Opposition Brief — David Lowery, Petitioner v. Benjamin Joffe, et al.

Supreme Court briefAug 22, 2022

Ask Donna

What actually matters in this document.

Text

No. 21-1535

In the Supreme Court of the United States

__________

DAVID LOWERY, PETITIONER

BENJAMIN JOFFE, ET AL.

__________

ON PETITION FOR A WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

__________

BRIEF FOR GOOGLE LLC IN OPPOSITION

__________

STEFFEN N. JOHNSON

PAUL N. HAROLD

Wilson Sonsini

Goodrich & Rosati, PC

1700 K Street, NW

Washington, DC 20006

(202) 973-8800

BRIAN M. WILLEN

Counsel of Record

ELI B. RICHLIN

Wilson Sonsini

Goodrich & Rosati, PC

1301 Ave. of the Americas

New York, NY 10019

(212) 999-5800

bwillen@wsgr.com

Counsel for Respondent Google LLC

QUESTIONS PRESENTED

1. Whether, as every court of appeals to address

the question has recognized, a district court properly

exercises its discretion under Federal Rule of Civil

Procedure 23(e)(2) in approving a settlement that provides both injunctive and cy pres monetary relief,

where—as the district court found and the circuit

court affirmed—“it [is] not feasible to distribute funds

directly to class members.” App. 2a, 20a.

2. Whether, under Rule 23(b)(3), a district court

must find that individual class members can be identified without significant difficulty or expense before

certifying a class for purposes of settlement.

CORPORATE DISCLOSURE STATEMENT

Google LLC is a subsidiary of XXVI Holdings Inc.,

which is a subsidiary of Alphabet Inc., a publicly

traded company. No publicly held company owns 10%

or more of Alphabet Inc.’s stock.

iii

TABLE OF CONTENTS

Page

QUESTIONS PRESENTED ....................................... i

CORPORATE DISCLOSURE STATEMENT ........... ii

TABLE OF AUTHORITIES....................................... v

INTRODUCTION ....................................................... 1

STATEMENT ............................................................. 4

A. Factual background ....................................... 4

B. The complaint and the initial

proceedings below .......................................... 5

C. Jurisdictional discovery ................................ 6

D. The class settlement ...................................... 8

E. The district court’s approval decision ........... 9

F. The court of appeals’ decision ..................... 11

REASONS FOR DENYING THE PETITION ......... 13

The circuits uniformly permit cy pres relief

where direct distributions are infeasible. ......... 13

A. The decision below rests entirely on the

infeasibility of a direct distribution to

class members.............................................. 14

B. Every circuit recognizes infeasibility as

grounds for upholding cy pres

settlements. .................................................. 15

II. This case is a poor vehicle for addressing cy

pres settlements. ................................................ 20

A. The decision here rests on uncontested

and unique facts demonstrating

infeasibility. ................................................. 20

B. None of the cy pres abuses cited by

Petitioner occurred here. ............................. 21

C. There are serious questions about

Petitioner’s class membership and

standing. ...................................................... 23

D. Cy pres-only settlements are now

extremely rare. ............................................ 24

III. The decision below is correct. ............................ 25

A. Petitioner offers no credible method for

distributing funds to class members. ......... 26

B. The settlement is “fair, reasonable, and

adequate.” .................................................... 27

C. Cy pres awards do not violate the First

Amendment. ................................................ 28

IV. The second question presented does not

independently warrant review. ......................... 29

CONCLUSION ......................................................... 31

TABLE OF AUTHORITIES

Page(s)

Cases

Adashunas v. Negley,

626 F.2d 600 (7th Cir. 1980) ................................ 30

Amchem Prods., Inc. v. Windsor,

521 U.S. 591 (1997) .............................................. 30

In re Aqua Dots Prods. Liability Litig.,

654 F.3d 748 (2011).............................................. 30

In re Baby Products Antitrust Litig.,

708 F.3d 163 (3d Cir. 2013) ..................... 16, 19, 26

In re BankAmerica Corp. Sec. Litig.,

775 F.3d 1060 (8th Cir. 2015) ........ 2, 16, 18, 19, 27

Briseno v. ConAgra Foods, Inc.,

844 F.3d 1121 (9th Cir. 2017) .............................. 12

Carrera v. Bayer Corp.,

727 F.3d 300 (3d Cir. 2013) ................................. 30

Cherry v. Dometic Corp.,

986 F.3d 1296 (11th Cir. 2021) ............................ 30

Drazen v. Pinto,

41 F.4th 1354 (11th Cir. 2022) ............................ 27

In re EasySaver Rewards Litigation,

906 F.3d 747 (9th Cir. 2018) .......................... 14, 27

Exxon Co., U.S.A. v. Sofec, Inc.,

517 U.S. 830 (1996) .............................................. 15

Feder v. Electronic Data Systems Corp.,

248 F. App’x. 579 (5th Cir. 2007)......................... 24

Frank v. Gaos,

139 S. Ct. 1041 (2019) ................ 1, 3, 15, 20, 24, 25

In re: Google Inc. Cookie Placement

Consumer Priv. Litig.,

934 F.3d 316 (3d Cir. 2019) ................................. 19

In re Google Inc. St. View Elec.

Commc’ns Litig.,

794 F. Supp. 2d 1067 (N.D. Cal. 2011) .................. 6

In re Google Referrer Header Privacy

Litigation,

869 F.3d 737 (9th Cir. 2017) ............................ 3, 14

Graver Tank & Mfg. Co. v. Linde Air

Products Co.,

336 U.S. 271 (1949) .............................................. 15

Janus v. State, Cnty., and Mun. Emp.,

138 S. Ct. 2448 (2018) .......................................... 29

Joffe v. Google, Inc.,

746 F.3d 920 (9th Cir. 2013) .................................. 6

Jones v. Monsanto Co.,

38 F.4th 693 (8th Cir. 2022) .......................... 19, 29

Kamal v. J. Crew Grp., Inc.,

918 F.3d 102 (3d Cir. 2019) ................................. 25

Klier v. Elf Atochem N.A., Inc.,

658 F.3d 468 (5th Cir. 2011) ................ 2, 16, 17, 18

Lane v. Facebook, Inc.,

696 F.3d 811 (9th Cir. 2012) .......................... 11, 14

Marcus v. BMW of North Am. LLC,

687 F.3d 583 (3d Cir. 2012) ................................. 30

vii

Marek v. Lane,

571 U.S. 1003 (2013) .................................. 3, 11, 24

Masters v. Wilhelmina Model Agency, Inc.,

473 F.3d 423 (2d Cir. 2007) ................................. 16

In re Mexico Money Transfer Litig.,

267 F.3d 743 (7th Cir. 2001) ................................ 30

Mullins v. Direct Digital, LLC,

795 F.3d 654 (7th Cir. 2015) ................................ 30

Pearson v. NBTY, Inc.,

772 F.3d 778 (7th Cir. 2014) ................ 2, 16, 17, 26

In re Pharm. Indus. Average Wholesale

Price Litig.,

588 F.3d 24 (1st Cir. 2009) .................................. 16

Phillips Petroleum Co. v. Shutts,

472 U.S. 797 (1985) .............................................. 29

Spokeo, Inc. v. Robins,

578 U.S. 330 (2016) .............................................. 25

TransUnion LLC v. Ramirez,

141 S. Ct. 2190 (2021) .......................................... 27

Trichell v. Midland Credit Mgmt., Inc.,

964 F.3d 990 (11th Cir. 2020) .............................. 25

Constitutional Provisions

U.S. Const. amend. I ........................................... 28, 29

U.S. Const. art. III............................. 21, 23, 24, 25, 27

Other Authorities

Am. Law Inst. (ALI), Principles of the

Law of Aggregate Litigation § 3.07

(2010) .................................................................... 18

viii

Katherine Cienkus, Note, Privacy Class

Action Settlement Trends: Industry

Practice or Improper Incentives?,

Rev. of Litig., Spring 2021 ................................... 25

Fed. R. Civ. P. 23 ..................................... 12, 29, 30, 31

Fed. R. Civ. P. 23(b)(3) .............................................. 12

Fed. R. Civ. P. 23(e)(2) .......................................... 1, 19

Fed. R. Civ. P. 23(e)(5)(A) ......................................... 23

Fed. R. Civ. P. 23(f) ................................................... 30

Kaveh Pahlavan & Prashant

Krishnamurthy, Evolution and

Impact of Wi-Fi Technology and

Applications: A Historical

Perspective, 28 Int’l J. of Wireless

Info. Networks 3 (2021) ....................................... 21

INTRODUCTION

The petition rests principally on the premise that

this case is “functionally identical” to Frank v. Gaos,

139 S. Ct. 1041 (2019), where this Court took up (but

did not resolve) the circumstances in which cy pres

class action settlements satisfy Rule 23(e)(2). Pet. i.

That premise is false. The unique circumstances that

led to the settlement here distinguish this case from

Gaos and underscore that the ruling below neither

breaks new ground nor conflicts with any decision of

another court of appeals. Petitioner ignores the undisputed facts and portions of the decisions below that

illustrate why this is so, and none of his other arguments justify review.

This case arises from the unwanted acquisition by

Google Street View vehicles of transmissions from

open and unencrypted Wi-Fi networks—acquisitions

that ceased more than 12 years ago. In 2010, a group

of plaintiffs filed suit under the Wiretap Act, asserting

claims both on their own behalf and for a class comprising all persons who had Wi-Fi “payload” data collected between 2007 and 2010. Petitioner “does not

dispute” that “verifying that a person has a valid

claim would require making three determinations:

(1) the [claimant] had maintained an unencrypted WiFi network in the relevant period; (2) a Street View

vehicle passed within range of that network; and

(3) substantive communications * * * were transmitted” at “the precise fraction of a second when the

Street View vehicle passed by.” Pet. 18a-19a. It is

also undisputed that “‘[t]he only evidence’ of class

membership ‘is the intercepted data,’” which are “not

in the class member’s possession or readily accessible.”

App. 20a.

That is why it “took three years of intensive investigation and analysis” to analyze the standing claims

of eighteen named plaintiffs. App. 19a. That is why

“meaningful forensic verification” of 60 million claims

“would be prohibitively costly and time-consuming.”

Ibid. That is why Petitioner’s cure-all—having class

members self-identify—“would be pure speculation.”

App. 20a. And that is why the district court “f[ound]

that it was not feasible to distribute funds directly to

class members” (App. 2a)—an unchallenged finding

that distinguishes this case from all of Petitioner’s

cases rejecting cy pres settlements.

Once it becomes clear that there is no “viable way”

to “verify any claimant’s entitlement to settlement

funds” (App. 18a), the asserted circuit split—and any

need for this Court’s intervention—evaporates. The

circuits not only agree on the general standards governing cy pres relief, but uniformly recognize that

such relief is lawful “‘when it is not feasible to make

further distributions to class members.’” E.g., In re

BankAmerica Corp. Sec. Litig., 775 F.3d 1060, 1064,

1065 (8th Cir. 2015) (quoting Klier v. Elf Atochem N.A.,

Inc., 658 F.3d 468, 475 (5th Cir. 2011)). In every one

of the cases that Petitioner cites in asserting a split,

the settlement gave cy pres recipients residual funds

that could “feasibly be awarded” to “class members.”

E.g., Pearson v. NBTY, Inc., 772 F.3d 778, 784 (7th Cir.

2014). Indeed, every case involved residual cy pres—

cy pres distribution of settlement funds left unclaimed

after an initial monetary payment to class members.

Those cases by definition involve no issues with identifying class members.

This case is different. As the court of appeals explained, none of those cases “involved the sort of technical challenges to identifying class members present

here” (App. 18a), much less holds that cy pres settlements are unlawful when class members cannot be

identified. And the petition’s allegedly conflicting decisions on the second question presented involved the

standards for certifying classes for litigation, not settlement—an important difference that the petition

glosses over.

Numerous other factors support denying review.

Since Gaos and Marek v. Lane, 571 U.S. 1003 (2013)

(Roberts, C.J.), cy pres settlements have become the

even-rarer “exception, not the rule.” App. 15a. The

facts here, recognized by both courts below, make this

a uniquely fitting case for a form of settlement that all

agree should be used sparingly. Petitioner’s standing

to object to the settlement is questionable. The settlement here included meaningful injunctive relief and a

cy pres distribution that, as both courts below recognized, provided real benefits to the class. And it involved none of the potentially troubling features—

such as self-dealing by class counsel or funneling

money to preferred organizations—that have led

courts to raise concerns about cy pres class action settlements.

In short, the district court acted well within its discretion in finding that the settlement here lawfully

and sensibly brought an end to an unusual, long-running case that otherwise might have been impossible

to settle. There is no conflict and nothing else that

warrants this Court’s review. Certiorari should be denied.

STATEMENT

A. Factual background

In 2007, Google launched a feature called Street

View, which provides users with panoramic, streetlevel photographs of roads around the United States.

Street View images are taken by cameras mounted on

cars that drive down public roads while photographing their surroundings. For a time, Street View

vehicles were outfitted with off-the-shelf radio equipment and open-source software that passively collected network-identifying information openly broadcast by Wi-Fi networks along the roads they traveled,

which could enable Google to provide users with enhanced location-aware services.

In May 2010, Google learned that its Street View

vehicles had also acquired so-called “payload data”—

fragments of information being transmitted across

Wi-Fi networks that were configured to be open (i.e.,

networks that were not password-protected or encrypted). But the acquisition of payload data was limited. First, data were acquired only if transmitted

over an unencrypted Wi-Fi network at the precise moment that a Street View vehicle happened to pass by.

Second, Google’s software for identifying networks cycled through available Wi-Fi channels at the rate of

five times per second—meaning that data would only

be acquired if transmitted during the one-fifth of a

second when the software could see that specific WiFi network. App. 46a.

Google had no interest in acquiring these payload

data, and it has never used the data in any of its products or services. See App. 4a. Upon learning of the

unwanted data collection, Google promptly grounded

its Street View cars, segregated the acquired payload

data, made the data inaccessible, and hired a third

party to review what had happened. Ibid. Google also

publicly described these events on its official blog,

apologized for collecting payload data, and put procedures in place to prevent it from happening again.

Ibid.

Shortly after Google’s public disclosures, federal

and state agencies began investigating its conduct.

Although the federal agencies ultimately declined to

take action against Google, a joint investigation by 38

state attorneys general was resolved in 2013 with an

Assurance of Voluntary Compliance. C.A. App. 167179. That agreement required Google to: (1) delete or

destroy the payload data it had collected; (2) not collect and store payload data for use in any product or

service without notice and consent; (3) maintain a privacy program; and (4) implement a public-service and

educational campaign. C.A. App. 171-173. Google

also agreed to pay $7 million. Ibid.

B. The complaint and the initial proceedings

below

Beginning in May 2010, shortly after Google disclosed its collection of payload data, more than a dozen

putative class-action lawsuits challenging that activity were filed in courts around the country. The Judicial Panel on Multidistrict Litigation eventually

transferred those cases to the Northern District of

California for pretrial coordination.

Respondent Benjamin Joffe and the other named

plaintiffs alleged that, sometime between 2007 and

2010, payload data transmitted over their unencrypted Wi-Fi networks were collected by Google.

Plaintiffs sought to represent a class consisting of all

individuals whose Wi-Fi payload data were collected

during that time period. The plaintiffs’ Consolidated

Class Action Complaint, filed in November 2010, asserted claims under the federal Wiretap Act, 18 U.S.C.

§§ 2510 et seq., various state wiretap laws, and California’s unfair competition law.

Google moved to dismiss. The district court

granted the motion as to the state law claims, ruling

on preemption and standing grounds, but held that

the complaint stated a claim under the Wiretap Act.

In re Google Inc. St. View Elec. Commc’ns Litig., 794

F. Supp. 2d 1067 (N.D. Cal. 2011). The court certified

an interlocutory appeal on that claim and stayed further proceedings. The Ninth Circuit affirmed. Joffe

v. Google, Inc., 729 F.3d 1262 (9th Cir.), amended and

superseded on reh’g, 746 F.3d 920 (9th Cir. 2013).

C. Jurisdictional discovery

On remand, the district court ordered limited jurisdictional discovery into the plaintiffs’ standing. As

Google explained, a plaintiff would have standing only

if their WiFi payload data were actually acquired by

Google. At a minimum, that would require that:

(1) the plaintiff had maintained an unencrypted WiFi network during the relevant period; (2) a Street

View vehicle passed within range of that network; and

(3) payload data were transmitted within the precise

fraction of a second when the Street View vehicle

passed by.

The district court appointed a special master to

oversee the “intensive” three-year-long process of examining the collected data and to assess the named

plaintiffs’ standing. App. 46a. The special master

first had “to organize the data into a searchable database.” Ibid. That required recovering and forensically

preserving the data on the hundreds of individual

hard drives used by Google’s Street View vehicles. He

then had to develop custom software for processing

the raw data before organizing the data in a database.

App. 6a. Further complicating matters, the networking information that could identify the plaintiffs’ WiFi networks had been segregated from the payload

data, so the special master had to match the frames

containing that networking information with the corresponding frames of payload data. In addition, he

had to convert the raw, machine-readable data into

something people could read.

That was just the beginning. The special master

then spent two years “design[ing] and conduct[ing]”

the “complex technical searches” needed to determine

whether the payload data contained any communications intercepted from the plaintiffs. App. 46a. The

plaintiffs turned over personal information and forensic evidence relating to their wireless network equipment, including media access control (“MAC”) addresses, email addresses, and service set identifiers

(“SSIDs”). 1 Ibid. After developing a master search

protocol, the special master used a variety of methods

to search the data set. Ibid. He implemented three

search methodologies: for email addresses, for MAC

addresses, and for SSIDs near certain GPS coordinates. Ibid. Hits on these searches would indicate

that basic networking information had been collected,

1 A MAC address is a unique 12-digit hexadecimal iden-

tifier assigned by manufacturers to the computer hardware

component that connects a computer to a computer network. An SSID is the name of a wireless network, which

is required to connect to the network. SSIDs are customizable by the user and not necessarily unique.

but further inquiry was needed to determine if payload data from these networks had also been collected.

At the end of the special master’s intensive threeyear process, he filed a report with the district court,

which was “still not entirely conclusive on whether

Google had intercepted payload data from the named

plaintiffs.” App. 6a. In addition to disputes about

what the forensic examination of the data showed,

other significant questions remained in the case, including whether Google had “violated the [Wiretap

Act], whether Plaintiffs’ data was ‘readily accessible

to the general public,’ and whether, even if Plaintiffs

won, the Court would award statutory damages.” App.

68a. And as the district court explained, delay resulting from continued litigation over these legal issues

would make it more difficult for individual class members to recover: “every year that passes makes it increasingly likely that class members would replace

and dispose of the Wi-Fi routers they used between

2007 and 2010, which are critical to demonstrating

that Google actually intercepted their data.” Ibid.

D. The class settlement

In June 2018, facing these uncertainties and more

—and having already spent eight years litigating a

case involving events that took place between 2007

and 2010—the parties settled. App. 88a-110a. The

settlement class comprised “all persons who used a

wireless network device from which Acquired Payload

Data was obtained” from January 1, 2007, through

May 15, 2010. App. 92a.

The settlement included both monetary and injunctive relief. Google would pay $13 million into a

non-reversionary settlement fund. App. 95a-96a. After attorneys’ fees and costs, incentive awards to

named plaintiffs, and claims administration costs, the

remainder of the fund would be divided among cy pres

recipients, selected by the district court, who were

dedicated to promoting and protecting class members’

privacy interests. App. 97a-98a. Google had no role

in the selection or approval of the cy pres recipients.

The settlement also included multi-pronged injunctive relief, including requirements that Google

“destroy all Acquired Payload Data”; refrain from “collect[ing] and stor[ing] for use in any product or service

Payload Data via Street View vehicles, except with notice and consent”; host and maintain educational

webpages about configuring wireless networks securely; and extend for at least two additional years

Google’s obligations under its Assurance of Voluntary

Compliance. App. 98a-99a. While the Assurance of

Voluntary Compliance had similar provisions, the settlement agreement extended the time of Google’s obligations and expanded its obligations to host and

maintain educational webpages. Indeed, because of

the settlement agreement, Google has already significantly revised and expanded its educational webpages

to make them clearer, more detailed, and better able

to inform the public about how to protect their homeWi-Fi networks and opt out of certain location-based

services. See D. Ct. Dkt. 210 at 20.

E. The district court’s approval decision

The plaintiffs moved for preliminary approval of

the settlement. They proposed eight cy pres recipients,

and a ninth entity separately petitioned the district

court to receive cy pres funds. The district court

(Breyer, J.) granted preliminary approval, adding the

ninth group as a cy pres recipient.

Before the final approval hearing, only two putative class members objected. Petitioner David Lowery

objected to the settlement approval, cy pres recipients,

class certification, and the fee request. App. 111a153a. Petitioner’s claim of standing rested solely on a

declaration stating “[o]n information and belief” that

“Google surreptitiously collected, decoded, and stored

data from [his] WiFi connection, including payload

data,” during the class period. Dkt. 188-1 at 2. Another objector submitted a similar one-page letter. A

group of state attorneys general filed an amicus brief

objecting to the cy pres relief.

After holding a fairness hearing in early 2020 that

included arguments from the plaintiffs, Google, Petitioner, and the Arizona Attorney General’s Office, the

district court approved the settlement. App. 43a-84a.

In a comprehensive decision, the court rejected Petitioner’s argument that some claims process was feasible, holding that a cy pres distribution best benefited

the class because the settlement fund was otherwise

“non-distributable.” App. 69a-76a. The court expressly found that it would be impossible for class

members to self-identify: “[U]nlike a case in which a

class member could self- identify as having bought, for

example, a particular brand of cereal during the class

period, no member of the class here can know whether

Google intercepted his or her data” because “[t]he only

evidence is the intercepted data,” which “is not in the

class member’s possession.” App. 71a-72a.

Informed by its experience with the three-year

Special Master process, the court found that examining the data to determine class membership would be

prohibitively costly and time-consuming, requiring

the parties to “comb[] through nearly 300 million

frames of collected payload data and try[] to associate

it with individual Class Members.” App. 71a. “Even

assuming that * * * process would work,” the court

explained, it would not be “desirable.” App. 72a. Only

a small fraction of a class would be “able to file a claim”

—because the overwhelming majority would not, in

2018, have the information relating to the Wi-Fi systems they used in 2007-2010—which “would leave

99% of the class with no benefit from the Settlement

Fund.” App. 73a.

A cy pres settlement, on the other hand, would benefit the class by “increas[ing] the funding” for “some of

the most effective advocates for internet privacy in the

country” and “likely yield actual improvements to internet privacy.” App. 74a. The district court also

found that the settlement provided for “adequate” injunctive relief beyond the mandates imposed by the

Assurance of Voluntary Compliance. App. 76a-77a.

F. The court of appeals’ decision

The Ninth Circuit unanimously affirmed. App. 1a35a. In an opinion authored by Judge Bade, the court

began by explaining that settlements “provid[ing]

monetary relief only in the form of cy pres payments

to third parties” are not categorically unlawful. App.

16a. Full cy pres settlements are permissible so long

as “settlement funds are not distributable,” the “disbursements ‘account for the nature of the plaintiffs’

lawsuit, the objectives of the underlying statutes, and

the interests of the silent class members,’” and the settlements “satisfy the appropriate standards for fairness.” App. 16a-17a (quoting Lane v. Facebook, Inc.,

696 F.3d 811, 821 (9th Cir. 2012), cert. denied sub.

nom. Marek v. Lane, 571 U.S. 1003 (2013))).

In approving the settlement, the court of appeals

affirmed the district court’s finding that direct distributions to class members were “not feasible” here, as

“self-identification would be pure speculation, and

any meaningful forensic verification of claims would

be prohibitively costly and time-consuming.” App.

20a. Petitioner, the court observed, failed to identify

“a viable way” to “verify any claimant’s entitlement to

settlement funds.” App. 18a. He “d[id] not dispute”

that the only verification process would be “the

method the special master used—a process that took

three years of intensive investigation and analysis to

verify the claims of eighteen named plaintiffs.” App.

19a.

Petitioner pointed to other class settlements with

direct payments, but “none of the examples [he] cite[d]

involved the sort of technical challenges to identifying

class members present here.” App. 18a n.5. Nor was

self-identification viable, as “‘[t]he only evidence” of

class membership “is the intercepted data, and that

evidence is not in the class member’s possession’ or

readily accessible to the claims administrator.” App.

20a (quoting App. 72a (district court)).

The court also rejected Petitioner’s “Rule 23(b)(3)

‘superiority’ argument” that for the class device to be

superior to alternatives, settlement funds must be

“distribut[able] * * * to class members.” App. 21a.

This argument, the court reasoned, “essentially repackage[d]” Petitioner’s already-rejected argument

for “a blanket prohibition” on cy pres-only settlements

and was “similar” to the argument rejected by Briseno

v. ConAgra Foods, Inc., 844 F.3d 1121 (9th Cir. 2017),

which held that Rule 23 did not require that identifying class members be “administratively feasible.” App.

21a. After reaffirming those precedents, the court

also rejected Petitioner’s premise “that it is impossible

to provide meaningful relief to a class when there is

no feasible way of identifying class members,” explaining that a cy pres award with the required nexus

to the class would “particularly ‘benefit the plaintiff

class’” and “necessarily prioritize[] class members’ interests, even if it also provide[d] a diffuse benefit to

society at large.” App. 22a-23a.

Judge Bade concurred in her own opinion for the

court, writing separately “to express some general

concerns about cy pres awards”—though not about the

particular settlement in this case. App. 36a. And

Judge Bade did not claim that Ninth Circuit’s approach to cy pres conflicted with the approaches of

other circuits. To the contrary, in her opinion for the

court, she observed that “other circuits have generally

taken a similar approach to * * * approving cy pres

settlements.” App. 17a.

The Ninth Circuit denied rehearing and rehearing

en banc, with “no judge * * * request[ing] a vote on

whether to rehear the matter en banc.” App. 87a.

REASONS FOR DENYING THE PETITION

I. The circuits uniformly permit cy pres relief

where direct distributions are infeasible.

Petitioner’s lead argument for certiorari is that

“the Ninth [Circuit] stands alone on cy pres.” Pet. 17.

But his petition distorts both the Ninth Circuit’s decision and the decisions of other circuits. There is no

split, and the Ninth Circuit’s ruling is fully aligned

with every other circuit that has ruled on the propriety of cy pres class settlements.

A. The decision below rests entirely on the

infeasibility of a direct distribution to

class members.

The decision below stands for a limited proposition

—that cy pres relief is permissible where there is no

“viable way for a claims administrator to verify any

claimant’s entitlement to settlement funds.” App. 18a.

Aware that the court below ruled narrowly, petitioner

says the Ninth Circuit has blessed cy pres relief in

other cases where it may have been feasible to distribute cy pres funds directly to class members. Pet. 1718 (citing Lane v. Facebook, Inc., 696 F.3d 811 (9th Cir.

2012); In re Google Referrer Header Privacy Litigation,

869 F.3d 737 (9th Cir. 2017); and In re EasySaver Rewards Litigation, 906 F.3d 747 (9th Cir. 2018)). But

this case does not present that question—and Petitioner’s authorities all predate this Court’s consideration of the cy pres issue in Gaos.

The explicit premise of the decision below is that

“it was not feasible to distribute funds directly to class

members given the class size and the technical challenges to verifying class members’ claims.” App. 2a.

That conclusion was based on the parties’ and the district court’s experience with jurisdictional discovery,

which “took three years of intensive investigation and

analysis to verify the claims of eighteen named plaintiffs.” App. 19a. As the Ninth Circuit explained, each

class member’s claim requires “meaningful forensic

verification,” which would be “prohibitively costly and

time-consuming.” App. 20a. And since potential

claimants do not possess the data needed “to determine with any degree of probability whether they are

class members,” allowing them to self-identify by submitting declarations “would be pure speculation.”

Ibid.

Petitioner has never challenged these factual findings; nor could he. “‘A court of law, such as this Court

is, rather than a court for correction of errors in fact

finding, cannot undertake to review concurrent findings of fact by two courts below in the absence of a

very obvious and exceptional showing of error.’”

Exxon Co., U.S.A. v. Sofec, Inc., 517 U.S. 830, 841

(1996) (quoting Graver Tank & Mfg. Co. v. Linde Air

Products Co., 336 U.S. 271, 275 (1949)). The undisputed record here—which was both the reason that

the parties adopted a cy pres settlement framework

and the reason it was approved by both courts below—

disposes of his assertions that this case is “functionally identical” to Gaos (Pet. i), and that the court below made no “inquiry about whether cy pres is distributable to some class members” (Pet. 21). It also forecloses any suggestion that the decision below conflicts

with decisions of circuits that purportedly take another “view of feasibility” (Pet. 18). In short, the petition rests on a string of demonstrably false factual

premises.

B. Every circuit recognizes infeasibility as

grounds for upholding cy pres settlements.

The Ninth Circuit’s holding is fully aligned with

how other circuits have addressed these issues. No

circuit has categorically barred cy pres settlements or

adopted a legal standard for evaluating them that diverges from the standard applied below.

To begin with, every decision that Petitioner cites

as “categorically reject[ing] the Ninth Circuit’s test”

(Pet. 18) involved a cy pres distribution of residual

funds—where, by definition, there had been a previous distribution to class members, such that class

members could be identified and “distribution to the

class was clearly feasible.” E.g., BankAmerica, 775

F.3d at 1064; accord Klier, 658 F.3d at 478 (“it was

feasible to allocate the funds”); Pearson, 772 F.3d at

784 (funds could “feasibly be awarded” to “class members”); In re Baby Products Antitrust Litig., 708 F.3d

163, 169-170 (3d Cir. 2013) (reversing award to “cy

pres recipients in lieu of fully compensating class

members”). Even in that context, moreover, the

courts reject the absolute bar on cy pres settlements

that Petitioner urges.

Most importantly, the courts uniformly agree that

cy pres awards are lawful “‘when it is not feasible to

make further distributions to class members’”—either

because the amounts involved are “too small to make

individual distributions economically viable” (e.g.,

BankAmerica, 775 F.3d at 1064, 1065 (8th Cir.) (quoting Klier, 658 F.3d at 475 (5th Cir.)), or because “class

members cannot be identified” (e.g., In re Pharm. Indus. Average Wholesale Price Litig., 588 F.3d 24, 3334 (1st Cir. 2009)); accord Masters v. Wilhelmina

Model Agency, Inc., 473 F.3d 423, 436 (2d Cir. 2007)

(asking whether “it would be onerous or impossible to

locate class members” or “each class member’s recovery would be so small as to make an individual distribution economically impracticable”); Pearson, 772

F.3d at 784 (7th Cir.); Baby Prods., 708 F.3d at 173

(3d Cir).

Petitioner’s cases follow the same pattern. Start

with Pearson (Pet. 18-19), where the Seventh Circuit

reversed a settlement designating $2 million for class

counsel (69% of the settlement’s value) and $1.13 million for a cy pres award, compared with “$865,284”—

“7 cents apiece”—distributed to a multi-million-member class. 772 F.3d at 781, 783-784. Most of the class

was known: records showed that 4.72 million class

members bought pills from the defendant, which

“could have mailed $3 checks to all 4.72 million.” Id.

at 783. For the rest, a “sworn statement” sufficed.

Ibid. Here, by contrast, the factual premise of the decision below is that it is “not feasible to distribute

funds directly to class members”; they lack the information needed “to ‘self-identify’”; and there is “no alternative way for claimants to determine with any degree of probability whether they are class members”

that is not “prohibitively costly and time-consuming.”

App. 2a, 19a-20a & n.6.

On these facts, the present case would have come

out the same way in the Seventh Circuit, which explained that cy pres recipients may “receive money intended to compensate victims” where “it’s infeasible to

provide that compensation to the victims[].” Pearson,

772 F.3d at 784; see ibid. (“[a] cy pres award is supposed to be limited to money that can’t feasibly be

awarded to * * * class members”). The difference between this case and Pearson is simply that infeasibility “ha[d] not been demonstrated” there. Ibid. (emphasis added).

Fifth Circuit precedent is the same. In Klier (Pet.

19), the settlement allowed a subclass whose members

were exposed to the defendant’s arsenic, but without

becoming ill, to opt for medical monitoring rather than

compensation (which other subclasses received). The

demand for medical monitoring later waned, however,

leaving $830,000 left over, and the district court disbursed it to cy pres recipients. 658 F.3d at 472-473.

The Fifth Circuit reversed, explaining that cy pres

distribution “is permissible ‘only when it is not feasible to make further distributions to class members.’”

Id. at 475 (quoting Am. Law Inst. (ALI), Principles of

the Law of Aggregate Litigation § 3.07 cmt. a (2010)).

While it was “not feasible” to allocate more funds to

the medical monitoring subclass (Subclass B), “it was

feasible to allocate the funds to Subclass A,” “the most

seriously injured class members,” and one that had received a prior distribution. Id. at 477, 471 (emphasis

added); see id. at 478 & n.28 (the decision fits “comfortably” within “prior decisions” of “sister circuits”).

That is “not feasible” here. App. 2a, 19a-20a.

The Eighth Circuit “agree[s] with the Fifth Circuit.”

BankAmerica, 775 F.3d at 1064. In BankAmerica (Pet.

19), the court reversed a settlement distributing to cy

pres recipients $2.4 million left over after two distributions from a $490 million global securities settlement. As the court explained, cy pres relief is permissible if “the amounts involved are too small to make

[further] individual distributions economically viable.”

775 F.3d at 1065 (quoting ALI, Principles of the Law

of Aggregate Litigation § 3.07(a) (2010)). But the residual $2.4 million could have been distributed to the

class at an administrative cost of “$27,000.” Id. at

1064. Thus, “further distribution to the class was

clearly feasible,” and cy pres distribution was invalid.

Ibid.

This standard accords with the Ninth Circuit’s approach here. And any suggestion that the Eighth Circuit takes a harder line was rebuffed in a recent ruling

upholding a residual cy pres award and explaining

“that unclaimed funds may only be distributed cy pres

where existing class-member claimants have been

fully compensated and further distribution to remaining class members is not feasible.” Jones v. Monsanto

Co., 38 F.4th 693, 698-699 (8th Cir. 2022) (citing

BankAmerica, 775 F.3d at 1064).

Petitioner also asserts (at 20) a conflict with In re

Baby Products Antitrust Litigation, yet another case

involving cy pres distribution of residual funds. 708

F.3d at 173. Like the Fifth and Eighth Circuits, the

Third Circuit “agree[d] with the [ALI] that cy pres distributions are most appropriate where further individual distributions are economically infeasible,” but

“decline[d] to hold that cy pres distributions are only

appropriate” when further distributions are infeasible.” Id. at 169 (invalidating a settlement for allocating funds to “cy pres recipients in lieu of fully compensating class members”). A more recent Third Circuit

precedent expressly rejected petitioner’s view that “cy

pres-only settlements are unfair per se under Rule

23(e)(2).” In re: Google Inc. Cookie Placement Consumer Priv. Litig., 934 F.3d 316, 326 (3d Cir. 2019).

Nor is Petitioner aided by Judge Bade’s concurrence, which expressed “some general concerns” about

cy pres. App. 36a; see Pet. 14-15. Judge Bade authored the court of appeals’ unanimous opinion, and

her concurrence does not suggest that this settlement

should have been rejected, let alone that Ninth Circuit

precedent is out-of-step with that of other circuits.

See App. 36a (Bade, J., concurring). Her opinion for

the court in fact said the opposite. App. 17a.

In short, the legal standard that the Ninth Circuit

used to evaluate and uphold the settlement here is no

different from the standard applied in the other circuit court decisions that Petitioner cites. The difference is just that this case involves a set of “technical

challenges to identifying class members” that has no

analog in those cases. App. 18a. There is no circuit in

which the cy pres settlement here would have been invalidated. This Court’s review is not needed to create

uniformity or to bring an outlier circuit in line.

II. This case is a poor vehicle for addressing cy

pres settlements.

This Court previously granted certiorari to consider the permissibility of cy pres awards, and was unable to resolve the issues because of “substantial questions” about the plaintiffs’ standing. Gaos, 139 S. Ct.

at 1043-1044. This case, however, is an even worse

vehicle for addressing any questions about cy pres

class settlements, and developments since Gaos have

only diminished any need for review.

A. The decision here rests on uncontested

and unique facts demonstrating infeasibility.

As discussed, Petitioner’s broadside attack on cy

pres awards ignores the unique facts and circumstances of this case. The alternative to cy pres here

was a claims process in which claimants from a class

estimated to include 60 million people—the precise

number is impossible to know—would attempt to

show that their payload data were collected by

Google’s Street View vehicles between January 2007

and May 2010. But as the courts below found (and

Petitioner does not dispute), that process would have

been administratively infeasible, “prohibitively costly,”

and “time-consuming.” App. 20a; see also App. 6a, 46a,

71a-72a.

The nature of that burden was demonstrated by

“the three-year forensic investigation” preceding the

settlement, in which the parties worked with a special

master to determine whether the 18 named plaintiffs

had their own payload data collected. App. 2a. Making that showing was a threshold requirement for Article III standing, class membership, and any right to

recover under the Wiretap Act. And the jurisdictional

discovery process for the named plaintiffs alone required an “intensive” effort costing nearly $1 million.

App. 19a.

It thus became clear that setting up a settlement

claims process for potential class members would

have “requir[ed] a lengthy process, akin to the Special

Master’s process.” App. 72a. Moreover, it would have

required potential claimants to have “possession of

the Wi-Fi router [that they] used between 2007 and

2010”—something many class members no longer possess. Ibid. Four new Wi-Fi standards have been released since 2010, making it unlikely that anyone still

uses their obsolete router. See Kaveh Pahlavan &

Prashant Krishnamurthy, Evolution and Impact of

Wi-Fi Technology and Applications: A Historical Perspective, 28 Int’l J. of Wireless Info. Networks 3, 8

(2021). And “every year that passes makes it increasingly likely” that class members “dispose of the Wi-Fi

routers” they used more than a decade ago. App. 68a.

It was only against this unusual backdrop that the

parties and both courts below determined that it was

infeasible to establish a process to identify class members eligible to recover as part of a settlement. Very

few, if any, class actions involve such a record, and it

makes the settlement here especially fact-bound and

unlikely to recur.

B. None of the cy pres abuses cited by Petitioner occurred here.

This case is also a poor vehicle for review of cy pres

awards because none of Petitioner’s concerns about

abuse of cy pres are implicated here. Indeed, the bulk

of the petition concerns alleged abuses in other cases.

Petitioner complains that “cy pres can facilitate an

early settlement” under which “class attorneys are rewarded for selling their putative clients down the

river.” Pet. 27. But this case was litigated through

“vigorous and capable advocacy” for nearly a decade

(App. 53a)—including through an early interlocutory

appeal and intensive discovery into the named plaintiffs’ standing, with jurisdictional discovery overseen

by the special master. App. 5a-6a. It settled only after

“years of litigation and five months of settlement negotiations,” and only after the special master process

revealed the difficulty of ascertaining the standing of

even a few named plaintiffs—which confirmed that

many more years of litigation lay ahead. App. 67a.

Petitioner similarly complains about “the appearance of impropriety for district court judges.” Pet. 28.

But this case involves no suggestion that the district

court had any connection to any of the cy pres recipients. Nor is there any concern that Google steered cy

pres payments to favored charities. See App. 28a, 79a.

Under the settlement’s agreement’s express terms,

Google played no role in the selection of the cy pres

recipients (App. 97a), many of which have criticized

Google or filed amicus briefs adverse to the company

(see C.A. Google Ans. Br. 42-43). Given that these

charities are “some of the most effective advocates for

internet privacy in the country” (see App. 74a), it

should come as little surprise that a few of them previously received cy pres payments or donations from

Google. But in no sense did this settlement amount to

a give-away to Google’s preferred charities.

Nor does this case illustrate the danger of plaintiffdriven “forum shopping” (Pet. 33): it was assigned to

a district court in the Ninth Circuit by the Judicial

Panel on Multidistrict Litigation. App. 5a. And in any

event, as discussed above, the circuits all apply the

same standard in reviewing cy pres awards. This settlement would not have received a more skeptical review in any other circuit.

C. There are serious questions about Petitioner’s class membership and standing.

Questions concerning Petitioner’s class membership and Article III standing present yet more obstacles to the Court’s review. To be a member of the class,

Lowery must have “used a wireless network device

from which Acquired Payload Data was obtained.”

App. 92a. But as his own affidavit confirms, he lacks

personal knowledge that Google obtained his payload

data. See D. Ct. Dkt. 188-1 at 2 (stating “[o]n information and belief” that “Google surreptitiously collected, decoded, and stored data from my WiFi connection, including payload data”).

If Google did not obtain Petitioner’s payload data

during the class period—something he cannot know

without the type of investigation that took the special

master below three years to complete for the named

plaintiffs—he is not a “class member” with the right

to object to the settlement. Fed. R. Civ. P. 23(e)(5)(A)

(“Any class member may object * * * ”).

Nor would Petitioner have the injury-in-fact required to support the Article III standing required to

seek review in the court of appeals and this Court. See

Gaos, 139 S. Ct. at 1046. In contrast to the named

Plaintiffs, whose claims of standing were subjected to

the Special Master’s “extensive forensic analysis”

(App. 19a n.6), Petitioner did not even try to prove his

standing or offer up his router or its network infor-

mation. Instead, he came forward with merely an “information and belief” assertion that Google intercepted his data. D. Ct. Dkt. 188-1 at 2.

At a minimum, the “quantum of proof necessary to

establish class membership for the purpose of objecting to a settlement” is an open question that this

Court would have to resolve before considering the

merits, which would further complicate any review of

the questions presented (see Feder v. Electronic Data

Systems Corp., 248 F. App’x. 579, 581 (5th Cir. 2007)).

D. Cy pres-only settlements are now extremely rare.

Even apart from the unique circumstances here,

full cy pres settlements are “exceedingly rare”—there

have been “likely fewer than 20, ever.” Br. for Professor W. Rubenstein as Amicus Curiae in Support of Respondents 6, Frank v. Gaos, 139 S. Ct. 1041 (2019) (No.

17-961). One “extensive review of several thousand

class action cases” revealed only “18 cases in which a

federal court has ever approved full cy pres settlements.” Id. at 12; id. at 1a-2a (listing cases).

Further, cy pres-only settlements have grown

rarer over the past decade, as members of this Court

have registered concerns about undue reliance on cy

pres. See Marek v. Lane, 571 U.S. at 1005 (2013) (Roberts, C.J., statement respecting denial of certiorari);

Gaos, 139 S. Ct. at 1048 (Thomas, J. dissenting) (opining that a “cy pres-only” settlement whereby the class

“received no settlement fund, no meaningful injunctive relief, and no other benefit whatsoever” should

not have been approved). These concerns have been

noted by the bench, bar, and other commentators. See,

e.g., App. 36a, App. 74a-75a; Katherine Cienkus, Note,

Privacy Class Action Settlement Trends: Industry

Practice or Improper Incentives?, Rev. of Litig., Spring

2021, at 1, 19 (“Since 2012, no privacy settlement in

the data set achieved entirely cy pres relief for the

class”).

Moreover, many cases that might previously have

led to cy pres settlements have instead been dismissed

based on Gaos (139 S. Ct. at 1046) and Spokeo, Inc. v.

Robins, 578 U.S. 330 (2016), which clarified that Article III requires federal courts to weed out putative

class actions seeking statutory damages in the absence of actual harm. See, e.g., Trichell v. Midland

Credit Mgmt., Inc., 964 F.3d 990, 1005 (11th Cir. 2020)

(no Article III standing for violation of Fair Debt Collection Practices Act); Kamal v. J. Crew Grp., Inc., 918

F.3d 102, 117 (3d Cir. 2019) (no Article III standing

for violation of the Fair and Accurate Credit Transactions Act of 2003).

All of this confirms that the concerns about cy pres

settlements expressed by members of this Court have

been heeded. That such a settlement was used and

approved here—along with meaningful injunctive relief—simply reflects the undisputed, unusual circumstances of this case, in which identifying individual

class members (especially after the passage of so

much time) was “not feasible.” App. 2a, 20a. In sum,

Petitioner is mistaken that “the urgency of guidance

from this Court is undiminished” since Gaos. Pet. 16.

Empirical data confirm the Ninth Circuit’s observation that settlements such as this are the “exception,

not the rule.” App. 15a.

III. The decision below is correct.

Review is also unwarranted because the Ninth Circuit’s decision is correct. The district court did not

abuse its discretion in approving a cy pres settlement

tailored to the unique circumstances of this case. App.

2a.

A. Petitioner offers no credible method for

distributing funds to class members.

Lowery does not challenge—and has no basis to

challenge—the district court’s factual “finding,” affirmed by the court below, “that it was not feasible to

distribute funds directly to class members.” App. 2a;

see App. 17a-19a (affirming the district court’s finding); see also App. 69a-76a. Nor does he dispute that

replicating the process used by the special master to

assess the named plaintiffs’ claims “would be prohibitively costly and time-consuming.” App. 20a. Such a

process would have resulted in nearly all settlement

funds going to claims administrators rather than individual class members. See D. Ct. Dkt. 186 at 23

(stating that plaintiffs’ half of the Special Master’s

fees was “$487,476.05”); D. Ct. Dkt. 121-1 at 3 (ordering that the fees be split “50/50”).

Lowery’s cure-all is self-identification. Pet. 9 (class

members “could self-identify as he did”). That might

be feasible in other cases—say, where members of the

class can swear that they purchased a particular product, or bought a particular stock, during the class period. E.g., Pearson, 772 F.3d at 786 (“consumers who

have purchased Rexall’s glucosamine pills”); Baby

Prods., 708 F.3d at 170 n.4 (purchasers of “certain

baby products”); Easysaver Rewards, 906 F.3d at 753

(consumers “enrolled in the rewards program”);

BankAmerica, 775 F.3d at 1062 (“shareholders”). But

it could not work here. Lowery does not (and cannot)

dispute that “‘[t]he only evidence’ of class membership”

is “not in the class member’s possession’ or readily accessible to the claims administrator.” App. 20a. That

means “self-identification would be pure speculation”

—it would amount to giving away money to people

who simply have no way of knowing whether they are

class members or have standing to recover. Ibid.

That is not a feasible alternative in a case like this.

Cf. TransUnion LLC v. Ramirez, 141 S. Ct. 2190,

2204-2205 (2021) (“Every class member must have Article III standing in order to recover individual damages.”); Drazen v. Pinto, 41 F.4th 1354 (11th Cir. 2022)

(applying the same principle to settlement classes).

B. The settlement is “fair, reasonable, and

adequate.”

Nor is there any reason to doubt the conclusions of

the courts below that the settlement’s award of cy pres

and injunctive relief, viewed in context and as a whole,

was “fair, reasonable, and adequate.” App. 23a-26a;