Iowa Public Employees' Retirement System v. Merrill Lynch, Pierce, Fenner & Smith Inc

UNITED STATES DISTRICT COURT

SOUTHERN DISTRICT OF NEW YORK

IOWA PUBLIC EMPLOYEES’ RETIREMENT

SYSTEM; LOS ANGELES COUNTY

EMPLOYEES RETIREMENT ASSOCIATION;

ORANGE COUNTY EMPLOYEES RETIREMENT

SYSTEM; SONOMA COUNTY EMPLOYEES’

RETIREMENT ASSOCIATION; and TORUS

CAPITAL, LLC, on behalf of themselves and all

others similarly situated, 17 Civ. 6221 (KPF) (SLC)

Plaintiffs, OPINION AND ORDER

ADOPTING REPORT AND

-v.- RECOMMENDATION

MERRILL LYNCH, PIERCE, FENNER & SMITH

INC.; MERRILL LYNCH L.P. HOLDINGS, INC.;

and MERRILL LYNCH PROFESSIONAL

CLEARING CORP.,

Defendants.

KATHERINE POLK FAILLA, District Judge:

On June 30, 2022, Magistrate Judge Sarah L. Cave issued a 71-page

Report and Recommendation (the “Report”), in which she recommended that

the motion for class certification (the “Motion”) brought by Plaintiffs in this

antitrust class action be granted in part and denied in part. Plaintiffs filed a

modest objection to one portion of the Report, while the remaining defendants

in the case (collectively, “Defendants”) pursued broader challenges to her

decision. For the reasons that follow, the Court adopts the Report nearly in its

entirety, with a limited exception concerning the end date of the relevant class

period (the “Class Period”).

BACKGROUND1

A. Factual Background

Both this Court and Judge Cave have previously recounted the facts

underlying this action, in which Plaintiffs allege that a handful of banks

conspired to prevent the U.S. stock loan market from transitioning to a

transparent, direct electronic exchange. See Iowa Pub. Employees’ Ret. Sys. v.

1 This Opinion draws its facts from Plaintiffs’ Amended Complaint (“AC” (Dkt. #73)), the

parties’ submissions in connection with Plaintiffs’ Motion (Dkt. #411, 412, 413, 414,

415, 468, 469, 470, 513), and the Report (Dkt. #563).

The parties’ submissions include the following documents and the exhibits attached

thereto: (i) Declaration of Michael B. Eisenkraft in Support of Plaintiffs’ Motion for Class

Certification and Appointment of Cohen Milstein Sellers & Toll PLLC as Co-Lead Class

Counsel, dated February 22, 2021 (Dkt. #413); (ii) Declaration of Daniel L. Brockett in

Support of Plaintiffs’ Motion for Class Certification and Appointment of Class Counsel,

dated February 22, 2021 (Dkt. #414), including the expert report of Dr. Haoxiang Zhu

(“Zhu Report” (Dkt. #414-9)) and the expert report of Dr. Paul Asquith and Dr. Parag

Pathak (“A&P Report” (Dkt. #414-10)); (iii) Declaration of Michael A. Paskin in Support

of Defendants’ Memorandum of Law in Opposition to Plaintiffs’ Motion for Class

Certification, dated June 29, 2021 (Dkt. #432), including the expert report of Dr.

Terrence Hendershott (“Hendershott Report” (Dkt. #432-1)), the expert report of Dr.

Justin McCrary (“McCrary Report” (Dkt. #432-2)), and the expert report of Fabio

Savoldelli (“Savoldelli Report” (Dkt. #432-4)); (iv) Reply Declaration of Daniel L. Brockett

in Support of Plaintiffs’ Motion for Class Certification, dated October 5, 2021 (Dkt.

#470), including the reply expert report of Dr. Zhu (“Zhu Reply” (Dkt. #470-1)) and reply

expert report of Dr. Asquith and Dr. Pathak (“A&P Reply” (Dkt. #470-2)); (v) Declaration

of John S. Playforth in Support of Defendants’ Sur-Reply in Opposition to Plaintiffs’

Motion for Class Certification, dated November 22, 2021 (Dkt. #496), including the reply

expert report of Dr. McCrary (“McCrary Reply” (Dkt. #496-2)); and (vi) Sur-Sur-Reply

Declaration of Daniel L. Brockett in Further Support of Plaintiffs’ Motion for Class

Certification and Appointment of Class Counsel, dated January 18, 2022 (Dkt. #514).

For ease of reference, the Court refers to Plaintiffs’ memorandum of law in support of

the Motion as “Pl. Br.” (Dkt. #415), and notes that docket entry 415 is the corrected

version of the original memorandum of law that was filed on February 22, 2022 (Dkt.

#412). The Court refers to Defendants’ memorandum of law in opposition to Plaintiffs’

Motion as “Def. Br.” (Dkt. #431); and to Plaintiffs’ reply memorandum of law as “Pl.

Reply” (Dkt. #469).

Further, the Court refers to Plaintiffs’ objection to the Report as “Pl. Obj.” (Dkt. #573);

to Defendants’ opposition to Plaintiffs’ objection as “Def. Opp.” (Dkt. #595); and to

Plaintiffs’ reply to Defendants’ opposition as “Pl. Reply” (Dkt. #615). Similarly, the

Court refers to Defendants’ objections to the Report as “Def. Obj.” (Dkt. #576); to

Plaintiffs’ opposition to Defendants’ objections as “Pl. Opp.” (Dkt. #597); and to

Defendants’ reply to Plaintiffs’ opposition as “Def. Reply” (Dkt. #617).

Merrill Lynch, Pierce, Fenner & Smith Inc., 340 F. Supp. 3d 285, 297-310

(S.D.N.Y. 2018) (“IPERS I”) (denying Defendants’ motion to dismiss); Iowa Pub.

Employees’ Ret. Sys. v. Bank of Am. Corp., No. 17 Civ. 6221 (SLC), 2022 WL

2829880, at *1-13 (S.D.N.Y. June 30, 2022) (Report). The Court assumes

familiarity with these prior opinions, and adopts and employs the defined terms

from IPERS I. The Court provides here only a brief discussion of those facts

that are relevant to resolving the instant objections to the Report.

1. The U.S. Stock Loan Market

A stock loan transaction occurs when stock is temporarily transferred

from one investor to another. (AC ¶ 2). The owner of the shares, referred to as

a “stock lender” or “beneficial owner,” lends its stock to the borrower. (Id. ¶¶ 2,

97).2 Beneficial owners are typically entities such as pension funds, mutual

funds, or insurance companies that own U.S. equities. (Zhu Report ¶ 23).

Borrowers typically include entities like hedge funds. (See AC ¶¶ 19, 136, 164;

Zhu Report ¶¶ 190, 191). In exchange for lending stock to a borrower, the

beneficial owner receives cash or noncash collateral and is paid a “borrowing

fee.” (AC ¶¶ 2, 97). The borrower holds the stock for a period of time, and later

returns the stock to the beneficial owner when the borrowing period has ended.

(Id. ¶ 2). The period of time during which the borrower holds the stock is not

always preset; the loan can be for a fixed term, or it can be an “open” loan that

2 The stock lender transfers legal title to the stock to the borrower, but retains the

economic benefits of the stock, such as dividends. (Zhu Report ¶ 20).

can be terminated at any time by either the beneficial owner or the borrower.

(Zhu Report ¶ 21).

Notably, beneficial owners generally do not transact directly with

borrowers. (Zhu Report ¶ 28). Rather, in the above-described transaction, it is

typical that an “agent lender,” usually a custodian bank, would act as an

intermediary agent on behalf of the beneficial owner. (AC ¶ 97). The agent

lender would interact with a broker-dealer, such as the Prime Broker

Defendants in this case. (Id.).3 The broker-dealer would also serve as an

intermediary, borrowing the stocks from the beneficial owner, through the

agent lender, and lending the stocks to the borrower. (Zhu Report ¶ 28; see

Report 5 (outlining the typical steps of a stock loan transaction facilitated by a

broker-dealer)). In exchange for doing so, the broker-dealer would collect a fee

from the borrower for the transaction, a portion of which the broker-dealer

would then keep before passing the remainder of the fee to the lender. (AC

¶ 97). Because the broker-dealer typically would not inform the lender (acting

through the agent lender) or the borrower of the amount of the fee that the

borrower paid, or the amount of that fee that the lender received, neither the

borrower nor the lender would ever become aware of the broker-dealer’s portion

of the transaction fee. (Id.). “The differences in the price between the[se] two

3 As defined in IPERS I, the Prime Broker Defendants are large banks that “engaged in

securities lending and stock lending transactions with class members, either directly or

through Defendants’ affiliates.” Iowa Pub. Employees’ Ret. Sys. v. Merrill Lynch, Pierce,

Fenner & Smith Inc., 340 F. Supp. 3d 285, 298 (S.D.N.Y. 2018). The Prime Broker

Defendants act as “prime brokers,” which are broker-dealers “who offer an array of

services that include stock lending.” Id. at 300 (citing AC ¶ 102).

sides of the market” is called the “spread,” and it “is a measure of the

transaction cost for end users and a measure of prime broker revenues.” (Zhu

Report ¶ 28).

As Defendants’ experts explain, broker-dealers justify the spread by

explaining that broker-dealers provide other services to facilitate these

transactions, the cost of which is “bundled” with the cost of matching the

lender and borrower. (See McCrary Report ¶ 22; Savoldelli Report ¶¶ 25-26).

Those services include establishing relationships with agent lenders and

promoting stability of transactions by managing the risks of “recalls” of stocks

and “rerate[s]” of stock transactions. (Report 9-10). A “recall” occurs when a

stock lender ends the loan. (Zhu Report ¶ 21). A “rerate” occurs when one

party to the loan demands a different rate or threatens to terminate the loan if

the counterparty does not accept the new rate. (Id.).

Plaintiffs posit that the practice of stock lending is advantageous because

it promotes liquidity in financial markets by (i) allowing institutional investors

to earn returns while holding stable interests in a given company and

(ii) facilitating “short-selling,” a trading strategy. (AC ¶ 3). But despite the

importance of the stock loan market in the United States, it remains an “over-

the-counter” (“OTC”) market, in that there is no “central marketplace” where

trades can take place. (Id. ¶ 98). As a result, market participants have little

opportunity to “price shop” among competing broker-dealers and remain

unaware of the pricing terms on which other market participants are

transacting, all of which makes it difficult to negotiate with broker-dealers. (Id.

¶¶ 99, 103). For a variety of reasons, Plaintiffs argue that “th[is] OTC market

structure benefits intermediaries at the cost of customers.” (Zhu Report ¶ 32).

Further, “[t]he broker-dealer market is highly concentrated, and the six Prime

Broker Defendants control the majority of stock lending volume.” (A&P Report

¶ 115).

2. Defendants’ Anticompetitive Conduct

The Report discusses the evidence that Plaintiffs proffer as proof that the

Prime Broker Defendants conspired to prevent new market entrants, including

AQS and SL-x Markets (“SL-x”), from bringing innovations to the stock loan

market that would have disrupted their market position, as well as the

evidence Defendants proffer to refute those claims. (Report 11-19). For

example, the evidence shows that in 2001, Defendants Bank of America,4

Goldman Sachs, Morgan Stanley, JP Morgan, and UBS formed an entity called

“EquiLend” explicitly to address the “[t]hreat of disintermediation,” i.e., the

threat that intermediaries would become less ubiquitous in the domestic stock

lending market. (See Dkt. #414-4 at 44 (Presentation for a 2015 EquiLend

Board of Directors meeting)).

4 As stated in IPERS I,

due to Bank of America Corporation’s 2009 acquisition of Merrill

Lynch & Co., and to Bank of America Securities LLC’s 2010 merger

with Defendant Merrill Lynch, Pierce, Fenner & Smith Inc., the

Amended Complaint uses the term “Bank of America” to refer to

Defendants Merrill Lynch, Pierce, Fenner & Smith Inc., Merrill

Lynch Professional Clearing Corp., Merrill Lynch L.P. Holdings,

Inc., and their parents, subsidiaries, and affiliates.

IPERS I, 340 F. Supp. 3d at 298 n.2.

By 2009, members of EquiLend had formed a “CCP Working Group,”

precisely because they feared that “CCP”5 would be “forced on [them] by

regulators in some form[.]” (Dkt. #414-27 at 2). From the perspective of

EquiLend’s members, it was preferable to find a solution through EquiLend

that would be satisfactory to regulators while avoiding potential

disintermediation. (Id.). In consequence, the working group resolved to “stay

close to CCP market evolvement and understand CCP models but not formally

engage or develop links to any CCP[.]” (Id. (emphasis added)). What is more,

the Board members of EquiLend were instructed to report to EquiLend “if their

view or perception of CCPs change[d]” or if they “decide[d] to participate in CCP

bi-laterally.” (Id.; see also Dkt. #414-35 at 2 (managing director of EquiLend

explaining that EquiLend recommended that “owner firms inform [them] …

if/when they start doing any business through a CCP” because EquiLend did

not “want to be blindsided if [in] six months … half of the firms [had started]

using one of these CCPs in some capacity)).

At the end of January 2009, a company called AQS launched an

electronic trading platform that would allow trades to “be executed and

centrally cleared at transparent prices.” (See AC ¶ 9). The first Bank of

America trades were successfully executed on the AQS platform on January 31,

5 CCP, or central counterparty clearing, refers to the use of “clearinghouses” as

intermediaries between parties to a loan, protecting them from counterparty risk, i.e.,

the risk that the counterparty will default. (AC ¶¶ 9 n.3, 147). Clearinghouses do so by

becoming the “borrower to every lender and the lender to every borrower.” (Id. ¶ 9 n.3).

If “one party fails to meet its obligations, the clearinghouse steps in and assumes the

obligation.” (Id.).

2009 (Dkt. #414-40 at 4), and Bank of America initially worked to promote AQS

to hedge funds (Dkt. #414-39 at 755). But Plaintiffs have put forth evidence

that Bank of America later shifted resources away from that effort (Dkt. #414-

39 at 755-756), and claim that this shift occurred because Bank of America

needed to “f[a]ll in line with” the other Prime Broker Defendants (Pl. Br. 29).

The Prime Broker Defendants “[r]eally want[ed] to squash th[o]se muppets,”

namely AQS, and worked to do so by threatening supporters of AQS. (Dkt.

#414-48 at 2; see Report 13 (compiling evidence that the Prime Broker

Defendants pushed firms not to work with AQS)). By 2016, AQS had failed and

EquiLend purchased its assets. (Dkt. #414-39 at 12-14; see also Dkt. #414-62

at 3 (Credit Suisse employees discussing the topic of central clearing and

stating that “AQS learned the hard way that [neither Credit Suisse] nor any of

the major [prime brokers] will support a [h]edge fund model”). As the Report

acknowledges, Defendants argue that AQS failed, not because of any

interference by Defendants, but because the platform did not offer key services

that broker-dealers offer in stock loan transactions. (Report 14).

In 2011, another electronic stock lending platform emerged: SL-x. In

marketing itself, SL-x recognized that prime brokers were aware of “the

advantages of a central counterparty model for stock lending, but [were]

concerned about its impact on their market position and relationships.” (Dkt.

#414-63 at 3). SL-x therefore offered a model that was “[d]ealer focused,” in

that it “d[id] not try to change the existing buy-side/sell-side ecosystem,” “d[id]

not disintermediate dealers,” and “[m]aintain[ed] relationships as [a] key barrier

to entry.” (Id. at 27). The fact that SL-x would allow prime brokers to maintain

their market position distinguished it from other “electronic stock lending

platforms … [that had] failed because of lack of support from prime brokers

and agent lenders.” (Id. at 3). As detailed in the Report, Plaintiffs advance

evidence suggesting that the Prime Broker Defendants still ultimately, and as a

group, decided not to use SL-x. (Report 15; see also Dkt. #414-77 at 2

(contemporaneous email exchange stating that the EquiLend Board was

“unanimous in not seeing much benefit in the SL[-]x proposal”)). To refute

Plaintiffs’ suggestions of collusive and/or anticompetitive activity, Defendants

point to the testimony of SL-x’s Chairman, in which he stated that SL-x never

received regulatory approval in the United States and therefore did not have an

operational platform in the United States. (Dkt. #432-17 at 75-77).

Finally, the Report addressed evidence regarding the Prime Broker

Defendants’ response to Data Explorers, which was the leading provider of

securities lending data and which provided “transparency and benchmarking

to one of the last obscure frontiers in financial markets.” (Report 16 (quoting

Dkt. #414-81 at 5 (Markit company presentation))). Plaintiffs’ evidence

suggests that the Prime Broker Defendants not only discouraged funds from

participating in Data Explorers, but also decided to develop their own data

product, Markit. (Id. at 17). Ultimately, in 2012, Data Explorers was acquired

by Markit — a firm that was majority-owned by banks including the Prime

Broker Defendants. (Id.). As the Report noted, Defendants argue that they

acquired Data Explorers, not to stop the service, but because it was a market

leader in that sector, and that they did not stop providing data to Data

Explorers before the acquisition. (Id.). In point of fact, however, Markit has

never issued a data product that provides market-wide transparency. (Id.).

B. Procedural Background

1. Plaintiffs’ Claims and Class Certification Briefing

Plaintiffs filed an initial complaint on August 16, 2017 (Dkt. #1), and the

Amended Complaint on November 17, 2017 (Dkt. #73), bringing claims of

(i) conspiracy to restrain trade in violation of Section 1 of the Sherman Act; and

(ii) unjust enrichment in violation of New York law. Those defendants then in

the case jointly moved to dismiss; the motion was denied in IPERS I; and the

case proceeded to discovery.

On February 22, 2021, Plaintiffs filed the instant Motion, seeking class

certification and appointment of lead counsel. (Dkt. #411). At the parties’

request, this Court permitted briefing on class certification to be filed over a

period of seven months and directed Plaintiffs to refile the notice of motion for

class certification at the time they filed their reply brief. (See Dkt. #417). After

an additional extension, Defendants filed their memorandum of law in

opposition to the Motion on June 29, 2021. (Dkt. #431). On October 5, 2021,

Plaintiffs filed the renewed notice of motion and reply memorandum of law in

further support of the Motion. (Dkt. #468-469). Defendants then filed a sur-

reply on November 22, 2021 (Dkt. #495), followed by Plaintiffs’ sur-sur reply on

January 18, 2022 (Dkt. #513).

This Court referred the Motion to Judge Cave for a report and

recommendation. (Dkt. #471). On April 28, 2022, Judge Cave heard a full day

of oral argument from the parties on the Motion. (See April 28, 2022 Minute

Entry; see also Dkt. #535, 555-1, 556-1, 556-2, 557-1, 559-1 (parties’

presentations and oral argument transcript)).

2. The Report and the Objections to the Report

On June 30, 2022, Judge Cave issued the Report, recommending that

the Court grant Plaintiffs’ Motion in part and deny the Motion in part. After a

comprehensive, well-reasoned analysis of the parties’ arguments and evidence,

Judge Cave recommended that the following class be certified (the “Class”):

All persons and entities who, directly or through an

agent, entered into at least 100 U.S. Stock Loan

Transactions as a borrower from the prime brokerage

businesses of the U.S.-based entities of the Prime

Broker Defendants, or at least 100 U.S. Stock Loan

Transactions as a lender of Hard-to-Borrow stock to the

U.S.-based entities of the Prime Broker Defendants,

from January 1, 2012 until August 16, 2017.

Excluded from the Class are: Defendants, as well as

Citadel LLC, Two Sigma Investments, PDT Partners,

Renaissance Technologies LLC, TGS Management,

Voloridge Investment Management, and the D.E. Shaw

Group and their corporate parents, subsidiaries, and

wholly owned affiliates, as well as any federal

governmental entity, any judicial officer presiding over

this action, and any juror assigned to this action.

(Report 69-70). Judge Cave further recommended that the following

management subclasses be utilized (collectively, the “Subclasses”):

The “End-User Subclass”: All persons and entities

within the class who, directly or through an agent,

entered into at least 100 U.S. Stock Loan Transactions

as a borrower from the prime brokerage businesses of

the U.S.-based entities of the Prime Broker Defendants

during the Class Period; and

The “Beneficial Owner Subclass”: All persons and

entities within the class who, directly or through an

agent, entered into at least 100 U.S. Stock Loan

Transactions as a lender of Hard-to-Borrow stock to the

U.S-based entities of the Prime Broker Defendants

during the Class Period.

(Id. at 70). Judge Cave also recommended that Plaintiffs IPERS, LACERA,

OCERS, SCERA, and Torus be appointed as co-Lead Plaintiffs and that Cohen

Milstein and Quinn Emanuel be appointed as Class Counsel. (Id. at 65-66,

70).

On the particular issue of Plaintiffs’ request to extend the Class Period

end date from August 16, 2017, to February 22, 2021, or later, Judge Cave

expressed concern that adopting such a proposed end date would require the

reopening of fact discovery, which had closed in October 2020. (See Report 68;

see also Dkt. #298 (Fourth Amended Case Management Plan)). In Part IV.D. of

the Report, Judge Cave found that Plaintiffs had “failed to show the ‘good

cause’ required … for further extending fact discovery to encompass more than

four years of transactional data.” (Report 68). Accordingly, Judge Cave

recommended that this Court reject Plaintiffs’ proposed Class Period, and

instead set a Class Period of January 1, 2012, to August 16, 2017. (Id. at 69).

On August 15, 2022, Plaintiffs filed an objection to Part IV.D. of the

Report. (Dkt. #573). On the same date, Defendants filed their objections to the

Report. (Dkt. #576). Both Plaintiffs and Defendants filed oppositions to the

other side’s objections and later filed replies to those oppositions. (Dkt. #595,

597, 615, 617).6

DISCUSSION

A. Applicable Law

1. Review of a Magistrate Judge’s Report and Recommendation

When deciding whether to adopt a report and recommendation, the

district court “may accept, reject, or modify, in whole or in part, the findings or

recommendations made by the magistrate judge.” 28 U.S.C. § 636(b)(1)(C); see

also Fed. R. Civ. P. 72(b)(3). The Court conducts a de novo review of those

portions of a report and recommendation to which a party submits a timely

objection. See United States v. Romano, 794 F.3d 317, 340 (2d Cir. 2015).

“To accept those portions of the report to which no timely objection has

been made, ‘a district court need only satisfy itself that there is no clear error

on the face of the record.’” King v. Greiner, No. 02 Civ. 5810 (DLC), 2009 WL

2001439, at *4 (S.D.N.Y. July 8, 2009) (quoting Wilds v. United Parcel Serv.,

Inc., 262 F. Supp. 2d 163, 169 (S.D.N.Y. 2003)), aff’d, 453 F. App’x 88 (2d Cir.

6 On September 11, 2024, the Court granted final approval to two settlement agreements

and dismissed the following Defendants from the case: Credit Suisse Group AG; Credit

Suisse AG; Credit Suisse Securities (USA) LLC; Credit Suisse First Boston Next Fund,

Inc.; Credit Suisse Prime Securities Services (USA) LLC; Goldman, Sachs & Co. LLC;

and Goldman Sachs Execution & Clearing, L.P. (merged into Goldman, Sachs & Co.

LLC as of June 12, 2017); J.P. Morgan Securities LLC; J.P. Morgan Prime, Inc.; J.P.

Morgan Strategic Securities Lending Corp.; JPMorgan Chase Bank, N.A.; Morgan

Stanley; Morgan Stanley Capital Management, LLC; Morgan Stanley & Co. LLC; Morgan

Stanley Distribution, Inc.; Prime Dealer Services Corp.; Strategic Investments I, Inc.;

UBS AG; UBS Americas Inc.; UBS Securities LLC; UBS Financial Services Inc.;

EquiLend LLC; EquiLend Europe Limited; and EquiLend Holdings LLC. (Dkt. #683,

684). Accordingly, at present, Plaintiffs’ claims remain pending against only Merrill

Lynch, Pierce, Fenner & Smith Inc.; Merrill Lynch L.P. Holdings, Inc.; and Merrill Lynch

Professional Clearing Corp. The Clerk of Court is directed to modify the caption on the

docket accordingly.

2011) (summary order); accord Galeana v. Lemongrass on Broadway Corp., 120

F. Supp. 3d 306, 310 (S.D.N.Y. 2014). Indeed, a party’s failure to object timely

to a report and recommendation, after receiving clear notice of the

consequences of such a failure, operates as a waiver both of the party’s right to

object to the report and recommendation and of the right to challenge the

report and recommendation on appeal. See Frank v. Johnson, 968 F.2d 298,

300 (2d Cir. 1992) (“We have adopted the rule that failure to object timely to a

report waives any further judicial review of the report.”); accord Park v. State

Farm Mut. Auto. Ins. Co., No. 23-7230, 2024 WL 4834935, at *1 (2d Cir.

Nov. 20, 2024) (summary order) (collecting cases).

2. Class Certification

“[A] district judge may not certify a class without making a ruling that

each Rule 23 requirement is met[.]” In re Initial Pub. Offerings (“IPO”) Sec. Litig.,

471 F.3d 24, 27 (2d Cir. 2006). The party seeking certification must establish

that the proposed class meets all the requirements of Rule 23, by a

“preponderance of the evidence.” See Myers v. Hertz Corp., 624 F.3d 537, 547

(2d Cir. 2010). Under Rule 23(a), the moving party must show that:

[i] the class is so numerous that joinder of all members

is impracticable; [ii] there are questions of law or fact

common to the class; [iii] the claims or defenses of the

representative parties are typical of the claims or

defenses of the class; and [iv] the representative parties

will fairly and adequately protect the interests of the

class.

Fed. R. Civ. P. 23(a). In addition to the four express Rule 23(a) requirements of

numerosity, commonality, typicality, and adequacy, an implicit

“ascertainability” requirement must be met. See In re Petrobras Sec., 862 F.3d

250, 257 (2d Cir. 2017). Ascertainably is a “threshold requirement” that

necessitates that a class be “defined using objective criteria that establish a

membership with definite boundaries.” Id.

If the requirements of Rule 23(a) are met, “a plaintiff must also establish

that the proposed class falls into one of the three categories set forth in Rule

23(b).” Azor-El v. City of New York, No. 20 Civ. 3650 (KPF), 2024 WL 4326921,

at *7 (S.D.N.Y. Sept. 27, 2024). Here, Plaintiffs seek to certify a class under

Rule 23(b)(3). (Pl. Br. 16-50). Rule 23(b)(3) provides that a class action “may

be maintained if … the court finds [i] that the questions of law or fact common

to class members predominate over any questions affecting only individual

members, and [ii] that a class action is superior to other available methods for

fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3).

“Ultimately, the district court has broad discretion in deciding how and

whether to certify a class, arising from its ‘inherent power to manage and

control pending litigation.’” In re Aluminum Warehousing Antitrust Litig., 336

F.R.D. 5, 37 (S.D.N.Y. 2020) (quoting Myers, 624 F.3d at 547). This includes

the power to “alter or modify the class, create subclasses, and decertify the

class whenever warranted.” Sumitomo Copper Litig. v. Credit Lyonnais Rouse,

Ltd., 262 F.3d 134, 139 (2d Cir. 2001).

B. Analysis

As discussed, both Plaintiffs and Defendants timely filed objections to the

Report. Plaintiffs’ objection is limited to Part IV.D. of the Report, and thus each

of Plaintiffs’ arguments relates to the Class Period for the class action. In

contrast, Defendants’ objections encompass broader arguments regarding the

Report’s ultimate recommendation to certify the Class in the first instance.

Specifically, Defendants argue that Judge Cave erroneously concluded that

Plaintiffs satisfied the adequacy of representation requirement of Rule 23(a), as

well as the predominance and superiority requirements of Rule 23(b).7 The

Court begins by analyzing Defendants’ arguments, because Plaintiffs’ failure to

satisfy any of the Rule 23 criteria would be fatal to class certification. The

Court then proceeds to consider Plaintiffs’ Class Period arguments.

Ultimately, after de novo review of the challenged portions of the Report,

the Court finds that Plaintiffs have demonstrated that the Class recommended

by Judge Cave satisfies the adequacy, predominance, and superiority

requirements of Rule 23. Furthermore, the Court finds that the appropriate

end date for the Class Period is November 17, 2017. Accordingly, the Court

certifies the Class in the form recommended by Judge Cave, except that the

Class Period shall encompass January 1, 2012, through November 17, 2017.

1. The Requirements of Rule 23

a. Adequacy

Rule 23(a)(4) requires that “the representative parties [in a class action]

will fairly and adequately protect the interests of the class.” Fed. R. Civ.

P. 23(a)(4). In conducting the adequacy inquiry, courts must consider whether

(i) “plaintiff[s’] interests are antagonistic to the interest of other members of the

7 Neither side challenges the numerosity requirement.

class” and (ii) “plaintiff[s’] attorneys are qualified, experienced and able to

conduct the litigation.” In re Flag Telecom Holdings, Ltd. Sec. Litig., 574 F.3d

29, 35 (2d Cir. 2009).

The adequacy analysis is intended “to uncover conflicts of interest

between named parties and the class they seek to represent.” Amchem Prod.,

Inc. v. Windsor, 521 U.S. 591, 625 (1997). That said, “[n]ot every conflict

among subgroups of a class will prevent class certification — the conflict must

be ‘fundamental’ to violate Rule 23(a)(4).” In re Literary Works in Elec.

Databases Copyright Litig., 654 F.3d 242, 249 (2d Cir. 2011). Even where a

conflict is fundamental, because it goes to the “very heart of the litigation” and

is not speculative or hypothetical, see id. at 259 (Straub, J., concurring in part,

dissenting in part (quoting Cent. States Se. & Sw. Areas Health & Welfare Fund

v. Merck-Medco Managed Care, L.L.C., 504 F.3d 229, 246 (2d Cir. 2007))), “it

can be cured by dividing the class into separate ‘homogeneous subclasses ...

with separate representation to eliminate conflicting interests of counsel,’” id.

at 249-50 (quoting Ortiz v. Fibreboard Corp., 527 U.S. 815, 856 (1999)). Where

no fundamental conflict exists, the same class representatives and counsel

may represent multiple classes or subclasses in the same class action. See

Loc. 2507, Uniformed EMTs, Paramedics & Fire Inspectors v. City of New York on

behalf of Fire Dep’t of City of New York, No. 22 Civ. 10336 (AT), 2024 WL

4276495, at *7-8 (S.D.N.Y. Sept. 24, 2024) (finding that the lead plaintiffs had

no conflicts with the class or subclasses and that plaintiffs and lead counsel

were adequate representatives of the class and subclasses); see also Azor-El,

2024 WL 4326921, at *10-11 (finding that the representative parties were

adequate to represent the three proposed classes).8

As discussed above, Plaintiffs propose a class comprised of borrowers

and lenders who transacted with the U.S.-based entities of the Prime Broker

Defendants. (Pl. Br. 7-8). Plaintiffs also propose two management subclasses:

the “End-User Subclass,” consisting of borrowers, to be represented by named

Plaintiffs SCERA and Torus, and the “Beneficial Owner Subclass,” consisting of

lenders, to be represented by named Plaintiffs IPERS, OCERS, SCERA, and

LACERA. (Id. at 8, 13). Plaintiffs contend that the named Plaintiffs are

adequate class representatives, inasmuch as they “assert the same causes of

action, raise the same liability issues, and seek the same relief as all class

members.” (Id. at 14). Finally, Plaintiffs seek appointment of the Quinn

Emanuel and Cohen Milstein firms as co-lead class counsel. The Motion does

not contemplate separate representation of the subclasses. (Id. at 50).

In their objections, Defendants argue that the Report erred in finding

that Plaintiffs and their counsel are adequate class representatives. (Def.

Obj. 8-18). These arguments are threefold: (i) the Report failed to recognize

that the End-User Subclass and the Beneficial Owner Subclass are divided by

fundamental conflicts of interest; (ii) the Report failed to require separate

representation of the two Subclasses; and (iii) the Report erred in finding that

8 Rule 23(c)(5) explicitly grants courts the authority to divide a class into subclasses,

where it is appropriate. Fed. R. Civ. P. 23(c)(3)(5); see also Sumitomo Copper Litig. v.

Credit Lyonnais Rouse, Ltd., 262 F.3d 134, 139 (2d Cir. 2001) (stating that district

courts are best positioned to “assess the propriety of the class” and can “create

subclasses … whenever warranted”).

Torus and SCERA are adequate representatives of the End-User Subclass.

(Id.). This Court considers each argument in turn.

i. Fundamental Conflicts of Interest

To begin, Defendants argue that there are fundamental conflicts of

interest between the End-User Subclass and Beneficial Owner Subclass,

because “[l]enders and borrowers stand on opposite sides of the stock-lending

market” and “[t]hese opposing market positions create inherent conflicts of

interest over whether lenders or borrowers sustained most or all of the alleged

injury here[.]” (Def. Obj. 9). As a preliminary matter, the presence of entities

that are on opposite sides of a market in a class does not create a per se

fundamental conflict. See In re NASDAQ Mkt.-Makers Antitrust Litig., 169

F.R.D. 493, 513-15 (S.D.N.Y. 1996) (finding no fundamental conflict where the

class included both buyers and sellers, because both buyers and sellers had an

interest in proving the existence of the alleged conspiracy and in maximizing

the total amount of class-wide damages); In re Auction Houses Antitrust Litig.,

193 F.R.D. 162, 165 (S.D.N.Y. 2000) (finding the adequacy requirement met

where the class consisted of both buyers and sellers); see also In re Sumitomo

Copper Litig., 182 F.R.D. 85, 92 (S.D.N.Y. 1998) (stating that “it is settled in

this Circuit that … the presence of both purchasers and sellers … will not

defeat class action certification when plaintiffs allege that the same unlawful

course of conduct affected all members of the proposed class”).

Defendants claim in their objection (Def. Obj. 9-12), as they did before

Judge Cave (Def. Br. 39-43), that analyses of three issues — “but-for world”9

platform prices, platform fees, and search costs — illustrate a fundamental

conflict between the subclasses. Defendants first argue that the subclasses are

fundamentally at odds regarding but-for world pricing because borrowers have

an interest in demonstrating that anonymous trading platforms would have

yielded lower prices, while lenders have a converse interest in demonstrating

that such platforms would have yielded an ability to lend stock at higher

prices. (Def. Obj. 9-11). Relatedly, Defendants claim that borrowers and

lenders are at odds regarding the fees that would be charged on anonymous

trading platforms in the but-for world: Borrowers are alleged to have an

interest in showing that the fees charged by anonymous trading platforms

would be allocated equally between borrowers and lenders, while lenders have

an interest in showing that fees would be allocated entirely to borrowers. (Id.

at 11).

Both examples relate to damages allocation and therefore, as other

courts in this Circuit have found, do not preclude certification. In NASDAQ

Market-Makers, for example, the court found that potential conflicts between

parties on opposite sides of relevant transactions (there, buyers and sellers)

regarding damages did not necessitate denial of class certification. 169 F.R.D.

9 See AC ¶ 332 (referring to the “‘but for’ world where the role of the Prime Broker

Defendants is significantly reduced and their cut in large part flows instead to the

borrowers and lenders, minus the small fees charged by an efficient platform and the

clearing broker who provides the borrower and lender access to the central

clearinghouse”).

at 514-15. Similarly, in Sjunde AP-Fonden v. General Electric Company, the

court rejected an argument that the proposed class representatives were not

adequate to represent a class of members who purchased securities at different

times, because the court found that issues related to damages did not warrant

denial of class certification. 341 F.R.D. 542, 548-49 (S.D.N.Y. 2022).

Intra-class conflicts defeat adequacy only where the conflicts “extend …

beyond the apportionment of damages.” In re Foreign Exch. Benchmark Rates

Antitrust Litig., 407 F. Supp. 3d 422, 439 (S.D.N.Y. 2019) (“Forex III”). The

court in Forex III found such a conflict. 407 F. Supp. 3d at 439. There, the

class included members who transacted directly with each other, and the class

definition did not specify the days on which the alleged spread manipulation

occurred or the direction of the manipulation. Id. As a result, the class

members had “directly conflicting incentives to establish whether spread

manipulation occurred on certain dates and the extent to which it affected their

transactions.” Id. The court explicitly distinguished the circumstances of its

case from the circumstances in NASDAQ Market-Makers, wherein there was “no

conflict given [the] allegation that spread manipulation inflated prices for

purchasers and depressed prices for sellers, giving both an incentive to

establish spread manipulation.” Id.

Here, the same logic holds as in NASDAQ Market-Makers. Members of

both Subclasses allege that Defendants’ conspiracy resulted in inflated spreads

and costs that were borne by both borrowers and lenders. (AC ¶¶ 341-345,

376-377, 394). The evidence put forth by Plaintiffs supports the same.

Members of both Subclasses were harmed by Defendants because the

conspiracy resulted in an opaque market that deprived them all of “reduc[ed]

bid/offer spreads[,]” as well as “lower operational costs, improved liquidity, and

decreased risk[,]” that would have resulted from anonymous multilateral

trading. (Zhu Report ¶ 11; see also A&P Report ¶ 255 (providing that

centralized platforms would result in “more favorable prices for the End-User

and Beneficial Owner Subclasses”)). Both Subclasses were harmed by the

same conspiracy and have a shared interest in maximizing recovery for the

class. At the class certification phase, potential conflicts over the allocation of

damages do not defeat adequacy.

Attempting to provide a third example of a fundamental conflict between

the Subclasses, Defendants argue that only borrowers would benefit from

reduced search costs in the but-for world of anonymous trading platforms.

(Def. Obj. 12). They argue further that lenders would actually be harmed by

reduced search costs because it would lessen their ability to charge high

lending prices. (Id.). Again, the evidence does not support that argument. To

the contrary, Plaintiffs proffer evidence that anonymous platform trading would

decrease search costs for both the End-User and Beneficial Owner Subclasses.

(See A&P Report ¶ 87 (providing that the OTC market structure leads to “higher

search costs, because, in order to transact, both the borrower and lender must

contact and work through broker-dealers” (emphasis added)); see also Zhu

Report ¶ 11 (concluding that search costs would have been decreased for all

class members by anonymous multilateral trading)).

ii. Separate Representation of Subclasses

Defendants next argue that because a fundamental conflict of interest

divides the Subclasses, the Report erred in failing to require that different

counsel represent each subclass. (Def. Obj. 12-15). They point to two class

actions in which there were fundamental conflicts as support for this

contention. (Id. (citing Literary Works, 654 F.3d at 253-57; In re Joint E. & S.

Dist. Asbestos Litig., 982 F.2d 721, 742-43 (2d Cir. 1992)). Because this Court

has found that no fundamental conflict of interest exists at this phase of the

litigation (see supra D.1.a.i.), Defendants’ second argument necessarily fails.

iii. Adequacy and Typicality of Torus and SCERA

In their third and final adequacy challenge, Defendants argue that the

Report erred in finding that Torus and SCERA are adequate representatives of

the End-User Subclass. (Def. Obj. 17-18). According to Defendants, SCERA is

inadequate to represent the End-User Subclass of borrowers because, in

addition to being a borrower, it was a lender. (Id. at 17). But, as the Report

found, the fact that SCERA was both a borrower and lender actually supports

the lack of fundamental conflict between the Subclasses and makes SCERA an

appropriate representative of both. (See Report 47). See also E. Texas Motor

Freight Sys. Inc. v. Rodriguez, 431 U.S. 395, 403 (1977) (“[A] class

representative must be part of the class and possess the same interest and

suffer the same injury as the class members.” (internal quotation marks

omitted)).

Defendants next argue that Torus is an inadequate representative

because it is not a member of the End-User Subclass. Defendants ground this

argument in their claim that Torus was not a “client of a Defendant’s prime

brokerage business,” because it borrowed stock from Goldman Sachs

Execution & Clearing, L.P. and Merrill Lynch Professional Clearing Corp. (Def.

Obj. 18). As it happens, however, both Goldman Sachs Execution & Clearing,

L.P. and Merrill Lynch Professional Clearing Corp were named as Defendants in

this case and provided prime brokerage services to Torus. (See AC ¶¶ 48, 50;

see also Dkt. #560, Tr. 208:1-6).

Finally, Defendants renew their argument that Torus is an atypical

representative because it is a “tiny proprietary trading firm that bears no

resemblance to the large hedge funds that make up the vast majority of the

borrower subclass, and its trading volume was far too small to justify the

substantial costs it would have incurred for joining a trading platform.” (Def.

Obj. 18). Despite its comparatively smaller size, Torus met the threshold

number of transactions required for membership in the End-User Subclass,

and provided testimony that it would have joined a trading platform. (Dkt.

#470-10, Tr. 77:2-11). The Court therefore rejects Defendants’ arguments and

finds that Torus is a member of the End-User Subclass and an adequate

representative of that class.

b. Predominance

In addition to the Rule 23(a) requirements, the proposed class must also

satisfy the Rule 23(b)(3) predominance and superiority requirements. The

predominance inquiry is intended to “test[] whether proposed classes are

sufficiently cohesive to warrant adjudication by representation.” Amchem, 521

U.S. at 594. “[T]he requirement is satisfied ‘if resolution of some of the legal or

factual questions that qualify each class member’s case as a genuine

controversy can be achieved through generalized proof, and if these particular

issues are more substantial than the issues subject only to individualized

proof.’” Mazzei v. Money Store, 829 F.3d 260, 272 (2d Cir. 2016) (quoting

Myers, 624 F.3d at 547).

When assessing predominance, a court must examine “[i] the elements of

the claims and defenses to be litigated, [ii] whether generalized evidence could

be offered to prove those elements on a class-wide basis or whether

individualized proof will be needed to establish each class member’s

entitlement to relief, and [iii] whether the common issues can profitably be tried

on a class[-]wide basis, or whether they will be overwhelmed by individual

issues.” Scott v. Chipotle Mexican Grill, Inc., 954 F.3d 502, 512 (2d Cir. 2020)

(quoting Johnson v. Nextel Commc’ns Inc., 780 F.3d 128, 138 (2d Cir. 2015)

(internal quotation marks omitted)). “The three required elements of an

antitrust claim are [i] a violation of antitrust law; [ii] injury and causation; and

[iii] damages.” In re Currency Conversion Fee Antitrust Litig., 264 F.R.D. 100,

114 (S.D.N.Y. 2010) (quoting Cordes & Co. Fin. Servs. v. A.G. Edwards & Sons,

Inc., 502 F.3d 91, 105 (2d Cir. 2007) (internal quotation marks omitted)).

In their Motion, Plaintiffs argued that they satisfied the predominance

requirement because they can prove each of the elements of their antitrust

claim by common evidence, can prove class-wide injury at trial, and have put

forth a damages model that sufficiently demonstrates that damages can be

measured by a common methodology. (Pl. Br. 16-49). When evaluating those

arguments and issuing the Report, Judge Cave agreed with Plaintiffs and

determined that the predominance requirement was satisfied. (Report 49-63).

Defendants counter that the Report erred in making that determination

because, rather than being able to rely on common evidence to establish class-

wide injury at trial, “the jury would have to consider extensive individualized

evidence to determine whether individual class members were injured and to

assess whether particular transactions and class members satisfy the [Foreign

Trade Antitrust Improvements Act (“FTAIA”)]’s requirements.” (Def. Obj. 19).

In other words, Defendants’ predominance arguments focus on whether

Plaintiffs have established that both injury and damages can be proven by

common evidence. The Court discusses these arguments in greater detail in

the remainder of this section.

i. Antitrust Injury

In their Motion, Plaintiffs summarized the evidence that they argued

could prove class-wide impact, including the economic search model and

yardstick analyses conducted by their expert Dr. Zhu and the separate impact

analysis conducted by expert witnesses Dr. Asquith and Dr. Pathak. (Pl.

Br. 34-41). In cases where plaintiffs put forth an expert’s model as the basis

for their claim of class-wide impact, “a court is obliged to rigorously examine

the soundness of that model at the class certification stage[,]” and “may certify

a class under these circumstances only where the Court finds the model

methodologically sound.” In re Aluminum Warehousing Antitrust Litig., 336

F.R.D. at 46. That being said, “the Court only evaluates whether the method

by which plaintiffs propose to prove class-wide impact could prove such impact,

not whether plaintiffs in fact can prove class-wide impact.” In re Magnetic

Audiotape Antitrust Litig., No. 99 Civ. 1580 (LMM), 2001 WL 619305, at *4

(S.D.N.Y. June 6, 2001) (emphasis added).

Defendants argue that the Report erred in determining that the proffered

common evidence is capable of demonstrating class-wide injury. (Def. Obj. 38-

45). As detailed below, this Court has examined the proposed models and

other evidence challenged by Defendants and determines that they constitute

common evidence capable of demonstrating class-wide injury.

(a) Economic Search Model

Plaintiffs’ expert, Dr. Zhu, first used an economic search model, adopted

from his peer-reviewed independent research, to analyze the question of class-

wide impact. (Zhu Report ¶ 11). As explained by Dr. Zhu, in a “canonical OTC

market structure,” investors must search for prices by individually contacting

dealers. (Id. ¶ 32). Dealers then offer a price that investors cannot

simultaneously compare against other quoted prices. (Id.). Such a structure

necessarily “benefits intermediaries at the cost of customers,” because there is

no “easy access to market-wide price information or competing offers[.]” (Id.).

Dr. Zhu’s search model employs the concepts of “slow” and “fast”

customers to illustrate the impact that the adoption of trading platforms would

have on an OTC market structure. (Zhu Report ¶¶ 264-265). “Slow”

customers in the model are those that “must visit dealers one by one, and each

visit, including the first, incurs a search cost.” (Id. ¶ 264). Dr. Zhu explains

that search costs, in practice, include “the costs of making the operational,

technological, and legal arrangements with a dealer so that trades can take

place.” (Id.). “Fast” customers, in contrast, are those that “can observe all

dealers’ quotes simultaneously, at zero cost, and pick the best one.” (Id.

¶ 265).

Dr. Zhu posits that the results of the search model show that the

introduction of a trading platform benefits all customers. Certain customers

that change from “slow” to “fast” benefit because they gain the ability to receive

“[m]ore quotes from more dealers without the need to search[.]” (Zhu Report

¶ 281). The outputs of the model also demonstrate that spreads are reduced

for those customers. (Id.). Dr. Zhu’s results further demonstrate that even

customers who do not change from “slow” to “fast” experience benefits. (Id.

¶ 282). When trading platforms are introduced, dealers have imperfect

information about which customers are “fast” and which are “slow.” (Id.).

Because they do not know with certainty which customers have access to

pricing information, dealers price lower on average. (Id.). Therefore, Dr. Zhu

concluded, “[t]he model shows that the introduction of a multilateral trading

platform would benefit all or virtually all class members on both sides of the

market, including not only those traders that begin using the platform but also

those traders that continue to trade with prime brokers on a bilateral basis.”

(Id. ¶ 11).

In their objections, Defendants argue that Dr. Zhu’s search cost model is

incapable of proving injury to all borrowers and lenders. (Def. Obj. 39). First,

Defendants claim that the search cost model cannot show that all borrowers

would have benefitted from the introduction of a multilateral trading platform

because the model “assumes away all of the real-world reasons why many class

members would not have benefitted [from such trading platforms].” (Id. at 40).

For example, the model assumes that borrowers pay similar prices, despite

evidence that there is wide price dispersion in the stock lending market,

because borrowers have different revenue, asset sizes, and investment

strategies, among other variables. (Id. (citing McCrary Report ¶¶ 226-243;

McCrary Reply ¶¶ 63-65)). But Dr. Zhu cogently responded to this critique in

his reply, defending his model as using the “standard economic technique” of

“condition[ing price distribution] on all observables.” (Zhu Reply ¶¶ 274-275,

278). Dr. Zhu acknowledged that dispersion in real-world pricing is caused in

part by observable factors like a fund’s revenue or asset size. (Id. ¶ 278). As a

result, his model intentionally uses a “conditional distribution” (id. ¶ 283),

which allows it to isolate the “cause-and-effect relationship” between platform

usage and pricing, rather than introducing other factors (id. ¶ 278).

Defendants also argue more broadly that the model “assumes what it

purports to prove” — that all borrowers would benefit from lower search costs.

(Def. Obj. 41). Again, Dr. Zhu thoughtfully addressed that argument. Dr. Zhu

did not assume that all borrowers would benefit. Instead, he explained why

even customers who continue to operate in the OTC market, rather than on

multilateral platforms, would benefit from the entry of those platforms: a

platform-using segment of the stock loan market would both attract additional

market participants, generating more liquidity that would then move OTC

segment prices closer to platform segment prices, while also putting pressure

on the OTC segment of the market to retain their market share by reducing

their own profit margin. (Zhu Reply ¶¶ 238-252). Finally, as they did before

Judge Cave, Defendants argue that Dr. Zhu’s model is flawed because he failed

to “compare the actual world to the but-for-world[.]” (Def. Obj. 41; see McCrary

Report ¶¶ 244-247). But again, Dr. Zhu defended his model by explaining that

a perfect “before/after” analysis cannot be conducted because the conspiracy is

alleged to have occurred from the early 2000s through to the present. (Zhu

Report ¶ 253).

Second, Defendants argue that the economic search model cannot prove

an adverse impact on lenders because Dr. Zhu did not build a search cost

model for lenders and instead “speculated” that the results for such a model

would be the same as for borrowers. (Def. Obj. 42-44). But Dr. Zhu did not

speculate; rather, he opined that “the model is qualitatively the same for the

other side of the market (where beneficial owners enter into ‘lend’ transactions

with dealers).” (Zhu Report ¶ 262). That comports with Dr. Zhu’s prior

academic work on OTC markets. In presenting another version of his economic

search model in his joint paper with Professors Darrell Duffie and Piotr

Dworczak, Dr. Zhu reported the results of the model for buyers, but

acknowledged that the clients of financial intermediaries could be “buying or

selling” assets and, significantly, that “[t]he [seller] case is effectively the

same[.]” (See Dkt. #556-2 at Slide 30 (Plaintiffs’ Presentation Slides); see also

Zhu Report ¶ 45)). Further, in discussing the results of his academic research,

Dr. Zhu provided additional support for his finding that as search costs are

reduced, the prices charged to end users, which include borrowers and

lenders, are reduced. (Zhu Report ¶ 45).

Dr. Zhu concluded that the economic search model shows benefits for

“all or virtually all class members on both sides of the market.” (Zhu Report

¶ 11). After examining the model, the Court finds that it is methodologically

sound and capable of proving class-wide impact. That is all that is required at

this stage. Dial Corp. v. News Corp., 314 F.R.D. 108, 119 (S.D.N.Y. 2015), as

amended, No. 13 Civ. 6802 (WHP), 2016 WL 690895 (S.D.N.Y. Feb. 9, 2016)

(determining that proposed methodology was sufficient and did not defeat

predominance).

(b) Yardstick Analysis

Dr. Zhu also employed a yardstick analysis to examine the class-wide

effects of Defendants’ alleged conspiracy. (Zhu Report ¶¶ 290-315). In

conducting that analysis, Dr. Zhu selected several comparators, including the

stock market, the corporate bond market, the government bond market, and

the oil market. Each is a market in which Defendants were not alleged to have

conspired to prevent a transition from OTC trading to anonymous multilateral

trading. (See, e.g., id. ¶ 292 (explaining the similarities between the stock

market and the stock lending market); Zhu Reply ¶ 333 (same)). He examined

the transitions in those markets and observed empirical data that

demonstrated that each of the comparators experienced market-wide benefits

because of the introduction and adoption of anonymous multilateral trading.

For example, Dr. Zhu observed that “[m]ost transactions” in the stock market

today occur on multilateral exchanges and that the “bid/offer spreads [have]

nearly collapsed” in the electronic trading environment, benefitting all stock

traders. (Zhu Report ¶¶ 293, 299). He therefore concluded, based on all

examined comparable markets, that “multilateral trading mechanisms in the

U.S. stock loan[] market would similarly bring widespread benefits for all or

virtually all Class members.” (See id. ¶ 315).

Defendants’ objections to the yardstick analysis begin with their

argument that Dr. Zhu should have selected different comparators, such as

various overseas stock lending markets, rather than other U.S.-based markets

in different sectors. (Def. Obj. 44-45). But Dr. Zhu provided a compelling

defense of his model in his reply report, explaining that overseas stock lending

markets do not make perfect comparators, in part because the markets in

other countries are significantly smaller than the stock lending market in the

United States. (Zhu Reply ¶¶ 293-294). Moreover, Dr. Zhu found that “central

clearing and multilateral trading of stock loans have been implemented

successfully in many other countries with significantly smaller capital markets”

and that this “serve[s] as a lower bound of what is possible for the U.S.”

because it is a larger market. (Id. ¶ 294).

Courts in this Circuit have repeatedly stated that a “yardstick”

methodology is an accepted method to measure antitrust impact and damages.

See, e.g., In re Elec. Books Antitrust Litig., No. 11 MD 2293 (DLC), 2014 WL

1282293, at *25 (S.D.N.Y. Mar. 28, 2014) (collecting cases that state common

methods of quantifying antitrust damages, including the yardstick method); see

also In re Restasis (Cyclosporine Ophthalmic Emulsion) Antitrust Litig., 335

F.R.D. 1, 21 (E.D.N.Y. 2020) (finding that the decision to use a yardstick

approach to demonstrate class impact was “sound and workable”); SourceOne

Dental, Inc. v. Patterson Companies, Inc., No. 15 Civ. 5440 (BMC), 2018 WL

2172667, at *4 (E.D.N.Y. May 10, 2018) (stating that the yardstick methodology

is a “generally accepted method for measuring antitrust damages”). The Court

has examined the yardstick analysis conducted by Dr. Zhu, as well as his

thorough reply in further support of that analysis, and concludes that the

analysis is methodologically sound.

The Report found that Dr. Zhu’s evidence is capable of showing class-

wide impact, leaving for the jury the decision of whether that evidence in fact

shows class-wide impact. (Report 59). In doing so, Judge Cave did not “punt[]

to the jury.” (Def. Obj. 39). She applied the correct legal analysis — analyzing

whether the proposed methodology was sound and concluding in the

affirmative. After conducting its own analysis of the proposed methodology, the

Court agrees and finds that common evidence is capable of proving injury.

(c) Individual Inquiries

Finally, Defendants argue that even if Plaintiffs attempt to use common

proof to prove class-wide injury, common proof will not predominate at trial

because Defendants will need to conduct individual inquiries to contest the

existence of class-wide injury. (Def. Obj. 19-33). Defendants presented the

same arguments to Judge Cave (Def. Br. 23-29), and Judge Cave found that

they did not preclude predominance (Report 49-59). This Court sees no reason

to conclude differently.

Defendants first argue that individual inquiries would be necessary to

determine which class members valued certain aspects of OTC trading to the

extent that they would not have used multilateral trading platforms. (Def.

Obj. 20-32). Defendants identify two “unique features” of OTC trading that

they claim are valued by members of the class: “recall protection” and “rerate

management.” (See id. at 21). In particular, Defendants contend that prime

brokers provide recall protection to members of the Class when a security is

recalled prematurely by locating new shares of recalled stock for those clients,

thereby allowing the clients to maintain their positions and avoid losses. (See

Def. Br. 6, 14; Def. Obj. 21). They contend that prime brokers also provide

rerate management for members of the class by paying a portion of the

difference between the original rate and an increased rate, in instances when

an agreed-upon rate changes over time. (See Def. Br. 6; Def. Obj. 22).

Dr. Zhu examined the support for these arguments as put forth by

Defendants’ experts and reiterated his conclusion that “all Class members are

better off in the but-for world.” (See Zhu Reply ¶¶ 37, 59-60). More to the

point, Dr. Zhu explicitly rejected the premise that there are fundamental

differences between platform and OTC loans, opining instead that “[t]he

differences raised by Defendants — primarily the theoretical possibility of

implied ‘recall’ or ‘rerate’ protection that prime brokers build into the price of

OTC stock loans — are overstated, and … can be provided in the but-for world

by broker-dealers sponsoring clients on platforms.” (Id. ¶¶ 60, 73). Further,

he opined that (i) “there is scarce evidence in the record that shows the

presence and value of these purported protections” and (ii) “these purported

services are not offered for free, but sold in an opaque manner as a bundle in

the OTC segment without any transparency of the cost of each item.” (Id. ¶

73).

The Court acknowledges, as it must, that Defendants have offered limited

evidence to support the notion that certain end users value recall and rerate

protections. For instance, James O’Connor, the former Head of Portfolio

Finance and Head of Business Management at Och Ziff Capital Management,

maintained that prime brokers helped to protect the asset-manager from “risk

of recall or rerate.” (Dkt. #432-6 ¶ 16). O’Connor further stated that he did

not find “AQS’s anonymous platform to be a viable alternative to [their prime

broker] relationships.” (Id. ¶ 17).

But as Plaintiffs highlighted at oral argument before Judge Cave,

Defendants failed to provide any academic research to support their claim that

prime brokers provide these protections, and, if so, to what extent. (See Dkt.

559-1, Tr. 42:2-16; Dkt. #470-6, Tr. 326:2-9). As but one example,

Defendants’ experts analyzed a strikingly small amount of data in attempting

to prove the extent of these benefits. (See, e.g., Hendershott Report ¶ 88

(summarizing the exhibit that attempts to illustrate “how brokers may provide

re-rate protection to their short seller clients” based on one bank’s borrowing

and lending prices for one company’s stock)). Further, despite Defendants’

claims that these protections are extremely valuable, none of the prime

brokerage agreements between broker-dealers and end users included a

provision requiring (or even laying out the terms of) those services. (See Dkt.

#559-1, Tr. 43:1-6). Plaintiffs argue that this “highlights how opaque and

inefficient the market is for class members” (id. at Tr. 43:9-11), and the Court

agrees. The evidence shows that these benefits, to the extent they are valued

by Class members at all, can be contracted for when trading on multilateral

platforms. (Zhu Reply ¶¶ 95-100).

Moreover, Dr. Zhu fully expects that the OTC segment would remain in

the but-for world, and therefore that the benefits allegedly provided by prime

brokers would continue to be offered. (Zhu Reply ¶¶ 37, 62-67, 73). As Dr.

Zhu explained, “[t]he key question from an impact perspective” is not whether

some Class members benefit from conducting OTC trades, but whether all

Class members “in the but-for world would be better off than they were in the

real world.” (Id. ¶ 64).

In the Second Circuit, the existence of individual defenses “does not

compel a finding that individual issues predominate over common ones.”

Nextel Commc’ns Inc., 780 F.3d at 138 (internal quotation marks omitted).

Rather, courts, in determining whether to certify a Rule 23(b)(3) class, must

decide whether the issues “that qualify each class member’s case as a genuine

controversy … are more substantial than the issues subject only to

individualized proof.” Id. at 139 (internal quotation marks omitted). Having

weighed those issues, the Court finds that the issues raised by Defendants

regarding antitrust injury do not defeat predominance.

As a final point on this issue, the Court agrees with Plaintiffs that

Defendants’ purported plan to call “thousands” of class members, who have

neither been deposed nor subpoenaed, is unlikely to be carried out (or, for that

matter, be permitted by this Court). (See Pl. Opp. 36-39). Defendants may be

able to “pick off” some class members “through individualized rebuttal,” but

even those modest successes would not defeat predominance. Halliburton Co.

v. Erica P. John Fund, Inc., 573 U.S. 258, 276 (2014).

ii. Antitrust Damages

In this Circuit, “[a]ll that is required at class certification is that the

plaintiffs must be able to show that their damages stemmed from the

defendant’s actions that created the legal liability.” Sykes v. Mel S. Harris &

Assocs. LLC, 780 F.3d 70, 88 (2d Cir. 2015) (internal quotation marks omitted);

see also Nextel Commc’ns Inc., 780 F.3d at 138 (“Common issues — such as

liability — may be certified, consistent with Rule 23, even where other issues —

such as damages — do not lend themselves to classwide proof.”). Additionally,

under Supreme Court precedent, “certification pursuant to Rule 23(b)(3) [does

not] require[] a finding that damages are capable of measurement on a

classwide basis.” Roach v. T.L. Cannon Corp., 778 F.3d 401, 402 (2d Cir. 2015)

(applying Comcast Corp. v. Behrend, 569 U.S. 27 (2013)). The Supreme Court

has also expressed “willingness to accept a degree of uncertainty” in instances

where “[t]he vagaries of the marketplace … deny us sure knowledge of what

plaintiff’s situation would have been in the absence of the defendant’s antitrust

violation[.]” J. Truett Payne Co. v. Chrysler Motors Corp., 451 U.S. 557, 566

(1981) (comparing damages resulting from an antitrust violation to those

resulting from personal injury or condemnation of land, which are more

concrete).

The instant case is unlike Comcast Corp. v. Behrend, an antitrust class

action in which the Supreme Court clarified that “damages questions should be

considered at the certification stage when weighing predominance issues” and

ultimately held that certification was not appropriate. See Roach, 778 F.3d at

408 (explaining the scope of Comcast). In Comcast, the model “d[id] not even

attempt” to “measure only those damages attributable to” the impact theory.

569 U.S. 27, 35 (2013). Here, by contrast, Plaintiffs’ “theory of liability and

antitrust impact” is that “Defendants conspired to maintain an inflated spread

between the price at which they borrow from the Beneficial Owner Subclass

and the price at which they lend to the End-User Subclass.” (Pl. Br. 42).

Consistent with that theory, Plaintiffs’ damages model allows for the

comparison of real-world prices to estimated prices that would exist in a world

in which Defendants’ alleged conspiracy had not taken place, generating an

estimate of the amounts that borrowers were overcharged and that lenders

were underpaid by Defendants because of the anticompetitive conduct. (A&P

Report ¶¶ 25, 29; see generally id., Sections IX-XI). To demonstrate that the

damages methodology is a “common, formulaic approach [that] can reliably be

applied to measure the damages suffered on each stock loan transaction

without the need of individualized enquiry[,]” Drs. Asquith and Pathak have

already “compute[d] the damages incurred by each named Plaintiff on exemplar

stock loans transacted with the Prime Broker Defendants during the Class

Period.” (Id. ¶¶ 541-567).

In their objections, Defendants argue that Plaintiffs’ “model of but-for

prices” undercuts Plaintiffs’ impact showing, because Defendants’ expert, Dr.

McCrary, applied the model and found that it “predicts that over 30% of all

stock-loan transactions occurred at better prices in the actual world than those

that allegedly would have existed in the but-for world.” (Def. Obj. 33 (citing

McCrary Reply ¶¶ 18, 56, 59)). Judge Cave explicitly addressed this argument

in her Report, finding that Plaintiffs’ evidence had demonstrated that the 30%

calculation in fact “arose from … data errors.” (Report 54). Her finding was

not factual error.

Plaintiffs’ experts explain that Dr. McCrary’s calculations were “driven by

his inaccurate processing of data from Defendant UBS and grossly overstate[d]

the percentage of undamaged class members.” (A&P Reply ¶ 348). Drs.

Asquith and Pathak found that only 0.4% of all Beneficial Owner Subclass

accounts and 0.2% of all End User Subclass accounts lacked positive damages

on a transaction. (Id. ¶ 344; see also id. ¶ 348 (discussing that only 2% of

short-seller accounts, rather than 21% of short-seller accounts, are

undamaged when correcting Dr. McCrary’s data processing errors); Dkt. #556-

1 at 73-74 (showing the effects of the data processing errors on model output)).

Defendants argue further that the Report erroneously failed to resolve

disputes about the costs of platform trading, and that unrealistic cost

assumptions are baked into Plaintiffs’ damages model. (Def. Obj. 35-37). But

Plaintiffs’ experts rebuffed the critiques of Defendants’ experts regarding the

estimated costs of platform trading in the but-for world. (See A&P Reply ¶¶ 12-

13; see also Zhu Report ¶¶ 335-340). Drs. Asquith and Pathak found that the

costs proposed by Defendants’ expert, Dr. Hendershott, lacked support and

would render “all current stock lending … entirely uneconomical.” (A&P

Reply ¶ 13). Further, even when Drs. Asquith and Pathak made adjustments

to their calculations to address Dr. Hendershott’s concerns, they “continue[d] to

find positive damages for virtually all class members[.]” (Id. ¶ 12 (emphasis

added)).

Defendants next argue that, in making the finding that less than 0.5% of

the class was harmed, the Report “implicitly and erroneously adopted Plaintiffs’

argument that class members that suffered no net harm on a single stock loan

were nevertheless ‘harmed’ if the A&P model estimates that the but-for price

was better than the actual-world price on a single day of a multi-day loan.”

(Def. Obj. 37-38). They claim that this is error because this means that each

day of a multi-day stock loan is treated as a separate transaction, when in fact,

a multi-day stock loan is a single economic transaction. (Id. at 37). Instead,

Defendants argue, the damages for each day of a multi-day loan should be

netted against each other. But, as explained by Plaintiffs’ experts, “[m]ost

stock loans in the U.S. are ‘on-demand’ (or open) loans” which means that they

can be terminated at any time, for any reason, and therefore, “the loan is

effectively ‘re-rated’ every day; that is, the loan fee is reset daily.” (A&P Report

¶ 58). Further, the Court has already addressed Defendants’ argument that

some “class members suffered no net injury because actual-world prices often

were better than but-for prices.” (Def. Obj. 38). As discussed, Dr. Zhu

demonstrated that prices would have improved for all members of the class,

even those that currently receive preferable pricing. See supra D.1.b.i.a.

Finally, Defendants argue that the Report erred in finding that individual

damages inquiries do not defeat predominance. (Def. Obj. 45-46). The Court

notes that “[t]he Supreme Court [has not] foreclose[d] the possibility of class

certification under Rule 23(b)(3) in cases involving individualized damages

calculations.” Roach, 778 F.3d at 408. Further, here, the inquiries raised by

Defendants, namely whether Plaintiffs’ experts correctly applied filters to the

data and whether damages should be netted (Def. Obj. 45-46), are common

legal questions.

In sum, Plaintiffs have proposed a common methodology that purports to

measure only those damages attributable to the impact theory. The Report

thus correctly found that the issue of antitrust damages does not defeat

predominance. (Report 56).

iii. The Impact of the FTAIA

Defendants’ final predominance argument is that the Report erred in

finding that the FTAIA does not bar certification, because “innumerable

individual inquiries” will be required to determine which transactions must be

excluded from Sherman Act liability under the FTAIA. (Def. Obj. 46-49). The

FTAIA states, in relevant part, that the Sherman Act does “not apply to conduct

involving trade or commerce … with foreign nations.” 15 U.S.C. § 6a. By its

express terms, it thus “excludes from the Sherman Act’s reach much

anticompetitive conduct that causes only foreign injury.” F. Hoffmann-La

Roche Ltd. v. Empagran S.A., 542 U.S. 155, 158 (2004) (emphasis added); see

also Lotes Co. v. Hon Hai Precision Indus. Co., 753 F.3d 395, 413-14 (2d Cir.

2014) (“[T]he FTAIA generally excludes wholly foreign conduct from the reach of

the Sherman Act[.]”).

Of potential note, however, the FTAIA contains two exceptions: “[i] the

import exclusion, which applies to ‘conduct involving ... import trade or import

commerce,’ and [ii] the domestic effects exception, which applies to other

foreign conduct that has a direct, substantial, and reasonably foreseeable effect

on import or domestic commerce and that gives rise to a Sherman Act claim[.]”

Biocad JSC v. F. Hoffmann-La Roche, 942 F.3d 88, 94 (2d Cir. 2019) (internal

citations omitted). When either exception applies, the Sherman Act applies.

In the instant matter, Plaintiffs allege a domestic conspiracy involving the

U.S. stock loan market. To be included in the class, a person or entity must

have entered into the requisite number of “U.S. Stock Loan Transactions,”

those that involve the “lending or borrowing of a stock listed on [a U.S.

exchange,]” with “the U.S.-based entities of the Prime Broker Defendants[.]”

(Pl. Br. 7, 8 n.10). The present case is therefore distinguishable from the two

main cases on which Defendants rely, which are two decisions from the In re

Foreign Exchange Benchmark Rates Antitrust Litigation, No. 13 Civ. 7789 (LGS),

2016 WL 5108131, at *11 (S.D.N.Y. Sept. 20, 2016) (“Forex II”), and 407 F.

Supp. 3d 422 (S.D.N.Y. 2019) (“Forex III”). (Def. Obj. 46-48).10

In Forex II, a sister court in this District concluded that the FTAIA barred

“[p]laintiffs’ claims arising from OTC transactions where the plaintiff was

operating abroad and transacted with a foreign desk of a defendant, and

bar[red] claims arising from transactions conducted on a foreign exchange.”

2016 WL 5108131, at *15. Because of the class definition in the instant case,

however, none of Plaintiffs’ claims arises from OTC transactions where the

plaintiffs transacted with a foreign desk of a Defendant or from OTC

transactions arising from transactions conducted on a foreign exchange.

Further, even if some of the transactions involved foreign conduct, the

domestic effects exception would apply to them. As discussed, the conduct

ascribed to Defendants — conspiring to prevent the U.S. stock loan market

from transitioning into a transparent, direct, all-to-all electronic exchange —

clearly had “a direct, substantial, and reasonably foreseeable effect” on

10 Defendants do not in their objections reference the first opinion and order issued in the

Forex litigation, In re Foreign Exchange Benchmark Rates Antitrust Litigation, 74 F.

Supp. 3d 581 (S.D.N.Y. 2015) (“Forex I”), but the Court discusses Forex I below in

distinguishing the Forex litigation from the instant matter.

domestic commerce. 15 U.S.C. § 6a(1)(A). That effect was the continued

opacity of the U.S. stock lending market, and the concomitant lack of

alternative options for transactions in that market, all of which sustained high

spreads on U.S. transactions to the benefit of the Prime Broker Defendants and

to the detriment of Plaintiffs. Forex II does not require this Court to reach a

different conclusion. There, the plaintiffs argued that the “domestic effect of

[d]efendants’ conduct” had caused their foreign injuries, because there was a

“single [foreign exchange] market” that meant that antitrust conduct in the

U.S. affected foreign prices. Forex II, 2016 WL 5108131, at *14-15. While the

court held that the domestic effects exception did not apply, id., the facts of

that case are clearly distinguishable from the instant action.

Finally, the import exception applies to transactions involving even

foreign beneficial owners. In Allianz Global Investors GmbH v. Bank of America

Corp., the district court explicitly held that claims arising from instances where

“foreign-domiciled or operating Plaintiffs transacted with domestically-

operating Defendants … fall within the FTAIA imports exception.” 463 F. Supp.

3d 409, 423 (S.D.N.Y. 2020). Here, any foreign-domiciled beneficial owners

necessarily would have been transacting with domestic desks of the Prime

Broker Defendants, because of the way the Class is defined. On this point as

well, the Forex litigation cited by Defendants is inapposite. In Forex I, the

plaintiffs alleged that manipulation had occurred that “impacted currencies

worldwide,” and did not plead that the manipulation “was directed at the U.S.

import market.” 74 F. Supp. 3d at 599; see also Forex III, 407 F. Supp. 3d at

431-32 (finding that “the location of [p]laintiffs’ trading activities [wa]s highly

material to [their] class claims,” where the plaintiffs alleged an antitrust

conspiracy in a foreign exchange market and where class members traded with

foreign desks of the defendants). As discussed, the Class definition here

ensures that only transactions involving U.S.-operating Defendants are

included, and the evidence shows that the conspiracy was directed at the U.S.

stock loan market. (Pl. Br. 7, 8 n.10).

The Report correctly concluded that the FTAIA does not bar certification.

Having considered Defendants’ objections to the Report regarding

predominance, the Court finds that Plaintiffs have satisfied the predominance

requirement of Rule 23(b)(3) and proceeds to consider the superiority

requirement.

c. Superiority

Rule 23(b)(3) requires the moving party to establish that “a class action is

superior to other available methods for fairly and efficiently adjudicating the

controversy.” Fed. R. Civ. P. 23(b)(3). Rule 23(b)(3) provides non-exhaustive

factors that bear on the superiority determination, including: (i) “the class

members’ interests in individually controlling the prosecution or defense of

separate actions”; (ii) “the extent and nature of any litigation concerning the

controversy already begun by or against class members”; (iii) “the desirability

or undesirability of concentrating the litigation of the claims in the particular

forum”; and (iv) “the likely difficulties in managing a class action.” Id.;

Amchem, 521 U.S. at 615-16. By utilizing those factors, courts “consider the

interests of individual members of the class in controlling their own litigations

and carrying them on as they see fit.” Amchem, 521 U.S. at 616 (internal

quotation marks omitted).

The Report found that Plaintiffs have demonstrated superiority, and that

each of the Rule 23(b)(3) factors weighs in favor of that finding. (Report 63-65).

Objecting to that finding, Defendants argue that two of the factors, “the class

members’ interests in individually controlling the prosecution or defense of

separate actions” and “the likely difficulties in managing a class action,”

warrant a different conclusion. (Def. Obj. 49-50). As explained herein, those

factors do not weigh against certification, and the Court therefore adopts the

Report’s recommendation regarding superiority.

First, on the issue of the class members’ interests in individual

prosecution, Defendants’ objection primarily rests on their contention that the

Report “disregarded evidence showing that class members are both able and

willing to represent their own interests in this litigation[,]” despite its

acknowledgement that the putative class includes sophisticated hedge funds

and financial institutions. (Def. Obj. 50). Defendants are correct that the class

representatives in this case are sophisticated hedge funds and financial

institutions that are better positioned to carry out an individual lawsuit than

less sophisticated parties. But “authority in the Second Circuit holds that the

presence of sophisticated institutional [entities] capable of pursuing their

claims independently ‘is no bar to a class when the advantages of unitary

adjudication exist.’” N.J. Carpenters Health Fund v. Royal Bank of Scotland

Grp., PLC, No. 08 Civ. 5310 (DAB), 2016 WL 7409840, at *11 (S.D.N.Y. Nov. 4,

2016) (quoting Bd. of Trs. of the AFTRA Ret. Fund v. JPMorgan Chase Bank,

N.A., 269 F.R.D. 340, 355 (S.D.N.Y. 2010)); see also Cromer Fin. Ltd. v. Berger,

205 F.R.D. 113, 134 (S.D.N.Y. 2001) (finding that superiority was established

where “[r]elatively few of the investors” indicated a desire to litigate their claims

outside of the class action, even where they were “all concededly ‘sophisticated’

investors”). That is because while the superiority requirement exists in part to

ensure that the interests of individual members of the class have been

considered by a reviewing court, it is also “designed to avoid repetitious

litigation and possibility of inconsistent adjudications.” In re Payment Card

Interchange Fee & Merch. Disc. Antitrust Litig., 330 F.R.D. 11, 57 (E.D.N.Y.

2019) (internal quotation marks omitted); see also In re U.S. Foodservice Inc.

Pricing Litig., 729 F.3d 108, 130 (2d Cir. 2013) (recognizing that substituting a

single class action for numerous trials achieves economies of time, effort, and

expense, and promotes uniformity of decision).

Such advantages exist here, where there is a class that includes

thousands of members. (See A&P Report ¶ 25). Individual actions would prove

costly and inefficient in comparison to this class action. See Pub. Employees’

Ret. Sys. of Miss. v. Merrill Lynch & Co., 277 F.R.D. 97, 120 (S.D.N.Y. 2011).11

11 In Public Employees’ Retirement System of Mississippi v. Merrill Lynch & Co., the court

found that plaintiffs, several pension funds including Plaintiff Los Angeles County

Employees Retirement Association, had established superiority, in part because “there

[wa]s no overwhelming interest by class members to proceed individually.” 277 F.R.D.

97, 120 (S.D.N.Y. 2011). The court noted that a class action would be more efficient

and cost-effective than individual actions, where there were over 1,600 geographically-

dispersed class members. Id.

Additionally, “[m]ultiple actions by multiple plaintiffs could also significantly

reduce the prospects for recovery as it would decrease [P]laintiffs’ bargaining

power” (id.), which is an important consideration where, as here, Plaintiffs have

alleged that the Prime Broker Defendants “dominate[]” the stock loan market

and conspired to crush any efforts by individual actors to increase competition

in the stock loan market (see AC ¶ 1). And while the class representatives are

sophisticated entities alleging large individual damages, smaller institutional

investors may not be willing or able to maintain individual lawsuits against the

Prime Broker Defendants. See In re NASDAQ Mkt.-Makers Antitrust Litig., 172

F.R.D. 119, 130 (S.D.N.Y. 1997) (“[S]maller institutional investors may not be

willing and able to hire counsel to battle against the collective resources of the

nation’s largest financial industry firms.”).

Second, on the issue of manageability, Defendants argue that the Report

erred because it “summarily conclud[ed]” that the “individual issues” discussed

would not make a class action unmanageable. (Def. Obj. 50). But the Report

is correct that predominance and manageability can be related findings. See

Royal Park Invs. SA/NV v. Wells Fargo Bank, N.A., No. 14 Civ. 9764 (KPF) (SN),

2018 WL 1831850, at *9 (S.D.N.Y. Apr. 17, 2018) (finding that because

predominance was not satisfied due to necessary individual inquiries, the

Court and litigants would have great difficulty managing the litigation). Here,

the Court has found that individual issues will not predominate. Further,

“there are no apparent difficulties that are likely to be encountered in the

management of this action as a class action apart from those inherent in any

hard fought battle where substantial sums are at issue and all active parties

are represented by able counsel.” Pub. Employees’ Ret. Sys. of Miss., 277

F.R.D. at 120 (quoting Cromer Fin. Ltd., 205 F.R.D. at 134).

Accordingly, the Court finds that Plaintiffs have shown that litigation of

this case as a class action is superior to other methods of adjudication.

Moreover, having carefully considered Defendants’ adequacy, predominance,

and superiority arguments, the Court finds that Plaintiffs have satisfied

Rule 23 in the Class as defined. In its remaining analysis, the Court considers

Plaintiffs’ objections to the Report, which relate to (i) the end date of the Class

Period and (ii) the recoverability of damages arising after 2017.

2. The Class Period

In both the Complaint, filed on August 16, 2017, and the Amended

Complaint, filed on November 17, 2017, Plaintiffs proposed a class period that

spanned from January 7, 2009, “through the present.” (See Dkt. #1 at 1;

AC 1). At the stage of certification briefing, Plaintiffs proposed a class period of

January 1, 2012, through February 22, 2021,12 the date of the filing of the

original motion for class certification. (Pl. Br. 8; see also Pl. Reply 34-35).

In their opposition to certification, Defendants argued that the Class

Period should not extend beyond 2017 and that the phrase “through the

present” should be understood to mean November 17, 2017, the date of the

12 The Court notes that the original motion for class certification proposed the date

February 22, 2021, and that the reply memorandum proposed the date February 21,

2021. (Pl. Br. 8; Pl. Reply 35). The Court uses the date February 22, 2021, as that is

the date on which the certification motion was filed.

filing of the Amended Complaint. (See Def. Br. 49). Defendants maintained

that (i) there was no evidentiary basis for certifying a class that extends beyond

2017 because there has been no discovery covering the additional four-year

period encompassed in Plaintiffs’ proposed class period; and (ii) certifying a

class beyond 2017 would waste resources, because it would require additional

data productions, fact discovery, expert reports, and supplemental briefing.

(Id. at 49-50).

In their certification reply, Plaintiffs continued to argue for an end date of

February 22, 2021, offering that they had used the phrase “through the

present” in the Amended Complaint to indicate that the “conspiracy was still

causing ongoing harm” at the time of that filing. (Pl. Reply 34). Therefore, they

argued, the use of the phrase “through the present” in the Amended Complaint

was consistent with the proposed end date in the Motion (i.e., the harm was

ongoing at the time the Amended Complaint was filed and continued through

the date of the filing of the Motion). (Id.).

As a threshold matter, “[i]t is well-established that a certifying court ‘is

not bound by the class definition proposed in the complaint.’” In re Namenda

Direct Purchaser Antitrust Litig., 331 F. Supp. 3d 152, 210 (S.D.N.Y. 2018).

While this concept is typically cited in support of narrowing, rather than

expanding, a proposed class definition, courts in this District have

demonstrated that it is not “categorically improper” to expand a class definition

“beyond that which was proposed in the complaint[.]” Id. at 210-11 (collecting

cases). For example, in Menking, the plaintiff first proposed a citywide class in

her complaint and later proposed a statewide class in her motion for class

certification. Menking ex rel. Menking v. Daines, 287 F.R.D. 174, 181 (S.D.N.Y.

2012). The court ultimately concluded that the larger, statewide class

definition was proper, after ensuring that such a definition met the

requirements of the Rule 23(a). Id.

Further, there is no specific point in a case’s life cycle that courts in this

District use consistently as the end date for a class period. See, e.g., Decastro

v. City of New York, No. 16 Civ. 3850 (RA), 2019 WL 4509027, at *7 (S.D.N.Y.

Sept. 19, 2019) (selecting the date the complaint was filed as the end date for

the class period); Balverde v. Lunella Ristorante, Inc., No. 15 Civ. 5518 (ER),

2017 WL 1954934, at *5, 12 (S.D.N.Y. May 10, 2017) (certifying a class

“through the present,” the date of the decision on class certification); Hart v.

Rick’s NY Cabaret Int’l, Inc., No. 09 Civ. 3043 (PAE), 2013 WL 11272536, at *5-

6 (S.D.N.Y. Nov. 18, 2013) (modifying the class period end date from the date of

“the entry of judgment in this case” to the date on which fact discovery closed);

Ansoumana v. Gristede’s Operating Corp., 201 F.R.D. 81, 85 n.2 (S.D.N.Y.

2001) (considering the end date proposed in the complaint, namely the date of

entry of judgment, and the end date proposed in the certification motion,

namely the date of the filing of the complaint, and fixing the end date as the

date of the decision of the class certification motion). But though they may

differ in the date selected, courts do recognize that a definite end date is

advantageous, as it allows for ascertainability of class members, see Decastro,

2019 WL 4509027, at *7, and affords putative class members a better

understanding of whether their interests will be represented in the pending

lawsuit, see Rick’s NY Cabaret Int’l, Inc., 2013 WL 11272536, at *5-6.

As discussed above, Judge Cave ultimately recommended that Plaintiffs’

proposed class period be rejected in part, and that this Court set a class period

of January 1, 2012, to August 16, 2017, the date of the initial complaint.

(Report 69). Plaintiffs, understandably, object to the proposed end date and

argue that the Court should modify the Report to extend the class period to run

until at least November 17, 2017, the date that the Amended Complaint was

filed, or December 31, 2017, the date up to which Defendants have already

produced transactional data. (Pl. Obj. 16-17).

“[C]onsistent with the certifying court’s broad discretion over class

definition[,]” Namenda, 331 F. Supp. 3d at 212, the Court sets a Class Period of

January 1, 2012, to November 17, 2017, the date of the Amended Complaint.

The Court selects November 17, 2017, because the Amended Complaint is the

operative complaint in this matter, and because the evidence presented by

Plaintiffs extends throughout that period.

The remainder of Plaintiffs’ arguments pertain to whether Plaintiffs may

seek (i) damages arising after the end of the Class Period and (ii) additional

productions relevant to post 2017-damages. (Pl. Obj. 3-16). The Court notes

that in antitrust cases, provable damages can include future damages. See

Zenith Radio Corp. v. Hazeltine Rsch., Inc., 401 U.S. 321, 339 (1971) (“If a

plaintiff feels the adverse impact of an antitrust conspiracy on a particular

date, a cause of action immediately accrues to him to recover all damages

incurred by that date and all provable damages that will flow in the future from

the acts of the conspirators on that date[,]” unless the fact of their accrual of “is

speculative or their amount and nature unprovable.” (emphasis added)). In

other words, the certified Class members, who took part in the requisite

number of stock loan transactions between January 1, 2012, and

November 17, 2017, could seek to prove damages that flow from acts

conducted by the conspirators during the Class Period.

But, as the Report stated, fact discovery has closed in this case. (Report

68 (citing Dkt. #298 (Fourth Amended Case Management Plan) (stating that “all

fact discovery” was to be completed by October 16, 2020))). The Report also

correctly pointed out that “good cause” would be required to amend the Fourth

Amended Case Management Plan, which governs discovery in this case. (Id.).

See also Fed. R. Civ. P. 16(b) (providing that scheduling orders, like the Fourth

Amended Case Management Plan, shall not be modified except upon a showing

of “good cause” and by leave of the district judge)); cf. Sacerdote v. New York

Univ., 9 F.4th 95, 116 (2d Cir. 2021) (“District courts wishing to evaluate

motions for leave to amend under Rule 16 after a particular date need only

write their scheduling orders consistent with that intent, and state that no

amendment will be permitted after that date in order to trigger the Rule 16

standard.”). The Court declines, in ruling on a class certification motion, to

reach the question of whether supplemental production of transactional data is

permissible in this case to prove damages. To the extent that Plaintiffs wish to

seek such supplemental productions to update damages calculations, Plaintiffs

will be required to do so through a separate motion to the Court.

CONCLUSION

For the foregoing reasons, Defendants’ objections to the Report are

OVERRULED and Plaintiffs’ Motion is GRANTED IN PART and DENIED IN

PART. Plaintiffs’ objection to the Report regarding the Class Period is

OVERRULED IN PART. The Court certifies the Class, in the form

recommended by Judge Cave, except that the Class Period shall be January 1,

2012, through November 17, 2017. The Clerk of Court is directed to terminate

the motion pending at docket entry 468. The Clerk of Court is further directed

to modify the caption as discussed supra, at note 6.

The Court extends its appreciation to Judge Cave for her meticulous

Report. The parties remaining in this litigation are directed to submit a joint

letter to this Court on or before January 6, 2025, advising the Court as to how

they wish to proceed.

SO ORDERED.

Dated: December 6, 2024

New York, New York __________________________________

KATHERINE POLK FAILLA

United States District Judge