Reynaldo Reyes v. Netdeposit

PRECEDENTIAL

UNITED STATES COURT OF APPEALS

FOR THE THIRD CIRCUIT

_____________

No. 14-1228

_____________

REYNALDO REYES,

on behalf of himself and all others similarly situated,

Appellant

v.

NETDEPOSIT, LLC; MP TECHNOLOGIES d/b/a Modern

Payments; TELEDRAFT, INC.; NATIONAL PENN BANK;

WELLS FARGO BANK, N.A.; WACHOVIA BANK, N.A.;

ZIONS FIRST NATIONAL BANK

________________

APPEAL FROM THE UNITED STATES DISTRICT

COURT FOR THE EASTERN DISTRICT OF

PENNSYLVANIA

(D.C. No. 2-10-cv-00345)

District Judge: Honorable Juan R. Sánchez

________________

Argued

September 11, 2014

____________

Before: MCKEE, Chief Judge, SMITH and SHWARTZ,

Circuit Judges

(Opinion Filed: September 2, 2015)

______________

Howard I. Langer, Esq. (ARGUED)

John J. Grogan, Esq.

Edward Diver, Esq.

Irv Ackelsberg, Esq.

1

Peter E. Leckman, Esq.

LANGER, GROGAN & DIVER, P.C.

1717 Arch Street, Suite 4130

Philadelphia, PA 19103

Attorneys for Appellant

Jason A. Snyderman, Esq. (ARGUED)

Grant S. Palmer, Esq.

William R. Cruse, Esq.

Bridget E. Mayer, Esq.

Blank Rome LLC

One Logan Square

Philadelphia, PA 19103

Attorneys for Appellees

Charles L. Becker, Esq. (ARGUED)

Kline & Specter, PC

1525 Locust Street, 19th Floor

Philadelphia, PA 19102

Attorney for Amici Curiae

Public Interest Law Center of Philadelphia and

Community Legal Services

Julie Nepveu, Esq.

AARP Foundation Litigation

Room B4-245

601 E Street, N.W.

Washington, DC 20049

Attorney for Amici Curiae AARP, National

Association of Consumer Advocates, National

Consumer Law Center and Consumer Federation of

America

Richard B. Tucker, III, Esq.

Tucker Arensberg

1500 One PPG Place

Pittsburgh, PA 15222

Attorney for Amici Curiae American Bankers

2

Association and Independent Community Bankers of

America

Robert J. LaRocca, Esq.

Kohn, Swift & Graf

One South Broad Street

Suite 2100

Philadelphia, PA 19107

Attorney for Amici Curiae Richard Blumenthal,

Robert Casey, Edward J. Markey and Allyson Y. Schwartz

Francis X. Riley, III, Esq.

Charles M. Rowan, Jr., Esq.

Saul Ewing

650 College Road East

Suite 100

Princeton, NJ 08540

Attorney for Amicus Curiae Risk Management

Association

______________

OPINION OF THE COURT

______________

McKEE, Chief Judge.

Reynaldo Reyes appeals the District Court’s denial of

his motion to certify a class to sue for alleged civil violations

of the Racketeer Influenced and Corrupt Organizations Act

(“RICO”), 18 U.S.C. §§ 1962(c), (d). The defendants are

Zions First National Bank (“Zions Bank”) and its payment-

processor subsidiaries, Netdeposit, LLC and MP

Technologies (together, “Modern Payments”).

Reyes alleges that the defendants conspired to conduct

a fraudulent telemarketing scheme that caused unauthorized

debits from bank accounts owned by himself and members of

the proposed class.

3

The District Court held that class certification was

inappropriate because there were no issues common to the

class and Reyes could therefore satisfy neither the

commonality requirement of Federal Rule of Civil Procedure

23(a), nor the predominance requirement of Rule 23(b)(3).

This interlocutory appeal followed.1

I. BACKGROUND

Reyes alleges that the defendants conspired to conduct

a fraudulent scheme whereby certain telemarketing firms

would contact unsuspecting individuals and offer them

something of little or no value. Reyes alleges that, during

unsolicited phone calls with unsuspecting consumers, the

telemarketers would obtain bank account information which

was used to make unauthorized debits from the the

consumers’ bank accounts.2 In Reyes’ case, in an unsolicited

phone call in November 2007, telemarketer NHS Systems

told Reyes that he qualified for a free government grant.

NHS Systems then requested Reyes’ bank account

information, which he provided.

Telemarketers such as NHS Systems cannot readily

obtain funds directly from consumers’ bank accounts because

most banks are extremely reluctant to allow them to debit

accounts. Accordingly, telemarketers usually contract with

payment processing entities that debit bank accounts on the

1

See 28 U.S.C. § 1292(e); Fed. R. Civ. P. 23(f). The

District Court also granted various amici curiae leave to file

briefs in support of Reyes’ motion for class certification.

2

Reyes alleges that, besides phone-based marketing,

the scheme included deceptive mailers, Internet solicitation,

and “slamming,” a process whereby a telemarketer acquires a

consumer’s bank account information from another entity and

either contacts the consumer under the pretense of attempting

to verify the information or simply transfers money from the

consumer’s account. JA 513-15.

Because Reyes alleges that the merchants operated in

the same way and as part of a fraudulent enterprise, Appellant

Br. at 18, we refer to all of them as “telemarketers” for the

sake of simplicity.

4

telemarketer’s behalf. In Reyes’ case, NHS Systems did

exactly that. It provided Reyes’ bank account information to

Modern Payments, a third-party payment processing agency

and subsidiary of Zions Bank. Modern Payments then caused

Zions Bank to initiate an Automated Clearing House

(“ACH”) debit of Reyes’ bank account at Commerce Bank.

Pursuant to Zions Bank’s request, Reyes’ funds on deposit

with Commerce Bank were transferred to Modern Payments’

account at Zions Bank, and ultimately transferred to NHS

Systems, the Originator. Two debits were processed from

Reyes’ account using this ACH debit process, one for $29.95

and another for $299.95.3 Reyes v. Zions First Nat’l Bank, ---

F. Supp. 2d ---, Civ. Action No. 10-345, 2013 WL 5332107,

at *1 (E.D. Pa. Sept. 23, 2013).

The ACH debit process is an alternative to traditional

checking which is based on the transfer of paper instruments.

When an ACH transfer occurs, an Originating Depository

Financial Institution, like Zions Bank, initiates an ACH entry

at the request of an Originator (such as NHS Systems). Such

requests can also be made through third parties at the request

of the Originator. As we have just explained, Reyes alleges

that Modern Payments caused Zions Bank to take ACH debits

from bank accounts owned by him and other members of the

putative class at the request of the telemarketers, like NHS

Systems.

The Originating Depository Financial Institution (here,

Zions Bank) aggregates payments from customers and

transmits the payments in batches to an ACH Operator (either

the Federal Reserve or the Clearing House). The Operator

receives and sorts the batched payments and makes the funds

available to the Receiving Depository Financial Institution.

That institution, in turn, debits or credits the relevant account

based on the ACH entry. Here, Reyes alleges that Modern

3

Although many (if not all) of the debits involved in

this appeal are sufficiently small to be considered de minimis

by some, the cumulative amount is substantial. In fact, the

conduct alleged here has attracted the attention of the

impressive array of amici who have filed briefs in this matter,

including the AARP, four members of the United States

Congress, and the American Bankers Association.

5

Payments credited the accounts of each of the telemarketers

in the amount of funds fraudulently debited from

unsuspecting victims. Both Modern Payments, as payment

processor, and Zions Bank, as the processor’s bank, would

have collected a fee and then deposited the balance of

amounts debited from consumers directly into the account of

the telemarketer.4

Reyes alleges that his bank account and the accounts

of other consumers subjected to this fraudulent enterprise did

not have sufficient funds to satisfy many of the unauthorized

debits.5 Reyes called NHS Systems to complain about the

two withdrawals from his account, but he alleges NHS

Systems provided him with a misleading audio recording.6

Thereafter, Reyes initiated this action on behalf of

himself and a class of all individuals in the United States as to

whom ACH debit entries or [remotely created check] drafts

on their accounts were prepared by defendants Netdeposit,

Modern Payments, or Teledraft during the four-year period

immediately preceding the filing of this action and finally

charged to the class members’ bank accounts by a

Telemarketer, or pursuant to information provided to

defendants [Netdeposit, Modern Payments, or Teledraft] by

4

Though this sounds cumbersome, it is nothing more

than a series of computer entries or lines of computer code;

the entire series of transactions can easily be completed, and

the funds transferred and received, in one to two business

days.

5

It is not difficult to understand why the accounts did

not have a sufficient balance to pay the amount of the

attempted debits. If the depositors had never authorized the

debits, they would have had no way of knowing that the

balance in their accounts was not sufficient to satisfy the

amounts of the attempted withdrawals and no reason to take

steps to ensure that their balances were sufficient to cover

those debits.

6

According to NHS Systems, the recording

established that Reyes consented to the disputed debits.

6

the Telemarketers, or who otherwise incurred any bank

charges as a consequence of such ACH debit entries or

[remotely-created checks].

JA 526 ¶ 63.

The legal theory underlying this suit is that the

defendants were operating a RICO enterprise that was a total

sham.7 The proof that Reyes offers to support that claim

includes the inordinately high “return rates” of the

telemarketers who did business with Modern Payments and

Zions Bank. “Return rates” refer to how often an ACH debit

cannot be completed. There are many reasons why a

transaction may not be completed. These include (but are not

limited to) insufficient funds or a customer complaint

regarding the transaction. These complaints can result in the

debited funds being credited back to the consumer’s bank

account. Return rates are collected by NACHA (previously

known as the “National Automated Clearing House

Association”), a not-for-profit association that administers

and manages the ACH Network and other electronic

transactions.

Reyes stresses that “the lowest return rate at issue

[here] was 25 times the national average of 1.25%.”

Appellant Br. at 26 (emphasis in original). Moreover, that

was the telemarketer with the lowest rate. As Reyes notes, he

produced testimony that “[m]ost were over 50 times that

average.” Id. at 25-26 (citing JA 661, 663). According to

Reyes, the high return rates establish that Zions Bank and

Modern Payments had reason to know that the telemarketers

were engaged in fraudulent conduct.

Reyes produced additional evidence to support his

argument. That evidence included internal e-mails wherein

Zions Bank, Netdeposit, and Modern Payments discussed the

7

As noted, Reyes alleges that this fraudulent scheme

constituted an “enterprise” under RICO, 18 U.S.C. § 1962(c).

However, since the only issue before us is the District Court’s

refusal to certify a class under Rule 23, we need not discuss

the elements needed to prove a RICO conspiracy except

insofar as they may bear upon issues of class certification.

7

“staggering” return rates, JA 686, and the “alarm” that their

high return rates had caused NACHA. Id. at 814-15.

According to Reyes, “[w]ithin the first three months of

opening Modern Payments, for example, Zions was notified

of the likelihood of fraud by the Federal Trade Commission

(“FTC”), NACHA, and its own officers.” Appellant Br. at

28. There was also evidence that these defendants

communicated concern that they may be at risk for some of

their business lines being used for money laundering. JA

692.

Reyes claims the evidence establishes that “[e]ach of

the[] frauds operated in the same way, luring consumers with

some kind of ‘card’ or ‘benefit’ to obtain their bank account

information.” Appellant Br. at 18.8 These included

“promises of un-secured loans, pre-paid credit cards,

merchandise-club cards, identity-theft protection plans, and

health-care discount plans.” Meyer Decl., JA 1639-40 ¶ 19.

Thus, although there is some variation in the “benefits”

offered and the particular way each fraud was perpetrated,

Reyes asserts that there is no need to examine each individual

transaction “[b]ecause the business model of these schemes is

inherently fraudulent[.]” Appellant Br. at 19.

A. REYES’ EXPERT WITNESSES

Reyes offered several witnesses at the hearing on class

certification, including three experts. Amelia H. Boss, a

Drexel Law School Professor, testified as an expert on

banking and banking practices. Robert J. Meyer, a Wharton

Business School Professor, and Barbara A. Blake, a former

investigator for the Iowa Attorney General, both testified as

experts on fraudulent marketing practices. Although we refer

8

Reyes also cites numerous proceedings against some

of the telemarketers involving the FTC and other entities that

led to conclusions that the telemarketers were engaged in

fraud. See Appellant Br. at 4, 16, 20-21. The defendants

point out, and Reyes does not dispute, that the government

enforcement actions involving the telemarketers in this case

were initiated after the defendants stopped doing business

with them, and the defendants were not themselves parties to

any of those actions. Appellee Br. 49.

8

to their testimony throughout our discussion, it is helpful to

provide an initial summary of the testimony of each.

1. Amelia H. Boss

Professor Amelia Boss repeatedly highlighted “the

excessively high rates of returns generated by Modern

Payments and its clients, of which Zions was well aware,”

and asserted that the rates provided clear evidence of

fraudulent activity. Boss Decl., JA 582-92 ¶¶ 66-87.

According to Professor Boss, most “banks never (or rarely)

have debits returned as ‘unauthorized.’” Id., JA 582 ¶ 66.

Indeed, Zions Bank did not have any unauthorized returns

until it began working with Modern Payments. Boss pointed

out that when Modern Payments and Zions Bank began

working together in September 2006, the “number of

unauthorized returns immediately began to increase

dramatically from zero . . . to 799” in two months, and up to

2,632 unauthorized returns by February of the following year.

Id., JA 587-88 ¶ 78 (emphasis in original). According to

Professor Boss, when adjusted for size, Zions Bank had the

worst return rate of any bank in the country. Id.

Professor Boss also testified that Zions Bank had

numerous NACHA violations, knew that many of its

customers were high risk, and yet it completed inadequate

due diligence. Id., JA 587 ¶ 77. She explained that NACHA

“prohibits the use of the ACH system for telemarking

payments initiated by outbound telemarketing.” Id., JA 575 ¶

46.9 NACHA imposed this prohibition because it recognized

that telemarketers have a high likelihood of fraud. Id., JA

575 ¶ 46.

NACHA requires that transmission of debit and credit

entries and entry of data over the ACH Network be entered

with an appropriate label or code. Id., JA 561-63 ¶¶ 15-16,

9

See also Appellant Br. at 11-12 (discussing “rules

that prohibit banks from entering transactions derived from

outbound telemarketing (i.e., the type of cold-calling that Mr.

Reyes experienced).”) (citing JA 672 (2008 ACH Rules,

NACHA, Subsection 14.1.63)); Boss Decl., JA 597-601 ¶¶

98-107.

9

18. For instance, transactions originating on the Internet, like

those from the allegedly fraudulent Internet merchants here,

are labeled “WEB,” while certain telephone-initiated debits,

like those from telemarketers, are labeled “TEL.” Id., JA

562-63 ¶ 18. According to Professor Boss, Zions Bank and

Modern Payments failed to adhere to this policy because

Modern Payments often labeled TEL or WEB transactions as

“PPD” transactions. “PPD” refers to transactions that are

“obtained in pre-existing, signed writings.” Id., JA 597-98 ¶

99. They are, therefore, the safest category of transaction.

Mislabeling made it appear that the consumer had actually

authorized his or her bank account to be debited even though

there were no such written agreements.

“Zions knew that th[ese] representation[s] w[ere]

false.” Id., JA 597-98 ¶ 99. Yet, “[d]espite being advised of

[its] use of improper codes[,] Zions continued to attach the

false codes and continued to misrepresent the warranties.”

Id., JA 598 ¶ 101. According to Reyes, the mislabeling made

it possible to “hid[e] . . . the true source of the debits[]” “from

victims’ banks and regulators[.]” Appellant Br. at 30 (citing

Boss Decl., JA 597-99 ¶¶ 98-102). If accepted by a

factfinder, this testimony could obviously establish that the

PPD code was used to conceal the true nature of these

transactions. Accordingly, Professor Boss concluded that

Zions Bank

knowingly allowed the system to be used by

High Risk originators[, like Modern Payments,]

whom it had to know were engaged in fraud in

clear disregard of the governing rules and

standards, . . . made knowing

misrepresentations to the other participants in

the ACH system (including the consumer

receivers), . . . facilitated the misuse of the

system by Modern Payments and its customers,

and . . . did all this knowing of facts from

which, as a banking institution, it had to know

fraud was taking place.

Boss Decl., JA 601 ¶ 107.

2. Robert J. Meyer

10

Professor Robert Meyer testified that the Federal

Reserve Bank has concluded that 10% return rates are prima

facie evidence of fraud.10 Meyer Decl., JA 1636 ¶ 11.b.

Here, the return rates “rang[ed] from 30% to almost 90%[.]”

Id., JA 1636 ¶ 11.a. In addition, Professor Meyer pointed to

FTC enforcement actions and these telemarketers’ products

and services to conclude that the telemarketers operated

“schemes designed to obtain victims’ bank account

information by deception and to use that information to debit

victims’ accounts as quickly [and] as often as possible before

the victim understood what had occurred.” Id., JA 1634-35

¶¶ 8-9.

According to Professor Meyer, telemarketers

NHS/PHS, the Platinum Benefit Group, Vexeldale (also

known as Market Power Marketing Solutions, Sourdale, and

ZaZoom), RxSmart, Low Pay, and Group One Networks

offered a variety of products that were either valueless or

significantly devalued. Id., JA 1633, 1639-40 ¶¶ 6, 19.

Although there was some variation in the manner in which

the unauthorized debits were accomplished, Professor Meyer

concluded that they had the same components of deception:

(1) each was obtained from a first party at a small or

10

The defendants contest this statement, noting that

“[t]here is no law or regulation supporting this contention.”

Appellee Br. at 47 (emphasis in original). Professor Meyer,

like Investigator Blake and Reyes, who also refer to the

Federal Reserve Bank’s prima facie evidence of fraud, fails to

include a source. The record presented, however, does

contain an unrelated complaint quoting an unnamed Federal

Reserve official as writing that a 10% return rate “would

likely be regarded by bank supervisory agencies and/or law

enforcement agencies as prima facie evidence that your bank

knew or should have known that your [third-party payment

processors and/or merchants] had engaged in fraudulent

activities.” JA 1695 ¶ 64 (alteration in original). Despite

whatever dispute the parties have regarding the Federal

Reserve Bank’s position on return rates, neither party disputes

that the vast majority of U.S. banks have rates of returns that

are close to (or actually are) zero, while here, that number is

30% to 90%.

11

negligible cost; (2) all were used to debit consumer accounts

for exorbitant amounts, (3) none of these items conveyed any

net value to the consumers, and (4) all were sufficiently

complex, initially concealing their fraudulent nature to the

victim. Id., JA 1639-40 ¶ 19. Professor Meyer based his

opinion on his review of “telephone scripts and web landing

pages used in the marketing of the products and/or programs,

internal e[-]mails, and tables of ACH return rates for the

programs[.]” Id., JA 1633 ¶ 6.

3. Barbara A. Blake

Reyes’ expert Investigator Barbara Blake examined

each scheme in detail and also stated that return rates over

10% were prima facie evidence of fraud. Blake Decl., JA

1614-15 ¶ 19.

Investigator Blake looked at the operations of each

entity and other evidence before concluding that the

telemarketers operate “fundamentally fraudulent schemes[.]”

Id., JA 1627-28 ¶ 85. She noted that the “[h]igh return rates

are the plainest hallmark of mass-marketing fraud conducted

through the banking system. This has been recognized in

publications of the Comptroller of the Currency, FinCen,

NACHA and the FDIC.” Id., JA 1614 ¶ 18.

Investigator Blake concluded that “each of the

schemes at issue was a totally fraudulent mass-marketing

fraud” because each of them “uses purported ‘products’ and

‘services’ that are routinely used as part of mass-marketing

fraud schemes.” Id., JA 1613 ¶ 15. “Th[e] strong

presumption of fraud [evidenced by the high return rates] is

confirmed by the nature of the schemes . . . , which, once they

are understood, are plainly fraudulent on their face.” Id., JA

1615 ¶ 21. Investigator Blake’s conclusion was based on the

telemarketers’ sales scripts, the worthless nature of the

purported “products,” and the history of regulatory actions

taken against the telemarketers and their principals. Id., JA

1613-27 ¶¶ 15-84.

B. FACT WITNESSES

12

The District Court also heard the testimony of two fact

witnesses whose testimony will be discussed in greater detail

below.11 They were: Wayne D. Geisser, who was installed as

the NHS Systems receiver after the FTC obtained a

preliminary injunction against NHS Systems, and Jeanette A.

Fox, who was a senior director of risk investigation at

NACHA. Appellant Br. at 12, 26. Geisser and Fox testified

that high return rates are not dispositive of fraud. Fox Dep.,

JA 2987-2991; Geisser Dep., JA 1155, 3055. Fox explained

that an unauthorized return could be the result of a dispute

over an agreement, the amount charged, or the timing of the

charge. Fox Dep., JA 2975. Nevertheless, Fox did believe

that the rate of returns here was high enough to warrant an

investigation. Id., JA 1145. Even she agreed that the rates

strongly suggested fraud. Id., JA 2977 (stating there is

“[p]robably not” a legitimate reason for excessive return

rates). Geisser testified that the NHS Systems health discount

program brochures, fulfillment packages, purchase orders,

and contracts with call centers appeared to be legitimate,

Geisser Dep., JA 3049-50, even though he agreed that NHS

Systems was totally fraudulent. Id., JA 1161.

C. DEFENDANTS’ EXPERT WITNESSES

The defendants presented two experts: Kathleen O.

Milner, an expert in the ACH Network and NACHA rules and

guidance, who worked for many years in private financial

services, and Peter G. Djinis, a lawyer who specializes in

compliance concerning federal anti-money laundering and

Bank Secrecy Act laws and regulations with experience in the

Department of the Treasury.

1. Kathleen O. Milner

Kathleen Milner provided an overview of the ACH

system and NACHA. She explained that, although “NACHA

Operating Rules are contractually binding on parties to ACH

payments,” like the NACHA Operating Guidelines, they do

not “have the force of law[.]” Milner Decl., JA 3781-82 ¶ 11.

11

Fox and Geisser were called by plaintiff’s counsel as

“fact witnesses.” JA 2696; Fox Decl., JA 2972.

13

Milner also discussed high return rates, noting that

high returns rates “may be indicative of fraud,” but believed

that they are not “conclusive evidence of fraud.” Id., JA 3784

¶ 17. In support of this assertion, Milner summarized various

return codes that are used to identify transactions that are not

properly authorized, highlighting the ones used in this case

that are not dispositive of fraud. Id., JA 3785-89 ¶¶ 20-29.

She believed that, “without investigating the facts and

circumstances of each purported fraudulent transaction, you

are left simply with conjecture[]” about whether fraud is

involved. Id., JA 3789 ¶ 29. She also testified that, “[t]o

conclude that the merchants were in fact complete frauds and

to further conclude that the Zions Defendants knew them to

be so would require an individualized analysis of each

merchant and the Zions Defendants’ relationship with each

merchant.” Id.

Milner rejected Reyes’ assertion that the mislabeled

codes (labeling “TEL” and “WEB” transactions as “PPD”

transactions) indicate fraud. Rather, she concluded that the

mislabeled classifications were the result of a problem with

Modern Payments’ software. Id., JA 3791 ¶ 33. According

to Milner, using the TEL code liberally, not solely when there

was a preexisting relationship or consumer-initiated call was

not fraud, but due, in part, to the fact that “the requirements of

the TEL rule were broken up throughout the NACHA

rulebook and were, therefore, difficult to locate and confusing

to follow.” Id., JA 3794-95 ¶ 39.

Finally, Milner asserted that charging a fee for

returned transactions was also not indicative of fraud because

it is “accepted industry practice” to do so “as a disincentive

for Originators having returned transactions and an incentive

to obtain proper authorizations.” Id., JA 3795 ¶ 40.

2. Peter G. Djinis

Peter Djinis discussed various laws and regulations not

at issue here, including Office of the Comptroller of the

Currency bulletins, the Bank Secrecy Act, and the U.S.

Patriot Act, Djinis Decl., JA 3767-70 ¶¶ 13, 16, 18-20, since

Reyes alleges a RICO violation. Djinis also opined that

Modern Payments and Zions Bank conducted adequate due

14

diligence of the telemarketers at issue. Id., JA 3772-76, ¶¶

25-36.

D. REYES’ USE OF THE EVIDENCE

Reyes relies on this record to allege that the defendants

operated a RICO enterprise that was a “complete sham”

lacking any legitimate business substance. He draws upon

this theory to overcome potential challenges to the

predominance requirement for class certification. Under the

“complete sham” theory, the reviewing court can focus on the

defendant’s or defendants’ conduct as a whole in order to find

proof of elements that normally require evidence about each

plaintiff, like plaintiff’s reliance. Rosario v. Livaditis, 963

F.2d 1013 (7th Cir. 1992); Cullen v. Whitman Med. Corp.,

188 F.R.D. 226 (E.D. Pa. 1999); see also In re Cmty. Bank of

N. Va. Mortg. Lending Practices Litig., --- F.3d ---, Civ.

Action No. 13-4273, 2015 WL 4547042, at *19, *21 (3d Cir.

July 29, 2015). “[T]he causation element of fraud and breach

of contract can be satisfied through objective circumstantial

evidence on a classwide basis.” Elizabeth J. Cabraser, Trends

and Developments in the Filing, Certification, Settlement,

Trial and Appeal of Class Actions, SE99 A.L.I.-A.B.A. 743,

821 (2000) (“[I]t is unnecessary and unfair to impose

modalities of proof that are specific to such nonexistent

personal relationships to insulate defendants from classwide

liability to those with whom they related on a classwide

basis.”).

E. THE DISTRICT COURT’S RESOLUTION OF

REYES’ LEGAL THEORY

The District Court acknowledged the “complete sham”

theory, explaining that in order for Reyes to succeed on this

theory under Rule 23, he must demonstrate that the

“defendant’s conduct is so wrought with fraud as to be a

complete sham[.]” Reyes, 2013 WL 5332107, at *6. Thus,

the District Court realized that, in an appropriate case, “the

class members’ participation or involvement with the

defendant is sufficient evidence that each class member

suffered damages, rendering an analysis of individual

transactions unnecessary.” Id. As the District Court also

recognized, pursuant to the “complete sham” theory,

15

misrepresentations by the defendant resulting in the plaintiffs

experiencing common damages can prove common injury in

a RICO class action. Id. (citing Cullen, 188 F.R.D. at 235).12

In fact, the District Court even went so far (with some

justification) as to compare this to the “fraud on the market”

theory. Id. (citing Lester v. Percudani, 217 F.R.D. 345, 352

(M.D. Pa. 2003).13

However, the District Court concluded that class

treatment was inappropriate here because Zions Bank and

Modern Payments collaborated with separate mass-marketing

firms and did so in different ways. The District Court

12

As discussed further below, the Supreme Court has

held that predominance requires that plaintiffs show that their

individual injuries are capable of proof at trial through

common evidence and that their damages are measurable on a

classwide basis. Comcast Corp. v. Behrend, 133 S. Ct. 1426,

1430, 1432-33 (2013); but see In re Nexium Antitrust Litig.,

777 F.3d 9, 21 (1st Cir. 2015) (“But it is well-established that

the individuation of damages in consumer class actions is

rarely determinative under Rule 23(b)(3). Where common

questions predominate regarding liability, then courts

generally find the predominance requirement to be satisfied

even if individual damages issues remain.” (internal quotation

marks, alterations, and citation omitted)).

13

The fraud on the market theory is based on

the hypothesis that, in an open and developed

securities market, the price of a company’s

stock is determined by the available material

information regarding the company and its

business. . . . Misleading statements will

therefore defraud purchasers of stock even if the

purchasers do not directly rely on the

misstatements. . . . The causal connection

between the defendants’ fraud and the

plaintiffs’ purchase of stock in such a case is no

less significant than in a case of direct reliance

on misrepresentations.

Basic Inc. v. Levinson, 485 U.S. 224, 241-42 (1988)

(internal quotation marks omitted).

16

stressed that “Reyes almost completely relies on the high

return rates as his proof of the complete sham, [but] the

returns are different for each telemarketer, [therefore] he

cannot prove this complete sham theory on evidence common

to the class since members of the class interacted with

different telemarketers.” Id. at *8. The District Court also

emphasized that, although the high return rates were common

to the telemarketers, they are insufficient to prove fraud.

With this background as our analytical compass, we

will now discuss class certification within the context of

Reyes’ claim that he produced sufficient evidence of a sham

enterprise to satisfy the requirements for class certification

under Rule 23(b)(3).

II. DISCUSSION

“The class action is an exception to the usual rule that

litigation is conducted by and on behalf of the individual

named parties only.” Wal-Mart Stores, Inc. v. Dukes, 131 S.

Ct. 2541, 2550 (2011) (quotation marks omitted).

Federal Rule of Civil Procedure 23 sets out

requirements for bringing a class action. All potential classes

must initially satisfy four prerequisites to be certified: (1)

numerosity, (2) commonality, (3) typicality, and (4) adequacy

of representation. Fed. R. Civ. P. 23(a). Additional

requirements must then be satisfied depending on whether a

plaintiff seeks certification under Rule 23(b)(1), (2), or (3).

Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 590 (3d Cir.

2012). Reyes is attempting to proceed under Rule 23(b)(3).

Accordingly, he must satisfy the additional requirements of

predominance and superiority.14 Fed. R. Civ. P. 23(b)(3).

14

When an action proceeds as a class action under

Rule 23(b)(3), “the court finds that the questions of law or

fact common to class members predominate over any

questions affecting only individual members, and that a class

action is superior to other available methods for fairly and

efficiently adjudicating the controversy.” Fed. R. Civ. P.

23(b)(3).

17

The District Court did not reach all of the prerequisites

of Rule 23(a) or the superiority requirement in Rule 23(b)(3)

because it concluded that Reyes had not established

commonality as required under Rule 23(a) or predominance

under Rule 23(b)(3).

“Commonality” is a consideration of whether there are

“questions of law or fact common to the class[.]” Fed. R.

Civ. P. 23(a)(2). Commonality is satisfied when there are

classwide answers. Wal-Mart Stores, Inc., 131 S. Ct. at 2551-

52; Sullivan v. DB Invs., Inc., 667 F.3d 273, 298-300 (3d Cir.

2011) (en banc) (considering “whether the defendant’s

conduct was common as to all of the class members[]” and

common questions led to common answers such that the

“alleged misconduct and the harm it caused would be

common as to all of the class members[]”).

The predominance inquiry then focuses on whether

“the questions of law or fact common to class members

predominate over any questions affecting only individual

members[.]” Fed. R. Civ. P. 23(b)(3). “Predominance tests

whether proposed classes are sufficiently cohesive to warrant

adjudication by representation[.]” In re Hydrogen Peroxide

Antitrust Litig., 552 F.3d 305, 310-11 (3d Cir. 2008)

(quotation marks omitted). Though related, this standard is

“‘far more demanding’ than the commonality requirement of

Rule 23(a),” id. (quoting Amchem Prods., Inc. v. Windsor,

521 U.S. 591, 623-24 (1997)), and requires “more than a

common claim.” Newton v. Merrill Lynch, Pierce, Fenner &

Smith, Inc., 259 F.3d 154, 187 (3d Cir. 2001).

Reyes seeks to represent a class that may include tens

of thousands of claimants with potential civil RICO claims

arising from the defendants’ operation of a sham enterprise.

“Establishing liability under [§ 1962(c)] of the RICO statute

requires (1) conduct (2) of an enterprise (3) through a pattern

(4) of racketeering activity, plus an injury to business or

property.” In re Ins. Brokerage Antitrust Litig., 579 F.3d 241,

269 (3d Cir. 2009) (internal quotation marks and citations

omitted).

Reyes must show that the racketeering activity was the

“but for” cause as well as the proximate cause of the injury

18

purportedly suffered by the members of the proposed class.

Holmes v. Sec. Investor Prot. Corp., 503 U.S. 258, 268

(1992). Proximate cause requires “some direct relation

between the injury asserted and the injurious conduct

alleged.” Id.

Accordingly, we must determine if Reyes produced

sufficient evidence to show that he could prove his RICO

claim based on facts or evidence common to the class, and

whether such proof predominates over facts and

circumstances that are particular to given individuals within

the proposed class.

The District Court concluded that Reyes established

neither but-for cause, nor proximate cause because “Reyes

would not be able to establish there was an impact that

affected each class member in the same way.” Reyes, 2013

WL 5332107, at *8. In reaching that conclusion, the District

Court focused on the fact that potential class “[m]embers

were contacted by telemarketers through different mediums

about different products, and some, according to Reyes, were

not contacted at all because the telemarketers bought their

account information from other telemarketers.” Id.

We must determine whether the District Court (1)

applied the proper standard for assessing predominance and

commonality, (2) appropriately reviewed Reyes’ experts’

opinions and the other evidence on the record, and (3)

properly determined that commonality and predominance

were not established based on the evidence presented. Much

of our inquiry is guided by our analysis in In re Hydrogen

Peroxide Antitrust Litigation.

A. STANDARD OF REVIEW

In reviewing a denial of class certification, we subject

the District Court’s legal rulings to de novo review. In re

Hydrogen Peroxide Antitrust Litig., 552 F.3d at 312; see also

In re Initial Pub. Offerings Sec. Litig., 471 F.3d 24, 32 (2d

Cir. 2006).

We review the District Court’s findings of fact, its

application of law to facts, and its decision regarding class

19

certification for an abuse of discretion. In re Hydrogen

Peroxide Antitrust Litig., 552 F.3d at 312, 320 (citations

omitted). “The district court abused its discretion if its

decision rests upon a clearly erroneous finding of fact, an

errant conclusion of law or an improper application of law to

fact.” Newton, 259 F.3d at 165-66 (footnote and quotation

marks omitted). “To illustrate . . . using the example of

numerosity, review of the factual finding as to the size of the

proposed class would be for clear error, review of the judge’s

articulation of the legal standard governing numerosity would

be de novo, and review of the ultimate ruling that applied the

correct legal standard to the facts as found would be for abuse

of discretion.” In re Initial Pub. Offerings Sec. Litig., 471

F.3d at 41.

B. RULE 23 LEGAL STANDARDS

“Class certification is proper only if the trial court is

satisfied, after a rigorous analysis, that the prerequisites of

Rule 23 are met.” In re Hydrogen Peroxide Antitrust Litig.,

552 F.3d at 309 (footnote and quotation marks omitted). In

conducting its inquiry, a district court must rigorously assess

“the available evidence and the method or methods by which

plaintiffs propose to use the evidence to prove impact at

trial.” Id. at 312 (citations omitted); see also Comcast Corp.

v. Behrend, 133 S. Ct. 1426, 1432 (2013) (“[T]he court’s duty

[is] to take a ‘close look’ at whether common questions

predominate over individual ones.” (quoting Amchem

Products, Inc., 521 U.S. at 615). On appeal, we do not

“speculate as to what the District Court must have intended

[regarding a Rule 23 requirement]. We cannot just assume

the District Court conducted the appropriate analysis under

Rule 23. ‘Rigorous analysis’ requires more of the District

Court than that[.]” Neale v. Volvo Cars of N. Am., LLC, ---

F.3d ---, Civ. Action No. 14-1540, 2015 WL 4466919, at *15

(3d Cir. July 22, 2015).

We have also explained that the Rule 23 inquiry may

require the district court to “delve” behind the pleadings. In

re Hydrogen Peroxide Antitrust Litig., 552 F.3d at 316

(quotation marks omitted). This means that, at the

certification stage, the district court “cannot be bashful. It

must resolve all factual or legal disputes relevant to class

20

certification, even if they overlap with the merits—including

disputes touching on elements of the cause of action.”

Marcus, 687 F.3d at 591 (quotation marks omitted).

Without this searching inquiry, a district court cannot

determine if class certification is appropriate. Certification

“calls for findings by the court, not merely a ‘threshold

showing’ by a party, that each requirement of Rule 23 is met.

Factual determinations supporting Rule 23 findings must be

made by a preponderance of the evidence.” In re Hydrogen

Peroxide Antitrust Litig., 552 F.3d at 307.

1. Reyes’ Burden of Proof

The District Court imposed a burden of absolute proof

on Reyes at the certification stage. Reyes now correctly

argues that “the ‘absolute proof of fraud’ standard . . .

exceeds this Court’s holding that factual issues bearing on the

class certification ‘must be made by a preponderance of the

evidence.’” Appellant Br. at 35 (quoting In re Hydrogen

Peroxide Antitrust Litig., 552 F.3d at 320).

The District Court realized that “[t]o obtain class

certification, a plaintiff must demonstrate the proposed class

satisfies all four elements of Federal Rule of Civil Procedure

23(a), along with one of the three requirements under Rule

23(b)[,]” and that “the court’s findings as to the Rule 23

requirements must be supported by factual determinations

made by a preponderance of the evidence[.]” Reyes, 2013

WL 5332107, at *3 (citations omitted). However, the District

Court’s analysis suffered from the same malady that troubled

us in In re Hydrogen Peroxide Antitrust Litigation. “[S]ome

statements in [the District Court’s] opinion depart from the

standards we have articulated.” 552 F.3d at 321. The District

Court stated the correct standard but misunderstood the

burden of proof placed on a plaintiff seeking class

certification.

In concluding that there was insufficient commonality,

the District Court relied on the fact that “Reyes’s experts have

[not] testified that the return rates are . . . absolute proof of

fraud[.]” Reyes, 2013 WL 5332107, at *8. The District Court

also believed that proof must be so persuasive that the

21

certifying court “cannot have any doubt as to whether the

Rule 23 requirements are met.” Id. at *7.

The fact that the District Court imposed a burden of

absolute proof is illustrated by the following statement: “since

Reyes’s experts have testified that the return rates are only

‘red flags’ of fraud and not absolute proof of fraud, the

factfinder would have to analyze each telemarketer separately

to determine whether or not it is a complete sham.” Id. at *8

(emphasis added). The District Court also stated that it “must

assess how Reyes will use the evidence at trial to prove

impact on a classwide basis, and the Court cannot have any

doubt as to whether the Rule 23 requirements are met.” Id. at

*7 (emphasis added) (citation omitted).

In re Hydrogen Peroxide Antitrust Litigation requires

that a district court “not suppress doubt as to whether a Rule

23 requirement is met[.]” 552 F.3d at 321 (internal quotation

marks omitted). While a district court must consider how the

evidence will be used at trial, id. at 311 & n.8, the plaintiff’s

burden is not proof beyond “any doubt” as the District Court

required, but whether the claims are supported by a

preponderance of the evidence. Id. at 320.

We agree with the District Court’s conclusion that “it

is possible that not every telemarketer named in the Amended

Complaint is a complete sham.” Reyes, 2013 WL 5332107, at

*7. However, the standard is not whether it is mathematically

or scientifically possible that one of these telemarketers is not

a complete sham. Rather, Reyes must establish that it is more

likely than not that each of the telemarketers and the

defendants operated a complete sham as alleged. Reyes

“must affirmatively demonstrate his compliance” with Rule

23, Wal-Mart Stores, Inc., 131 S. Ct. at 2551, and “satisfy

[the trial court] through evidentiary proof [of] at least one of

the provisions of Rule 23(b).” Comcast Corp., 133 S. Ct. at

1432; see also Byrd v. Aaron’s Inc., 784 F.3d 154, 163 (3d

Cir. 2015). However, “the Rules and our case law have

consistently made clear that plaintiffs need not actually

establish the validity of claims at the certification stage.”

Sullivan, 667 F.3d at 306 (footnote omitted).

22

Here, as in In re Hydrogen Peroxide Antitrust

Litigation, in remanding this matter to the District Court, we

“recognize that the able District Court did not have the

benefit of the standards we have articulated.” 552 F.3d at

322. Nevertheless, it is now clear that the District Court must

(1) conduct rigorous analysis, (2) review all avenues of

inquiry in which it may have doubts (even if it requires

reviewing the merits) in order to (3) be satisfied and (4) make

a definitive determination on the requirements of Rule 23, or

even (5) require that a plaintiff demonstrate actual, not

presumed conformance with Rule 23 requirements. We

stress, however, that the preponderance of the evidence

standard governs. The perfection of proof that the District

Court demanded here is simply not required. Moreover, it is

important for the District Court to remember that an inability

to calculate damages on a classwide basis will not, on its own,

bar certification. Neale, 2015 WL 4466919, at *17 & n.10

(collecting cases). A district court errs when it holds a

plaintiff seeking class certification to a higher standard of

proof than proof by a preponderance of the evidence, and

remand is thus appropriate. See In re Hydrogen Peroxide

Antitrust Litig., 552 F.3d at 322 (“To the extent that the

District Court’s analysis reflects application of incorrect

standards, remand is appropriate.”).

2. Evidence of Commonality and Predominance

It is often appropriate to discuss commonality and

predominance together because the commonality inquiry is

subsumed into the predominance inquiry. See, e.g., Danvers

Motor Co. v. Ford Motor Co., 543 F.3d 141, 148 (3d Cir.

2008) (“[W]here an action is to proceed under Rule 23(b)(3),

the commonality requirement is subsumed by the

predominance requirement.” (internal quotation marks

omitted)). We have also concluded that “[r]eading the

District Court’s commonality and predominance analyses

together . . . is appropriate in [the RICO] context[.]” In re

Ins. Brokerage Antitrust Litig., 579 F.3d at 266-67 (citation

omitted).

Therefore, we do not fault the District Court for

combining its discussions of commonality and predominance.

Reyes, 2013 WL 5332107, at *5-9. However, in the interest

23

of clarity, we will address each requirement separately. We

will first discuss commonality, and then discuss

predominance. In re LifeUSA Holding, Inc., 242 F.3d 136,

145 (3d Cir. 2001) (“[C]ommon questions (commonality)

must be established before predominance can be found . . . .”

(emphasis added)); see generally Marcus, 687 F.3d at 597,

600-12.

a. Commonality

As explained above in relation to the different return

rates, the District Court concluded that because Reyes’

evidence “result[s] in individual inquiries as to each

telemarketer,” class certification is precluded. Reyes, 2013

WL 5332107, at *8. The focus of the District Court’s inquiry

was whether the proposed class would “generat[e] common

answers apt to drive the resolution of the litigation, such that

a determination of the truth or falsity of the contention will

resolve an issue that is central to the validity of each one of

the claims in one stroke.” Id. at *4 (emphasis in original)

(internal quotation marks omitted).

Federal Rule of Civil Procedure 23(a)(2) merely

requires that there be “questions of law or fact common to the

class[.]” Commonality does not require perfect identity of

questions of law or fact among all class members. Rather,

“even a single common question will do.” Wal-Mart Stores,

Inc., 131 S. Ct. at 2556 (quotation marks and alterations

omitted); Baby Neal v. Casey, 43 F.3d 48, 56 (3d Cir. 1994)

(discussing how Rule 23(a)(2) “is easily met[]”). “A putative

class satisfies Rule 23(a)’s commonality requirement if the

named plaintiffs share at least one question of fact or law with

the grievances of the prospective class.” Rodriguez v. Nat’l

City Bank, 726 F.3d 372, 382 (3d Cir. 2013) (internal

quotation marks omitted). A court’s focus must be “on

whether the defendant’s conduct [is] common as to all of the

class members[.]” Sullivan, 667 F.3d at 298. “Again, th[e]

bar is not a high one.” Rodriguez, 726 F.3d at 382.

When a party seeks to certify a class to bring a RICO

claim, the focus is on the defendant’s conduct. As we said in

In re Insurance Brokerage Antitrust Litigation:

24

[p]roving the first element of a RICO violation

in this case would involve common questions

about the activities of the . . . Defendants and, in

particular, whether the . . . Defendants

participated or engaged in conduct with other . .

. Defendants. The second element also involves

common questions of law and fact, namely

whether an enterprise of . . . Defendants existed

. . . either as an association in fact or as a more

formal organization or entity. Proving the third

and fourth elements would encompass common

questions of law and fact as well, including

whether activities that constitute racketeering

were taking place through the enterprise (such

as mail or wire fraud) and whether these

racketeering activities were recurring such that

a pattern could be established.

579 F.3d at 269-70 (emphasis added). These elements focus

on the defendants’ joint conduct vis-à-vis the plaintiffs.

Likewise, the District Court in this case noted that “the first

four elements of a RICO claim . . . focus[] on the defendants’

conduct and the effect of that conduct.” Reyes, 2013 WL

5332107, at *5 (citation omitted). Thus, a properly supported

RICO allegation will often contain common issues because,

like “commonality[, a RICO allegation,] is informed by the

defendant’s conduct as to all class members and any resulting

injuries common to all class members[.]” Sullivan, 667 F.3d

at 297; see also In re Ins. Brokerage Antitrust Litig., 579 F.3d

at 269-70. “Whether the evidence presented proves a RICO

conspiracy[]” is a question “common to each class member

and will generate common answers[.]” In re Cmty. Bank of

N. Va. Mortg. Lending Practices Litig., 2015 WL 4547042, at

*13-14.

We also note that the elements of a RICO claim fit

Wal-Mart’s commonality requirement because determining

the “truth or falsity” of a “common contention[,]” here an

element of RICO, “will resolve an issue that is central to the

validity of each one of the claims in one stroke.” Wal-Mart

Stores, Inc., 131 S. Ct. at 2551; see In re Cmty. Bank of N.

Va. Mortg. Lending Practices Litig., 2015 WL 4547042, at

25

*13 (distinguishing a putative RICO class action from the

pitfalls of the class action rejected in Wal-Mart).

Reyes has presented evidence which, if accepted,

could establish by a preponderance of the evidence that (1)

“Zions and Modern Payments were processing transactions

for a number of entities -- in addition to NHS Systems -- that

government agencies [later] determined were fraudulent[,]”

Appellant Br. at 15; Appellee Br. at 48-49, (2) Zions Bank,

Netdeposit, and Modern Payments sent e-mails

communicating a sense of shock regarding the “staggering”

return rates, JA 686, and concerns that they may be at risk for

some of their business lines being used for “money

laundering[,]” id. at 692, (3) Zions Bank was aware that its

high return rates “alarm[ed]” NACHA,” id. at 814-15, and (4)

Modern Payments was “afraid of” a probe by the FTC

regarding potential fraud, id. at 693, and Zions Bank received

warnings from other banks that they “will disput[e] all

charges” generated by NHS Systems. Id. at 1061-62. On

remand, the District Court should consider whether this

evidence, if accepted as credible, is sufficient to conclude that

there is “actual, not presumed, conformance” with Rule 23.

Marcus, 687 F.3d at 591 (quotation marks omitted); see also

Wal-Mart Stores, Inc., 131 S. Ct. at 2551 (requiring “[a] party

seeking class certification [to] affirmatively demonstrate”

compliance with Rule 23).

We are not persuaded that the evidence here is similar

to that which was deemed insufficient for class certification in

Wal-Mart. Wal-Mart affirmed that commonality can be

satisfied when there is “significant proof” that the defendant

“operated under a general policy of discrimination.” Id. at

2553-54 (quotation marks omitted).

In Wal-Mart, the defendant operated in a discretionary,

and thus arguably individualized way toward the members of

the proposed class. The record did not establish a national

policy or any single individual or group of individuals

responsible for the exercise of discretion.

Here, there are slight variations in the telemarketers’

and defendants’ conduct underlying the putative class

members’ claims. For example, some plaintiffs were offered

26

government grants, while others were offered healthcare

discount cards, and still others had no direct involvement at

all as their account information was purchased by a

defendant. However, unlike Wal-Mart, we are not concerned

with damages that result from the exercise of anyone’s

discretion.

Further, in Wal-Mart, the plaintiffs’ statistical and

anecdotal evidence failed to demonstrate a “common mode of

exercising discretion that pervades the entire company[.]” Id.

at 2554-55. Without that, commonality could not be

established.

In stark contrast, the sham theory used here relies on a

“common mode” of behavior and a “general policy” of fraud.

See In re Cmty. Bank of N. Va. Mortg. Lending Practices

Litig., 2015 WL 4547042, at *13 (distinguishing Wal-Mart

and reasoning that “the Plaintiffs in this case have alleged that

the class was subjected to the same kind of illegal conduct by

the same entities, and that class members were harmed in the

same way, albeit to potentially different extents[]”).

Reyes alleges that the defendants operated together in

a common, fraudulent scheme that was a complete sham

masquerading as a legitimate business undertaking. When a

plaintiff alleges that a defendant’s disputed conduct is a

complete sham, the relevant inquiry is whether there was an

ongoing scheme to defraud or deceive, Rosario, 963 F.2d at

1017-19, and whether the defendant fails to meet the most

basic standard of its purported commercial undertaking,

Cullen, 188 F.R.D. at 235. If so, a common harm will be

presumed from the mere fact that class members were all

similarly injured by the sham. A court can then conclude that

there are common issues and that those common issues will

predominate over individual issues.15

When viewed at this macroscopic level, rather than the

microscopic level that may be required when allegations

implicate a defendant’s exercise of discretion, it is clear that

we are not faced with the individual circumstances that were

15

Because the “complete sham” theory also applies to

the predominance inquiry, it will be discussed further below.

27

fatal to certification in Wal-Mart. As said, it is for the

District Court to determine in the first instance whether Reyes

has presented evidence that demonstrates commonality.

b. Predominance

Reyes also contends that the District Court erred in

finding that class claims did not predominate over individual

claims. He argues that the District Court mistakenly focused

on the return rates instead of considering that evidence in

context with all of the testimony of his three experts, the FTC

proceedings,16 and all other evidence that the defendants

knew that they were operating a fraudulent enterprise.

“The Rule 23(b)(3) predominance inquiry tests

whether proposed classes are sufficiently cohesive to warrant

adjudication by representation.” Amchem Products, Inc., 521

U.S. at 623 (citation omitted). It is important to note that

Reyes is alleging a kind of consumer fraud. “[T]he Supreme

Court has observed that predominance is a test readily met in

certain cases alleging consumer or securities fraud or

violations of the antitrust laws.” In re Hydrogen Peroxide

Antitrust Litig., 552 F.3d at 321-22 (quotation marks

omitted). One relevant “guidepost[] that direct[s] the

predominance inquiry[]” is “that commonality is informed by

the defendant’s conduct as to all class members and any

resulting injuries common to all class members[.]” Sullivan,

667 F.3d at 297. “[Wal-Mart v.] Dukes actually bolsters [this]

position, making clear that the focus is on whether the

defendant’s conduct was common as to all of the class

members, not on whether each plaintiff has a ‘colorable’

claim.” Id. at 299. Rule 23 does not require the absence of

all variations in a defendant’s conduct or the elimination of

all individual circumstances. Rather, predominance is

satisfied if common issues predominate. In re Linerboard

Antitrust Litig., 305 F.3d 145, 162 (3d Cir. 2002). “[T]he

16

As noted supra note 8, the proceedings came after

the defendants’ involvement here. But, if admissible, it

would be for the factfinder to determine if the record, taken as

a whole, supports the conclusion (or an inference) that the

defendants were aware of the facts underlying the

proceedings, or the proceedings themselves.

28

focus of Rule 23(b)(3) is on the predominance of common

questions[.]” Amgen Inc. v. Connecticut Ret. Plans & Trust

Funds, 133 S. Ct. 1184, 1195 (2013) (emphasis in original).

A district court “analyze[s] predominance in the context of

Plaintiffs’ actual claims.” Neale, 2015 WL 4466919, at *14.

When Rule 23 is the mechanism to redress alleged

RICO violations, predominance and commonality are

satisfied if each element of the alleged RICO violation

involves common questions of law and fact capable of proof

by evidence common to the class. In re Ins. Brokerage

Antitrust Litig., 579 F.3d at 269-70. This is true even if

“establishing an injury is not as conducive to common

proof[,]” so long as a court is “satisfied that the plaintiffs

have presented a plausible theory for proving a class-wide

injury as a result of the racketeering activities of the alleged

enterprises at issue[.]” Id. at 270; see also Neale, 2015 WL

4466919, at *17 & n.10 (collecting cases regarding how an

inability to calculate damages on a classwide basis will not,

on its own, bar certification). “The question is not what valid

claims can plaintiffs assert; rather, it is simply whether

common issues of fact or law predominate.” Sullivan, 667

F.3d at 305 (citation omitted); In re Hydrogen Peroxide

Antitrust Litig., 552 F.3d at 311-12.

The predominance inquiry seeks to resolve whether

there are “reliable means of proving classwide injury[.]” In

re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d 244,

252-53 (D.C. Cir. 2013); see also In re Nexium Antitrust

Litig., 777 F.3d 9, 21 (1st Cir. 2015). This is often done with

the assistance of experts.

The District Court relied on Johnston v. HBO Film

Management, Incorporated, 265 F.3d 178, 190 (3d Cir.

2001), and In re LifeUSA Holding, Incorporated, 242 F.3d

136, 146 (3d Cir. 2001), to conclude that, because these

“telemarketers . . . interacted with consumers in different

ways (by telephone, internet, and slamming),” Reyes, 2013

WL 5332107, at *7, and because return rates are not absolute

proof that the defendants are a complete sham, “the only way

to determine if a class member was defrauded was to

individually examine the interaction between the defendant

29

and each class member.” Id. at *6. Neither case persuades

us.

In In re LifeUSA Holding, Incorporated, the fraud

involved a product that “was not sold according to standard,

uniform, scripted sales presentations.” 242 F.3d at 146. We

concluded that simply focusing on whether all potential class

members were misled by the same product was inappropriate.

Id. We also stated that when “the record is uncompromising

in revealing non-standardized and individualized sales

‘pitches’ presented by independent and different sales agents,

all subject to varying defenses and differing state laws, . . .

certification of individualized issues [is] inappropriate.” Id.

at 147; cf. Moore v. PaineWebber, Inc., 306 F.3d 1247, 1255

(2d Cir. 2002) (“[M]aterial uniformity in the

misrepresentations may be established without the use of a

standardized sales script[, for example.]”).

In Johnston, we again held that oral misrepresentation

must be uniform in order to establish predominance under

Rule 23(b)(3). See 265 F.3d at 190. Thus, “it has become

well-settled that, as a general rule, an action based

substantially on oral rather than written communications is

inappropriate for treatment as a class action.” Id. (internal

citations omitted).

However, both of those cases involved legitimate

businesses and Reyes relies upon that to distinguish the

circumstances here, arguing instead that the defendants and

telemarketers here were not legitimate business entities. See

In re Cmty. Bank of N. Va. Mortg. Lending Practices Litig.,

2015 WL 4547042, at *19, *21 (affirming the District Court’s

finding that individual issues of reliance will not predominate

when the plaintiffs “allege[] that [the defendant] performed

absolutely no services to earn the . . . fees[]”). Rather, Reyes

asserts that they all acted in concert to perpetrate fraud

through a sham enterprise and that each of the defendants had

to have known the real nature of that fraudulent enterprise.

He also claims that any variations in the manner in which

various individuals were defrauded is irrelevant and not

sufficient to preclude common issues of law from

predominating. The argument is based on his claim that the

defendants “engaged in systematic fraud and operated solely

30

to debit victims’ accounts[]” and “there is no evidence of any

cohort of class members who were not victims.” Appellant

Br. at 64.

Likewise, in Cullen, plaintiffs claimed that, while the

“defendants represented that they were providing a program

that would prepare students for careers as ultrasound

sonographers[,]” they “misrepresent[ed] the nature of the

ultrasound program, and failed to provide the education

represented.” 188 F.R.D. at 228. Even though the institution

was accredited, the program that the plaintiffs were enrolled

in was not. That prevented the program’s graduates from

sitting for the relevant registration examination. Id. at 228-

29. The plaintiffs contended that the “defendants’ operation

was a ‘complete sham,’ and did not provide even a minimal

education.” Id. at 228. Cullen held that the sham theory

satisfied the predominance requirement and it certified a

RICO class. Cullen explained: “plaintiffs’ proof will be

focused on the defendants’ conduct; plaintiffs’ case will

revolve around evidence that the school did not meet the most

basic standards of an educational program. It need not

involve time consuming proof of individual causation or

reliance.” Id. at 235. Instead, at trial, “[i]f the plaintiffs can

prove that [the defendant] was a complete sham, then a fact

finder can infer from the evidence that anyone who paid

tuition and attended the school suffered damage.” Id.

Cullen relied on Rosario, which had similar facts and

involved a similar fraudulent scheme. The trial evidence in

Rosario established that the defendant failed to provide

students with the skills, equipment, or environment to prepare

for a cosmetology career for which the school was supposed

to train students. Rosario, 963 F.2d at 1016-17. The Court of

Appeals for the Seventh Circuit affirmed the District Court’s

certification of the class. It held that the operative question

was whether the “schools operated pursuant to an ongoing

scheme to defraud and deceive prospective students[,]” thus

applying a “complete sham” theory. Id. at 1018. The

defendants attempted to negate that approach by relying on

evidence that some students were satisfied with their

education and two students who graduated did become

licensed and were working in the field. The Seventh Circuit

refused to conclude that the defendant-institution was not a

31

complete sham merely because “some class members were

satisfied with their teachers, and at least two class members

graduated from [the defendant’s] schools and passed the state

licensing exam.” Id. at 1017. Rather, the Court concluded

that “[t]he fact that there is some factual variation among the

class grievances will not defeat a class action.” Id. at 1017-18

(citation omitted).

While the District Court here recognized Reyes’ theory

of a sham enterprise, it nevertheless concluded that the

evidence Reyes offered in support of that theory was

insufficient to satisfy predominance because different sales

pitches were used and different products were pitched.

However, if absolute conformity of conduct and harm were

required for class certification, unscrupulous businesses could

victimize consumers with impunity merely by tweaking the

language in a telemarketing script or directing some (or all) of

the telemarketers not to use a script at all but to simply orally

convey a general theme designed to get access to personal

information such as account numbers.

We do not believe that our discussion of predominance

in our prior cases intended to either license such behavior by

placing it beyond the reach of Rule 23 or to supply a roadmap

that would guide the unscrupulous in designing fraudulent

schemes that would be beyond the reach of Rule 23.

Otherwise, although such subtle but irrelevant variations in

the manner of defrauding members of the public would not

insulate unscrupulous marketers from liability in individual

suits, it would -- for all practical purposes -- insulate them

from class actions. An interpretation of Rule 23 that places

class actions beyond the reach of consumers who have been

victimized by fraudulent schemers who are wise enough to

adopt schemes with subtle (but meaningless) variations would

invite the kind of consumer fraud that Reyes is alleging here.

Class actions are often the only practical check against

the kind of widespread mass-marketing scheme alleged here.

The individual claims arising from such conduct are usually

too small to justify suit unless aggregated in a class action.

This is particularly true when, as is often the case, the scheme

targets unsophisticated consumers with little disposable

income and without the means or wherewithal to seek

32

assistance of legal counsel. As a practical matter, the average

victim of such a scheme nearly always finds it far easier --

and much cheaper -- to reluctantly accept any loss and move

on than to undertake the expense and inconvenience endemic

in the protracted process of trying to recover a few dollars

years later.

A class action “permit[s] the plaintiffs to pool claims

which would be uneconomical to litigate individually. . . .

[M]ost of [such] plaintiffs would have no realistic day in

court if a class action were not available.” Phillips Petroleum

Co. v. Shutts, 472 U.S. 797, 809 (1985). “[C]onsumers have

little interest in litigating their claims individually because of

the small amount of money per plaintiff that is at stake.”

Cabraser, supra, at 822 (quotation marks omitted).

Moreover, obtaining counsel to pursue such a claim is usually

the height of impracticality -- even for those who can afford

to do so. “What rational lawyer would have signed on to

represent the [party] in litigation for the possibility of fees

stemming from a $30.22 claim?” AT&T Mobility LLC v.

Concepcion, 131 S. Ct. 1740, 1761 (2011) (Breyer, J.,

dissenting).

In such cases, the class action can “create[] greater

access to judicial relief[.]” Marcus, 687 F.3d at 594. “[I]t is[,

in fact,] possible to think of consumer class actions as

providing an indispensable mechanism for aggregating claims

when the individual stake is low and the similarity of the

challenged conduct is high.” Samuel Issacharoff, Group

Litig. of Consumer Claims: Lessons from the U.S.

Experience, 34 Tex. Int’l L.J. 135, 149 (1999).17 Thus, class

actions have the practical effect of allowing plaintiffs who

have suffered relatively de minimis loss to nevertheless

function as private attorneys general and thereby deter fraud

in the marketplace.

17

See Fed. R. Civ. P. 23 advisory committee’s note to

the 1966 amendment (“[A] fraud perpetrated on numerous

persons by the use of similar misrepresentations may be an

appealing situation for a class action, and it may remain so

despite the need, if liability is found, for separate

determination of the damages suffered by individuals within

the class.”).

33

The “complete sham” theory, if supported by an

appropriate record, can satisfy the predominance prong of

Rule 23(b)(3) by focusing on the overarching material and

defining aspects of a defendant’s conduct. This allows a

court to certify a class despite subtle (but meaningless)

variations that may occur while perpetrating a fraudulent

scheme. Thus, as Reyes contends “[u]ltimately, the question

is the extent to which fraud can be shown to have been

committed on the class by common evidence.” Appellant Br.

at 64. On remand, the District Court should consider the

entire record to determine if it supports that contention.

Investigator Blake testified “each of the schemes at

issue was a totally fraudulent mass-marketing fraud.” Blake

Decl., JA 1613, ¶ 15. She concluded that these were all

“precisely the kind of fundamentally fraudulent schemes that,

in [her] experience, have been routinely adjudicated on the

basis of common evidence of fraud.” Id., JA 1627-28, ¶ 85.

Professor Meyer reviewed the evidence and concluded that

the proof of fraud was classwide. He identified four key

features common to the way each telemarketer obtained bank

account information and initiated debits from bank accounts.

He concluded that although “these ‘products and programs’

appear to vary, all had [those] four common features essential

to their use as props in the scheme of deception[.]” Meyer

Decl., JA 1639-40 ¶ 19. The court-appointed receiver for

NHS Systems, Geisser, testified that NHS Systems was

“totally fraudulent and that no consumer who . . . had money

taken from their account was not injured in the amount of

money taken from their account[.]” Geisser Dep., JA 1161.

The District Court may consider that the defendants’

experts offer little reason to conclude that predominance

cannot be satisfied here. As noted above, Milner testified for

the defendant. Milner did not agree that a return rate in

excess of 10% was prima facie evidence of fraud. However,

she also explained that pursuant to its Operating Rules,

“NACHA gathers specific information concerning individual

Originators with high rates of unauthorized returns . . . [and]

may send the bank of an originator producing unauthorized

returns in excess of 1% a written request for information

34

about specific originators.” Milner Decl., JA 3787, ¶ 25

(citation and footnote omitted).

Professor Boss testified that “the overwhelming

majority of banks in the United States have no unauthorized

returns at all . . . .” Boss Decl., JA 590 ¶ 82. Reyes focuses

our attention on testimony that Zions Bank, like most U.S.

banks, had no unauthorized returns in August of 2006. It then

started working with its wholly-owned subsidiary, Modern

Payments, and its unauthorized returns began to “immediately

. . . increase dramatically from zero in August 2006 to 799 in

October, [and] 2,095 to 2,632 by February 2007.” Id., JA

587-88 ¶ 78 (emphasis in original). As noted earlier, when

adjusted for its size, Zions Bank had the highest rate of

returns of any U.S. bank.

Reyes also offers the following chart summarizing the

evidence regarding the return rates of the telemarketers:

Total Return Rates

First Total Return Rate as a

Average Month Months Multiple of 2007

Return Return With National Average

Fraud Scheme Rate* Rate* Zions* (1.25%)**

NHS 64.79% 51.90% 15 52

RX Smart 54.21% 42.00% 3 43

Market Power Entities

Market Power 86.73% 85.10% 11 69

Payday loan 73.46% 70.98% 4 59

Get Your Credit Rept 62.48% 55.94% 4 50

Vexeldale 74.09% 70.52% 4 59

Platinum Benefit Group 30.83% 37.98% 39 25

Group 1 Entities 51.67% 65.41% 21 41

Low Pay Inc 68.83% 65.41% 10 55

* JA 661 ** JA 663

Appellant Br. at 25 (footnote omitted).18

18

We are not, of course, suggesting that the District

Court had to accept the conclusions in the chart or the expert

testimony that the scheme here is susceptible to class

treatment. That determination remains left to the sound

discretion of the District Court. However, the District Court

35

The crux of the Rule 23(b)(3) predominance inquiry

here is whether a preponderance of the evidence supports

Reyes’ proposition that the defendants operated a complete

sham, and whether an affirmative answer to that inquiry

would establish the required element of predominance. Reyes

claims that if the evidence supports such a finding by a

preponderance of the evidence, the predominance

requirement of Rule 23(b)(3) is satisfied, even though the

telemarketers did not read from a uniform script or use the

same method of defrauding each of the members of the

putative class. Appellant Br. at 62-63 (distinguishing this

case from In re LifeUSA Holding, Incorporated and Johnston

because there is a “question about the legitimacy of the

[entity] itself[]” in which Reyes alleges the existence of one

“systematic fraud”).

As we have reiterated, Reyes argues the District Court

erred in relying only on high return rates to deny class

certification. As detailed above, Reyes also introduced proof

of the structure of each of the schemes and FTC

investigations. In particular, Reyes points to “broader

eviden[ce] . . . includ[ing] three experts (all of whom opined

that the underlying mass-marketing schemes were completely

fraudulent), the related government proceedings, Mr.

Geisser’s testimony, and a wealth of documentary evidence

and deposition testimony reflecting Defendants’ knowledge

of the fraud they were furthering.” Id. at 54.19 According to

is not free to simply ignore such testimony without explaining

why it is rejecting it. That is the antithesis of the thorough

and rigorous inquiry that the law demands at the certification

stage.

19

See also Appellant Reply Br. at 9-12 (discussing

additional evidence, including “internal bank documents

showing that Defendants were aware that they were providing

services to frauds,” warnings from regulators, other banks,

and employees “directly inform[ing] [the defendants] that

they were carrying out transactions for frauds[,]” expert

testimony, testimony from Geisser, “documents and

declarations from both regulatory actions that identified each

of the mass-marketing entities as frauds and those that

36

Reyes, the “evidence [presented about the fraudulent

companies] applies to the class as a whole, [therefore] the

jury’s consideration of th[e] evidence would apply to each

class member’s claim.” Id. at 37. If accepted, the District

Court may conclude that this evidence supports a finding that

there was a single fraudulent RICO enterprise, that each

defendant participated in that enterprise, and that all members

of the proposed class were damaged in the amount of the

funds debited from their bank accounts pursuant to the

fraudulent scheme. We leave it for the District Court to

assess this on remand.

Reyes further asserts that the District Court could have

specified subclasses for each defendant if it were concerned

about proving fraud with respect to each entity. Id.;

Appellant Reply Br. at 6 n.4. On remand, to the extent that

ordered complete restitution of their victims[,]” data

demonstrating the “explo[sion]” of unauthorized returns

following the formation of the alleged RICO enterprise, and

“documents showing that Defendants closely monitored the

entities’ extreme return rates—which they openly

acknowledged were staggering[.]” (citations and quotation

marks omitted)).

In mentioning this evidence, we in no way suggest that

we agree with the District Court’s conclusion that evidence of

high return rates cannot establish fraud on this record. We

refrain from concluding that return rates can never be prima

facie evidence of fraud or knowledge of fraud, no matter how

much they vary from industry norms. See Blake Decl., JA

1612-13 ¶ 14.d (“The schemes at issue here involve unusually

high numbers of returns even compared to other fraudulent

schemes I have investigated. It is therefore highly likely that

most of these schemes engaged in what is commonly known

as ‘slamming.’”); Boss Decl., JA 587-88, 590 ¶¶ 78, 82

(“Anyone in possession of the[] facts [about the return rates]

would have to have known fraud was involved. . . . It is

obvious that the high rates of unauthorized returns are

indicative of fraudulent activity by these companies.”); Meyer

Decl., JA 1637-38 ¶¶ 12.a, 14 (“For there to be such high

return rates there had to be fraud involved. . . . I conclude

with reasonable certainty that the return rates alone establish

that each of the entities at issue was fraudulent.”).

37

the District Court is concerned with the commonality and

predominance requirements because of variations in the way

the affairs of the allegedly fraudulent enterprise was

conducted, it can consider whether Reyes’ proposed

subclasses would adequately address those concerns. Fed. R.

Civ. P. 23(c)(5), (d); see also Neale, 2015 WL 4466919, at

*15 (“[T]he District Court should evaluate the relevant claims

(grouping them where logical and appropriate) and rule on the

predominance [and commonality] question[s] in light of the

claims asserted and the available evidence.” (citation

omitted)); Tobias Barrington Wolff, Discretion in Class

Certification, 162 U. Pa. L. Rev. 1897, 1898 (2014)

(“[D]istrict courts sometimes exercise discretion in defining

the parameters of the class definition and deciding when

subclasses are necessary, often acting independently of any

proposals made by the parties.” (footnote omitted)).20 We

note, however, “[t]he court has no sua sponte obligation so to

act.” U.S. Parole Comm’n v. Geraghty, 445 U.S. 388, 408

(1980). “[I]t is not the District Court that is to bear the

burden of constructing subclasses. That burden is upon the

[plaintiff] . . . who is required to submit proposals to the

court.” Id.

As we have already explained, Reyes is correct in

asserting that the District Court erred when it “examined one

piece of evidence -- the mass-marketers’ extreme return rates

-- and concluded that these rates alone were not ‘absolute

proof’ of fraud.” Appellant Br. at 37. The law does not

permit a district court to only consider and analyze the “most

significant evidence[.]” Reyes, 2013 WL 5332107, at *2.

Rather, the trial court must consider “all relevant evidence

and arguments, including relevant expert testimony of the

parties [unless there is a reason to reject certain testimony].”

In re Hydrogen Peroxide Antitrust Litig., 552 F.3d at 325

(emphasis added); see also In re Blood Reagents Antitrust

Litig., 783 F.3d 183, 187 (3d Cir. 2015). Moreover, as we

have also explained, we do not require “absolute proof.”

Indeed, such a burden would be prohibitive, as few things are

capable of absolute proof that removes all possibility of

20

For a very thorough development of the evolution of

discretion under Rule 23, see Wolff, Discretion in Class

Certification, 162 U. Pa. L. Rev. at 1911-39.

38

doubt. Here, the District Court’s analysis singled out the

“most significant evidence” (return rates) and then improperly

applied a heightened standard to that evidence while ignoring

all of the other testimony.

However, there is an even more fundamental flaw in

the District Court’s analysis. The District Court relied on fact

witnesses while ignoring expert testimony, and confused the

testimony of the witnesses it did consider.21 As noted above,

Reyes presented declarations from three expert witnesses:

Professors Boss and Meyer, and Investigator Blake.

However, the District Court failed to mention, let alone

closely scrutinize, any of Reyes’ experts. See Amchem

Products, Inc., 521 U.S. at 615-16 (referring to the trial

court’s “close look” before finding predominance and

superiority); In re Hydrogen Peroxide Antitrust Litig., 553

F.3d at 320 (“[A] district court errs as a matter of law when it

fails to resolve a genuine legal or factual dispute relevant to

determining the requirements.”). The District Court then

referred to Geisser and Fox as Reyes’ expert witnesses.

Reyes, 2013 WL 5332107, at *7. Geisser and Fox were not

expert witnesses, but fact witnesses. Geisser was installed as

the NHS Systems receiver after the FTC obtained a

preliminary injunction against NHS Systems; Fox was a

senior director of risk investigation at NACHA. Appellant

Br. at 10, 12, 26.

Perhaps because it confused their role and expertise,

the District Court relied on a concession by the fact witnesses

to justify limiting its analysis to evidence of high return rates.

Reyes, 2013 WL 5332107, at *7 (relying on Geisser’s

concession and Fox’s silence to conclude that the high return

rates are not dispositive of fraud). However, not only did

Reyes’ expert witnesses fail to make this concession, Blake

Decl., JA 1612-13 ¶ 14.d; Boss Decl., JA 587-88, 590 ¶¶ 78,

82; Meyer Decl., JA 1637-38 ¶¶ 12.a, 14, they also based

their conclusions on the additional evidence discussed above.

21

Much of the confusion here is understandable. The

District Court inherited the rather voluminous record in this

case from another judge.

39

The defendants correctly remind us that “[d]istrict

courts are not required under Hydrogen Peroxide, or any

other authority, to cite specifically to the declarations of every

expert or every other piece of evidence relevant to class

certification merely to prove that they were considered in its

rigorous analysis of the Rule 23 requirements.” Appellee Br.

at 32-33 (emphasis in original) (citations omitted). That is

undoubtedly true. However, a “rigorous analysis” is

nevertheless mandated, and the defendants do not contest

that.

Quite obviously, an analysis of testimony that refers to

individuals as expert witnesses when they are merely fact

witnesses and confuses one party’s fact witnesses with

experts, while not correctly citing any expert testimony, is

inconsistent with the demanding scrutiny required under Rule

23. “[T]he court’s obligation to consider all relevant

evidence and arguments extends to expert testimony, whether

offered by a party seeking class certification or by a party

opposing it.” In re Hydrogen Peroxide Antitrust Litig., 552

F.3d at 307 (emphasis added).

We have not required a district court to refer to, or cite,

every expert that either party presents and we do not do so

now.22 Rather, we merely apply the holding of In re

Hydrogen Peroxide Antitrust Litigation: “[T]he district court

must make whatever factual and legal inquiries are necessary

and must consider all relevant evidence and arguments

presented by the parties.” 552 F.3d at 307 (emphasis added)

(citation omitted).23

The defendants argue that the District Court’s

mistaken reference to fact witnesses as Reyes’ experts did not

22

See, e.g., Marcus, 687 F.3d at 603 (failing to

interpret In re Hydrogen Peroxide Antitrust Litigation as

requiring a district court to explicitly discuss every expert

opinion).

23

As we have explained above, In re Hydrogen

Peroxide Antitrust Litigation merely “emphasize[s] the need

for a careful, fact-based approach, informed, if necessary, by

discovery.” 552 F.3d at 326.

40

result in an abuse of discretion because the District Court

addressed the key arguments that each of Reyes’ experts

made. This record belies that argument. The District Court

merely focused on high return rates, without more. Reyes,

2013 WL 5332107, at *7-8. Not only did the District Court

minimize the probative value of that evidence by requiring

absolute proof of fraud, thereby undermining its relevance to

the requirements of Rule 23(a) and Rule 23(b)(3), it also

failed to confront Reyes’ experts’ arguments on the

importance of the return rates. See, e.g., Boss Decl., JA 582-

92 ¶¶ 66-87; Meyer Decl., JA 1637-38 ¶ 12.b. Further, as

discussed above, the District Court failed to confront Reyes’

experts’ arguments of “clear indications of fraud[]” based on

more than return rates. Boss Decl., JA 587 ¶ 77. See, e.g.,

Blake Decl., JA 1613 ¶ 15; Boss Decl., JA 592 ¶ 87; Meyer

Decl., JA 1639 ¶ 18.24

A district court errs as a matter of law when it confuses

testimony, as the District Court did here, and fails to carefully

scrutinize the relevant, disputed testimony.25

III. CONCLUSION

For the reasons set forth above, we will vacate the

order denying class certification and remand for proceedings

consistent with this opinion.

24

Amici remind us that Milner testified that

“transactions may be returned to a merchant under 68

different return codes, each of which has a different definition

and covers different return scenarios. None of these codes are

defined to identity fraud.” American Bankers Ass’n & Indep.

Cmty. Bankers of America Br. at 5. That does not negate the

force of the evidence of rate of return of each of the defendant

merchants here. It is fair to assume that similarly situated

banks would generally have roughly equivalent return rates.

Indeed the testimony here supports that assumption.

25

We may have a different situation if the District

Court quoted the expert witnesses, but mistakenly referred to

them as fact witnesses. That is not what happened. Here, we

have the fact witnesses’ arguments and names, not the

experts’.

41