Visa. Suppose, as is common in mobile payments, the three-party system
wanted to partner with a bank to enter the EU or a Member State. It could not guarantee its partner
an exclusive deal and, if it entered into a relationship in any country, it would potentially have to
extend that deal to all banks and payment institutions in that country and the other EU countries as
well and be price capped when doing so.
B.
SOFTENING FOUR-PARTY SYSTEM COMPETITION
The European Parliaments payment legislation weakens competition among the four-party
systems.
To begin with it softens competition between MasterCard and Visa. They wont be able to
use the interchange fee to compete for issuers, consumers, or merchants. No longer would one of
these companies be able to lower their interchanges fees for a particular type of merchant to secure
acceptance, to increase their interchange fees to persuade banks to switch card volume to them, or
to increase their interchange fees to promote benefits that could attract more cardholders.
Four-party systems have used selective reductions in interchange fees to promote new
technologies such as chip-and-pin cards and contactless cards. They would lose that ability under the
payments legislation: with such a substantial reduction in these fees it is unlikely they could persuade
banks to accept an even lower fee. If one of the systems came up with a technology for accepting
payment at the point of salefor example related to mobile paymentsit would lose one of its
main tools for persuading merchants to invest in the necessary changes. MasterCard and Visa will of
course continue to compete but will do so with one hand tied behind their backs.
The payments legislation weakens competition among the global four-party systems and the
independent domestic systems for these same reasons. None of these systems will be able to agree
to higher interchange fees to compete for banks and consumers. And, with the drastic reductions,
few if any could risk bank defections if they wanted to lower interchange fees further to promote
innovative technologies or business practices by the merchants.
The threat to domestic competition though is actually much worse. An independent
domestic system could not offer banks a somewhat higher interchange fee to induce them to switch
20
from the domestic MasterCard or Visa network. That eliminates an important competitive tool.
Once the payments legislation makes all systems exactly the same when it comes to interchange fees
the advantage of switching to a system that lacks the scale economies, brand recognition, and
marketing prowess of MasterCard and Visa is lessened. The proposed legislation places the survival
of the domestic systems at risk and it is conceivable that they will wither over time or simply become
appendages of the global four-party networks.
Lastly, the very low interchange fee caps proposed by the European Parliament largely gut
the business models of new four-party entrants. As Ive noted this isnt mere conjecture. We have
the dead body to prove it. A substantial viable pan-European entrant gave up when the European
Commission wouldnt relent on its insistence that four-party systems have not only low interchange
fees but interchange fees that cant be any higher for any country, industry, product, transaction
type, or anything else.
C.
THE ANTICOMPETITIVE PAYMENTS LEGISLATION
The European Commission and the European Parliament have put forward legislation that
is anti-competitive. It fixes the interchange fee that MasterCard and Visa use to compete with each
other and independent domestic schemes. It places independent domestic schemes that would be
less able to differentiate themselves at a disadvantage. After the Commission helped destroy a major
pan-European entrant, the proposed legislation raises a barrier to further entry by any potential new
payment systems. As a finishing touch it hobbles all of MasterCard and Visas three-party system
rivals.
Almost two decades ago, when Visa proposed rules to prohibit three-party systems from
pursuing arms length licensing agreements with banks that were members of Visa, the European
Commission claimed competitive harm and acted swiftly to prevent it. It seems perverse now that
the EU is at risk of delivering the same outcome for Visa and legitimizing the endeavor that it
previously claimed was anticompetitive.29
As per the following press release the Commission's Directorate General for Competition reached the view that
Visa's proposed rule prohibiting its member banks from partnering with American Express would have infringed the
EC competition rules because it would have restricted competition between international cards systems as well as
between banks which issue cards riding on those systems. Moreover, at the time, the Commission confirmed its
determination to ensure that access to the payments card market by new competitors such as three-party schemes should
not be impeded. http://europa.eu/rapid/press-release_IP-96-585_en.htm.
29
21
VI. HOW THE EU PAYMENTS LEGISLATION WILL AFFECT CONSUMERS
The European consumer is ultimately the big loser from the proposed EU payments
legislation.
Studies have found that consumers end up paying more in countries that have capped
interchange fees.30 This result is obvious. Competition forced banks to offer low fees (either on the
cards themselves or the current account held by the consumers) when they were getting interchange
fee revenues. When that revenue is reduced sharply banks have to increase fees (either on the cards
themselves or on the current account held by the consumers).
Defenders of interchange fee caps, in my experience, do not deny this relationship between
interchange fees and consumer prices. Instead, they counter the well-documented fact that
consumers pay banks more when interchange fees decline with the claim that merchants pass the
entirety of their interchange fee cost savings back to consumers in the form of lower prices and that
therefore consumers come out ahead when these lower merchant prices are considered.31 They base
this claim on pure speculation and provide no evidence that any merchant has passed any savings
whatsoever on.
Their assertions are not supported by economic theory and are roundly rebutted by
empirical evidence.32 My study in the US, for example, found that merchants kept about half of the
savings from debit card fee reduction for themselves and that consumers will end up losing more
than $22 billion as a result of shifting the costs on to them.33 Large retailers with revenues of
hundreds of millions of dollars a year and market valuations north of $1 billion were the chief
beneficiaries.
Evans, David S. and Rosa Abrantes-Metz (2013) The Economics and Regulation of the Portuguese Retail Payments
System. Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2375151.
31 Evans, David S., Howard Chang and Steven Joyce (2013) The Impact of the U.S. Debit Card Interchange Fee
Regulation on Consumer Welfare: An Event Study Analysis. Coase-Sandor Institute for Law and Economics Working
Paper No. 658 (2D Series). Available at: http://www.law.uchicago.edu/files/file/658-dse-hj-sj-impact-fixed.pdf.
Forthcoming, Journal of Competition Law and Economics.
32 Evans, David S. and Abel M. Mateus (2011) How Changes in Payment Card Interchange Fees Affect Consumers
Fees and Merchant Prices: An Economic Analysis with Applications to the European Union. Available at SSRN:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1878735.
33 There is no serious dispute that merchants do not pass on 100 percent of the cost savings nor could there be given
economic theory and empirical evidence. Notably, a study in the US estimated that merchants kept about a third of the
interchange fee cost savings as profits. Shapiro, Robert J. (2013) The Costs and Benefits of Half a Loaf: The Economic
Effects of Recent Regulation of Debit Card Interchange Fees. SONECON. Available at:
https://nrf.com/sites/default/files/The_Costs_and_Benefits_of_Half_a_Loaf.pdf.
30
22
Figure 2: Average Debit Interchange Fees versus Per Capita GDP in EU Countries
90
80
Luxembourg
70
GDP per Capita (Euro 000s)
60
50
Denmark
40
30
Netherlands
Ireland
Sweden
Finland
Belgium
Austria
United Kingdom
France
Germany
Italy
Spain
Greece
20
Hungary
Slovenia
Portugal
Malta
10
Cyprus
Czech Republic
Estonia
Latvia
Slovakia
Lithuania
Croatia
Poland
Romania
Bulgaria
0
0.00%
0.20%
0.40%
0.60%
0.80%
1.00%
1.20%
1.40%
Debit Interchange Fee (%)
Source: European Central Bank.
The extent of the redistribution from merchants to consumers varies enormously across
countries. Consumers in high-interchange fee countries such as Romania will lose much more
relatively speaking than consumers in low-interchange fee countries such as Denmark. Figure 2
shows the relationship between the average debit card interchange fee in each country and GDP per
capita. The graph shows that the debit card interchange fee is higher in countries with lower per
capita GDP. The countries that will have the largest reductions in interchange fees, and therefore
the greatest harm to consumers, are the poorest countries; the countries that will have the lowest,
and in some cases no, reductions in interchange fees and therefore the least harm to consumers are
the richest countries. There are exceptions, of course, but that is the general rule. The interchange
fee caps take from the consumer and give to the merchant, and they take the most from the poorest
consumers. As I said, this is Alice-in-Wonderland public policy.
The EU payments legislation harms European consumers in other ways. Approximately 88
billion was spent in 2012 by Europeans using cards from three-party systems in the Member States
23
that account for more than 90 percent of GDP and population. Consumers will also face occasional
opportunistic surcharging by merchants on such purchases. A consumer also risks having merchants
that advertise that they accept a three-party system brand turn around and reject the particular card
the consumer presents. In the longer run consumers are also likely to find that they have less choice
than they have now as independent domestic card systems and three-party systems are forced to
withdraw from the payment card market in countries across the EU. Consumers are also likely to
have even less choice than they would get in the absence of the payments legislation. Thats because
the legislation will reduce entry into payments cards in the European Union.
VII. CONCLUSION
The European Commissions proposal and the common position reached by the European
Parliament in April 2014 is ill-conceived and poorly thought through. This is no surprise, given how
woefully inadequateand in some areas, completely lackingthe impact assessment undertaken by
the Commission is. The proposed legislation destines Europe to having a payment card industry
operated largely by banks and run by two global brands. The prospect of low caps on interchange
fees has already killed or chilled the prospects for the emergence of a new pan-European payment
system. Those caps will temper competition between MasterCard and Visa and may drive
independent domestic systems out of business altogether over time. The bizarre restraints on the
three-party systems in European countries, all of them much smaller than their four-party rivals
across Europe, will make these companies less vibrant competitors and may drive them out of many
countries in Europe. For European consumers the proposed payments legislation would lead to a
hefty price tag, diminished choice, and depressed innovation.
The European Council would be wise to discard this legislation as currently drafted. Any
new legislation should completely abandon restraints on the three-party systems that are essentially
fringe competitors in Europe. There is simply no sensible rationale for these restraints and none has
been offered. New legislation should also drop the caps on interchange fees. These caps weaken
competition between MasterCard, Visa, and independent domestic card systems. They also shift the
costs of payment cards to consumers and will cost European consumers billions of euros in added
fees.
24
APPENDIX
CALCULATION OF INTERCHANGE FEES BY COUNTRY
This appendix describes the calculation of interchange rates by country.
We started by obtaining Visa and MasterCards current intra-country interchange fee
schedules.34 Visa and MasterCard typically have many different interchange rates that depend on the
type of card used, how the transaction is processed, and the merchants sector. Separately for credit
and debit, for each country, for each system, and for each merchant sector, we limited attention to
one interchange rate. In each case, we picked an interchange rate that applied to intra-country
transactions that used non-premium consumer cards in a face-to-face transaction. For cases where
there was a separate rate for EMV or Chip+PIN transactions, we used that rate. In countries
without that distinction, we used the rate for electronically authorized transactions. There are
interchange fee rates that are higher or lower but the rates we used were typically in the middle and
reflect the most common type of transactions.
We then converted each interchange rate into a percentage of the transaction amount. In
most cases the fees were already expressed in percentage terms, so this involved no additional
calculations. In other cases, the interchange fees were expressed as a flat fee plus a percentage of the
transaction amount. In these cases, we needed to make an assumption about the average transaction
size. We used data from the European Central Bank for 2012 to calculate an EU-wide average
transaction size for debit cards (48.15) and credit cards (67.82).35 We used these average
transaction sizes to calculate the average total interchange fee. For example, in Belgium, the
interchange fee for Visa debit is 0.15% + 0.015. At an average transaction size of 48.15, this
works out to (0.0015*48.15 + 0.015)/48.15 = 0.18%.
Next, for each country, system (Visa or MasterCard), and product (credit or debit), we
calculated the median interchange rate across all merchant sectors. This gives us two debit
interchange rates and two credit interchange rates for each country.
Next, we obtained estimates of interchange fees for the large domestic card systems. Table A
summarizes the data.
Visa
Europe,
Our
Fees.
Available
at
http://www.visaeurope.com/en/about_us/our_business/fees_and_interchange.aspx;
MasterCard
International,
MasterCard
Intra-Country
Interchange
Fees.
Available
at
http://www.mastercard.com/us/company/en/whatwedo/interchange/Country.html.
35
European
Central
Bank,
Statistical
Data
Warehouse.
Available
at
http://sdw.ecb.europa.eu/reports.do?node=1000001431.
34
25
Table A: Interchange Fees for Large Domestic Card Systems

Country
Denmark
France
System
Dankort
Cartes Bancaires
Type
Debit
Credit
Germany
Italy
Portugal
Spain
Spain
Spain
Spain
Spain
Spain
ZKE
PagoBancomat
MB
Servired
Euro 6000
4b
Servired
Euro 6000
4b
Debit
Debit
Debit
Credit
Credit
Credit
Debit
Debit
Debit
Interchange Fee (Actual)

0.20%36
0.28% + fraud adjustment
averaging no greater than 0.03%37
0.30%38
0.1309% + 0.1039
0.50%40
0.76%41
0.79%42
0.75%43
0.3344
0.3245
0.3046
Interchange Fee (%)

0.20%
0.31%
0.30%
0.34%
0.50%
0.76%
0.79%
0.75%
0.69%
0.66%
0.62%
Cases with a flat interchange fee were converted to percentages of the transaction using the
same method used for Visa and MasterCard.
Denmarks Dankort pays issuers a flat fee per transaction, where the level of the flat fee depends on the issuers
annual number of transactions. In general, this fee is less than 0.20 percent, although it may be higher on small-value
transactions. Ministry of Growth and Business Denmark, Interchange Fee Regulation and Domestic Debit Card
Schemes,
June
2,
2014.
Available
at
http://www.euoplysningen.dk/upload/application/pdf/ca7ff3c0/201305502.pdf?download=1. The exact average interchange fee
appears to be non-public.
37 Cartes Bancaires, Current CB Multilateral Interchange Fees and Tariffs. Available at http://www.cartesbancaires.com/IMG/pdf/CB_Interchange_Fees_and_Tariffs.pdf; PaySys, French Anti-Trust Authority Decision on
MIF, PaySys SEPA Newsletter, June-July 2011. Available at http://www.paysys.de/download/SepaJunJul11.pdf.
38
Der
Handel,
Bundeskartellamt
kippt
Girocard-Gebhr,
April
8,
2014.
Available
at
https://www.derhandel.de/news/finanzen/pages/Bundeskartellamt-kippt-Girocard-Gebuehr-10503.html.
39
Conzorzio
Bancomat,
Commissioni
Interbancarie.
Available
at
http://www.bancomat.it/it/consorzio/commissioni.html.
40 David S. Evans and Rosa Abrantes-Metz (2013), The Economics and Regulation or the Portugese Retail Payments
System. Available at http://www.sibs.pt/export/sites/sibs_fps/pt/documentos/The-Economics-and-Regulation-ofthe-Portuguese-Retail-Payments-System_2013.pdf.
41 Servired, Tasas de Intercambio: Intra-Sistema. Available at http://www.servired.es/tasas-de-intercambio/intrasistema/.
42 Euro 6000, Tasas de intercambio: Intra-Sistema. Available at http://www.euro6000.com/informacioncorporativa/tasas/intrasistema.
43 4B, Tasas de intercambio. Available at http://www.4b.es/productos-y-servicios/comercios/tasas-de-intercambio.
44 Servired, Tasas de Intercambio: Intra-Sistema. Available at http://www.servired.es/tasas-de-intercambio/intrasistema/.
45 Euro 6000, Tasas de intercambio: Intra-Sistema. Available at http://www.euro6000.com/informacioncorporativa/tasas/intrasistema.
464B, Tasas de intercambio. Available at http://www.4b.es/productos-y-servicios/comercios/tasas-de-intercambio.
36
26
Next, we calculated the median debit and credit interchange rate for each country, taking the
median across all systems. Then we calculated an overall average interchange rate as a weighted
average of the debit and credit interchange rates, using each product types share of all payment card
spending in the country as the weights. Data on the value of debit and credit card payments were
taken from 2012 data from the European Central Bank.47
European
Central
Bank,
Statistical
http://sdw.ecb.europa.eu/reports.do?node=1000001431.
47
Data
Warehouse.
Available
at
1201_1250_EVANS_WEB_110612 (DO NOT DELETE)
11/6/2012 5:38 PM
GOVERNING BAD BEHAVIOR BY USERS OF

MULTI-SIDED PLATFORMS
David S. Evans
ABSTRACT
Multi-sided platforms such as exchanges, search engines, social networks, and software
platforms create value by assembling and serving communities of people and businesses.
They generally solve a transaction problem that prevents agents from coming together to
exchange value. An essential feature of these platforms is that they promote positive
externalities between members of the community. But as with any community, there are
numerous opportunities for people and businesses to create negative externalities, or engage
in other bad behavior, that can reduce economic efficiency and, in the extreme, lead to the
tragedy of the commons. Multi-sided platforms, acting selfishly to maximize their own
profits, often develop governance mechanisms to reduce harmful behavior. They also
develop rules to manage many of the same kinds of problems that beset communities
subject to public laws and regulations. They enforce these rules through the exercise of
property rights and, most importantly, through the Bouncers Right to exclude agents
from some quantum of the platform, including prohibiting some agents from the platform
entirely. Private control is likely to be more efficient than social control (through laws and
regulation) in dealing with negative externalities on platform communities because the
platform owner can monitor bad behavior more closely and deal with this behavior more
expeditiously than a public regulator. Therefore, the courts and antitrust authorities should
exercise caution in finding anti-competitive exclusion when that exclusion is conducted as
part of a private governance mechanism for dealing with bad behavior of some platform
users that harm other users.
2012 David S. Evans.

Chairman, Global Economics Group; Lecturer, University of Chicago Law School;
Visiting Professor, Faculty of Laws, University College London. I would like to thank
Richard Epstein and Richard Schmalensee for helpful comments; Lauren Chiang, Jacqueline
Murphy, Steven Joyce, Margaret Schilt, and Nikhil Tuladhar for excellent research help; and
Google for research funding. None of these individuals or entities necessarily agrees with me
and I retain sole ownership of any errors.
1202
11/6/2012 5:38 PM
BERKELEY TECHNOLOGY LAW JOURNAL [Vol. 27:1201
TABLE OF CONTENTS
I.
INTRODUCTION .......................................................................................... 1202
II.
MULTI-SIDED PLATFORM STRATEGIES FOR

INCREASING VALUE.................................................................................. 1207
A.
B.
C.
III.
BAD BEHAVIOR AND PLATFORM COMMUNITIES .................. 1212

A.
B.
C.
D.
E.
IV.
I.
SOCIAL NETWORKS ................................................................................ 1226

STOCK EXCHANGES ............................................................................... 1231
SEARCH ENGINES ................................................................................... 1235
PRIVATE VS. PUBLIC GOVERNANCE IN THE

REGULATION OF BAD BEHAVIOR ON PLATFORMS .............. 1240
A.
B.
VII.
THE PROPERTY RIGHTS FRAMEWORK ............................................... 1220

SELECTIVE EXCLUSION ......................................................................... 1222
INFORMATION AND TRANSPARENCY ................................................. 1225
GOVERNANCE REGIMES FOR KEY PLATFORMS ..................... 1226

A.
B.
C.
VI.
IT OUGHT TO BE A CRIME .................................................................... 1214

POOR AND ASYMMETRIC INFORMATION ........................................... 1215
CONGESTION AND OPTIMIZING PHYSICAL SPACES ........................ 1216
CASE IN POINT: DECORMYEYES ........................................................ 1216
BAD BEHAVIOR AND PLATFORM VALUE ........................................... 1218
DEALING WITH BAD BEHAVIOR ....................................................... 1218

A.
B.
C.
V.
THE PLATFORM TOOLKIT ..................................................................... 1209

VALUE CREATION AND EXTERNALITIES ........................................... 1210
VALUE DISTRIBUTION AND COORDINATION ................................... 1211
THE BENEFITS OF PRIVATE REGULATION ........................................ 1241

ANTI-COMPETITIVE EXCLUSION: A PROPOSED THREE-STEP
ANALYSIS .................................................................................................. 1243
CONCLUSION ................................................................................................ 1249
INTRODUCTION
If you win an auction on eBay but do not get the good, or the good is
not what was advertised, you can complain to the e-commerce site in
addition to giving the merchant a low rating. The site may decide to punish
the merchant, including prohibiting them from ever selling again on eBay.
Merchants receive protections too. Consumers are required to pay for any
item they win in an auction and can bid only if they intend to buy the item
that they win. Among other things, these rules prohibit consumers from
2012]
11/6/2012 5:38 PM
GOVERNING MULTI-SIDED PLATFORMS
1203
bidding in several auctions for similar items and then only paying for the
cheapest item they have won.1 Consumers that have too many unpaid items
can lose their buying privileges.2 eBay has a system that governs bad behavior
by the consumers and merchants that use its website. Many other businesses
that are multi-sided platforms, like eBay, also have governance systems for
dealing with bad behavior that creates negative externalities3 across platform
participants. This Article examines these governance systems and explores
the relationship between the public and private control of negative
externalities.
Multi-sided platforms create value by helping two or more different types
of users, who could benefit from getting together, find and interact with each
other, and exchange value.4 They include software platforms (e.g., Apples
iOS), financial exchanges (e.g., NASDAQ), search engines (e.g., Microsofts
Bing), social networks (e.g., LinkedIn), shopping malls (e.g., Water Tower
Place in Chicago), advertising-supported media (e.g., CNN), and e-commerce
sites that connect businesses and shoppers (e.g., Amazon).5 Multi-sided
platforms solve a transaction problem6 that prevents these different types of
users from getting together on their own to exchange value. There are
positive externalities between the multiple types of users. Platforms provide
1. For guidance relating to the rules for buyers and sellers on eBay, see Rules & Policies,
(last visited Aug. 1, 2011).
2. For eBays policies on unpaid items see Unpaid Item Policy, EBAY,
http://pages.ebay.com/help/policies/unpaid-item.html (last visited Apr. 18, 2012).
3. Externalities are costs and benefits that agents impose on each other and are not
transmitted through the price system. Generally, externalities involve a lack of direct
agreement between the agents to bear the cost or provide the benefit and a lack of direct
compensation for bearing costs or providing benefits. Negative externalities, such as air
pollution and barking dogs, involve costs. Positive externalities, such as restoring historical
buildings, researching new technologies, and pursuing education, involve benefits. See N.
GREG MANKIW, PRINCIPLES OF ECONOMICS 20407 (3d ed. 2004).
4. See Bernard Caillaud & Bruno Jullien, Chicken and Egg: Competition Among
Intermediation Service Providers, 34 RAND J. OF ECON. 309, 309310 (2003); Jean-Charles
Rochet & Jean Tirole, Platform Competition in Two-Sided Markets, 1 J. OF THE EUR. ECON.
ASSN 990, 990 (2003).
5. See generally David S. Evans, Some Empirical Aspects of Multi-sided Platform Industries, 2
REV. OF NETWORK ECON. 191 (2003); DAVID S. EVANS & RICHARD SCHMALENSEE,
CATALYST CODE: THE STRATEGIES BEHIND THE WORLDS MOST DYNAMIC COMPANIES
(2007); Thomas Eisenmann, Geoffrey Parker & Marshall W. Van Alstyne, Strategies for TwoSided Markets, 84 HARV. BUS. REV. 92 (2006).
6. As Rochet & Tirole observe, the inapplicability of the Coase Theorem is a
necessary condition for the existence of a multi-sided platform. Jean-Charles Rochet & Jean
Tirole, Two-Sided Markets: A Progress Report, 37 RAND J. ECON 645, 649 (2006). When the
Coase Theorem holds, individual users would be able to engage in value-maximizing
exchange directly. See Ronald Coase, The Problem of Social Cost, 3 J. OF L. & ECON. 1 (1960).
EBAY, http://pages.ebay.com/help/policies/overview.html
1204
11/6/2012 5:38 PM
ways to promote these positive externalities and thereby create value for the
community of users they serve. 7
Whenever people and businesses get together, and in any community,
there are many opportunities for people and businesses to behave badly and
to thereby generate negative externalities. This bad behavior can reduce
economic efficiency and in the extreme lead to the tragedy of the commons.8
Multi-sided platforms such as eBay develop governance systems to reduce
this bad behavior and minimize negative externalities. This Article shows that
multi-sided platforms develop systems of rules and penalties to manage many
of the same kinds of problems that communities subject to public laws and
regulations face. These platforms enforce such rules by exercising their
property rights to exclude users from the platform. In some cases, the rules
and penalties imposed by the platform are similar to, and in some cases close
substitutes for, rules and penalties adopted by a public regulator.
Private control is likely to be more efficient than social control9 in dealing
with negative externalities in platform communities. The platform owner can
monitor bad behavior more closely and deal with this behavior more quickly
than can a public regulator. Multi-sided platforms face antitrust complaints
concerning reductions in service or denial of service by the platform. This
Article argues that the courts and antitrust authorities should exercise caution
in assessing these claims when the exclusion at issue is related to platform
governance for dealing with bad behavior. It proposes a three-step test for
anti-competitive exclusion in these cases in which the burden shifts to the
complainant when the platform has engaged in exclusion as part of an
established internal governance system.
Despite the pervasiveness of private governance systems for bad
behavior, there is little research on the topic and none that examines the
public policy issues addressed in this paper. Rochet and Tirole were the first
to identify the role of the platform as a regulator in their seminal paper on
two-sided platforms.10 Boudreau and Hagiu present a detailed analysis of
platform regulation and highlight the fact that platforms leverage a wide
variety, and nuanced set, of instruments to maximize value.11 However, they
treat all non-price instruments used by platforms as a form of regulation for
7. See David S. Evans & Richard Schmalensee, The Industrial Organization of Markets with
Two-Sided Platforms, 3 COMPETITION POLY INTL 151, 154 (2007).
8. See generally Garrett Hardin, The Tragedy of the Commons, 162 SCIENCE 1243 (1968).
9. Social control refers to the enforcement of public laws and regulations.
10. Rochet & Tirole, supra note 4. See also Evans & Schmalensee, supra note 7, at 163.
11. See Kevin J. Boudreau & Andrei Hagiu, Platforms Rules: Multi-Sided Platforms as
Regulators, in PLATFORMS, MARKETS, AND INNOVATION 163 (Annabelle Gawer ed., 2009).
2012]
11/6/2012 5:38 PM
1205
market failures and consider a multitude of strategies for addressing positive

and negative externalities. This Article focuses narrowly on the existence of
negative externalities among platform users and the use of governance
systems to deal with this bad behavior. It shows that these practices are
analogous to the governance systems for communities, clubs, and other
similar entities.
The analysis here is loosely related to the framework put forward by
Strahilevitz.12 He argues that property rights include several subordinate
rights that enable private businesses to deal with information asymmetries
and examines the extent to which private property rights and public
governance systems are substitutes. This Article adopts his framework of
subordinate property rights but then examines how these rights facilitate the
development of private governance systems for multi-sided platforms that,
like polities, must govern a community of members who may interact
positively or negatively with each other.
The governance of bad behavior among members of platform
communities is a worthy subject for several reasons.
First, although multi-sided platforms have existed for thousands of years,
they are becoming an increasingly important part of the fabric of the
economy. The development of the Internet has facilitated the creation of
these platforms and some of these platforms have become global businesses
quite rapidly. For example, Facebook, which started in 2004, has more than
845 million active monthly users worldwide,13 integrates more than 7 million
applications and websites,14 and had advertising revenue in 2011 of more
than $3.1 billion.15 Understanding key aspects of how these platforms work
helps in numerous contexts ranging from business to litigation.
Second, as shown below, many of these platforms have developed
private governance regimes. These private systems include rules, standards,
detection, penalties, adjudication, and other elements. They apply to
significant portions of the worlds population as a result of the global growth
of these platforms. For example, most businesses have websites that are
indexed by Google and many people have Facebook pages. Both of these
platforms have private governance systems that among other things can
expel users from the platform for violations.
12. Lior Strahilevitz, Information Asymmetries and the Rights to Exclude, 104 MICH. L. REV.
1835, 1838 (2006).
13. Facebook, Inc., Registration Statement (Form S-1), at 1 (Feb. 1, 2012), available at
http://sec.gov/Archives/edgar/data/1326801/000119312512034517/d287954ds1.htm.
14. Id. at 75.
15. Id. at 50.
1206
11/6/2012 5:38 PM
Third, the ability of platforms to enforce rules concerning negative

externalities rests on being able to penalize and ultimately exclude members
of the community. That naturally leads to disputes that sometimes end up
before courts or regulatory authorities. These issues arise particularly when
penalties are levied against platform users that provide services that compete,
or might compete, with those provided by the platform. Many of the current
antitrust complaints against Google, for example, involve companies that
protest that their businesses have been harmed as a result of reductions in
their search ranking and that Google gives preferential treatment to its own
vertical search or price-compensation services.16 Understanding the role of
rules in policing negative externalities can help distinguish pro-competitive
from anti-competitive business practices of platform owners.
After providing a brief introduction to multi-sided platforms, Part II
situates the governance of negative externalities in the larger set of practices
in which multi-sided platforms engage to maximize the value they generate
for their communities as well as for themselves in the form of profits. Part
III describes sources of negative externalities and relates the problems faced
by multi-sided platforms to polities as well as other businesses that must deal
with negative externalities created by their customers. Part IV examines the
governance methods platforms have developed to manage these problems. It
draws on research concerning the business practices of multi-sided platforms
in a diverse set of industries and over time. Part V provides detailed
examinations of four economically significant industries that highlight
platform governance: social networks, stock exchanges, search engines, and
software platforms. Part VI analyzes the legal and policy issues that arise
from disputes involving platform governance. It considers the distinction
between efficient regulation of negative externalities and anti-competitive
exclusion as well as the use of social versus private control over negative
externalities in platform communities.
16. See Leo Cendrowicz, The E.U. Probe: Is Google Rigging Its Search Results?, TIME, Dec.
2, 2010, available at http://www.time.com/time/business/article/0,8599,2034138,00.html;
Companies Ask EU Commission To Step in On Google Search Ranking Complaint, ITPROPORTAL
(Feb. 24, 2010), http://www.itproportal.com/2010/02/24/companies-ask-eu-commissionstep-google-search-ranking-complaint/; Claire Cain Miller, Texas Probes Google on Ranking
of Search Results, N.Y. TIMES, Sept. 3, 2010, available at http://www.nytimes.com/2010/
09/04/technology/04google.html?adxnnl=1&adxnnlx=1314113487-4SfkO0V/SuFxNfcRM
dHkbQ; Kinderstart Sues Google Over Lower Page Ranking, REUTERS, Mar. 19, 2006, available at
http://www.usatoday.com/tech/news/2006-03-19-google-kinderstart_x.htm.
2012]
II.
11/6/2012 5:38 PM
1207
MULTI-SIDED PLATFORM STRATEGIES FOR

INCREASING VALUE
eBay creates value through the well-known process of exchange. An

individual finds an antique sewing machine in their attic but places little value
on it. Another individual collects antique sewing machines. eBay provides a
platform for those two individuals to find each other and make a trade. The
collector pays money to the sewing machine owner and the collector gets the
sewing machine. They are both better off as a result. The sewing machine
owner could have sold his machine to an antique store and the collector
could have gone to antique stores to find the sewing machine. The ecommerce platform can provide a more efficient means of commerce in
antique sewing machines because it can aggregate the demand of many
antique sewing machine owners and many buyers and help bring them
together to engage in trades.
All multi-sided platforms exhibit the same basic features that we see with
eBay. Although platforms can have more than two distinct groups of users,17
it is helpful to describe these basic features for a two-sided platform.
First, the platform brings together two types of users that can generate
value by coming together. For example, a platform might bring together a
man and a woman who are looking for companionship, a sender and receiver
of money, a mobile software application developer and a mobile phone user,
a search engine user and an advertiser, or many other combinations. These
two types of users have interdependent demand functions for platform
services in the sense that the demand by members of one group for the
platform depends positively on the ability to access and engage in value
exchange with members of the other group.18
17. Facebook, for example, is a four-sided platform. It is a communications platform

for senders and receivers of information. This communications platform is also open to
advertisers who want to reach the people that are using Facebook to communicate with each
other. It is further made available to entrepreneurs who develop software applicationssuch
as social gamesthat run on Facebook. There are interdependencies among these four
groups of economic users. Facebook, Inc., Registration Statement, supra note 13, at 1.
18. The two types of users do not necessarily need to value each other. It is sufficient
that it is possible to create net value by putting them together. A company may value
presenting advertising to a consumer but the consumer may be indifferent to the advertising
or even willing to pay to avoid it. So long as the value of presenting advertising to the
consumer is greater than the cost to the consumer of receiving it, there are potential gains to
trade. The role of a platform for advertising-supported media is to pay the consumer to be
exposed to advertising by providing content. In effect, the media platform owner uses some
of the money that the advertiser is willing to pay to reach a user to fund content creation
that incentivizes users to be exposed to advertising in return for free media content.
1208
11/6/2012 5:38 PM
Second, the platform provides a number of services to reduce the

transactions costs for these users to come together and to exchange value.
Those services could include facilities to search and match users, to figure
out exchange values, and to settle transactions. A typical financial exchange,
for example, helps agents find counterparties to a trade and provides facilities
for them to consummate a trade in addition to many other complementary
services.
Third, the platform owner maximizes profit by choosing prices and other
strategic variables that recognize the interdependencies between the two
groups of users. The economic literature has focused on pricing policies.19
The profit-maximizing prices must solve a coordination problem between
the multiple sides. A group of users will usually place no value on a platform
unless one or more of the other groups of users are also on the platform.
The profit-maximizing prices may be at or below marginal cost, and may be
zero or negative, and therefore reflect a type of subsidy to one side.20 (The
social welfare maximizing prices have the same characteristics, although there
is no guarantee that the privately and socially optimal prices will coincide.21)
In practice, multi-sided platforms use a wide variety of mechanisms to
generate value for platform users and to structure how much net value each
group of users receives.22 Platforms simultaneously determine how to
maximize the overall value of the platform for the users and the allocation of
this value among both user groups and the platform owner. Slicing the pie
differently results in bigger or smaller pies because of the interdependencies
between the groups. The platform owner therefore needs to figure out how
to slice the pie in order to make the pie as big as possible.
This Part describes how multi-sided platforms maximize and allocate
value and thereby provides a framework for understanding the role of
governance systems in this process. Section II.A describes the set of tools
that platforms have available to them. Section II.B shows how these tools are
19. See, e.g., E. Glen Weyl, A Price Theory of Multi-Sided Platforms, 100 AM. ECON. REV.
1642, 164954 (2010) (discussing, in part, various pricing models as they relate to network
platforms).
20. Id. at 1652; Jean-Charles Rochet & Jean Tirole, supra note 4, at 991, 992.
21. The privately optimal prices set by a multi-sided platform may differ from the
socially optimal prices if the platform sets prices too high and output too low. But it may
also select a pricing structure that does not solve the coordination problem between the
groups of users as efficiently as a social planner would. See Alexander White & E. Glen Weyl,
Insulated Platform Competition 10, 19, 29 (NET Inst. Working Paper No. 10-17, 2011), available
at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1694317.
22. See EVANS & SCHMALENSEE, supra note 5, at 1316; Eisenmann, Parker & Van
Alstyne, supra note 5, at 37; Boudreau & Hagiu, supra note 11, at 175.
2012]
11/6/2012 5:38 PM
1209
used to create value and in particular to generate positive externalities.

Section II.C examines the role of value distribution in solving the
coordination problem.
A.
THE PLATFORM TOOLKIT
Multi-sided platform businesses have a number of tools for solving

coordination problems that generate value for users. To begin with, they can
devise a platform design that facilitates the interaction of the different types
of users. YouTube, for example, recognized a latent demand for online video
sharing. To meet this demand it designed a software and hardware platform
that enabled people to upload and view videos. It included features that
helped people who uploaded videos to find an audience and helped people
who wanted to view videos to find ones they would enjoy.23
Platforms also provide specific packages of products or services to each
type of user to facilitate coordination. As with a single-sided business, the
platform has to decide on the optimal combination of product attributes and
price. The difference in the case of a multi-sided platform is that offerings
that induce users on one side to join the platform and interact often provide
value to the users on the other side.24 Software platform designers frequently
provide software developers with software developer kits (SDKs) and other
assistance to facilitate the efficient development of software that works on
the platform.25 End users benefit from those quality enhancements indirectly.
Prices are an important element of the toolkit for solving coordination
problems. The platform may charge users fees for access to and use of that
platform. Those prices can be adjusted to achieve an optimal combination of
users, and intensity of use, given their demands. That may entail having a
higher incremental profit margin from one side than the other. It also may
lead to implicit subsidies to some or all users on one side. OpenTable, for
example, charges restaurants for participating in its platform and a fee for
each reservation made through the platform; individual patrons can make
23. See Glenn Chapman, YouTube Redesigns Website to Keep Viewers Captivated, AFP (Mar.
31, 2010), http://www.google.com/hostednews/afp/article/ALeqM5jfGfKKsiwbxNv8Xo
Ubm8ZlRZZWyw (listing design features such as personalized viewing recommendations
and modified playlist tools).
24. See Mark Armstrong, Competition in Two-Sided Markets, 37 RAND J. ECON. 668, 668
(2006); Rochet & Tirole, supra note 6, at 659.
25. See, e.g., SDKs & Tools Facebook Developers, FACEBOOK, http://developers.
facebook.com/docs/sdks (last visited Feb. 27, 2012); iOS Dev Center Apple Developer,
APPLE, https://developer.apple.com/devcenter/ios/index.action (last visited Feb. 27, 2012);
Microsoft SDKs, MICROSOFT, http://msdn.microsoft.com/en-us/dd299405 (last visited Feb.
27, 2012); SDKs and Downloads, PAYPAL, https://cms.paypal.com/us/cgi-bin/?cmd=_
render-content&content_ID=developer/library_download_sdks (last visited Feb. 27, 2012).
1210
11/6/2012 5:38 PM
reservations and access other additional services such as reviews, directions,

and restaurant suggestions for free.26 OpenTable presumably decided that in
order to coordinate the demands of restaurants and consumers it needed to
offer services free to patrons to attract enough consumers to make the
platform valuable to restaurants.
Finally, and a central subject of this Article, platforms can develop and
employ governance systems that regulate the actions of participants. These
systems can rely on implicit or explicit contracts, detection mechanisms, and
penalties. Governance systems facilitate coordination and generate value for
users by preventing some users from reducing and possibly destroying the
value of the platform. OpenTable, for example, can prevent people who have
a record of not showing up for restaurant reservations from making further
reservations. That increases the value of the platform to restaurants. But it
also increases the value of the platform to users since, if OpenTable was an
unreliable source of reservations, restaurants might not use it, and a valuable
service might therefore not be available to consumers.
B.
VALUE CREATION AND EXTERNALITIES
These tools are important for achieving the fundamental economic

purpose of a platform: to release value by bringing users together. eBay
succeeded in doing this by creating a website where sellers could post
products for sale, developing an auction mechanism that allowed buyers to
bid for those products, providing a convenient payment mechanism that
enabled sellers to receive funds from buyers, implementing a rating system
that enabled buyers to communicate information about the sellers they dealt
with to other buyers, and devising rules to ensure the integrity of the bidding
and selling process. One study of eBay found that the buyers obtained
consumer surplus that averaged about $4 per purchase and totaled more than
$7 billion in 2003.27 The sellers earned surplus as well, equal to the difference
between what they netted from buyers and their personal valuations of the
goods.28
Platforms rely on the tools described above to maximize the value they
create for users overall subject to various constraints, including costs. A core
challenge is enabling users that can engage in mutually beneficial exchanges
26. See Farhad Manjoo, As OpenTable Booms, Who Gets the Dough?, FAST COMPANY (Apr.
1, 2011), http://www.fastcompany.com/magazine/155/as-opentable-booms-who-gets-thedough.html.
27. See Ravi Bapna, Wolfgang Jank & Galit Shmueli, Consumer Surplus in Online Auctions,
19 INFO. SYS. RES. 400, 400 (2008).
28. Id. at 405.
2012]
11/6/2012 5:38 PM
1211
to find each other. That is partly addressed through platform design. Online
dating sites such as eHarmony rely on detailed questionnaires to find matches
for people and then have a process through which people can narrow their
searches.29
The platform also needs to ensure that there are high-quality matches
in which the users can split significant value. For many multi-sided platforms,
the likelihood of high-quality matches increases with the number of
participants. To develop thicker markets, platforms use pricing and other
tools to drive participation and positive feedback effects. Some stock
exchanges, for example, provide subsidies to providers of liquidity.30 More
liquidity providers attract more liquidity takers that, in turn, drive more
liquidity providers.
C.
VALUE DISTRIBUTION AND COORDINATION
As these examples illustrate, value creation is intimately connected to

value distribution. A platform has to secure the participation of each side in
sufficient numbers to generate value. That involves solving a coordination
problem. Members of each group of users would benefit from being on the
platform but they will not join the platform unless enough members of the
other groups join as well.
The economic literature on multi-sided platforms demonstrates the role
of the pricing structure and other strategic decisions by the platform in
solving this coordination problem.31 By tilting the pricing structure so that
one side contributes relatively more incremental margin and the other side
contributes relatively less incremental margin, the platform can potentially
entice enough members of each group to join.32 Once they do, positive
feedback effects can fuel growth.
These economic models focus, for simplicity, on the prices that agents
are charged.33 More generally, however, platforms provide net value to
members of each group where that net value is the difference between the
29. See eHarmony Membership Options, EHARMONY, http://www.eharmony.com/singles/
servlet/about/membership (last visited Feb. 22, 2012).
30. See NYSE Euronext, Annual Report (Form 10-K), at 22 (Feb. 28, 2011), available at
http://www.sec.gov/Archives/edgar/data/1368007/000095012311019827/y86275e10vk.htm.
31. See Daniel F. Spulber, Solving the Circular Conundrum: Communication and Coordination in
Internet Markets, 104 NW. U. L. REV. 537, 53839 (2010); Caillaud & Jullien, supra note 4, at
322.
32. See Wilko Bolt & Alexander F. Tieman, Heavily Skewed Pricing in Two-Sided Markets,
26 INTL J. INDUS. ORG. 1250, 125051 (2008); Rochet & Tirole, supra note 6, at 659; Marc
Rysman, The Economics of Two-Sided Markets, 23 J. OF ECON. PERSP. 125, 130 (2009).
33. See Rysman, supra note 32, at 129.
1212
11/6/2012 5:38 PM
total value received and the total cost incurred by that member. Platforms
solve the coordination problem through adjusting what users receive34 as well
as how much they pay. Apple provides many features on its phones that
users find attractive. It also provides software developers with an operating
system, tools, and a store for selling applications that developers find
appealing.35
In determining the relative benefits realized by each group of agents, the
platform necessarily makes decisions that allocate benefits between different
groups of users. All else equal, charging one group less means charging
another group more. The point extends beyond pricing. Platforms make
design and other decisions that shift the relative benefits between the two
sides. Shopping malls, for example, sometimes place anchor stores that
attract the most shoppers as far apart as possible, put up and down escalators
far apart, and make other physical design decisions to increase the foot traffic
in front of stores.36 Malls therefore convey an added benefit on the stores
who pay for space, while imposing some costs on shoppers who get in for
free.37 The stringency of the governance system for different types of users,
including the penalties imposed, determines, in part, the distribution of value.
III.
BAD BEHAVIOR AND PLATFORM COMMUNITIES
A negative externality arises when economic agent A imposes costs on

economic agent B that agent B has neither agreed to bear nor has received
compensation for.38 These costs arise in many ways and include many forms
of behavior that range from the seemingly innocuous to the obviously
egregious, as the following examples illustrate.
At some point when you enter a highway you increase congestion
and impose costs on all other drivers.
A fisherman increases the likelihood that the fishing grounds will
be depleted and therefore imposes costs on other fishermen.
A factory emits smoke that imposes costs on many people who
live in the surrounding area.
34. That is, the platform may decide to shift the demand schedule for a group of users
to the right or the left by providing more or fewer product attributes that these users value.
35. See iOS Dev Center Apple Developer, supra note 25.
36. See Malcolm Gladwell, The Terrazzo Jungle, THE NEW YORKER, Mar. 15, 2004, at
12224.
37. For a general analysis of strategies in which platforms increase consumer search
costs, see Andrei Hagiu & Bruno Jullien, Why do Intermediaries Divert Search?, 42 RAND. J.
ECON. 337 (2011).
38. See Mankiw supra note 3, at 20307.
2012]
11/6/2012 5:38 PM
1213
An investment manager engages in a Ponzi scheme that will

ultimately collapse and harm the later investors.
A local gang attacks its rivals.
The perpetrator does not bear any costs for imposing the externality and
indeed may benefit from it. In the absence of laws and regulations, agents
impose high costs on society. As is well known, left unchecked negative
externalities reduce consumer welfare. For the examples above, rules that
minimize negative externalities include:
Toll booths or congestion charges that reduce the incentives to
drive.
Pollution taxes or tradable emission permission taxes that limit
pollution.
Fishing limits, including penalties for exceeding ones quota.
Prison sentences and fines for engaging in investment fraud.
Numerous criminal penalties for gang violence.
Many of the rules, both civil and criminal, that communities adopt can be
viewed as solutions to negative externalities. Polities ban many sorts of
behavior on the grounds that they worsen life in the community. Changing
social norms alter the costs and benefits, and formal societal rules evolve as a
result (for instance, rules defining the boundaries of unlawful pornography).
Platforms create communities of users with shared interests who benefit
from interacting together. Successful platforms have identified positive
interdependencies between users, figured out how to reduce transactions
costs between these users, and determined price and non-price mechanisms
for bringing these users together, thereby coordinating them into a
community. Some users, however, may impose negative externalities in many
of the same ways and for many of the same reasons as in polities. They may
engage in behavior that reduces the value of the platform and that cannot be
mitigated through pricing structures. As with polities these actions may range
from the seemingly innocent to the obviously bad. The negative externalities
that occur on multi-sided platforms are similar to those that occur in polities,
as we show next.
1214
A.
11/6/2012 5:38 PM
IT OUGHT TO BE A CRIME
Platform users may engage in various forms of behavior that have no

redeeming virtues and that are analogous to practices that are criminal
violations in polities.39
Platform users may engage in myriad types of fraud and
misrepresentation. Fraud may occur when merchants sell counterfeit goods40
or accept payment but then do not ship the goods.41 Brokers on exchange
platforms could engage in front-runningthat is, profiting from information
on trades placed by clients possibly to the detriment of the client. Firms can
encourage their employees to click on a competitors ads on search engines
to impose costs on them.42 Misrepresentation may also occur when people lie
about their age or their weight on dating sites,43 the true nature of a
smartphone application,44 or the popularity ranking of their websites.45 This
list does not exhaust the ways in which opportunistic behavior can occur on
platforms.
Polities have rules to prevent people from assaulting other people,
engaging in bodily harm, committing libel, and causing mental distress. These
issues can arise on multi-sided platforms as well. Interactions in nightclubs
and other physical and virtual dating venues can result in bodily harm.
Craigslist, for example, has been used by sexual predators to meet, and in
some cases kill or rape, victims.46 Social networks can be used for inflicting
39. See Richard A. Posner, An Economic Theory of the Criminal Law, 85 COLUM. L. REV.
1193, 119597 (1985) (framing crimes as market bypasses creating net disutility).
40. See Rob Unsworth, Amazon Marketplace Offering Dangerous Goods, BBC (Dec. 8, 2008),
http://www.bbc.co.uk/blogs/watchdog/2008/12/amazon_marketplace_offering_da.html.
41. See Amazon Marketplace and Ebay Fraud, ANDRE GUNTHER PHOTOGRAPHY,
http://www.aguntherphotography.com/blog/amazon-marketplace-ebay-fraud.html (last visited
Aug. 23, 2011).
42. See Dan Shipe, Is AdWords Click Manipulation Taking Money From Your Pocket?,
ARTICLE/.COM (Aug. 29, 2008), http://www.articleslash.net/Internet-and-BusinessesOnline/PPC-Advertising/475634__Is-AdWords-Click-Manipulation-Taking-Money-FromYour-Pocket.html.
43. See Illusion or Deception?? Why do some people misrepresent themselves on dating sites?,
EHARMONY ADVICEMEMBER DISCUSSION FORUM, http://advice.eharmony.com/
boards/dating-advice/dating/50844-illusion-deception-why-do-some-people-misrepresentthemselves-dating-sites-3.html (last visited Aug. 1, 2011).
44. See Claudine Beaumont, Google remotely deletes Android apps, THE TELEGRAPH (Jun.
25, 2010), http://www.telegraph.co.uk/technology/google/7854560/Google-remotelydeletes-Android-apps.html.
45. See Tom Espiner, Google blacklists BMW.de, CNET NEWS (Feb. 6, 2006),
http://news.cnet.com/Google-blacklists-BMW.de/2100-1024_3-6035412.html.
46. See Abby Goodnough, Medical Student Is Indicted in Craigslist Killing, N.Y. TIMES, June
21, 2009, at A13, available at http://www.nytimes.com/2009/06/22/us/22indict.html.
2012]
11/6/2012 5:38 PM
1215
emotional distress.47 A highly publicized case involved the use of a social

network by Lori Drew to retaliate against a young girl, Megan Meier, who
had a disagreement with Drews daughter.48 Using a fake account under
which Drew assumed the false persona of a teenage boy, Drew orchestrated
an online romance with Meier, had the fake boy become hostile, and
eventually suggested that Meier kill herself. Soon after, Meier committed
suicide.49
Interactions on multi-sided platforms can involve behavior that some
users find offensive. This is no different than a regular community. People
may incur costs as a result of unwanted exposure to hate speech,
pornography, violent images, and other offensive content. Even if they are
not exposed to this content, they may dislike being part of a community in
which such behavior takes place.
B.
POOR AND ASYMMETRIC INFORMATION
A standard problem in the exchange of value is that one party has

information that the other party does not have. Lack of information imposes
costs on actual and potential trading partners. In the extreme this can lead to
market breakdowns as a result of the lemons problem.50 The collapse of
the videogame market in the United States in 1983 has been attributed to a
lemons problem. Consumers could not distinguish low quality from high
quality games before buying them. Producers therefore had incentives to
create cheaper low quality games that drove the high quality games out of the
market. But consumers did not want to buy video game consoles to run low
quality games.51
More generally, asymmetric information reduces market efficiency by
reducing the likelihood that users will find the matches that maximize the
47. See, e.g., Yancy v. U.S. Airways, Inc., No. Civ. A. 10-983, 2011 WL 2945758 (E.D.
La. July 20, 2011) (embarrassing pictures of employee were posted on Facebook); Maremont
v. Susan Fredman Design Group, Ltd., No. 10 C 7811, 2011 WL 6101949 (N.D. Ill. Dec. 7,
2011) (employer used employees personal Twitter and Facebook accounts to advertise while
employee was in the hospital).
48. United States. v. Lori Drew, 259 F.R.D. 449 (C.D. Cal. 2009).
49. See Mom: MySpace Hoax Led to Daughters Suicide, ASSOCIATED PRESS (Nov. 16, 2007),
http://www.foxnews.com/story/0,2933,312018,00.html.
50. See George A. Akerlof, The Market for Lemons: Quality Uncertainty and the Market
Mechanism, 84 Q. J. OF ECON. 488 (1970) (owners of cars that are lemons are more likely to
sell them; since consumers cannot distinguish lemons from good cars, the sale prices of
good cars are depressed; that leads owners of good cars not to sell; the market for reselling
quality cars is therefore destroyed).
51. See DAVID S. EVANS, ANDREI HAGIU & RICHARD SCHMALENSEE, INVISIBLE
ENGINES: HOW SOFTWARE PLATFORMS DRIVE INNOVATION AND TRANSFORM INDUSTRIES
12425 (2006).
1216
11/6/2012 5:38 PM
total value from trade. Several studies have found that requiring corporate
bond traders to disclose information on trading prices resulted in improved
efficiency and substantially lower trading costs.52
Opportunistic behavior resulting from asymmetric information may
increase the uncertainty for people and companies that are considering using
a platform. Traders may prefer platforms that have transparent pricing and
social network users may prefer platforms where information about people is
reliable.
C.
CONGESTION AND OPTIMIZING PHYSICAL SPACES
Negative externalities can result from increasing the number of users for
multi-sided platforms. Physical platforms face congestion problems. A
nightclub provides a trivial example. Too many people will make it harder for
people to mingle and enjoy themselves. A shopping mall provides a more
interesting example. An increase in the number of merchants may increase
search costs and therefore harm other sellers as well as buyers.
Multi-sided platforms have to design and manage their spaces to reduce
negative externalities as well as to promote positive externalities. Expanding
the square footage of a mall to accommodate more stores imposes costs on
shoppers who have to walk farther on average. Similar considerations apply
to virtual platforms. Search engines need to make decisions on how many
results and advertisements to display on each page and in what format. They
need to do this to promote positive externalities and also to limit congestion
and reduce search costs.53
D.
CASE IN POINT: DECORMYEYES
Vitaly Borkers strategies for selling eyewear on the web highlight almost
all these forms of bad behavior; they also provide a lesson in what happens
52. These studies concern the introduction of the TRACE (Trade Reporting and
Compliance Engine) system for corporate bonds in the United States. See Hendrik
Bessembinder, William Maxwell & Kumar Venkataraman, Market Transparency, Liquidity
Externalities, and Institutional Trading Costs in Corporate Bonds, 82 J. OF FIN. ECON. 251 (2006);
Amy Edwards, Lawrence Harris & Michael Piwowar, Corporate Bond Market Transaction Costs
and Transparency, 62 J. OF FIN. 1421 (June 2007); Michael Goldstein, Edith Hotchkiss & Erik
Sirri, Transparency and Liquidity: A Controlled Experiment on Corporate Bonds, 20 REV. OF FIN.
STUD. 235 (2007). For a summary, see Hendrik Bessembinder & William F. Maxwell,
Transparency and the Corporate Bond Market, 22 J. OF ECON. PERSP. 217 (2008).
53. Eye-tracking studies of how consumers use search engines reveal that both the
quality of the consumer experience and the attention paid to advertising varies substantially
with the layout and organization of the search results page. See, e.g., CatalystGroup, Google vs.
Bing: Search Engine Preference, June 2009, available at http://www.catalystnyc.com/cofactors/
wp-content/uploads/2009/06/catalyst-eye-tracking-bing-vs-google-may-2009.pdf.
2012]
11/6/2012 5:38 PM
1217
when platform governance is imperfect. Borker learned that search engines

did not distinguish between good and bad cites to his website.54 He
responded to complaints with highly offensive emails and posts that
generated even more complaints.55 He told a New York Times reporter, Ive
exploited this opportunity because it works. No matter where they post their
negative comments, it helps my return on investment. So I decided, why not
use that negativity to my advantage?56
Borker, who used the aliases Tony Russo and Stanley Bolds, received
many complaints because he engaged in fraudulent behavior, including
adding spurious charges to customers payment cards.57 When people
pursued their complaints against him, Borkersometimes using one of his
aliasesthreatened them with bodily harm including death or rape.58 He
threatened one customer with sexual violence when she said she was going to
have her credit card issuer reverse an overcharge.59 Borker later sent the
customer pictures of the outside of her apartment building in a further
attempt to intimidate her.60
Borker relied heavily on two web platforms for these practices. He
fulfilled his orders from sellers on eBay who were directed to ship to his
customers.61 He also used search engine rankings to drive business.62 Initially,
the governance systems of these platforms failed to thwart him. In the case
of search results, he had discovered a way to manipulate the search algorithm
as well as the detection methods then in place for identifying efforts to
distort results.63 After the story was reported, Google developed an algorithm
to detect efforts to increase search rankings by encouraging bad comments.64
54. See David Segal, A Bully Finds a Pulpit on the Web, N.Y. TIMES, Nov. 28, 2010,
available at http://www.nytimes.com/2010/11/28/business/28borker.html.
55. Id.
56. Id.
57. See id.
58. See Google hitman: Eyeglasses seller threatened customers with rape and murder to
elevate his website on search engines, DAILY MAIL, May 15, 2011, available at
http://www.dailymail.co.uk/news/article-1387310/Google-hitman-Vitaly-Borker-threatenedcustomers-rape-murder.html.
59. See Segal, supra note 54.
60. Id.
61. See id.
62. See id.
63. Id.
64. Alyson Shontell, Making Customers Hate You Makes Google Love You, BUS. INSIDER,
Dec. 3, 2010, http://www.businessinsider.com/making-customers-hate-you-makes-googlelove-you-2010-12.
1218
11/6/2012 5:38 PM
Meanwhile, public law took care of Borker, who pled guilty to wire fraud,
mail fraud, and sending threatening communications.65
E.
BAD BEHAVIOR AND PLATFORM VALUE
Many forms of bad behavior, such as fraud and misrepresentation, clearly

reduce the value of the platform. Like many forms of criminal behavior, this
behavior has no redeeming virtues. However, some behavior that imposes
costs on other participants in the platform and therefore seems bad also
provides benefits. When these benefits outweigh the costs, the behavior does
not generate negative externalities, on net, and therefore is not behavior that
the platform would necessarily want to discourage.
That could be the case, for example, with asymmetric information in
some cases. Users may need incentives to invest in acquiring information,
and the ability to capitalize on their control over that information may
provide those incentives. The benefits from increased dynamic efficiency
from investment in gathering information may outweigh the losses from
lower static efficiency as a result of not sharing that information.66 Greater
transparency on social networks has its costs too. Information that increases
the value of one relationship may decrease the value of another relationship.67
IV.
DEALING WITH BAD BEHAVIOR
Multi-sided businesses could simply rely on civil and criminal law and
government regulation to deal with the negative externalities that arise on
their platforms. Users have recourse to laws involving breach of contract,
fraud, market manipulation, assault and battery, and intentional infliction of
emotional distress to deal with many of the problems discussed above.68
Laws and regulations have tackled asymmetric information problems through
65. Kathy Kristof, Online Retailer Guilty of Fraud, Threats, CBS MONEYWATCH (Dec.
6, 2010), http://moneywatch.bnet.com/saving-money/blog/devil-details/onlineretailer-charged-with-fraud-threats/3478/.
66. Studies of increased transparency in bond markets found that transparency may
have reduced liquidity, leading to some markets being inefficiently thin. Hendrik
Bessembinder & William Maxwell, Markets: Transparency and the Corporate Bond Market, 22 J.
OF ECON. PERSP. 217, 22829 (2008).
67. See Danah Boyd, None of This Is Real: Identity and Participation in Friendster, in
STRUCTURES OF PARTICIPATION IN DIGITAL CULTURE 132, 145 (Joe Karaganis ed., 2007).
For the record, Ms. Boyd does not capitalize her first or last name. This imposes a negative
externality on people, like me, who then have to deal with editors and readers who are not in
on the deviation from grammatical rules.
68. See, e.g., Complaint at 8, Thompson v. Facebook, No. 1:09-cv-22927-KMM (S.D.
Cal. Sept. 29, 2009); First Amended Complaint, Yingling v. eBay, No. C-09-01733 (N.D. Cal.
June 16, 2009).
2012]
11/6/2012 5:38 PM
1219
mandatory disclosure laws, mandatory cooling-off periods, and return

policies.69 New multi-sided platform businesses can give rise to novel issues
such as cyber bullying. Governments can pass new laws in response to this,
as the State of Missouri did after the Megan Meier suicide and the
unsuccessful prosecution of the instigator.70
Some multi-sided platforms have, however, developed their own
mechanisms for dealing with bad behavior. They adopt rules for the users on
one or more sides of the platform, institute reporting and detection
mechanisms in order to uncover violations of these rules, require evaluations
of the evidence through mandatory arbitrations, impose penalties, and
sometimes even allow appeals from the initial evaluation as shown below.
There are two reasons for platforms not to rely entirely on the public
sector. The first is that the platforms generally are able to enforce rules to
reduce negative externalities more rapidly and efficiently than the public
sector.71 Search engines can develop algorithms for detecting efforts to
manipulate search rankings and de-list websites that are trying to take
advantage of users or demote them in the search rankings. An e-commerce
platform can decide after a few complaints to drop a merchant from its site,
and exchanges can debar traders. In addition to the inherent differences in
the efficiency of public and private actors, the fact that the public sector
must give people and businesses rights of due process (which society has
found necessary to check the governments enormous powers over its
citizens) necessarily makes the enforcement of laws and regulations by the
public sector more time consuming and expensive. A platform owner, for
example, can monitor a users behavior without showing probable cause. A
platform can therefore provide value to its community by providing an
efficient governance system.
The second reason for a platform to take action is that public laws and
regulations may be incomplete when it comes to policing negative
externalities on platforms. The government may not have recognized (or
recognized too slowly) a problem such as cyber bullying on social networks
or the opportunistic inflation of search rankings. It may have other objectives
or obligations, such as the protection of free speech that deter or preclude it
69. See, e.g., Truth in Lending Act, 15 U.S.C. 1601 (1968); Magnuson-Moss Warranty
Act, 15 U.S.C. 2301 (1975); FTC Cooling-Off Rule, 16 C.F.R. 429 (1995).
70. See Associated Press, Mo. Internet Harassment Bill Passed after MySpace Suicide, USA
TODAY, Jul. 1, 2008, available at http://www.usatoday.com/news/nation/2008-06-30internet_N.htm.
71. For a related discussion see Henry E. Smith, Property and Property Rules, 79 N.Y.U L.
REV. 1719 (2004); Henry E. Smith, Exclusion and Property Rules in the Law of Nuisance, 90 VA.
L. REV. 965 (2004).
1220
11/6/2012 5:38 PM
from enforcing rules that private parties may decide to adopt themselves
such as against hate speech or pornography. The government may also
decide not to pursue various problems simply because the cost of doing so
including unintended consequences that often result from government laws
and regulationsexceeds the likely benefits. The platform has more
information about the problems, can react more quickly to them, and can
modify its governance mechanisms more quickly if they are not working or
are having perverse effects.
A private platform does not, of course, have the same range of
instruments available to it as a public entity does. It cannot issue search
warrants, engage in wiretaps, conduct dawn raids, put people in jail, or debar
wrongdoers from anything other than participating on the platform itself.
Unless it invokes public laws, for example by filing a breach of contract
claim, it cannot compel the discovery of information for an investigation.
Nor can a private platform collect penalties unless it requires a bond or
collects it as a condition of the agent having continued access to the
platform.
In fact, all of the platforms powers, aside from contract and other rights
that it would have to enforce in public courts, rest in its property rights over
the platform including, most importantly, its ability to exclude. This section
elaborates on this point before turning to an overview of platform
governance concerning negative externalities.
A.
THE PROPERTY RIGHTS FRAMEWORK
Strahilevitz has presented a useful framework for analyzing how property

rights can be used to deal with information asymmetries.72 The right to
exclude has four subordinate rights: (1) the Hermits Right to keep all agents
off the owners property; (2) the Bouncers Right to admit agents selectively
to the property and therefore to eject agents selectively from the property; (3)
the Exclusionary Vibe which uses social and psychological sanctions to
discourage some agents from entering the property; and (4) the Exclusionary
Amenity which uses club goods to sort desirable and undesirable entrants.73
72. See Lior Jacob Strahilevitz, Information Asymmetries and the Rights to Exclude, 104
MICH. L. REV. 1835, 1837 (2006). Boudreau & Hagiu, supra note 11, at 169, emphasize the
Bouncers Right identified by Strahilevitz. However, in the context of their analysis of the
general solution by platforms of market failures, platforms also use the Exclusionary Vibe
(e.g., a magazine for a niche audience) and the Exclusionary Amenity (e.g., a discount
department store as an anchor in a mall).
73. Strahilevitz, supra note 72, at 1837.
2012]
11/6/2012 5:38 PM
1221
He argues that the last three of these rights are substitutable methods for
maximizing the value of the property.74
Strahilevitzs framework generally maps well into the tools that multisided platforms use to optimize the value of their property. Some platforms
exercise the Hermits Right through design decisions75 and start out life
single-sided. For example, the Palm Pilot created their own applications and
did not allow others to do so for about eighteen months after its launch.76
Almost all of the strategies for reducing negative externalities depend on the
exercise of the Bouncers Right. This is discussed in more detail in the next
Section. The Exclusionary Vibe and Exclusionary Amenity are used by many
platforms to attract a particular group of users on one side that is valuable to
a particular group of users on the other side. There is a blurry line between
the two strategies for multi-sided platforms. Niche magazines are an
example. Runners World is designed to attract runners and companies that
want to sell to them.77 The vibe and amenity go togethercompanies that
want to advertise to runners are attracted to the amenity by an aggregation of
runners created by the vibe generated by the content (not to mention the
title).
Smith, who Strahilevitz builds on, argues that property rights and
governance are substitutes from the standpoint of maximizing social
welfare.78 The idea is that there are some market failures that governments
can resolve precisely through laws and regulation. There are others that
private parties can solve through the blunt instrument of property rights
because they have better access to information.
Multi-sided platforms can be analyzed in this framework. They represent
the interests of a communityalbeit a private and voluntary onejust as a
government does, and are perhaps even more motivated than the
government to maximize, at least approximately, the social wealth of that
community. The platform owner also has incentives to take the long-run
interests of the community into account since it is maximizing the long-run
value for itself or its shareholders. The platform often uses, among other
74. Id. at 1861.
75. See EVANS, HAGIU & SCHMALENSEE, supra note 51, at 16869.
76. Id.
77. See RUNNERS WORLD, http://www.runnersworld.com (last visited Feb. 29, 2012).
The website homepage features a banner ad for the book MARATHON: The Ultimate
Training Guide; the Shoes & Gear section includes a Store Finder tool, and the Nutrition &
Weight Loss section displays a side ad for the book The Lean Belly Prescription: The fast and
foolproof diet and weight-loss plan from Americas top urgent-care doctor.
78. See Henry E. Smith, Exclusion Versus Governance: Two Strategies for Delineating Property
Rights, 31 J. OF LEGAL STUD. S453 (2002).
1222
11/6/2012 5:38 PM
things, a governance system for dealing with negative externalities among

platform users. But for the platform, property rightsand the bundle of
rights to exclude identified by Strahilevitz, and in particular the Bouncers
Rightare necessary for governance.
B.
SELECTIVE EXCLUSION
Platform governance generally consists of a set of rules for platform

agents that proscribe certain actions by these agents or compel certain other
actions. These rules can be used to increase positive externalities. For
example, card networks require banks to insert acceptance marks on cards
and merchants to post acceptance marks; this makes it easier for cardholders
and merchants who use the same payment method to find each other.79 More
commonly, though, rules are designed to eliminate or mitigate negative
externalities. These rules need to have consequences to be meaningful. Those
consequences can involve partial or full exclusion from the platform, or its
benefits, for some period of time, perhaps forever.80 The platform also needs
to be able to detect violations for these rules to be meaningful. That could be
a combination of proactive detection or response to complaints. And finally,
the platform may employ a process in which suspected wrongdoers can plead
their cases, or at least convey potentially useful information, and possibly an
appeals process.
The Portobello Road Antique Dealers Association in London provides
an example.81 A number of antique dealers have located on Portobello Road
in London. That is a common situation in the economic geography of
agglomeration and is an example of a platform that emerges naturally without
any necessary ownership. But some of these dealers decided to start an
association to address common issues. One of those issues involved creating
and maintaining a high quality brand. For this purpose they adopted a code
of ethics. Members are required to post the price and as much information as
possible about the item. The code also prohibits members from
misrepresenting antiques or misleading their customers. The association also
provides a dispute resolution service for customers who believe they have
gotten a bad deal. Members who violate the code can be bounced from the
79. See, e.g., Visa International Operating Regulations, VISA, October 15, 2011, at 15354
(ma r k s r e q u i r e d o n c a r d s ) a nd 4 34 ( ma r ks r e q u i r e d a t po i nt - of - s a le ) ,
http://usa.visa.com/merchants/operations/op_regulations.html.
80. See, e.g., Statement of Rights and Responsibilities, FACEBOOK, http://www.facebook.com
/legal/terms (last visited Feb. 29, 2012); Your User Agreement, EBAY, http://pages.ebay.com/
help/policies/user-agreement.html (last visited Feb. 29, 2012) [hereinafter eBay User
Agreement].
81. See EVANS & SCHMALENSEE, supra note 5, at 11011.
2012]
11/6/2012 5:38 PM
1223
association and therefore lose access to the credibility signal that the
association provides to customers.82
Other commerce platforms have similar rules. eBay has a detailed user
agreement for buyers and sellers.83 It tells users that eBay has the right to
restrict their access to the site in various ways, including full termination, if
the user abuses the site.84 The user agreement includes a mandatory dispute
resolution mechanism for buyers and sellers. eBay has detailed rules for
buyers85 and sellers86 that prohibit a variety of actions that could result in
negative externalities. A major concern is the integrity of the auction process.
For example, buyers are not allowed to bid on items offered by sellers they
know personally. Sellers who are banned from the site can appeal that
decision.87
These types of rules solve several possible externality problems. Consider
the case of payment cards. Card users likely value certainty over the prices
they will pay when they use their cards at accepting merchants; they also
likely value the certainty that merchants with a sign indicating they accept the
networks card actually do so. To the extent that merchants impose
surcharges or refuse cards, they impose costs not only on the cardholders
affected by these decisions but also cardholders generally through the
introduction of uncertainty. In addition, merchants that surcharge cards or
82. If it is reported that a member violates the code, we interview that member to
hear he/she has to say and take action accordingly. This may be a verbal or written warning
not to do it again or more severe action including expulsion from the Association. E-mail
from Leyla Ozyurt, Portobello Market Association, to Jacqueline Murphy, Consultant,
Market Platform Dynamics (Mar. 18, 2012, 13:05:00 CST) (on file with author).
83. See eBay User Agreement, supra note 80.
84. The user agreement says Without limiting other remedies, we may limit, suspend
or terminate our service and user accounts, prohibit access to our sites and their content,
services and tools, delay or remove hosted content, and take technical and legal steps to keep
users off the sites if we think that they are creating problems or possible legal liabilities,
infringing the intellectual property rights of third parties, or acting inconsistently with the
letter or spirit of our policies (for example, and without limitation, policies related to shill
bidding, conducting off-eBay transactions, feedback manipulation, circumventing temporary
or permanent suspensions or users who we believe are harassing our employees or other
users). Additionally, we may, in appropriate circumstances and at our discretion, suspend or
terminate accounts of users who may be repeat infringers of intellectual property rights of
third parties. We also reserve the right to cancel unconfirmed accounts or accounts that have
been inactive for a long time, or to modify or discontinue eBay sites, services or tools. Id.
85. See Rules for BuyersOverview, EBAY, http://pages.ebay.com/help/policies/buyerrules-overview.html (last visited Feb. 29, 2012).
86. See Rules for SellersOverview, EBAY, http://pages.ebay.com/help/policies/sellerrules-overview.html (last visited Feb. 29, 2012).
87. See HelpSuspended accounts, EBAY, http://pages.ebay.com/help/account/
suspended-accounts.html (last visited Apr. 10, 2012).
1224
11/6/2012 5:38 PM
refuse to accept cards selectively may impose costs on other merchants by

degrading the overall quality of this form of payment. There is a further
negative externality. Some merchants may use the desire to pay with a card as
a method for implementing price discrimination. On average, consumers that
want to pay with cards are less likely to have another equally convenient
payment method and may therefore be willing to pay a higher price to the
merchant. This may be a profit-maximizing strategy, especially when it is
unlikely the consumer will be a repeat customer (e.g., a tourist).
A common problem for dating sites involves preventing unwanted
approaches. Sites such as eHarmony check their users against lists of
registered sex offenders.88 They also do not allow users to search for profiles.
Instead the site matches profiles using its algorithm and pre-screens each
candidate. An introduction is made only if both parties agree to this. At that
point, individual identifying information is made available to both parties.89
eHarmony also provides a service whereby users can report problems and
eHarmony can take actions including removing offending individuals from
its service.90
While many multi-sided platforms have governance systems to limit
negative externalities, others do not or have quite limited ones. Advertisingsupported media tend to have very limited screening of ads. They often
prohibit advertisements that would be offensive to their readers.91 In the next
88. See Safety Tips, EHARMONY, http://www.eharmony.com/safety/tips (last visited

Feb. 29, 2012). The Do Your Own Research Section states that eHarmony employs
screening tools, including checking U.S. subscribers against sex offender registries in the
United States. See id.
89. See eHarmony Tour, EHARMONY, http://www.eharmony.com/tour (last visited Apr.
10, 2012); eHarmony is More than Traditional Dating Sites, EHARMONY,
http://www.eharmony.com/why/ (last visited Apr. 10, 2012).
90. Pam Holmgren, A Safety Reminder from eHarmony, EHARMONY BLOG (Apr. 18,
2011), http://advice.eharmony.com/blog/2011/04/18/a-safety-reminder-from-eharmony-2/
(If you ever are concerned about one of your matches for any reason, please send an email
to matchconcerns@eharmony.com so we can investigate and take appropriate action. We
have a team dedicated to our members safety and close accounts immediately when we
receive a credible complaint about someones suspicious behavior. Well notify you
whenever someone is removed from the service, so please pay attention to those emails if
you ever receive one and stop all communication with that person.).
91. The Chicago Tribune does not intentionally take advertisements for sex services,
but the paper does not police false advertising or provide readers with any mechanism for
complaining about advertisers they have interacted with as a result of seeing an
advertisement in the media. See Advertiser ServicesHow It Works, CHI. TRIB.,
http://www.chicagotribune.com/advertiser/how-it-works/ (last visited Mar. 1, 2012).
Craigslist had, in the past, accepted the advertising of adult services on its site, but it has
since removed the category. See Claire Cain Miller, Under Fire From Critics, Craigslist Blocks
2012]
11/6/2012 5:38 PM
1225
section we will see how MySpace, the leading social network site in the
United States in the mid 2000s, had a very limited governance system initially
and imposed some rules only in response to significant media and
governmental pressure.
The rules discussed above are generally enforced using the Bouncers
Right. Users that violate the rules can be ejected from the platform. Some
peoplesuch as known sexual predators on dating sitesare barred from
entering the platform in the first place.
C.
INFORMATION AND TRANSPARENCY
Multi-sided platforms also provide information to deal with negative

externalities. That has become increasingly common as a result of the
development of Internet and web technologies. eBay Motors has reduced the
lemons problem by providing ratings on automobile dealers. eBay provides a
mechanism for consumers to rate merchants after they have made a
purchase. They send consumers reminders to provide these ratings.92 A
consumer can minimize the likelihood of getting a car with undisclosed
problems by buying from an automobile dealer that has a very high rating.
Automobile dealers presumably know that a negative rating can have a
serious effect on their ability to make sales. The reviews limit the ability of
automobile dealers to take advantage of consumers by exploiting asymmetric
information. They also limit the ability of dealers to impose negative
externalities on each other since good dealers tend to drive out bad dealers as
consumers lower their expectations on the quality of cars they get from bad
dealers.93 Similar rating systems are common now on web-based platforms
that connect buyers and sellers.94
Access to Adult Services Pages, N.Y. TIMES, Sept. 4, 2010, available at

http://www.nytimes.com/2010/09/05/technology/05craigs.html.
92. See eBay Updates to Payment and Leave Feedback Reminder Emails Sent to Buyers, EBAY,
http://pages.ebay.com/sellerinformation/news/reputationemails.html (last visited Feb. 29,
2012) (Because the Feedback system has such an important role on eBay in measuring a
sellers quality of service, an email reminding buyers to leave Feedback is sent by eBay as a
consistent part of all transactions.).
93. For example, in recently purchasing an automobile on eBay, the author found that
it was possible to limit consideration to dealers with 100 percent satisfaction ratings. There
was no need to consider dealers that had complaints.
94. See, e.g., Rating a Seller, AMAZON, http://www.amazon.com/gp/help/customer/
display.html?nodeId=537806 (last visited Feb. 29, 2012); Feedback scores, stars, and your
reputation, EBAY, http://pages.ebay.com/help/feedback/scores-reputation.html (last visited
Feb. 29, 2012); How do I leave feedback?, ETSY, http://www.etsy.com/help/article/102 (last
visited Feb. 29, 2012).
1226
11/6/2012 5:38 PM
The provision of information is often an application of the Exclusionary

Vibe. The multi-sided platform is exercising its property rights when it
collects information from users on the platform about other users and makes
that information publicly available. It does not bounce users that engage in
opportunistic behavior, or users that create negative externalities as a result
of the lemons problem, but it does establish a mechanism that tends to drive
low quality users off of the platform.
The Exclusionary Vibe can be used to reduce negative externalities in
other ways than the direct provision of information. An example is JDate,
which advertises itself as the premier Jewish singles community.95 One would
expect that having Gentiles would impose negative externalities on Jews
looking for other Jews since it would increase their search costs and reduce
matching efficiency. It would not appear that the site has any way to verify
religious or ethnic background, so it could not specifically exclude non-Jews.
But it can establish a vibe that this is a site meant for Jewish men and women
to meet each other. It also encourages Gentiles to identify themselves.96
V.
GOVERNANCE REGIMES FOR KEY PLATFORMS
Three economically prominent platform typessocial networks, stock

exchanges, and search enginesillustrate the role of governance systems, the
methods that are chosen for these governance systems, and the tensions that
governance systems create between the platform sides.
A.
SOCIAL NETWORKS
The evolution of social networks from Friendster to MySpace to

Facebook shows the role of negative externalities in platforms that are
perhaps the closest to traditional communities and shows how different
governance systems can affect platform value. Negative externalities played a
role in the downfall of both Friendster and MySpace. Meanwhile,
Facebookwhich has almost a billion active monthly users around the
worldcontinues to thrive because it created and emphasizes a nice
community that prohibits foul language, fake identifies, and pornography.
95. See JDATE, http://www.jdate.com/ (last visited Aug. 1, 2011).
96. See Lisa Scherzer, Looking for Mr. Goodstein: When Gentile Singles Seek Jewish Mates,
INTERFAITHFAMILY. COM, http://www.interfaithfamily.com/relationships/interdating/
Looking_for_Mr_Goodstein_When_Gentile_Singles_Seek_Jewish_Mates.shtml (last visited
Feb. 29, 2012) (quoting Gail Laguna, vice president of communications at MatchNet.com,
owner of JDate: When a member sets up a profile on JDate, they are asked to put down
their religious affiliation; there is an option to pick another religious stream, she said. If they
are non-Jews, we encourage them to put down that theyre not Jewish and not pretend
theyre of the Jewish faith ).
2012]
11/6/2012 5:38 PM
1227
Although initially successful, Friendsters belated attempt to impose a

private governance regime ultimately led to its failure. Friendster launched in
2002 and grew rapidly.97 Jonathan Abrams, its founder, thought the dating
sites of the time were too anonymous and creepy.98 They also provided
inaccurate information. As he put it, [w]ith JDate, [for example,] a guy is
almost bound to be twenty pounds heavier or twenty years older than he is in
his photo.99 To solve this problem Abrams developed Friendster so that
people could link to friends and see friends of friends. Were trying to make
the process more accountable, he said.100 People will put a more accurate
picture of themselves on Friendster because you know your friends will see
it.101 By 2003 Friendster had more than three million users.102
However, the use of Friendster to create fake profiles quickly became
popular. Fakestersas they were calledmade up fictional personas for
themselves. According to Danah Boyd, Fakesters were created for famous
people, fictional characters, objects, places and locations, identity markers,
concepts, animals, and communities.103
Friendsters management soon recognized that Fakesters posed burdens
on the social networking site. Boyd writes, [a]lthough most participants
loved the playful aspect of Fakesters, it further complicated the network
structure and created an appearance of unreliability, which irritated both the
company and individuals intent on using Friendster for serious
networking.104 In addition, some of the Fakesters attracted massive traffic,
which caused congestion on the sites servers. Recognizing these negative
externalities, Friendsters owners decided to purge the Fakesters, who
consumed significant amounts of scarce server capacity and created noise.105
In response to Friendsters purge, the Fakesters organized themselves
and attempted to reinsert their profiles. They also sought revenge on
Friendster by having Fraudsters masquerade as real people. Friendsters
growth slowed considerably as a result of its continued efforts to exclude
97. See Boyd, supra note 67, at 133.
98. See Finding Love Online, Version 2.0, BLOOMBERG BUSINESSWEEK, Jun. 10, 2003,
http://www.businessweek.com/technology/content/jun2003/tc20030610_4294_tc104.htm.
99. JULIA ANGWIN, STEALING MYSPACE: THE BATTLE TO CONTROL THE MOST
POPULAR WEBSITE IN AMERICA 50 (2009).
100. Id.
101. Id.
102. See A Cautionary Tale, FAST COMPANY (Dec. 19, 2007), http://www.fastcompany.
com/magazine/115/open_features-hacker-dropout-ceo-cautionary-tale.html.
103. Boyd, supra note 67, at 148.
104. Id. at 150.
105. See id. at 151.
1228
11/6/2012 5:38 PM
people from the network. One could conclude that Friendsters downfall
resulted from imposing and enforcing a governance regime. As will soon be
clear, a more plausible interpretation is that Friendsters failure to impose
rules against fake identities at the outset resulted in significant negative
externalities.
Unlike Friendster, MySpace embraced a lax policy towards the reliability
of consumer information. This policy helped its early rise but led to the
websites ultimate downfall. MySpace was founded in 2003:106 its founders
thought that Friendster was making a mistake in preventing people from
having fake identities.107 The website quickly attracted people who were being
deleted from Friendster, including Tila Tequilaa Vietnamese model
whose real surname was Nguyen. Tila Tequila attracted a larger following on
Friendster in part by posting provocative photos of herself. Friendster
deleted her account several times, and as a result she moved to the more
welcoming MySpace.108 Others followed. MySpace grew very quickly,
overtook Friendster, and became one of the most heavily trafficked sites on
the Internet for a period of time.109
MySpaces laissez faire governance policies soon caused problems.
Because MySpace did not require or encourage people to provide reliable
information, it attracted child sex predators as well as minors who lied about
their ages. The site also did little to discourage people from having user pages
with partial nudity, obscenity, crude sexual jokes, and other objectionable
content.110 MySpace gained a reputation as a vortex of perversion111 and
as a site that was not very safelike a citys red light district.
As the popularity of the site attests, a large number of people liked the
risqu nature of MySpace. Yet, advertiserswho provided the principal
106. See Company Overview of MySpace, Inc.: Snapshot, BLOOMBERG BUSINESSWEEK

http://investing.businessweek.com/research/stocks/private/snapshot.asp?privcapId=120412
(last visited Mar. 1, 2012).
107. See Danah Boyd, Friends, Friendsters, and Top 8: Writing Community into Being on Social
Network Sites, FIRST MONDAY (Dec. 4, 2006), http://firstmonday.org/htbin/cgiwrap/
bin/ojs/index.php/fm/article/view/1418/1336.
108. See Lev Grossman, Tila Tequila, TIME, Dec. 16, 2006, http://www.time.com/
time/magazine/article/0,9171,1570728,00.html.
109. It was one of comScores top fifty web properties in the United States by April
2005. See JULIA ANGWIN, supra note 99, at 126.
110. See id. at 181.
111. See Felix Gillette, The Rise and Inglorious Fall of MySpace, BLOOMBERG
BUSINESSWEEK, Jun. 27, 2011, at 52, 58, available at http://www.businessweek.com/
magazine/content/11_27/b4235053917570_page_5.htm.
2012]
11/6/2012 5:38 PM
1229
source of revenue for MySpace112did not. Companies did not want to risk
displaying their brands on pages with objectionable content. With limited
interest from major brands in buying advertising inventory, MySpace moved
its advertising inventory to other advertising networks, including Googles
context-based advertising network.113 These networks inserted low-price
advertisements automatically into areas MySpace made available.114 Not
surprisinglygiven the content of the site, the low prices for the advertising
inventory, and some of the people attracted to MySpacea number of the
advertisements that were displayed were also related to things that some
people would find objectionable.
Facebookwhich started in February 2004took a very different
approach than either Friendster or MySpace. It adopted strict rules to
prevent bad behavior despite being initially targeted to, and started by,
college kids. Like Friendster it focused on creating a platform for people to
manage their relationships with friends. But Facebook required people to use
their real identities. It initially limited access to the site to people with valid
university email accounts ending in .edu, starting with harvard.edu. It then
expanded to selected groups, including businesses with identifiable email
addresses. When it opened to the world in September 2006, Facebook had
500 regional networks.115 Although this approach made it more difficult for
people to use fake identities on Facebook it was still possible. Facebook
like Friendsterdeletes the pages with fake identities that violate its terms of
service. For example, in 2007 it deleted the pages that people had set up to
represent brands that were not allowed at the time.116
Facebook has also taken active steps to limit negative externalities on its
site that would limit its appeal to new users and to advertisers who are
considering inserting messages on its pages. Its terms of service prohibit
112. I s MyS pa ce F re e?, M Y S P A C E . C O M , ( A u g . 2 2, 2 0 0 8 ) , ava ilabl e at
http://web.archive.org/web/20080822094142/http://www.myspace.com/Modules/Help/
Pages/HelpCenter.aspx?Category=1&Question=33.
113. See Google signs $900m News Corp Deal, BBC NEWS, Aug. 7, 2006,
http://news.bbc.co.uk/2/hi/business/5254642.stm.
114. See Kevin Kelleher, MySpace and Friends Need to Make Money. And Fast., WIRED, Mar.
24, 2008, available at http://www.wired.com/techbiz/it/magazine/16-04/bz_socialnetworks.
115. See Janet Kornblum, Facebook Will Soon be Available to Everyone, USA TODAY, Sept.
11, 2006, http://www.usatoday.com/tech/news/2006-09-11-facebook-everyone_x.htm.
116. See Niall Kennedy, Facebook Cleanses Pages of Supposed Fakesters, NIALL KENNEDYS
WEBLOG (Dec. 1, 2007, 8:54 PM), http://www.niallkennedy.com/blog/2007/12/facebookpages-deletions.html. Note, however, that Tila Tequila has a fan page (perhaps the name is
no longer viewed as fake) but with decidedly less provocative pictures than she has on
MySpace. See Tila Tequilas Albums, FACEBOOK, http://www.facebook.com/Tila/photos (last
visited Apr. 16, 2012).
1230
11/6/2012 5:38 PM
various actions including bullying, intimidating, or harassing any user, posting

content that is hateful, threatening, or pornographic, incites violence; or
contains nudity or graphic or gratuitous violence.117 As of April 2009, 150
(eighteen percent) of the companys 850 employees focused on policing the
website for offensive content.118 They would delete photos such as a girl
blowing an epic cloud of pot smoke that violated the social norms the
company wanted to promote.119
At Facebook, the range of policed activity is broad. A division
called User Operations looks at all content that users say is
harassing (via report this links spread liberally throughout the
site) or that shows drugs, nudity or pornography. It also maintains
an extensive blacklist of forbidden names that cannot be used to
make new profiles, like Batman. Some of this monitoring is quite
small beer: youre not allowed to call someone a jerk on
Facebook if someone reports it. Employees also vigorously enforce
their real-name culture; they even disabled the actress Lindsay
Lohans account in December after discovering that she was on the
site under an alias.120
The treatment of negative externalities is only one feature that has

influenced the relative fortunes of Friendster, MySpace, and Facebook.
However, popular accounts of their downfalls tend to highlight the difficulty
that Friendster had in dealing with the Fakester problem and the reputation
that MySpace acquired for having a seedy and unsafe community. Decisions
on private governance systems contributed to the financial results for these
social networks. MySpace was bought by News Corporation in 2005 for $580
million and sold in June 2011 for $35 million.121 Friendster was sold for $26
million in 2009 and closed down its original site and deleted user profiles in
2011.122 Facebook displaced MySpace as the leading social network measured
117. See Statement of Rights and Responsibilities, Facebook, Apr. 26, 2011,
http://www.facebook.com/terms.php.
118. See Nick Summers, Walking the Cyberbeat, THE DAILY BEAST, Apr. 30, 2009,
http://www.thedailybeast.com/newsweek/2009/04/30/walking-the-cyberbeat.html.
119. Id.
120. Id.
121. See Jessica E. Vascellaro, Emily Steel & Russell Adams, News Corp. Sells Myspace for a
Song, WALL ST. J., Jun. 30, 2011, available at http://online.wsj.com/article/SB1000142405270
2304584004576415932273770852.html.
122. See Julianne Pepitone, Friendster Plans to Nuke Its User Data, CNN MONEY (Apr. 26,
2011, 2:48 PM), http://money.cnn.com/2011/04/26/technology/friendster/index.htm.
2012]
11/6/2012 5:38 PM
1231
by users and page views in May 2008.123 Facebooks has a market value of
almost $100 billion based on trading in its stock in private markets.124
One of the major business risks that Facebook facesand a source of
continuing controversyconcerns how much control it gives users over the
dissemination of private data.125 Individual users can face adverse effects
from disclosure while other members of the Facebook platformsuch as
advertisers and application developersbenefit from greater access to data.
Its governance rules concerning privacy-related negative externalities, and its
choices relative to new competitors like Google+, are likely to be important
for striking the optimal balance between the competing interests of members
of its community.126
B.
STOCK EXCHANGES
Modern stock exchangeswhich have detailed rules and regulations that

are designed to ensure the integrity of their marketsprovide another
example of how communities govern negative externalities. As observers
note, [s]tock exchanges around the world invest considerable manpower,
technological effort and financial resources to curb market manipulation and
to promote market efficiency and integrity.127 They impose rules concerning
market manipulationdoing things to artificially affect market signals such
as disclosing false information or creating a false impression of trading
activity. They also impose rules concerning insider tradingusing material
non-public information. Nasdaq, for example, has detailed rules regarding
123. See Caroline McCarthy, Facebook Overtakes MySpace Globally, ZDNET (June 23, 2008,
9:13 AM), http://www.zdnet.com/news/facebook-overtakes-myspace-globally/207724.
124. See Lee Spears & Mohammed Hadi, Facebooks Implied Value Slips to $98 Billion in
Private Market Trading, BLOOMBERG, Feb. 16, 2012, http://www.bloomberg.com/news/
2012-02-16/facebook-s-implied-value-slips-to-98-billion-in-private-market-trading.html.
125. For a litany of controversies surround Facebook, see Criticism of Facebook,
WIKPEDIA, http://en.wikipedia.org/wiki/Criticism_of_Facebook (last visited Mar. 1, 2012).
126. There is perhaps no better antidote to professional prognostication than the history
of social networks. Friendster was a highly praised internet business for much of 2003 before
evolving into a famous case study of business mistakes. See Boyd, supra note 67; Mikolaj Jan
Piskorski & Carin-Isabel Knoop, Friendster (A) (Harvard Business School, Paper No. 9-707409, 2007). MySpace was then lauded for its brilliant effort, in part based on its willingness
to let anyone do anything on its site, in displacing Friendster. Commentators thought that it
had won the race for dominance in social networking. See Marc Gunther, News Corp.
(hearts) MySpace, CNNMONEY, Mar. 29, 2006, http://money.cnn.com/2006/03/28/
technology/pluggedin_fortune/. It remains to be seen whether Facebook will make the sorts
of mistakes in balancing the interests of its communityin particular juggling negative and
positive externalities that upended its predecessors or many other possible business mistakes
that could reverse its growth.
127. See Douglas Cumming, Sofia Johan & Dan Li, Exchange Trading Rules and Stock
Market Liquidity, 99 J. OF FIN. ECON. 651 (2011).
1232
11/6/2012 5:38 PM
wash trades, pre-arranged trading, fictitious orders, giving up priority,

churning, front-running, and a variety of other types of practices.128 In
addition to market manipulation rules, exchanges have rules for business
dealings among members including rules about payment and delivery.129
Exchanges enforce these rules in a variety of ways including expelling
members for violating them.130
While modern stock exchanges are subject to government laws and
regulation, modern exchanges adopt and enforce their own rules as well.131 In
the United States, for example, exchanges design their own rules. Although
the Securities and Exchange Commission (SEC) approves rules governing
U.S. exchanges, the SEC does not dictate the rules.132 In fact, a critical feature
in the development of stock exchanges in the eighteenth century was the
ability to exclude users for bad behavior as the early history of the London
Stock Exchangedescribed belowdemonstrates. Indeed, private control
over negative externalities among traders was critical to the emergence of the
modern stock exchange.
The securities market in London operated informally for a couple of
hundred years.133 Securities were traded bilaterally as far back as the sixteenth
century.134 It was convenient for traders to have places to congregate. They
initially did so at the Royal Exchange, where commodities were traded. After
being ejected for crowding the exchange, traders aggregated themselves in
some of the coffee housessuch as Jonathans Coffee Housein the
nearby Exchange Alley.135
128. See id. at 652.

129. NYSE Enforcement, NYSE, http://rules.nyse.com/NYSE/Rules/ (last visited
Aug. 1, 2011).
130. For the rules governing NYSE enforcement actions, see Disciplinary Rules (Rules
47577), NYSE, http://rules.nyse.com/nysetools/PlatformViewer.asp?SelectedNode=chp
_1_7&manual=/nyse/rules/nyse-rules/. For a list of disciplinary actions by the NYSE, see
http://www.nyse.com/DiscAxn/discAxnIndex.html.
131. See John Carson, Self Regulation in Securities Markets: International Trends and New
Directions after the Financial Crisis, C OMPLIAX C ONSULTING , I NC ., Mar. 2009,
http://compliax.com/app/download/1791228504/Self-Regulation+in+Securities+Markets.pdf.
132. See 15 U.S.C. 78s(b)(1) (2010) (statute empowering self-regulatory organizations,
including exchanges, to make their own rules, subject to approval by the SEC); SEC Release
34-50699, Part I.A, Nov. 18, 2004, available at http://www.sec.gov/rules/proposed/3450699.htm (background information on the self-regulation of exchanges, in the context of a
proposed rule to strengthen exchange governance).
133. See RANALD MICHIE, THE LONDON STOCK EXCHANGE: A HISTORY 15 (1999).
134. Id.
135. See Edward Stringham, The Emergence of the London Stock Exchange as a SelfPolicing Club, 17(2) J. PRIVATE. ENTERPRISE 1, 46 (2002).
2012]
11/6/2012 5:38 PM
1233
In the early days of the London Stock Exchange, traders faced the
problem of ensuring that exchange partners would honor agreements to
complete buy and sale orders when they came due. The Barnards Act, passed
in 1734, declared time-based bargains a form of gambling for which it was
not possible to enforce contracts.136 Because traders could not rely on
contracts, [i]t was . . . left to the market participants themselves to create a
code of conduct that enforced the conditions necessary for trade. Even
without the legal impediments . . . those who participated actively in the
market would seek to find a solution to their own problems among
themselves.137 In the mid eighteenth century, several groups of traders in
financial instrumentsincluding bankers and marine underwriters
organized themselves into exclusive associations in which members who
violated the stated or unstated rules of the association could be ejected.138
A group of stockbrokers, who had operated an informal market at
Jonathans Coffee House, tried to do the same in 1761. According to one
contemporary source, The gentlemen at this very period of time . . . . have
taken it into their heads that some of the fraternity are not so good as
themselves . . . . and have entered into an association to exclude them from
J-----s coffee-house.139
They paid the coffee house for the right to use the premises exclusively
for three hours a day.140 As required by that agreement, the master of the
coffee house, a Mr. Feres, apparently ejected a Mr. Renoux who then sued
for assault.141 According to the London Chronicle, on June 9, 1762:
It being proved at the trial that that house had been a market (time
out of mind) for buying and selling government securities, the Jury
brought in their verdict for the plaintiff, with one shilling damage;
by which means Jonathans Coffee-house is now a free and open
market, and all combinations there destroyed.142
In 1772, a group of London stockbrokers took another approach

towards creating an exclusive trading society. They funded the construction
of a new building, the Stock Exchange, for trading.143 Given the previous
136. MICHIE, supra note 133, at 31.
137. Id.
138. Id.
139. See C. F. Smith, The Early History of the London Stock Exchange, 19 AMER. ECON. REV.
206, 215 (1929) (quoting T. MORTIMER, EVERY MAN HIS OWN BROKER, xiv (2d ed. 1761)).
140. See MICHIE, supra note 133, at 31.
141. See James Oldham, Law Reporting in the London Newspapers, 17561786, 31 AM. J.
LEGAL HIST. 177, 18586 (1987).
142. Id. at 186.
143. See MICHIE, supra note 133.
1234
11/6/2012 5:38 PM
legal result, they made admission open for a daily fee.144 By the late 1790s,
with the growth in securities markets, the governing committee of the Stock
Exchange found that they did not have enough power to enforce discipline
and faced difficulties in funding the administration of the exchange. The
owners of the Stock Exchange decided in January 1801 to convert the open
exchange into a closed subscription room for which members paid an
annual fee.145 The London Stock Exchange started on March 3, 1801.146
The new exchange adopted regulations for conducting business.
[A]dherence to these rules and regulations was monitored and adjudicated
by a committee, including full-time administrative staff, and enforced by the
threat of expulsion from the market.147 Most of the regulations focused on
creating trust among members, particularly involving payment and delivery.
As a late nineteenth century treatise put it:
[The London Stock Exchanges] main objects appear to be the easy
and expeditious transaction of business, and the enforcement of
fair dealing among its members. To these ends . . . a set of results
formed for the admission and expulsion of members, and for the
control of their conduct both between themselves and towards the
public.148
The London Stock Exchange was also concerned with limiting negative
externalities that members could impose on each other through market
manipulation or asymmetric information. Writing about events in 1943, one
historian observes, [o]ne of the main functions of the Stock Exchange was
to ensure a level playing field for all its members in terms of equal access to
information. Consequently . . . it tried to ensure that price-sensitive
information, such as company results, were released simultaneously to all.149
The Exchange also treated very seriously any matter of insider trading,
whether accidental or deliberate.150 In 1943, it expelled one member who
received tips from a journalist on his stock recommendations.151 It also
warned members about doing business with non-members who raised
144.
145.
146.
147.
148.
Id.
Id. at 35.
Id.
Id.
See RUDOLPH EYRE MELSHEIMER & WALTER LAURENCE, THE LAW
CUSTOMS OF THE LONDON STOCK EXCHANGE 1 (Nabu 1879).
150. Id. at 295.
151. Id.
AND
2012]
11/6/2012 5:38 PM
1235
insider-trading concerns. At least at this time, the government had no interest

in outlawing insider trading.152
Stock exchangeslike other platformshave incentives to deal with
negative externalities among their members and to maintain the reputation of
the exchange with the public. That does not necessarily mean that they have
adopted the socially optimal governance structure. And governments,
especially after the Great Depression, have imposed regulations on stock
exchanges and also oversee the rules these exchanges adopt themselves.
There is a long-standing debate on the efficacy of government regulation
that is beyond the scope of this Article.153 The analysis of governance
mechanisms for multi-sided platforms indicates, however, that stock
exchanges have incentives to adopt rules and regulations to maximize the
value of the platform and to do so in part by mitigating negative externalities
among their members. The analysis of government intervention in securities
markets should consider whether the government could provide tools that
would help private enforcement, whether there are aspects of private
regulation that could be done more efficiently by the government, and
whether there are deviations between private and public incentives for
maximizing the value of the platform.
C.
SEARCH ENGINES
Search engines also provide a clear example of how multi-sided platforms

privately govern negative externalitiesand in particular how search engines
regulate the problem of companies manipulating search rankings. Strong
financial incentives exist for companies to rank as highly as possible in search
engine results. For example, a media website makes more money if it attracts
more viewers because advertising revenue is proportional to viewers and
because the website will attract more viewers if it ranks more highly.
Likewise, an e-commerce website makes more money if it attracts more
viewers. Because some fraction of viewers will buy products and services
from the website, the more viewers the website has the more sales it will
make.
Companies that run websites can make more money if they can use
knowledge of a search engines algorithm toin effecttrick the search
152. Id. at 296.
153. For an important contribution to the literature, see Rafael La Porta, Florencio
Lopez-de-Silanes & Andrei Shleifer, What Works in Securities Laws?, 61 J. OF FIN. 1 (2006).
They conclude from a study of securities markets in forty-nine countries that laws facilitating
private enforcement through disclosure and liability rules benefit stock markets and that
public enforcement plays a modest role at best. Id. at 20.
1236
11/6/2012 5:38 PM
engine into thinking their websites rank higher than they should. According
to one study, the top spot in a search ranking gets more than a third of the
clicks compared to about seventeen percent for the second spot and three
percent for the tenth spot.154 More clicks results in more viewers.
However, opportunistic behavior associated with the manipulation of
search results imposes significant costs on users. First, users receive distorted
and inaccurate search results. Second, strategies that manipulate search
rankings degrade the value of search results since users have no way of
knowing whether any particular search result is the product of a manipulated
or an unmanipulated ranking. The search engines deploy counterstrategies
including frequent changes in the algorithms as well as governance rulesto
counter efforts at manipulating.155
Googles ongoing efforts to detect and punish websites that manipulate
the Google search algorithm illustrate the role of governance rules for search
engines. Google has developed a sophisticated governance system for
mitigating negative externalities for its platform community of users,
websites, and advertisers. However, it faces some difficult tradeoffs. It has
developed guidelines that describe the good activities that it encourages
websites to engage in and the bad activities that are banned because they
distort the information-value of results.156 It provides recommendations to
webmasters on good technical, design, and content practices that will benefit
website users and also help the website signal to the search algorithm that it
is a high quality and relevant site.157 This is an example of trying to promote
positive externalities.
Google also describes deceptive and manipulative practices that could
result in the imposition of sanctions on the offending website.158 The basic
principle is that websites are not supposed to do things that are designed to
influence the search engine results as opposed to providing value to users.
Google identifies specific techniques that websites are not supposed to use,
including having hidden text or links, cloaking or sneaky redirects, loading
pages with irrelevant keywords, having multiple pages with substantially the
same content, and using doorway pages that are just created for search
154. Daniel Ruby, The Value of Google Result Positioning, INSIGHTS, CHITIKA (May 25,
2010), http://insights.chitika.com/2010/the-value-of-google-result-positioning.
155. See W e b ma s te r T ool s : S e a r ch E n gi n e O pt i m iz a ti o n ( S EO ), G O O G L E ,
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=35291&ctx=cb&sr
c=cb&cbid=mi5abtuzaoia (last visited Apr. 17, 2012).
156. See id.
157. See id.
158. See id.
2012]
11/6/2012 5:38 PM
1237
engines.159 The list is not intended to be exhaustive, and Google makes it

clear that it will take action for any effort to distort search results
artificially.160
These quality guidelines cover the most common forms of
deceptive or manipulative behavior, but Google may respond
negatively to other misleading practices not listed here (e.g. tricking
users by registering misspellings of well-known websites). Its not
safe to assume that just because a specific deceptive technique isnt
included on this page, Google approves of it. Webmasters who
spend their energies upholding the spirit of the basic principles will
provide a much better user experience and subsequently enjoy
better ranking than those who spend their time looking for
loopholes they can exploit.161
Google cannot disclose too much about how it detects violations because
that would enable websites to manipulate the system.162
For example, Google imposed sanctions on J.C. Penneyan American
department store chain that also sells merchandise onlinefor manipulating
search results. In late 2010, the company achieved the top search result
rankings for many of the products that it sells as a result of a highly
successful Search Engine Optimization (SEO) strategy.163 An SEO
consultant for the company inserted terms for J.C. Penney products in
thousands of websites along with links back to www.jcpenney.com.164 By
doing this, the SEO consultant fooled Googles search algorithm into
thinking that www.jcpenney.com was a more relevant website for the
159. According to Google, [d]oorway pages are typically large sets of poor-quality
pages where each page is optimized for a specific keyword or phrase. In many cases,
doorway pages are written to rank for a particular phrase and then funnel users to a single
destination. Webmaster Tools: Cloaking, sneaky Javascript redirects, and doorway pages, GOOGLE,
http://www.google.com/support/webmasters/bin/answer.py?answer=66355 (last visited
Mar. 1, 2012).
160. See Webmaster Tools: Search Engine Optimization, supra note 133.
161. See Webmaster Tools: Webmaster Guidelines, GOOGLE, http://support.google.com/
webmasters/bin/answer.py?hl=en&answer=35769&topic=2370419&ctx=topic (last visited
Mar. 19, 2012).
162. The New York Times quotes the editor of the Search Engine Land blog, which
covers the search industry: Google is just cagey about everything. That, he said, is because
the company is perpetually worried that the more it reveals about the vaunted mathematical
algorithm it uses to drive search results, the more people will try to game it. David Segal, A
Bully Finds a Pulpit on the Web, N.Y. TIMES, Nov. 28, 2010, at BU1, available at
http://www.nytimes.com/2010/11/28/business/28borker.html?pagewanted=all. A spammers
forum, www.blackhatworld.com, even provides tips for gaming the rules.
163. See David Segal, The Dirty Little Secrets of Search, N.Y. TIMES, Feb. 13, 2011, at BU1,
available at http://www.nytimes.com/2011/02/13/business/13search.html?pagewanted=all.
164. Id.
1238
11/6/2012 5:38 PM
inserted search terms than it really was. When it found out about J.C.
Penneys strategy, Google imposed a penalty on the company. It manually
reduced the search rankings for J.C. Penney for approximately 90 days.165 As
a result of the manual action J.C. Penneys rankings fell. For the search term
Samsonite carry on luggage, for instance, J.C. Penny fell from the first to
the seventy-first spot.166 Given the low click rate after the tenth spot on the
first page, downgrading J.C. Penney to the seventy-first spot had almost the
same effect as excluding it from that particular search query result altogether.
Googles governance system balances the value of providing users access
to websites, ensuring the accuracy of the rankings, and deterring websites
from manipulating the system.167 In some cases it appears that Google
subjects websites to manual actions that reduce their rankings for some
period of time.168 In other caseslike the case of J.C. Penneywebsites are
subjected to manual actions that reduce their rankings until they apply for
reconsideration. Still in other cases websites are delisted altogether, although
they have the possibility of applying for reconsideration. For example,
BMW.de was delisted in 2006 for using doorway pages.169
Googles manual process involves the use of algorithms to detect
possible violations as well as human decisions on how to respond and
whether to reconsider. As the web has expanded, however, it is not feasible
for Google to rely mainly on this process to ensure the quality of website
rankings and mitigate externalities. As of June 2011, more than 350 million
websites existed and about 150,000 new websites appeared each day.170
Consequently, Google modifies its search algorithm frequently both to
improve its performance and to counter efforts to game the algorithm.
Changes in the algorithm result in changes in rankings and in some cases
material changes in rankings. For example, a major change to the algorithm
in February 2011 affected the quality score (an estimate of the relevance of a
165. See Matt Rosoff, Google Has Stopped Punishing JC Penney, BUS. INSIDER, May 25, 2011,
http://articles.businessinsider.com/2011-05-25/tech/30017965_1_panda-google-jc-penney.
166. See Segal, supra note 163.
167. For further discussions of practices that Google has attacked, see Barry Schwartz,
Google Vows Renewed Look At Cloaking in 2011, SEARCH ENGINE LAND, Dec. 28, 2010,
http://searchengineland.com/google-vows-to-look-at-deceptive-cloaking-techniques-59802.
168. See Mitch Fournier, What a Google Penalty Looks Like, MITCH FOURNIER (July 21,
2011), http://mitchfournier.com/2011/07/21/what-a-google-penalty-looks-like.
169. See Tom Espiner, Google Blacklists BMW.de, CNET NEWS (Feb. 6, 2006, 6:58 AM),
http://news.cnet.com/Google-blacklists-BMW.de/2100-1024_3-6035412.html?tag=content
Main;contentBody;1n.
170. See Julie ODell, How Big Is The Web & How Fast Is It Growing?, MASHABLE (Jun. 19,
2011), http://mashable.com/2011/06/19/how-many-websites/#17199How-Big-Is-the-Web.
2012]
11/6/2012 5:38 PM
1239
website to a keyword) of 11.8 percent of the queries that Google received.171

As a cross check on the changes to the algorithm, Google looked at whether
its changes had detected websites that users (based on Chrome browser
users) had chosen to block. The changes eliminated eighty-four percent of
the several dozen most-blocked domains.172
Changes in the algorithm arein the short runa zero-sum game for
websites. Some websites rise in the rankings while others fall. Those that rise
in the rankings quietly celebrate while some of those who fall complain
loudly. Those who complain do not necessarily have strong grounds. For
instance, some of the websites that fall in the rankings do so because they
have tried to manipulate the algorithm and have been penalized by Google as
a consequence. Other websites fall in the rankings as collateral damage from
the attempts by some websites to artificially inflate their rankings. In dealing
with attempts to manipulate search rankings, Google is forced to repeatedly
change its algorithm in ways that may create uncertainty even for websites
that play by the rules. In other words, if all websites played by the rules they
would all have greater certainty about the ranking algorithm.
Googles efforts to deal with negative externalities generated by
opportunistic websites rest entirely on its ability to exercise property rights
over the platform. By delisting websites, Google preserves the quality of its
rankings and provides a significant disincentive to websites engaging in
manipulation of search results. Google achieves almost the same result by
manually reducing website rankings since there is a low value to being listed
far down in the rankings. And finally, by being able to change its algorithm,
Google possesses a scalable approach for mitigating negative externalities on
a massively large and exponentially expanding platform. The consequence is
that, over the longer run, the short run zero-sum game becomes a significant
positive-sum gain for the platform, its users, and also for websites, who
benefit from the enhanced quality signals that the platformhere Google
is able to provide.
Although Googles private governance regime may be beneficial, it is
also controversial. As a result of changes in Googles rankings, several
171. See Finding More High-quality Sites in Search, THE OFFICIAL GOOGLE BLOG (Feb. 24,
2011, 6:50 PM), http://googleblog.blogspot.com/2011/02/finding-more-high-quality-sitesin.html.
172. The changes in part increased the detection of scraper sites and content farms
that present shallow or low quality content by, for example, simply copying snippets from
websites that present the original content. See Danny Sullivan, Google Forecloses On Content
Farms with Panda Algorithm Update, SEARCH ENGINE LAND (Feb. 24, 2011),
http://searchengineland.com/google-forecloses-on-content-farms-with-farmer-algorithmupdate-66071.
1240
11/6/2012 5:38 PM
companies have filed antitrust suits against Google for allegedly engaging in
exclusionary practices or have encouraged governments to initiate antitrust
investigations.173 There have also been calls for government search neutrality
regulation174 under which search engines like Google would face legal
constraints on adjusting results or penalizing websites. This topic is
addressed further in the next Part, which looks at the distinction between
efforts to use exclusion to police negative externalities and efforts to use
exclusion to restrict competition.
VI.
PRIVATE VS. PUBLIC GOVERNANCE IN THE

REGULATION OF BAD BEHAVIOR ON PLATFORMS
Governments have developed extensive responses to bad behavior by

members of the community. Criminal law and the police powers of the state
deal with various actions by members of society that harm others.
Governments use these laws to prohibit practices ranging from fraud to
insider trading to murder. Common law also helps regulate negative
externalities in the community through property rights, contract rights, and
tort liability, which provides incentives to exercise care. Law and economics
scholars have argued that much of criminal and common law can be
interpreted as a rational attempt to maximize social welfare.175
173. See TradeComet.com LLC v. Google, Inc., 693 F. Supp. 2d 370 (S.D.N.Y. 2010);
Person v. Google, Inc., No. C 06-7297, 2007 WL 1831111 (N.D. Cal. June 25, 2007);
KinderStart.com v. Google, Inc., No. C 06-2057, 2007 WL 831806 (N.D. Cal. Mar. 16,
2007); Google, Inc. v. MyTriggers.com, Inc., No. 09CVH10-14836 (Franklin Cnty. Ct. of
Common Pleas, Ohio, Aug. 31, 2011); Leo Condrowicz, The E.U. Probe: Is Google Rigging Its
Search Results?, TIME (Dec. 2, 2010), http://www.time.com/time/business/article/
0,8599,2034138,00.html; Miguel Helft, Lawsuit Says Google Was Unfair to Rival Site, N.Y.
TIMES, Feb. 17, 2009, available at http://www.nytimes.com/2009/02/18/technology/
internet/18google.html; Claire Cain Miller, Texas Probes Google on Ranking of Search Results,
N.Y. TIMES, Sept. 4, 2010, at B3, available at http://www.nytimes.com/2010/09/04/
technology/04google.html?adxnnl=1&adxnnlx=1314113487-4SfkO0V/SuFxNfcRMdHkbQ;
Companies Ask EU Commission to Step in on Google Search Ranking Complaint, ITPROPORTAL
(Feb. 24, 2010), http://www.itproportal.com/2010/02/24/companies-ask-eu-commissionstep-google-search-ranking-complaint/; Kinderstart Sues Google Over Lower Page Ranking, USA
TODAY, Mar. 19, 2006, available at http://www.usatoday.com/tech/news/2006-03-19google-kinderstart_x.htm.
174. See Adam Raff, Search, But You May Not Find, N.Y. TIMES, Dec. 28, 2009, at A27,
available at http://www.nytimes.com/2009/12/28/opinion/28raff.html. The author is the
founder of a company that complained to the European Commission that Googles penalties
for his site, Foundem, violated the EUs competition laws. See also James Grimmelmann,
Some Skepticism About Search Neutrality, in ADAM MARCUS, THE NEXT DIGITAL DECADE:
ESSAYS ON THE FUTURE OF THE INTERNET 435 (Berin Szoka, eds, 2010).
175. See RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 375380 (4th ed. 1992).
2012]
11/6/2012 5:38 PM
1241
In addition to criminal law and common law, governments have also

adopted regulatory laws to deal with various actions that were not addressed
sufficiently by common law. These range from the regulation of pollution, to
lemon laws for the sale of used automobiles, to laws against cyber
harassment. Many of these laws and regulations have been rationalized along
the lines of modern welfare economics as solutions to market failures.176
The governance mechanisms for private multi-sided platforms mirror
many of these laws and regulations. Multi-sided platforms have developed
rules and enforcement mechanisms for dealing with negative externalities
created by agents on their platforms.177 These include efforts to keep
platforms safe from sexual predators, cyber harassment, pornography, underprovision of information, market manipulation, fraud, misrepresentation, and
opportunistic distortion of information.178 Efforts to police platforms likely
provide enormous social value, since some multi-sided platforms would
provide significantly poorer service and perhaps not even be viable in the
absence of efforts to control negative externalities.
The governance of negative externalities by multi-sided platforms
nevertheless raises two public policy issuesanti-competitive exclusion and
the benefits of private versus public regulation. Both issues result from the
exercise of bouncers rights to enforce rules to mitigate putative negative
externalities. As previously explained, in the early days of the London Stock
Exchange English courts prohibited the traders who rented the premises of
Jonathans Coffee House from excluding other traders.179 However, the issue
of exclusion is increasingly prominent as a result of several multi-sided
platforms having created highly successful global businesses. These include
Apple, eBay, Facebook, Google, Microsoft, NYSE/Euronext, and Visa.
A.
THE BENEFITS OF PRIVATE REGULATION
An important question is whether privately owned multi-sided platforms

should be allowed to operate private governance systems that include fines,
exile, and jail-like punishment.180 Facebook, for example, is regulating social
176. See W. KIP VISCUSI, JOSEPH E. HARRINGTON, JR. & JOHN M. VERNON,
ECONOMICS OF REGULATION AND ANTITRUST 13 (4th ed. 2005).
177. See supra Part V.
178. Id.
179. See supra Section V.B.
180. Googles policy of taking websites that have violated their policy far down in the
rankings is sometimes referred to as Google jail. See Dave Johnson, SEO Dirty Tricks That
Can Land Your Companys Website in Google Jail, CBS NEWS (Feb. 28, 2011),
http://www.cbsnews.com/8301-505143_162-28650615/seo-dirty-tricks-that-can-land-yourcompanys-website-in-google-jail.
1242
11/6/2012 5:38 PM
behavior for a community that has a population that is twice as large as the
population of the United States. Should Facebooka private enterprisebe
regulating the pictures that users show to their friends or what users can say
to each other?
This question highlights the tradeoff between public and private control
because both mechanisms are imperfect.181 On the one hand, public control
is subject to a myriad of breakdowns in political and legislative processes,
imperfections in the government institutions, and unintended consequences
of rigid methods of control mandated by law. On the other hand, private
control is problematic because for-profit firmsincluding multi-sided
platformsdo not necessarily have incentives to maximize social welfare and
may in fact have incentives to reduce social welfare.182 As discussed below, a
platform could even adopt rules as a pretext to exclude competitors.
Although for-profit firms do not necessarily have incentives to increase
social welfare, multi-sided platforms have incentives to maximize the value of
their platforms to the community because they obtain profit by extracting
value from the platform. They also have incentives to reduce negative
externalities. The review of multi-sided platforms above shows that many
platforms havein facterected sophisticated governance mechanisms to
do so.
In assessing whether public or private control should govern multi-sided
platforms, it is important to note that multi-sided platforms have several
advantages over public regulators. They have more information on practices
that may lead to negative externalities and the impact on the community. As
private firms, multi-sided platforms can make decisions quickly on how to
deal with negative externalities and modify practices quickly, especially if they
observe unintended consequences. They also face fewer constraintsfor
better or worsesince they are not subject to due process or administrative
procedure requirements. However, the platforms lack some of the
investigative methods and penalties that a public enforcer would have.
The issue of public versus private control has recently come up in
proposals for search neutrality.183 Proponents of search neutrality proposals
181. See Simeon Djankov et al., The New Comparative Economics, 31 J. COMPAR. ECON.
595, 598 (2003).
182. See Alexander White & E. Glen Weyl, Insulated Platform Competition (NET Inst.,
Working Paper No. 10-17, 2011).
183. See Making the Case for Search Neutrality, SEARCH NEUTRALITY,
http://www.searchneutrality.org/search-neutrality (Oct. 11, 2009); Can Search Discrimination
by a Monopolist Violate U.S. Antitrust Laws?, FAIRSEARCH, http://www.fairsearch.org/wpcontent/uploads/2011/07/Can-Search-Discrimination-by-a-Monopolist-Violate-U.S.Antitrust-Laws1.pdf (last visited Apr. 10, 2012).
2012]
11/6/2012 5:38 PM
1243
argue that the government should prohibit search engines from

manipulating search results and that all search results should be presented
based on a governmentally determined notion of relevance. These proposals
arenot surprisinglybeing advanced by websites that have had changes in
their rankings as a result of manual downgrading or algorithmic changes.184
Such proposals would limit multi-sided platforms use of bouncers rights to
penalize websites that are trying to manipulate search rankings. These
regulations would also limit search engine innovations to improve
algorithms, because anyone whose rankings change could claim that the
algorithm did not change neutrally. Thus government regulation could
impose substantial social costs both by requiring potentially slow and
expensive regulatory review and by making false-positive findings of
manipulation based on differences in complex and potentially subjective
judgments of relevance.185
B.
ANTI-COMPETITIVE EXCLUSION: A PROPOSED THREE-STEP

ANALYSIS
When a firm is excluded from participating in a multi-sided platform it

can incur significant harm. The excluded firm may decide to sue under
antitrust laws and argue that the multi-sided platform uses significant market
power to harm competition by excluding the firm from a relevant antitrust
market. The court or competition authority will then need to decide whether
the exclusion is anti-competitive or an efficient business response by the
multi-sided platform. The multi-sided platform may explain that it has
established efficient rules for reducing negative externalities and that it
excluded the firm because the firm violated the rules the platform. That
argument could be true as shown above, or it could be a pretext for anticompetitive exclusion. This article proposes a three-part test to help courts
and competition authorities distinguish between efficient private governance
and anti-competitive exclusion.
When a firmespecially one with significant market powerexcludes
another firm from the market, its actions may be subject to scrutiny under
antitrust laws.186 In the United States, exclusionary practices may violate
Section 2 of the Sherman Act, which prohibits attempts to create or maintain
184. See Adam Raff, Search, But You May Not Find, N.Y. TIMES, Dec. 27, 2009, available at
http://www.nytimes.com/2009/12/28/opinion/28raff.html.
185. See James Grimmelmann, Some Skepticism About Search Neutrality, in THE NEXT
DIGITAL DECADE: ESSAYS ON THE FUTURE OF THE INTERNET (Berin Szoka & Adam
Marcus eds., 2010).
186. See EINER ELHAUGE & DAMIEN GERADIN, GLOBAL ANTITRUST LAW AND
ECONOMICS (Foundation 2d ed. 2011).
1244
11/6/2012 5:38 PM
a monopoly. In certain circumstances, exclusionary practices may also violate

Section 1, which prohibits unreasonable agreements in restraint of trade.187 In
the European Union, exclusionary practices could be considered abuses of
dominance under the Article 102 TFEU.188
As antitrust has adopted a more economics-based approach, some
aspects of exclusionary abuse claims have been viewed as problematic or at
least controversial.189 For example, U.S. courts have expressed skepticism
over vertical restraint cases and monopoly leveraging.190 Courts have
generally rejected claims premised on the exclusion of a competitor from
access to property owned by the defendant.191 U.S. courts have also
expressed skepticism over claims involving the exclusion of direct
competitors such as those that stem from the improvement of a product.192
In fact, they have raised the bar for some practices sufficiently high that they
are effectively (if not literally) per se legal under the rule of reason. For
example, it is extremely difficult for a plaintiff to establish that a rival has
engaged in predatory pricingthat is the rival has priced low to drive
out a rival and thereby monopolize a market.193
The European Union has also moved in this direction. The European
Commission guidelines on enforcement priorities for Article 82 (now Article
102 TFEU) recognize that vertical and horizontal exclusionary practices
should be subject to the rule of reason and that depending on the factual
circumstances a firm with significant market power may not have the ability
or incentives to foreclose a rival.194 The European Union continues to
187. See Sherman Antitrust Act, 15 U.S.C. 12 (1890).
188. See Treaty on the Functioning of the European Union, art. 102, May 9, 2008, 51
OFFICIAL J. OF EUR. UNION 89 (2008).
189. The U.S. courts in particular have been influenced by the findings of economists
that firms engage in vertical foreclosure practices for pro-competitive reasons. See Leegin
Creative Leather Prods., Inc v. PSKS, Inc., 551 U.S 877 (2007); Brief for Economists as
Amici Curiae Supporting Petitioner, Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551
U.S. 877 (2007).
190. See HERBERT HOVENKAMP & PHILLIP E. AREEDA, FUNDAMENTALS OF
ANTITRUST LAW (Aspen 3rd ed. 2004); ELHAUGE & GERADIN, supra note 186.
191. See Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP, 540
U.S 398 (2003).
192. See Allied Orthopedic Appliances Inc., v. Tyco Health Care Group LP, 592 F.3d
991 (9th Cir. 2010).
193. Mark S. Popofsky, Defining Exclusionary Conduct: Section 2, The Rule of Reason, and the
Unifying Principle Underlying Antitrust Rules, 73 ANTITRUST L. J. 435, 44243 (2006).
194. See European Commission, Communication from the CommissionGuidance on
Its Enforcement Priorities in Applying Article 82 (EC) to Abusive Exclusionary Conduct by
Dominant Undertaking (Feb. 24, 2009), available at http://eur-lex.europa.eu/LexUriServ/
LexUriServ.do?uri=CELEX:52009XC0224(01):EN:NOT. See also Steven C. Sunshine,
2012]
11/6/2012 5:38 PM
1245
mandate access to essential facilities but imposes a significant burden of

persuasion on the parties demanding access.195
Nevertheless, it is possible that firms with significant market power may
engage in anti-competitive exclusion. The question addressed here is how to
handle exclusionary conduct cases that result from the imposition of
penalties on firms under established multi-sided platform governance
systems.
A platform could invoke negative externalities in excluding a possible
rival from access to its platform. If it does not have a governance system in
operation, and exclusion of the competitor is based on an idiosyncratic
decision rather than a systematic process for dealing with negative
externalities, there is no reasonbased on the analysis aboveto evaluate
the conduct differently than other exclusionary conduct allegations. The
court should take the efficiency explanation offered by the defendant into
account in weighing the pro-competitive and anti-competitive aspects of the
practice, or in applying the otherwise applicable legal framework.196
It is now widely accepted that antitrust rules should take into account the
costs and likelihood of making mistakes.197 Rules that tend to absolve firms
that have engaged in anti-competitive practices can encourage more firms to
engage in these practices. Rules that tend to condemn firms for engaging in
practices that are pro-competitive can deter firms from advancing social
welfare.
That balance between false positives and false negatives is especially
critical when the challenged action of the platform is taken pursuant to a preexisting governance system. The fact that a firm is a platform and maintains a
governance system for dealing with negative externalities provides a strong
presumption that the firm is increasing social welfare by policing bad
behavior pursuant to that governance system on its platform. That is likely to
Deputy Assistant Attorney General, U.S. Dept of Justice, Address at the American Bar
Association Section of Antitrust Law: Vertical Merger Enforcement Policy (Apr. 5. 1995),
available at http://www.justice.gov/atr/public/speeches/2215.pdf.
195. See Case C-7/97, Oscar Bronner GmbH & Co. KG v. Mediaprint Zeitungs, 1998
E.C.R. I-7791, 4147 (holding that refusing access does not constitute abuse of a
dominant position within the meaning of Article 86 of TFEU, even when there is only one
nationwide newspaper home-delivery scheme, and even when the small circulation of rival
newspapers renders it economically impossible to establish similar delivery schemes).
196. See the five part test in United States v. Microsoft, 253 F.3d 34 (D.C. Cir. 2001).
197. See ELHAUGE & GERADIN, supra note 186; HERBERT HOVENKAMP & PHILLIP E.
AREEDA, FUNDAMENTALS OF ANTITRUST LAW (Aspen 3d ed. 2004); David S. Evans, Why
Different Jurisdictions Do Not (and Should Not) Adopt the Same Antitrust Rules, 10 CHI. J. INTL. L.
161 (2009); Frederick Beckner III & Steven C. Salop, Decision Theory and Antitrust Rules, 67
ANTITRUST L. J. 41 (19992000).
1246
11/6/2012 5:38 PM
be an extremely valuable service to the consumers on the various sides of the

platform. The ability to exclude those who create negative externalities is
critical to the functioning of that governance system and ultimately for the
overall value of the platform.
Of course, it is possible that a platform has established or structured a
governance system as a pretext for excluding competitors. A platform could
establish rules that prohibit participants from relying on or interconnecting
with other platforms. Prior to its break-upfor exampleAT&T provided
local and long-distance telephone service under public utility rate regulation
and also had an unregulated equipment manufacturer.198 It established rules
concerning the interconnection of equipment and other services to its
networks that it claimed were designed to ensure the integrity of the
telephone system.199 It has been argued that AT&T sometimes applied these
rules to exclude competing equipment manufacturers and competing longdistance telephone service providers in violation of the antitrust laws.200 To
settle an antitrust case brought by the U.S. Department of Justice, AT&T
agreed to divest its local operating companies and its equipment
manufacturer.201 Therefore, the existence of a governance system should
notby itselfpreclude a finding of anti-competitive exclusion.
Nevertheless, the existence of a governance system increases the
likelihood that the practice that results in exclusion is pro-competitive.
Governance systems are common among platforms, are necessary for dealing
with negative externalities, and can increase consumer welfare. To the extent
there are serious negative externality problems on the platform, antitrust
decisions that prohibit firms from engaging in pro-competitive exclusion
would impose significant costs on the platform because the platform would
be forced to weaken its enforcement mechanisms. Moreover, other platforms
concerned with the risk of an antitrust suit will weaken their government
systems to the overall detriment of platform customers.
The 1762 decision concerning Jonathans Coffee House202 illustrates the
dangers of limiting the ability of private platforms to exercise bouncers
rights to regulate their platforms. As a result of the decision, the first London
Stock Exchange had to adopt ineffective methods for dealing with negative
externalities among members. It was not until forty years after the decision
198. See TIM WU, THE MASTER SWITCH: THE RISE AND FALL OF INFORMATION
EMPIRES 57, 18889 (2010).
199. Id. at 189.
200. Id. at 10809.
201. United States v. Western Elec. Co. Inc., 969 F.2d 1231 (D.C. Cir. 1992).
202. Oldham, supra note 141, at 18586.
2012]
11/6/2012 5:38 PM
1247
that the English stock exchange could effectively regulate bad behavior
among its members.203 The error-cost analysis of governance systems
indicates that the standard rule of reason approach should be modified in the
same wayand for similar reasonsit has been modified for other practices
that are likely to be pro-competitive.204 There should be a presumption that
exclusion that results from an established governance system for dealing with
negative externalities is pro-competitive. The plaintiff should bear the burden
of persuasion of showing otherwise.
Figure 1
Based on these considerations this Article proposes a three-step test

summarized in Figure 1for exclusionary behavior involving multi-sided
platforms. This Article assumes the plaintiff has already met the usual burden
204. See David S. Evans & A. Jorge Padilla, Designing Antitrust Rules for Assessing Unilateral
Practices: A Neo-Chicago Approach, 72 U. CHI. L. REV 72 (2005). In this Article, I am focusing
only on how existing antitrust rules should be modified given the existence of platformbased governance systems for dealing with negative externalities. I do not address whether
existing rules are sound.
1248
11/6/2012 5:38 PM
of establishing a relevant market andas applicablethe potential for the

challenged practice to raise price or restrict output. In the first step of the
analysisdefending against the complainants prima facie casethe
platform should have the opportunity to demonstrate that it has established a
governance system for dealing with negative externalities and that the
practice at issue results from the exercise of that governance system. If the
platform cannot do so, standard antitrust rules applicable to the practice at
issue should apply (see step three). If the platform can do so, the analysis
should move to the second step.
The second step should consider whether the exclusion is inconsistent
with the use of the governance system to deal with negative externalities or
whether the governance system itself has been established as a pretext for
excluding competitors. The plaintiff should bear the burden of
demonstrating that the governance system is not reasonably related to
enforcement of the goals the governance mechanism is designed to achieve.
The plaintiff may be able to show that the platform applied the rules
differentiallyexcluding a competitor when it would not have ordinarily
excluded the firm that allegedly violated the rulesor created a separate class
of offenses that results in the exclusion of competitors. The plaintiff may
also be able to show that the governance system is unrelated to the
correction of negative externalities or established as a pretext for exclusion.
The plaintiffs claim should be rejected unless the plaintiff can demonstrate
that the practice is inconsistent with the mitigation of negative externalities.
Otherwise the analysis should proceed to the third step.
The third steparrived at if the defendant fails the first step or the
plaintiff succeeds in the second stepwould follow the standard antitrust
analysis applicable to the challenged conduct at issue. In the United States,
the third step would involve analyzing whether the practice (i) forecloses or
raises rivals costs in a manner that enhances the defendants ability to raise
price or restrict market output to the detriment of consumers and then
(ii) weighing the anti-competitive and pro-competitive effects to determine
whether the practice is, on net, anti-competitive. Under many circumstances,
a successful showing by the complainant under step two would demonstrate
the lack of pro-competitive effects.
As with any error-cost based approach, this three-step analysis cannot
eliminate false positives and false negatives. The plaintiff might succeed in
showing that the application of an exclusionary penalty is a pretext even
though it is a valid attempt to eliminate a negative externality. A defendant
might succeed in showing that an exclusionary penalty is part of a procompetitive governance system when in fact it is designed to exclude a
competitor and harm consumers. Nevertheless, the approach reduces the
2012]
11/6/2012 5:38 PM
1249
likelihood and costs of errors in light of the role of governance systems in

reducing negative externalities that would otherwise harm consumers.
VII.
CONCLUSION
Although the business model has been around for millennia, multi-sided
platforms have become particularly prominent since the advent of the
commercial web.205 The Internet and web technologies facilitate the creation
of platforms for different types of users that would benefit from getting
together.206 As a result of scale economies and the ability to either replicate a
platform across geographies, or to connect global communities, some of
these platforms have become global players. These multi-sided platforms are
likely to attract increasing attention from policymakers because of their
economic and social significance. Several already have.
An essential feature of these platforms is that they promote positive
externalities between members of the community. But as with any
community, there are numerous opportunities for people and businesses to
generate negative externalities that can reduce economic efficiency andin
the extremelead to the tragedy of the commons.
Most discussions of these platforms have focused on how multi-sided
platforms create value by harnessing positive externalities and how positive
network effects can result in the emergence of dominant platforms in
particular categories. Much less attention has been given to the role these
platforms play in mitigating negative externalities. As it turns out, many of
these platforms have developed governance systems for dealing with bad
behavior. These governance systems ultimately depend on the ability of the
platform to exclude agents from some quantum of the platform, including
prohibiting them from the platform entirely.
Exercising these exclusionary rights is controversial. The platform has to
balance the interests of its multiple constituents. Rules concerning negative
externalitiesjust as those involving positive onesshift value between
different sides. Like a polity, a platform must balance competing values, such
as freedom of speech and protection from hate speech and other verbal
harassment. The exercise of exclusionary rights to enforce rules can also lead
to complaints by the excluded parties and in some cases lawsuits. The fact
that a platform is engaging in exclusion as part of a governance system for
dealing with negative externalities has important implications for the antitrust
analysis of exclusion. Exclusion of actors who diminish the value of the
205. See Evans & Schmalensee, supra note 7, at 152.
206. See Evans & Schmalensee, supra note 5, at 5.
1250
11/6/2012 5:38 PM
platform to all its constituents is prevalent, important, and beneficial.

Antitrust analysis should therefore use exclusion based on a governance
system for dealing with negative externalities as a shield for practices that are
likely to be pro-competitive, and recognize that condemning the
implementation of governance systemsas a general matteris likely to
create false positives. The three-step analysis proposed above would better
balance the costs of false positives and false negatives in light of the common
use of governance systems to mitigate negative externalities and increase
consumer welfare in platform communities.
While both private and public control of bad behavior on platforms have
their place, private control through a platform will often be superior to
public control. The platform can identify problems more easily, can correct
these problems with greater agility, and is in a better position to minimize the
unintended consequences of rules. Private governance of multi-sided
platforms increases platform value to constituent communities and facilitates
growth, development, and innovation. These benefits of private governance
should have an important place in the analysis of anti-competitive charges
against a platforms efforts to govern its community.
Public Authority Involvement in Payment Card Markets: Various Countries

August 2014 Update
Prepared by Fumiko Hayashi and Jesse Leigh Maniff

Payments System Research Department
Federal Reserve Bank of Kansas City
The original tables appeared in Bradford, Terri and Fumiko Hayashi. 2008. Developments in Interchange Fees in United States and Abroad, Federal Reserve
Bank of Kansas City Payments System Research Briefing, April issue. Jean Allix and Alan Frankel provided helpful comments for this update.
1. Interchange and Merchant Service Fees

a. Actions taken by public authorities
Country
Argentina
Australia
Austria
(European
Union (EU)
member since
1995)
Belgium
(EU 1952)
Brazil
Credit Card
Debit Card
1999: Law 25.065 for Credit Cards is enacted. The law establishes norms that regulate various aspects related to the credit, debit,
and retail card systems. Among these norms is the setting of limits on the ability to implement price discrimination in merchant
fees.
2005: Law 25.065 is amended with Law 26.010, which requires merchant acquirers to charge the same merchant discount rate to
businesses within the same category. The maximum merchant discount rate is set at 3% for credit cards and 1.5% for debit cards.
2003: The Reserve Bank of Australia (RBA) mandates
2006: RBA introduces interchange standards for the EFTPOS
Bankcard, MasterCard, and Visa to set interchange fees on a
and Visa Debit systems.
cost-based benchmark.
2009: RBA revises EFTPOS interchange fee standard for
multilateral interchange fees.
2009: RBA continues interchange regulation.
2013: RBA implements the new EFTPOS interchange fee
standards.
2003: The Austrian Cartel Court fines Europay Austria, who runs Maestro debit card payment system. The Court declares that
Europay formed an illegal cartel with almost all Austrian banks with respect to a provision in the payment card contract and held
interchange fees excessively high, which the Court views as an abuse of Europays dominant position.
2006: Following the European Commissions interim reports on the retail banking industry, Austrian banks agree to review
arrangements for setting interchange fees and announce that a reduction can be expected. They will also take steps to foster
genuine competition in acquiring between Europay Austria and Visa Austria.
2007: Europay appeals to the Austrian Supreme Court. The Supreme Court confirms the Cartel Courts judgment and increases
the fine from 5 million to 7 million, noting undue enrichment accrued to Europay Austria during the time of the
anticompetitive behavior.
2006: The Belgian Competition Council accepts commitments offered by Banksys to have the investigation (which began in
2000) of its allegedly abusive conduct in electronic payment services and debit card terminals markets closed. The commitments
include separate contracts for acquiring services and terminals, relaxation of exit terms for terminal rental agreements, and a
number of price reductions.
2006: In May, Banco Central do Brasil (the Central Bank of Brazil) issues Directive 1/2006, which focuses the payment cards
industry. The Directive does not establish either obligations or prohibitions and does not mandate any sanctions. In June, Banco
Central do Brasils Economic Law Office and the Secretariat for Economic Monitoring agree to cooperate in order to collect
payment card industry data and to coordinate public policy actions. In September, payment card industry data collection begins.
2009: The Brazilian competition authority investigates the anti-competitive behavior of Redecard and Visa-Cielo as a means of
reducing merchant discount fees.
2010: Banco Central do Brasil publishes Report on the Brazilian Payment Card Industry. The Secretariat of Economic Law
continues to investigate the possibility of a violation of the economic order due to the anti-competitive behavior of acquirers. An
Canada
Chile
China
Colombia
Denmark
(EU 1973)
European
Union
agreement is signed in which acquirers made a commitment to end exclusivity in their credit and debit card schemes.
2011: Banco Central do Brasil publishes Statistical Update on the Brazilian Payment Card Industry, which concludes that despite
the end of exclusivity, there was no significant change in the market share of the two largest acquirers (Visa-Cielo and Redecard).
2009: In March, the Senate Committee on Banking, Trade and
Mid 1990s: A consent order from the Competition Bureau of
Commerce announces that it will move forward with an
Canada allows Interac to set its own interchange fee, but since
investigation of Canadas credit and debit card system. In June, its establishment, Interac has set its interchange fee at zero.
the investigation results are published as a report.
2014: The Federal Budget includes provisions to help lower
credit card acceptance costs for merchants.
2005: The Chilean Antitrust Court admits a complaint filed by the National Economic Prosecutor alleging abuse of a dominant
position by Transbank, the acquirer of all credit and debit cards issued in the country. The Court imposes a fine of approximately
$56,000. The National Economic Prosecutor requests, among other things, the modification of Transbanks price structure in such
a way that it would be public, objective, and based on costs. A partial understanding between the parties, requiring Transbank to
reduce merchant fee ceilings and present a self-regulating plan for setting prices, finally settles the issue.
2002: The Peoples Bank of China sets the maximum merchant fee rates and the division of the merchant fee which consists of
the interchange fee, switch fee, and merchant acquirer fees (so called the 8:1:1 rule).
2011: The Chinese Ministry of Commerce proposes a cap on interchange fees 0.3% of the sale up to 100 yuan (US$15 or 12
euro). The proposal also includes a cap for switch fees: China UnionPay (the countrys only card network) cannot charge
merchants more than 0.05% on credit card sales with a maximum of 5 yuan per transaction.
2012: The State Council approves a change to the decade-old standards on merchant fees which will reduce most merchant fees
by one-fourth or more effective February 25, 2013.
2004: The Superintendent of Industry and Commerce, Colombias competition authority, passes the new Inter-bank Exchange
Tariff, allowing merchants to negotiate fee rates with merchant acquirers.
2006: Credibanco (a Visa issuer) is required to exclude certain costs included in its fee computation that were judged not to
correspond exclusively to payment card services offered to merchants.
1990: The Act of Certain Payment Instruments sets a cap on
1990: The Act of Certain Payment Instruments sets Dankort
merchant service charges (MSC) on internationally-branded
MSC to be zero.
credit/debit cards issued by Danish banks for domestic
2003: An amendment to the Act introduces a positive MSC to
transactions at 0.75% of transaction value or 1.25% of
Dankort transactions and reduces the fees on Maestro and Visa
transaction value with a minimum of DKK 1.95 on the Internet. Electron from 0.75% to 0.4%, with a maximum of DKK 4.
2005: An annual fee per retailer replaces Dankort MSC.
2002: The European Commission (EC) reaches an agreement with Visa to reduce its cross-border interchange fees by December
2007. The benchmark for its interchange fees is to be set at the level of the cost of supplying Visa payment services and cannot
exceed the cost of the services which issuing banks provide, wholly or partly, to the benefit of merchants.
2007: In December, EC rules that MasterCards interchange fees are illegal.
2008: In March, MasterCard files an appeal of the EC decision.
2009: In April, EC and MasterCard reach an interim agreement, setting MasterCard interchange rates at, on average, 0.3% for
France
(EU 1952)
credit cards and 0.2% for debit cards (effective July 1, 2009). EC also sends a Statement of Objections to Visa asserting its
preliminary view that multilateral interchange fees (MIFs) directly set by Visa violate European Antitrust rules (Article 81 EC
Treaty and Article 53 EEA Treaty).
2010: In April, Visa Europe proposes to cap the weighted
average MIF for consumer immediate debit card transactions at
0.2%. The cap is applicable to cross-border transactions within
EEA and, separately, to domestic transactions in each EEA
country where MIFs are either set directly by Visa Europe
(Greece, Hungary, Iceland, Ireland, Italy, Luxembourg, Malta,
the Netherlands, and Sweden) or the Visa Europe cross-border
rates would apply by default. In December, EC drops its
investigation into interchange fees for immediate debit.
2012: In May, the General Court of the EU confirms the ECs 2007 decision prohibiting MasterCards interchange fees.
MasterCard has appealed the ruling to the European Court of Justice and a decision is expected in September 2014. In July, the
European Commission submits a supplementary statement of objections to Visa Europe regarding its use of multilateral
interchange fees in the EEA. The Commission alleges that these MIFs restrict competition and put upward pressure consumer
prices.
2013: In April, the EC opens a formal investigation into whether several of MasterCards inter-bank fees and practices violate EU
antitrust rules. The proceedings identify MasterCards inter-bank fees on payments made by cardholders from non EEA countries
and its cross-border acquiring rules as items of particular concern. In July, the EC proposes a regulation to cap interchange fees
for four-party scheme consumer debit and credit card transactions at 0.2% and 0.3% respectively. These caps would initially apply
to cross-border transactions but after the transition period, they would apply to both cross-border and domestic transactions.
2014: In February, the EC renders legally binding the commitments offered by Visa Europe to cut MIF (to 0.3% for credit and
0.2% for debit transactions) and reform its rules in the 28 EU countries and Iceland, Norway and Liechtenstein. As of January 1,
2015, Visa Europe also commits to allow cross-border acquirers to offer either the domestic debit or credit MIF applicable at the
merchants location or an MIF rate of 0.2 % for debit card transactions and 0.3% for credit card transactions, subject to certain
conditions. In April, the European Parliament amends the ECs proposal to cap interchange fees by applying the rules to
commercial cards, removing the distinction between cross-border and domestic payments systems, and setting a cap of the lower
of .07 or 0.2% for debit card payments. The amendment also expands the scope of the proposal to include three-party schemes if
their volume exceeds a threshold set by the European Commission.
1990: The Paris High Court rules that methods for determining interchange fees could be accepted in accordance with the
Competition Councils statement of objections.
2011: In July, the French Competition Authority (FCA) closes its investigation concerning interchange fees by accepting the
commitments offered by the Groupement des Cartes Bancaires (a syndicate of banks issuing payment cards). Among other
things, the commitment includes a reduction in the interchange fees from 0.47% to 0.3% on average for all cards. The period of
the commitments is four years beginning October 1, 2011. During this period, a steering committee charged by the FCA will be in
charge of devising a methodology to revise fees at the expiration of the commitments. The FCA turns its attention to the
Germany
(EU 1952)
Greece
(EU 1981)
Hungary
(EU 2004)
India
Israel
interchange fees set by other payment card systems, including MasterCard and Visa.
2013: In September, Visa and Mastercard agree with FCA to cap their interchange fees at 0.28%.
2013: In May, the Bundeskartellamt (the competition authority of Germany) states its preliminary anticompetitive concerns to
leading bank associations who have jointly agreed to set interchange fees at 0.3% per electronic cash card ec card (debit)
transaction, with a minimum of .08.
2014: In April, the Bundeskartellamt renders legally binding the leading bank associations agreement to abandon the jointly set
fee.
2008: The Hellenic Competition Commission accepts
commitments offered by banks that aim to reduce the level of
interchange fees in response to objections expressed in the
Directorate General for Competitions Report. In addition, the
banks will take steps to reduce proportionally the level of
commissions charged to consumers for services rendered. The
commitments are effective until 1/08/2010.
2006: Gazdasagi Versenyhivatal (GVH, the competition authority of Hungary) considers intervening in the payment card market.
Interchange fees are regarded as too high compared to costs, especially in the case of debit cards. Price discrimination between
on-us (acquirer=issuer) and f
2008: GVH launches an antitrust probe against several credit
card companies, including MasterCard, Visa, and POS
operators under suspicion of collusion on prices and practices
to keep competitors at bay.
2009: GVH imposes fines upon Visa Europe, MasterCard and
top commercial banks, ruling they have inhibited competition
by forming an illegal bank card interchange-fee cartel.
2012-2013: The Ministry for National Economy (MNE), in cooperation with the Magyar Nemzeti Bank (MNB, the central bank
of Hungary) and GVH, drafts a bill regulating interchange fees. Additional amendments are submitted to Parliament that would
cap interchange fees at 0.2% for debit cards and 0.3% for credit cards.
2012: To promote the use of debit cards, the Reserve Bank of
India caps the merchant discount rate on debit card transactions
at 0.75% of the transaction amount for values at or below Rs
2000 (US$35) and at 1% of the transaction amount for values
above Rs 2000. These caps take effect September 1, 2012.
2006: The Antitrust Tribunal in Israel reaches an agreement
with banks to reduce interchange fees from 1.25% to 0.875%
by 2012.
2011: Credit card companies adopt the Tribunal's methodology
for calculating interchange fees and agree on a reduction in the
Italy
(EU 1952)
Latvia
(EU 2004)
Mexico
Netherlands
(EU 1952)
New Zealand
fees from 0.875% to 0.7% by July 2014.

2010: The Italian Competition Authority (ICA) fines MasterCard and eight banks for allegedly using licensing agreements to keep
interchange fees high and passing those charges on to merchants. The order requires MasterCard to provide economic justification
for its fees and banks to revisit the terms of their contracts with merchants. MasterCard and the banks involved are given 90 days
to show that the allegedly anti-competitive activities have ceased.
2010: The ICA accepts commitments offered by
PagoBANCOMAT (the dominant Italian network) in response
to an investigation opened in October 2009. The commitments
aim to reduce the level of multilateral interchange fees (MIFs)
for national transactions using national PagoBANCOMAT
branded debit cards and include: a 4% reduction of MIFs, a
pledge to not increase MIFs in the future, and a re-definition of
MIFs in accordance with ICA.
2014: In February, the ICA commences an investigation into
whether Consortium Bancomats decision to set interchange
fees for bill payment transactions made with a PagoBancomat
debit card at .10 per transaction is a violation of antitrust law.
2011: The Latvian Competition Council decides that 22 commercial banks have infringed the Competition Law by participating in
multilateral interchange fee agreements and imposes fines on those banks.
2006: The Bank of Mexico and the Mexican Bankers Association agree to reduce interchange fees.
2014: MasterCard promises the Netherlands Authority for
2004: The Netherlands Competition Authority (NMa) fines
Consumers and Markets (ACM) to reduce its interchange fee
Interpay, which operates the debit card system, and member
rate for domestic credit card payments from 0.9% to 0.7%
banks for charging excessive merchant fees for PIN debit
(June 1, 2014), 0.5% (January 1, 2015), and 0.3% (January 1,
transactions.
2016).
2005: NMa withdraws the accusation and the fine imposed on
Interpay but upholds the fine on the banks.
2007: Proceedings are initiated by the New Zealand Commerce
Commission against Visa, MasterCard and member institutions
of the two schemes, alleging price-fixing in the setting of
interchange fees.
2009: The Commission agrees with Visa (on August 12) and
MasterCard (on August 24) to settle credit card interchange fee
proceedings. The agreements require both networks to alter the
scheme rules in New Zealand, allowing merchants to surcharge,
nonbanks to become acquirers, and card issuers to individually
set interchange fees (the networks set the maximum
interchange fee rates).
Norway
Panama
Poland
(EU 2004)
Portugal
(EU 1986)
Romania
(EU 2007)
South Korea
The general position of authorities regarding the introduction of

new payment systems in Norway has been that payers should
cover costs. This position can be seen as early as the 1974
report from the Payment Systems Committee.
2003 - 2004: Under the 1998 banking law, the Superintendent
of Banks issues regulations for banks that issue and manage
credit cards. These regulations establish procedures for
approving a credit card and authorize the charges for
commissions and other related items.
2007: The Polish Office of Competition and Consumer Protection (OCCP) orders banks to discontinue their multilateral
interchange fee agreements.
2008: In November, the Court of Competition and Consumer Protection (CCCP) overturns the OCCPs decision on interchange
fees, holding that the participation of 20 banks in an agreement fixing the fee levels does not constitute an infringement of the
Competition Act in the European Union (i.e., Art 81.1 EC) nor equivalent national provision.
2010: In April, the Court of Appeal repeals the CCCPs decision and submits it back to CCCP for review.
2012: In March, the National Bank of Poland (NBP) publishes Program of card charges reduction in Poland with the aim of
lowering interchange fees via a non-regulatory compromise to 0.70% for debit cards and 0.84% for credit cards by 2017. Due to
insufficient participation, the proposal fails by July, and the NBP takes steps to lower rates through the legislature instead.
2014: In January, a law goes into effect that amends the Act on Payment Services to include a maximum interchange fee of 0.5%
on all Polish payment cards. Affected entities have six months to comply.
2006: Following the European Commissions (EC) interim reports on the retail banking industry, Portuguese issuers and acquirers
meet some of the ECs concerns by reducing domestic interchange fees and removing preferential bilateral domestic interchange
fees.
2014: The Romanian Ministry of Finance proposes capping on multilateral interchange fees at 0.2% for debit cards and at 0.3%
for credit cards.
2005: The Korean Fair Trade Commission rules that BC Cards
(South Koreas four-party scheme credit card) joint pricing of
merchant service charges is a cartel, imposes a fine of 10.092
billion Korean won, and issues corrective measures.
2011: The Financial Services Commission (FSC) announces
comprehensive measures to reform the structure of the credit
card market, including the merchant fee system. Card
companies are to prepare a reasonable system by the first half
of 2012.
2012: The National Assembly approves The Revision of the
Credit Finance Law (effective in January 2013). Among other
things, this revision requires credit card companies to apply
Spain
(EU 1986)
South Africa
Switzerland
Turkey
special merchant fee rates determined by FSC to merchants

with annual revenue under a certain level (determined by
presidential decree) and to provide relevant information to
authorities.
2005: The Spanish Competition Tribunal denies authorizing the interchange fee arrangements of the Spanish card schemes. In
December, Spanish card networks and merchants reach an agreement coordinated by the Spanish Ministry of Industry, Tourism
and Trade for interchange fees to be reduced immediately and progressively (effective in November 2006).
2009: The maximum limits for credit and debit card interchange fees are extended for the 2009/2010 period. The Council of the
National Competition Commission (CNC) concludes that applying the maximum limits derived from the cost studies to intrasystem interchange fees would not be appropriate.
2010: In December, the CNC Council declares the monitoring of the agreement closed to the extent that it expired on December
31, 2010. Since January 2011, Spanish card schemes have been free to decide upon the level of default interchange fees, while
still enduring maximum transparency.
2014: In July, the Government approves caps on interchange fees. For a 20 or less transaction, the cap is set at 0.1% for debit
cards and 0.2% for credit cards. For a higher value transaction (higher than 20), the cap is set at 0.2% or 0.07, whichever is less,
for debit cards and 0.3% for credit cards. These caps apply to four-party schemes only. The caps become effective on September
1, 2014.
2014: After the 2008 inquiry report that recommended interchange fee regulation, the South African Reserve Bank determines the
levels of debit and credit card interchange fees based on whether the issuer and the acquirer of a given transaction are a compliant
of EMV (for card-present) and 3D secure (for card-not-present). These rates become effective on January 1, 2015.
2005: The Swiss Competition Commission and credit card
2009: The Commission opens a preliminary investigation into
issuers agree to reduce interchange fees from 1.65-1.70% to
Maestros introduction of an interchange fee.
1.30-1.35%.
2010: The Commission opens a preliminary investigation into
Debit MasterCards introduction of a domestic fallback
2009: The Commission again opens an investigation into
interchange fee.
interchange fees for Visa and MasterCard credit cards.
2010: The Commission sets the maximum interchange fee for
2011: The Secretariat of the Competition Commission closes
2010 at 1.058%.
preliminary investigations. It concludes that an interchange fee
2011: The Commission reduces the maximum interchange fee
for Maestro card transactions could violate the Act on Cartels
to 0.990% for 2011.
while an interchange fee for Debit MasterCard might be
possible within certain limits, e.g., its market share remains
below 15% and the interchange fee is, on average, no more than
0.20 Swiss francs per transaction.
2005: The Turkish Competition Authority (TCA) makes a decision on Interbank Card Centre (BKM)s clearing commission rate
by member banks. The decision states that, in order to grant exemption to the clearing commission formula proposed by the
consultancy firm on behalf of BKM, the formula must be adjusted for certain cost items.
2009: BKM requests an indefinite exemption for setting joint interchange commission rates for credit card but the TCA decides
that an individual exemption might be granted to the joint rates for three years if certain conditions are met.
United States
Venezuela
2011: The Federal Reserve Board sets the debit card

interchange fee standards for regulated banks whose asset size
exceeds $10 billion (at the bank holding company level). Debit
cards issued by banks with less than $10 billion in assets and
reloadable prepaid cards are exempted from the interchange fee
standards.
(This regulation is the subject of current litigation).
2008: In December, Resolucin N 08-12-01 is passed (effective January 2009) which states that the Board of the Central Bank of
Venezuela will set limits on merchant discount rates and trade commissions for payments made by debit and credit for each
merchant category; these rates will be reviewed annually.
Credit
Debit
2012: The Estonian Competition Authority terminates the proceedings of the interchange fees for card payments after a number of
reductions in interchange fees made by banks.
Mid 2000s: The Finnish Competition Authority launches an
investigation into interchange fees on EMV cards (0.31%).
Traditional magnetic stripe cards charge merchants between
zero to 0.05 euro per transaction.
2006: The Bundeskartellamt (the competition authority in
Germany) receives a legal complaint from the German Retail
Association, alleging that fees charged to merchants for
MasterCard and Visa transactions, which average 150 basis
points, prevent widespread credit card acceptance in Germany.
2004: On the initiative of the Ministry of Finance,
Kredittilsynet (the financial supervisory authority) establishes a
project group to report on competitive conditions in the
Norwegian market for international payments and charge cards.
2005: Norges Bank (the central bank of Norway) states in its
2005 Annual Report that the regulation of interchange fees is
also being considered.
2011: The Romanian Competition Council (RCC) opens a sector inquiry, targeting four main areas, including setting the
interchange fees on payment cards.
2013: The RCC publishes the report of the inquiry and finds the interchange fees of Visa and MasterCard are higher in Romania
b. Investigations initiated
Country
Estonia
(EU 2004)
Finland
(EU 1995)
Germany
(EU 1952)
Norway
Romania
(EU 2007)
Singapore
South Africa
United
Kingdom
(EU 1973)
than those in other European countries.

2013: The Competition Commission of Singapore (CCS) concludes that Visas multilateral interchange fee scheme does not
violate Singapores Competition Act.
2004: The Task Group for the National Treasury and the South African Reserve Bank recommends that the Competition
Commission investigate the possibility of a complex monopoly in the governance and operation of the national payment system.
2006: Following the findings in the report The National Payment System and Competition in the Banking Sector, the Commission
begins a public inquiry into bank charges and access to the payment system.
2008: In December, the inquiry report is published, recommending regulation in the setting of interchange fees.
2005: In September, the Office of Fair Trading (OFT) finds that 2007: OFT expands the scope of its investigation into
MasterCards interchange fee arrangements are illegal. In
interchange fees to include immediate debit cards.
October, the OFT issues a statement of objections against Visa
regarding its agreement on multilateral interchange fees.
2006: In February, OFT launches a new investigation against
MasterCard. In June, the OFTs finding on MasterCard is
appealed and OFT consents to its decision being set aside by
the Competition Appeal Tribunal, due to a change made by
MasterCard in setting interchange fees. OFT refocuses on credit
card interchange fees set by MasterCard and Visa.
2012: The UK government submits a response to the Court of Justice in support of the European Commissions decision and the
General Court judgment (regarding MasterCard).
Annex. Zero interchange fee schemes

Current Zero Interchange Fee Scheme
Canada
Interac (debit)
Denmark
Dankort (debit)
(EU 1973)
New Zealand
EFTPOS (debit)
Norway
Bank-Axept (debit)
Previous Zero Interchange Fee Scheme
Belgium
2007: Bancontact/Mister Cash (debit) introduces explicit interchange fees.
(EU 1952)
Finland
2011: Pankkikortti (debit) phases out at the year-end.
(EU 1995)
Germany
2006: POZ (debit) phases out.
10
(EU 1952)
Luxembourg
(EU 1952)
Netherlands
(EU 1952)
2013: ELVs (direct debit) phase out in February 2016 is planned.

2011: Bancomat (debit) phases out at the year-end.
2006: PIN (debit) introduces explicit interchange fees.
2011: PIN brand is discontinued.
11
2011: In October, the prohibition on surcharging for

domestic credit cards is lifted.
As of 2011, surcharging is allowed.
1996: The ban on surcharging for Interac transactions is

lifted through a consent order by the Competition Bureau
of Canada.
2010: The Ministry of Finances code of conduct for credit and debit cards requests that payment card network rules ensure
that merchants are allowed to provide discounts for different methods of payment.
2013: The Competition Tribunal dismisses the case brought in 2010 by the Commissioner of Competition against
MasterCard and Visa over no-surcharge rule and notes that the proper solution to the issue is a regulatory framework.
As of 2011, surcharging is allowed, but the payment instruments for which surcharges may be requested are specified.
Credit
Debit
2003: Prohibition on surcharging is lifted.
2006: Prohibition on surcharging is lifted for Visa and
2012: The Reserve Bank of Australia changes the
MasterCard signature debit card transactions.
surcharging Standards, which allows credit and scheme debit
card networks to cap the amount of surcharges at amounts
reasonably related to merchants cost of card acceptance
(effective in March 2013).
As of 2011, surcharging is prohibited but offering discounts is allowed.
2. Surcharges and Discounts (Actions taken by public authorities)

Country
Australia
Austria
(EU 1995)
Belgium
(EU 1952)
Bulgaria
(EU 2007)
Canada
Cyprus
(EU 2004)
Czech Republic
(EU 2004)
Denmark
(EU 1973)
Estonia
(EU 2004)
European Union
2007: The Payment Services Directive (PSD) does not allow payment service providers to prevent the payee from
requesting from the payer a charge or from offering him a reduction for the use of a given payment instrument. However,
the PSD allows Member States to forbid or limit the right to request charges, taking into account the need to encourage
competition and promote the use of efficient payment instruments.
2009-2010: The PSD is implemented into national law.
2013: The European Commission proposes revisions to the Payment Services Directive (PSD2), which bans surcharges on
the interchange-fee-regulated cards but allows surcharges on non-regulated cards (e.g., corporate cards and three-party
12
Fiji
Finland
(EU 1995)
France
(EU 1952)
Germany
(EU 1952)
Greece
(EU 1981)
Hungary
(EU 2004)
Ireland
(EU 1973)
Israel
Italy
(EU 1952)
Latvia
(EU 2004)
Lithuania
(EU 2004)
Luxembourg
(EU 1952)
Malta
(EU 2004)
Mexico
Netherlands
(EU 1952)
New Zealand
Poland
(EU 2004)
Portugal
scheme cards).
2012: After several merchants were found to be applying surcharges to credit card users despite the practice being
prohibited by Fijis Merchant Services Agreement, the Reserve Bank of Fiji intervenes by upholding the No Surcharge
Rule for both credit and debit card payments effective November 1, 2012.
As of 2011, surcharging is allowed but the amounts of surcharges are required to be reasonable and not to exceed the
payees actual costs.
As of 2011, surcharging is prohibited, but offering discounts is allowed.
1993: The ban on surcharging is lifted.
As of 2005, most merchants do not surcharge; some deep discount retailers offer cash discounts.
As of 2011, surcharging is prohibited, and offering discounts is limited to certain payment instruments.
As of 2011, surcharging is not prohibited.
1993: The Mexican Competition Commission reaches an agreement with a number of banks, forbidding them from
prohibiting merchants from offering discounts for cash payments in their acquiring contracts.
2009: Agreements between the Commerce Commission and Visa/MasterCard require Visa/MasterCard to allow merchants
to surcharge.
As of 2011, surcharging is allowed, but the amount of surcharges is determined either by legislation or the payee.
13
(EU 1986)
Romania
(EU 2007)
Slovakia
(EU 2004)
Slovenia
(EU 2004)
Spain
(EU 1986)
Sweden
(EU 1995)
Switzerland
United Kingdom
(EU 1973)
United States

As of 2011, surcharging and offering discounts are limited to certain payment instruments.

As of 2011, surcharging is generally prohibited but offering discounts is allowed.
2011: In December, HM Treasury announces that the government will ban excessive surcharges on all forms of payment,
and extend the ban across most retail sectors. The announcement also states that the UK will become the first European
country to act by implementing forthcoming European legislation early to ban this practice before the end of 2012.
2012: In December, The Consumer Rights (Payment Surcharges) Regulations 2012 ban merchants from charging
consumers more than the cost borne to them for accepting a given means of payment (effective in April 2013).
2010: The Justice Department (DOJ) files a lawsuit against American Express, Visa, and Mastercard alleging that their
merchant fees and restrictions imposed on merchant practices violate antitrust law. The DOJ reaches a settlement with Visa
and MasterCard to eliminate rules preventing merchants from offering consumers discounts, rewards, and information
about card costs.
2011: In July, the settlement is approved by a federal judge.
2012: A preliminary settlement between merchants and Visa, MasterCard, and several large issuer banks requires Visa and
MasterCard to allow merchants to impose surcharges on credit card transactions, subject to a cap and other consumer
protection measures. The change would take effect in early 2013.
2013: In December, the settlement between merchants and Visa, MasterCard, and several large issuer banks is approved by
a federal judge.
2014: Merchants and their trade associations appeal the Visa/MasterCard settlement. In July, the trial of the case brought
by DOJ against American Express begins in federal court.
14
Sources:
Interchange and Merchant Service Fees
Argentina
http://www.oecd.org/dataoecd/31/19/38820123.pdf
http://www.protectora.org.ar/legislacion/ley-25065-tarjetas-de-credito/43/ (in Spanish)
http://www.1st-annapolis.com/interchange-world-difference
http://ir.americanexpress.com/phoenix.zhtml?c=64467&p=irol-reportsAnnualArchive
http://www.bcra.gov.ar/pdfs/marco/Marco%20Legal%20completo.pdf (in Spanish)
Australia
http://www.rba.gov.au/payments-system/reforms/review-card-reforms/pdf/review-0708-issues.pdf
http://www.rba.gov.au/payments-system/reforms/cc-schemes/cc-fees-benchmark.pdf
http://www.rba.gov.au/media-releases/2006/mr-06-02.html
http://www.rba.gov.au/payments-system/legal-framework/standards/interchg-visa-debit.pdf
http://www.rba.gov.au/payments-system/reforms/debit-card-systems/201211-reg-frmwrk-eftpos-sys/pdf/fin-ref-ris-112012.pdf
Austria
http://www.concurrences.com/article.php3?id_article=23935&lang=fr
http://www.concurrences.com/article.php3?id_article=14787
http://europa.eu/rapid/pressReleasesAction.do?reference=MEMO/07/40
Belgium
http://www.freshfields.com/publications/newsletters/newsletter.asp?typeid=6&newsletterid=31&contentid=74
http://www.concurrences.com/article.php3?id_article=12426#nb1
http://www.pseconsulting.com/pdf/articles/interchange/consequences_of_mif_mar05.pdf
http://ec.europa.eu/internal_market/finservices-retail/docs/finfocus/finfocus3/finfocus3_en.pdf
http://ec.europa.eu/competition/sectors/financial_services/information_paper_payments_en.pdf
Brazil
http://siteresources.worldbank.org/INTPAYMENTREMMITTANCE/Resources/JoseMarciano.pdf
http://www.reuters.com/article/idUSN1427355720090715
http://www.reuters.com/article/idUSN1918952520100419
15
Canada
http://www.kc.frb.org/publicat/pscp/2005/Weiner-Wright.pdf
http://www.kc.frb.org/publicat/pscp/2005/Regulatory_panel.pdf
http://www.parl.gc.ca/40/2/parlbus/commbus/senate/com-e/bank-e/rep-e/rep04Jun09-e.pdf
http://www.fin.gc.ca/n10/data/10-049_1-eng.asp
http://www.budget.gc.ca/2014/docs/plan/ch3-4-eng.html
Chile
http://www.tdlc.cl/noticias/detalle.php?id=8 (unavailable as of February, 2012)
China
http://www.kpmg.com/CN/en/IssuesAndInsights/ArticlesPublications/Documents/card-payments-aspac-0906.pdf
http://ebusiness.mit.edu/research/papers/212_jhausman_chinaepayment.pdf
http://www.paymentssource.com/news/Chinese-Ministry-Proposes-Capping-Merchant-Card-Acceptance-Fees-3008969-1.html
http://www.lexology.com/library/detail.aspx?g=ef4125ae-b93a-4fb3-9d72-3d36246157ce
Colombia
http://www.consumidoresint.cl/novedades/detallenovedad.asp?id=1148001800 (unavailable as of November, 2007)
Denmark
http://www.forbrug.dk/fileadmin/Filer/FO_English/UK-betalingsmiddellov.pdf (unavailable as of February, 2012)
European Payment Cards Yearbook 2005-6.
Estonia
http://www.konkurentsiamet.ee/?id=21231
European Union
http://www.kc.frb.org/publicat/pscp/2005/Friess.pdf
http://europa.eu/rapid/pressReleasesAction.do?reference=MEMO/06/260&type=HTML&aged=0&language=EN&guiLanguage=en
http://europa.eu/rapid/pressReleasesAction.do?reference=IP/10/462&format=HTML&aged=0&language=EN&guiLanguage=en
http://europa.eu/rapid/pressReleasesAction.do?reference=MEMO/10/224&format=HTML&aged=0&language=EN&guiLanguage=en
http://europa.eu/rapid/pressReleasesAction.do?reference=IP/10/1684
http://curia.europa.eu/jcms/upload/docs/application/pdf/2012-05/cp120069en.pdf
16
http://europa.eu/rapid/pressReleasesAction.do?reference=IP/12/871
http://europa.eu/rapid/press-release_IP-13-314_en.htm?locale=en
http://ec.europa.eu/internal_market/payments/docs/framework/130724_proposal-regulation-mifs_en.pdf
http://europa.eu/rapid/press-release_MEMO-13-719_en.htm?locale=en
http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-%2F%2FEP%2F%2FTEXT%2BREPORT%2BA7-20140167%2B0%2BDOC%2BXML%2BV0%2F%2FEN&language=EN
http://europa.eu/rapid/press-release_IP-14-197_en.htm
http://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:52014XC0516(01)&from=EN
http://curia.europa.eu/jcms/jcms/Jo1_6581/?dateDebut=11/09/2014&dateFin=11/09/2014
Finland
Conroy, Victoria. 2009. Finland plays its cards right, February 10, Cards International, VRL Financial News. www.vrl-financial-news.com
http://www.paymentcardyearbooks.com/country-profiles/finland.php
France
Judgment (Case A 318/02 SERVIRED Interchange fees)
http://www.autoritedelaconcurrence.fr/user/standard.php?id_rub=389&id_article=1657
http://www.cgsh.com/files/Publication/75ebe29a-10d7-4807-9972-537340418c50/Presentation/PublicationAttachment/815bf175-f645-449e-af6a5a6bea2d5f4f/National%20Competition%20Report%20Q2%202011.pdf
http://www.autoritedelaconcurrence.fr/user/standard.php?id_rub=482&id_article=2251 (in French)
http://www.reuters.com/article/2013/09/23/us-mastercard-visa-idUSBRE98M0LM20130923
Germany
http://english.zahlungsverkehrsfragen.com/subscribers/payments_at_the_pos.htm
http://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Meldungen News Karussell/08_4_2014_EC-Cash.html
http://www.lexology.com/library/detail.aspx?g=fedeafa3-90fd-4a22-808f-ff9063d10049
https://www.sepadeutschland.de/assetfile_51e7a6f84e9fc4f412000017_0f0.pdf
Greece
http://www.epant.gr/img/x2/news/news172_1_1236245010.pdf
Hungary
http://www.gvh.hu/domain2/files/modules/module25/pdf/bankkartyahasznalat_2006.pdf
http://www.gvh.hu/gvh/alpha?do=2&st=2&pg=133&m5_doc=6071
17
http://ec.europa.eu/competition/ecn/brief/01_2010/paymentcards_hu.pdf
http://english.mnb.hu/Root/Dokumentumtar/ENMNB/Kiadvanyok/report-on-paymentsystems/Jelentes_a_fizetesi_rendszerrol_2013_ENG_final.pdf
http://www.politics.hu/20130902/bill-capping-bank-credit-card-interchange-fees-submitted/
India
http://rbi.org.in/scripts/NotificationUser.aspx?Id=7304&Mode=0
http://rbi.org.in/scripts/BS_CircularIndexDisplay.aspx?Id=7422
Israel
http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=DAF/COMP/AR(2012)5&docLanguage=En
Italy
http://www.law360.com/articles/207330/italy-fines-mastercard-banks-over-interchange-fees
http://ec.europa.eu/competition/ecn/brief/05_2010/it_bancomat.pdf
http://www.lexology.com/library/detail.aspx?g=b88eefcc-4a39-4cd7-962e-ea7487a83dc7
http://www.agcm.it/trasp-statistiche/doc_download/4104-i773provvedimentodiavvioistruttoria.html (in Italian)
Latvia
http://www.kp.gov.lv/?object_id=1084&module=news
Luxembourg
http://www.abbl.lu/node/8441
Mexico
Negrn, Jos Luis. The regulation of payment cards: The Mexican experience, Review of Network Economics, 4:243-265, December 2005.
http://www.kc.frb.org/publicat/pscp/2005/Ortiz.pdf
Netherlands
http://www.nma.nl/en/competition/financial_and_business_services/banks_and_business_services_cases/default.aspx
https://www.acm.nl/en/publications/publication/12643/ACM-MasterCard-lowers-tariffs-for-credit-card-payments/
18
http://www.betaalvereniging.nl/en/fields-of-activity/debit-card-transactions-and-pos-terminals/
http://www.oecd.org/regreform/sectors/39347699.pdf
New Zealand
http://www.comcom.govt.nz/media-releases/detail/2009/commercecommissionandvisareachagre/
http://www.comcom.govt.nz/media-releases/detail/2009/commercecommissionandmastercardagr/
Norway
http://www.kredittilsynet.no/archive/f-avd_word/01/04/Regul011.doc
http://www.norges-bank.no/Upload/English/Publications/Economic%20Bulletin/2006-04/01-Payments%20history.pdf
http://www.norges-bank.no/Upload/import/front/rapport/en/bf/2005/hele_heftet.pdf
Panama
http://www.iib.org/associations/6316/files/gs2004.pdf
Poland
http://www.uokik.gov.pl/news.php?news_id=1004
http://www.uokik.gov.pl//news.php?news_id=2045
http://www.nbp.pl/homen.aspx?f=/en/system_platniczy/charges_reduction.html
http://www.warsawvoice.pl/WVpage/pages/articlePrint.php/21570/news
http://www.lexology.com/library/detail.aspx?g=1e5c3fd6-4226-42a1-b969-e50f5ff941b8
http://orka.sejm.gov.pl/proc7.nsf/ustawy/966_u.htm (in Polish)
Portugal
http://europa.eu/rapid/pressReleasesAction.do?reference=MEMO/07/40&format=DOC&aged=1&language=EN&guiLanguage=fr
Romania
http://www.consiliulconcurentei.ro/uploads/docs/items/id8334/utila_carduri_2013_english.pdf
http://www.aursf.ro/wp-content/uploads/2014/02/Speech-Alin.pdf
http://www.romania-insider.com/romanias-new-regulations-on-cash-payment-limits-and-interchange-fees-ceiling-need-parliamentapproval/114965/
19
Singapore
http://www.ccs.gov.sg/content/ccs/en/Media-and-Publications/Media-Releases/ccs-issues-a-clearance-decision-on-visas-mif-system.html
South Africa
http://www.compcom.co.za/assets/Banking/Nonconreport/9-Appendices.pdf
http://www.compcom.co.za/assets/Uploads/AttachedFiles/MyDocuments/Banking-Press-Statement.doc
http://www.compcom.co.za/technical-report/
http://www.compcom.co.za/assets/Banking/Nonconreport/8-Conclusions.pdf
https://www.resbank.co.za/Publications/Detail-Item-View/Pages/Publications.aspx?sarbweb=3b6aa07d-92ab-441f-b7bfbb7dfb1bedb4&sarblist=21b5222e-7125-4e55-bb65-56fd3333371e&sarbitem=6155
South Korea
Asia Africa Intelligence Wire. "FTC slaps 10.1 bln won fine on BC Card for cartel activities." June 30 2005.
http://www.accessmylibrary.com/article-1G1-134475090/ftc-slaps-10-1.html
http://www.paymentssource.com/news/korean-card-networks-cut-interchange-fees-3008219-1.html
Spain
http://ec.europa.eu/competition/sectors/financial_services/inquiries/sec_2007_106.pdf
http://www.rbrlondon.com/newsletters/b221e.pdf
Judgment (Case A 318/02 SERVIRED Interchange fees)
Judgment on individual exemption (Case no. A314/2002 SISTEMA 4B)
Proceedings in the case of amendment or revocation (Case no. A287/00 Euro 6000)
http://ec.europa.eu/competition/ecn/brief/01_2011/es_debit.pdf
http://www.cncompetencia.es/Inicio/Noticias/tabid/105/Default.aspx?Contentid=287318&Pag=1
http://www.ecb.int/pub/pdf/scpops/ecbocp131.pdf
http://www.lamoncloa.gob.es/lang/en/gobierno/councilministers/Paginas/2014/20140704-council-minister.aspx
https://www.boe.es/diario_boe/txt.php?id=BOE-A-2014-7064
Switzerland
http://www.news.admin.ch/NSBSubscriber/message/attachments/14452.pdf (in German)
Competition Commission. Annual Report 2005.
Competition Commission. Translation of the Decision of the Competition Commission from 25. January 2010. Concerning Preliminary
Injunctions in the Case of the Investigation under Article 27 of the Federal Act on Cartels and Other Restraints of Competition Relating to 220389: Credit Cards-DMIF II.
20
Competition Commission. To the Federal Council: Annual Report 2010 of the Competition Commission
http://www.globalcompetitionreview.com/reviews/37/sections/132/chapters/1405/
http://www.mondaq.com/x/86584/Trade+Regulation+Practices/National+Competition+Law+Report++Q3+2009
http://www.oecd.org/competition/PaymentSystems2012.pdf
Turkey
http://www.oecd.org/competition/PaymentSystems2012.pdf
United Kingdom
http://www.oft.gov.uk/shared_oft/ca98_public_register/decisions/oft811.pdf
http://www.oft.gov.uk/news/press/2006/97-06
http://www.kc.frb.org/publicat/pscp/2005/Vickers.pdf
http://www.oft.gov.uk/OFTwork/competition-act-and-cartels/ca98-current/interchange-fees/
United States
http://www.federalreserve.gov/newsevents/press/bcreg/20110629a.htm
http://www.gpo.gov/fdsys/pkg/FR-2011-07-20/pdf/2011-16861.pdf
http://www.cadc.uscourts.gov/internet/opinions.nsf/FE9EDC4B5E2C6D9E85257CA2004FB19A/$file/13-5270-1484753.pdf
Venezuela
http://www.bcv.org.ve/ley/reso081201.pdf (in Spanish)
http://www.tsj.gov.ve/gaceta/diciembre/041208/041208-39073-25.html# (in Spanish)
Surcharges and Discounts
Australia
http://www.rba.gov.au/publications/consultations/201106-review-card-surcharging/pdf/201106-review-card-surcharging.pdf
http://www.rba.gov.au/publications/consultations/201112-variation-surcharging-standards/pdf/201112-variation-surcharging-standards.pdf
http://www.rba.gov.au/payments-system/reforms/cards/201206-var-surcharging-stnds-fin-ref-ris/pdf/201206-var-surcharging-stnds-fin-ref-ris.pdf
http://www.rba.gov.au/payments-system/surcharging/index.html
21
Austria
http://ec.europa.eu/internal_market/payments/docs/framework/transposition/psd_transposition_study_report_en.pdf
Belgium
Bulgaria
Canada
http://www.ct-tc.gc.ca/CasesAffaires/CasesDetails-eng.asp?CaseID=333
http://www.ct-tc.gc.ca/CMFiles/CT-2010-010_Summary%20of%20Confidential%20Decision_317_38_7-23-2013_8408.pdf
Czech Republic
Cyprus
Denmark
Estonia
European Union
http://ec.europa.eu/internal_market/payments/docs/framework/130724_proposal-revised-psd2_en.pdf
Fiji
http://www.fijitimes.com/story.aspx?id=204526
http://www.reservebank.gov.fj/docs2/1%20Press%20Release%20No%2029%20%20RBF%20Upholds%20'No%20Surcharge%20Rule'%20for%20Fiji.pdf
22
Finland
France
Germany
Greece
Hungary
Ireland
Israel
http://www.ny.frb.org/research/conference/2005/antitrust/Spiegel.pdf
Italy
Latvia
Lithuania
Luxembourg
Malta
23
Mexico
Netherlands
New Zealand
http://www.comcom.govt.nz/media-releases/detail/2009/commercecommissionandvisareachagre/
http://www.comcom.govt.nz/media-releases/detail/2009/commercecommissionandmastercardagr/
Poland
Portugal
Romania
Slovakia
Slovenia
Spain
Sweden
http://www.iflr.com/Article/2713018/Implementation-of-the-Payment-Services-Directive.html
Switzerland
24
United Kingdom
http://www.hm-treasury.gov.uk/press_148_11.htm
http://www.legislation.gov.uk/uksi/2012/3110/made
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/175298/13-719-guidance-on-the-consumer-protection-paymentsurcharges-regulations-2012.pdf
United States
http://www.justice.gov/opa/pr/2010/October/10-at-1115.html
http://www.justice.gov/atr/cases/f262800/262864.htm
http://www.sutherland.com/files/News/6e8c1562-4474-47f2-85e9-82548465a8ef/Presentation/NewsAttachment/7c1450a9-a785-4b84-9d2784f4e1535322/CORP%20Alert%207.25.11.pdf
http://pressreleases.visa.com/phoenix.zhtml?c=215693&p=irol-newsarticlePR&ID=1714726&highlight=
http://newsroom.mastercard.com/press-releases/mastercard-enters-into-agreement-to-settle-u-s-merchant-litigations/
https://www.paymentcardsettlement.com/Content/Documents/Final%20Approval.pdf
http://www.topclassactions.com/lawsuit-settlements/lawsuit-news/31478-merchants-appeal-visamastercard-swipe-fee-class-action-settlement/
https://nrf.com/sites/default/files/Documents/NRF%20RILA%20Brief.pdf
http://www.justice.gov/atr/cases/americanexpress.html
25

Payment Card Networks Panel MaterialsPart 3 - 0

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Payment Card Networks Panel MaterialsPart 3 - 0

Diunggah oleh

Hak Cipta:

Format Tersedia

Case 1:96-cv-05238-JG-JO Document 1652 Filed 12/13/13 Page 1 of 55 PageID #: 10183

UNITED STATES DISTRICT COURT

IN RE PAYMENT CARD INTERCHANGE

KENNY NACHWALTER, P.A.

ROBINS, KAPLAN, MILLER & CIRESI L.L.P

SPERLING & SLATER

BERGER & MONTAGUE, P.C.

HANGLEY ARONCHICK SEGAL

Co-Lead Counsel for Class Plaintiffs

Counsel for the Individual Plaintiffs

ARNOLD & PORTER LLP

WILLKIE FARR & GALLAGHER LLP

HOLWELL SHUSTER & GOLDBERG LLP

PAUL, WEISS, RIFKIND, WHARTON &

Attorneys for Defendant Visa U.S.A. Inc.

Case 1:96-cv-05238-JG-JO Document 1652 Filed 12/13/13 Page 2 of 55 PageID #: 10184

MORRISON & FOERSTER LLP

SHEARMAN & STERLING LLP

Attorneys for Defendant Bank of America

Attorneys for Defendant Barclays Bank

OMELVENY & MYERS LLP

SKADDEN, ARPS, SLATE, MEAGHER &

Attorneys for Defendant Capital One Bank

Attorneys for Defendant JPMorgan Chase &

SIDLEY AUSTIN LLP

KEATING MUETHING & KLEKAMP PLL

Attorneys for Defendant Citibank, N.A.

Attorneys for Defendant Fifth Third Bancorp

KUTAK ROCK LLP

Attorneys for Defendant First National Bank

Attorneys for Defendant HSBC Finance

PULLMAN & COMLEY, LLC

Attorneys for Defendant National City

Attorneys for Defendant Texas Independent

ALSTON & BIRD LLP

PATTERSON BELKNAP WEBB &

Attorneys for Defendant SunTrust Banks,

Attorneys for Defendant Wachovia Bank,

Case 1:96-cv-05238-JG-JO Document 1652 Filed 12/13/13 Page 3 of 55 PageID #: 10185

FRIEDMAN LAW GROUP LLP

FREEDMAN BOYD DANIELS

Attorneys for Plaintiffs in No. 05-cv-5074

Attorneys for Plaintiff NATSO, Inc.

Attorneys for Objector Home Depot U.S.A.,

VORYS, SATER, SEYMOUR AND PEASE

EMERY, CELLI, BRINCKERHOFF &

Attorneys for Objector Target Corporation

Attorneys for Objector National Retail

SCHLAM, STONE & DOLAN, LLP

PERKINS COIE LLP

Attorneys for Amicus Curiae U.S. Public

Attorneys for Objector First Data

BOIES, SCHILLER & FLEXNER LLP

KIRKLAND & ELLIS

Attorneys for Objector American Express Co.

Attorneys for Objector Discover Financial

Case 1:96-cv-05238-JG-JO Document 1652 Filed 12/13/13 Page 4 of 55 PageID #: 10186

MILLER & CHEVALIER CHARTERED

THE LAW OFFICES OF JOHN J. PENTZ

Attorneys for Objector Blue Cross Blue Shield

Attorneys for Defendant Daviss Donuts and

LAW OFFICES OF EDWARD F. SIEGEL

THRASH LAW FIRM, P.A.