Lisa Kim, individually and on behalf of all others similarly situated v. Rich Allison

FOR PUBLICATION
UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT
LISA KIM, individually and on behalf
of all others similarly situated,
Plaintiff-Appellee,
v.
R ICH ALLISON; S TEVE FRYE,
Objectors-Appellants,
v.
TINDER , INC ., a Delaware
corporation; M ATCH GROUP , LLC, a
Delaware limited liability company;
M ATCH GROUP , INC ., a Delaware
corporation,
Defendants-Appellees.
No. 19-55807
D.C. No.
2:18-cv-3093-
JFW-AS
OPINION
Appeal from the United States District Court
for the Central District of California
John F. Walter, District Judge, Presiding
Argued and Submitted January 15, 2021
Pasadena, California
Filed August 17, 2021

-- 1 of 25 --

2 ALLISON V . TINDER
Before: Consuelo M. Callahan and Paul J. Watford, Circuit
Judges, and Jed S. Rakoff,* District Judge.
Opinion by Judge Rakoff;
Dissent by Judge Callahan
SUMMARY**
Class Settlement
The panel reversed the district court’s approval of a pre-
certification class settlement, vacated the district court’s
judgment and attorneys’ fees award, and remanded for the
district court to conduct the more probing inquiry required
for a pre-certification class settlement.
Plaintiff Lisa Kim brought suit against Tinder, Inc. in
federal court pursuant to the Class Action Fairness Act of
2005 (“CAFA”) for violations of California’s Unruh Civil
Rights Act and its unfair competition statute. Tinder
successfully compelled arbitration, and Kim and Tinder
reached a settlement, before class certification, that applied
to a putative class. Class members Rich Allison and Steve
Frye objected. The district court rejected the objections,
certified the class for settlement purposes, granted final
approval of the proposed settlement, and awarded Kim a
* The Honorable Jed S. Rakoff, United States District Judge for the
Southern District of New York, sitting by designation.
** This summary constitutes no part of the opinion of the court. It
has been prepared by court staff for the convenience of the reader.

-- 2 of 25 --

ALLISON V. TINDER 3
$5,000 incentive payment and her counsel $1.2 million in
attorneys’ fees.
Addressing the district court’s approval of the settlement
overall, the panel held that the district court correctly recited
the fairness factors under Fed. R. Civ. P. 23(e)(2), but that
the district court abused its discretion by underrating the
strength of the plaintiff’s case, overstating the settlement
value, and overlooking the suggestions of collusion present.
The panel held that independent of the district court’s
abuse of discretion in its overall evaluation of the settlement,
the approval of the attorneys’ fees was itself an abuse of
discretion. By adopting without any scrutiny the purported
value of the injunctive relief and failing to consider the likely
claims rate, the district court shirked its independent duty to
assess the value of the settlement
Judge Callahan dissented. She agreed with the majority
that the district court’s $24 million valuation of the
settlement agreement was to some degree overinflated, but
she dissented because the district court nevertheless
reasonably evaluated the settlement class’s relatively weak
claims. Because the settlement provided for fair, reasonable,
and adequate value for the release of the class’s claims, she
would affirm on the ground that the district court did not
abuse its discretion in approving the settlement. Judge
Callahan also disagreed with the majority’s opinion
discussion of whether the award of attorneys’ fees was an
abuse of discretion because the discussion was superfluous,
given the majority’s holding that the district court’s approval
of the settlement should vacated, and because the objectors
waived any challenge to the district court’s lodestar
calculations.

-- 3 of 25 --

4 ALLISON V . TINDER
COUNSEL
Danielle Leonard (argued) and Michael Rubin, Altshuler
Berzon LLP, San Francisco, California; Kimberly A.
Kralowec, Kralowec Law P.C., San Francisco, California;
Alfred G. Rava, Rava Law Firm, San Diego, California; for
Objectors-Appellants.
Adrian R. Bacon (argued) and Todd M. Friedman, Law
Offices of Todd M. Friedman P.C., Woodland Hills,
California; John P. Kristensen, Kristensen LLP, Los
Angeles, California; for Plaintiff-Appellee.
Donald R. Brown (argued), Robert H. Platt, and Benjamin
G. Shatz, Manatt Phelps & Phillips LLP, Los Angeles,
California, for Defendants-Appellees.
OPINION
RAKOFF, District Judge:
Beginning in 2015, the dating app Tinder began offering
reduced pricing for those under 30, later changed to those
under 29. In 2017, plaintiff Lisa Kim purchased a premium
version of the Tinder app, but because she was already in her
thirties, she paid more for her monthly subscription than
those in their twenties. Kim brought suit against Tinder in
federal district court pursuant to the Class Action Fairness
Act of 2005 (“CAFA”) for violations of California’s Unruh
Civil Rights Act and its unfair competition statute. Over
Kim’s opposition, Tinder successfully compelled
arbitration. After a daylong mediation session with a retired
judge, Kim and Tinder reached a settlement, before class
certification, that applied to a putative class.

-- 4 of 25 --

ALLISON V. TINDER 5
Specifically, the settlement class included all California-
based Tinder users who were at least 29 years old when they
subscribed to Tinder’s premium services and were charged
a higher price than younger subscribers. As part of the
settlement, Tinder agreed to eliminate age-based pricing in
California for new subscribers. Class members who
maintained or reactivated their Tinder accounts would
automatically receive 50 “Super Likes” (described below),
for which Tinder would ordinarily have charged $50.
Finally, class members who submitted a valid claim form
would also receive their choice of $25 in cash, 25 Super
Likes, or a one-month free subscription to the premium
Tinder service previously purchased.
Class members Rich Allison and Steve Frye, whose
attorneys represent the lead plaintiff in a competing age-
discrimination class action against Tinder in California state
court, were among six class members who objected to the
proposed settlement. These two objectors, in particular,
argued that Tinder offered too paltry a cash payout, as well
as Super Likes that premium subscribers did not need and
subscriptions that former subscribers did not want, all in
exchange for releasing valuable claims that had only been
strengthened by recent victories in related California actions.
Rejecting these objections, the district court certified the
class for settlement purposes, granted final approval of the
proposed settlement, and awarded Kim a $5,000 incentive
payment and her counsel $1.2 million in attorneys’ fees.
Allison and Frye now appeal.
We conclude that, while the district court correctly
recited the fairness factors under Fed. R. Civ. P. 23(e)(2), it
materially underrated the strength of the plaintiff’s claims,
substantially overstated the settlement’s worth, and failed to
take the required hard look at indicia of collusion, including

-- 5 of 25 --

6 ALLISON V . TINDER
a request for attorneys’ fees that dwarfed the anticipated
monetary payout to the class. We therefore reverse the
district court’s approval of the pre-certification class
settlement, vacate the judgment and attorneys’ fees award,
and remand for the district court to conduct the “more
probing inquiry” that a pre-certification class settlement
demands. See Hanlon v. Chrysler Corp., 150 F.3d 1011,
1026 (9th Cir. 1998).
BACKGROUND
Tinder is a dating app operated by Tinder, Inc., Match
Group, LLC, and Match Group, Inc. (collectively, “Tinder”)
that uses computer technology to match users with nearby
singles. Users “swipe right” on a dating profile to indicate
interest and left to signify a lack of it. If two users both swipe
right, the potential couple can message each other through
the app. To indicate heightened interest, users can send one
another a “Super Like.” Users receive one free Super Like
daily but can purchase additional Super Likes for $1 each.
In March 2015, Tinder launched Tinder Plus, an ad-free
premium service that offered new features: users could
swipe right unlimited times, change their minds about
matches initially rejected, see dating profiles of users in
other cities, and receive more than the one free “Super Like”
per day allotted to regular users. Tinder Plus operated on a
two-tiered pricing basis: subscribers thirty and under paid
$9.99 a month, and subscribers over thirty paid $19.99. In
March 2016, Tinder lowered the age cutoff for a reduced
subscription price from 30 years old to 29 years old. In 2017,
Tinder launched still another premium service, Tinder Gold,
which used the same two-tiered pricing scheme for
subscribers under and over 29.

-- 6 of 25 --

ALLISON V. TINDER 7
To create a Tinder account, a user must agree to the
Terms of Use. Since 2014, the Terms of Use have included
a ban on class actions and a binding arbitration provision for
all disputes arising from or relating to Tinder services that
cannot be brought in small claims court. In July 2015, Tinder
added a login screen disclosure (“the sign-in wrap
agreement”) informing users that continued use of the
service indicated consent to the Terms of Use.
Kim created her first Tinder account in October 2013 and
another in April 2015. She purchased a Tinder Plus
subscription on February 21, 2017, paying $19.99 as a user
over thirty. Two days later, after being presented with the
sign-in wrap agreement, she logged into the account.
On April 12, 2018, Kim sued Tinder, alleging age
discrimination in violation of the Unruh Civil Rights Act,
Cal. Civ. Code §§ 51 et seq. Kim later amended her
complaint to add a claim under California’s unfair
competition law. See Because the suit was a putative class
action, it was brought, pursuant to CAFA, in the U.S. District
Court for the Central District of California.
Tinder moved to compel arbitration, which Kim
opposed. Kim argued that her Unruh Act claim sought public
injunctive relief and that the arbitration agreement was
unenforceable to the extent it barred such relief. Kim also
sought discovery with respect to the evidence that she had
viewed and consented to the arbitration agreement. The
district court denied the discovery request as “vague.” The
court further determined that Kim consented to the sign-in
wrap agreement and that the agreement was enforceable, not
least because it still permitted Kim to seek injunctive relief
through arbitration. The court stayed the case and directed
Kim to arbitration.

-- 7 of 25 --

8 ALLISON V . TINDER
Kim appealed the district court’s arbitration ruling. But
while the appeal was pending, Kim and Tinder participated
in a full-day mediation with retired Judge Louis Meisinger
on November 29, 2018. The parties reached a settlement on
December 1, 2018 and entered an agreement memorializing
that settlement on December 31, 2018.
The Settlement Agreement
The settlement agreement defines the settlement class to
include “every California subscriber to Tinder Plus or Tinder
Gold during the Class Period who at the time of the
subscription was at least 29 years old and was charged a
higher rate than younger subscribers, except those who
choose to opt out of the Settlement Class.” The class
contains about 240,000 members. Under the agreement,
every class member who has or reactivates a Tinder account
will automatically receive 50 Super Likes, regardless of
whether the user files a claim. In addition, class members
who file a timely claim will receive their choice of:
“(1) $25.00 in cash; (2) 25 Super Likes (but only if the Class
Member has a current Tinder account); or (3) a one-month
subscription to Tinder Plus or Tinder Gold, depending on
which of those services the Class Member had previously
purchased (this option is not available to any Class Member
who has a current subscription to Tinder Plus or Tinder
Gold).” Finally, the settlement contained an injunctive
component. Defendants agreed to eliminate age-based
pricing for new subscribers in California, but “reserve[d] the
right to offer a youth discount to subscribers age 21 or
younger.”
As part of the settlement, Tinder further agreed not to
challenge an award of attorneys’ fees of $1.2 million plus
reasonable costs and expenses. Similarly, Tinder agreed not

-- 8 of 25 --

ALLISON V. TINDER 9
to oppose an incentive award of $5,000 to Kim “for services
performed in representing the Settlement Class.”
Objections
After the settlement was presented to the district court
for approval, Allison and Frye filed objections, arguing that
the plaintiff and class counsel were inadequate, that the
claims form was burdensome, that the risk to the class posed
by further litigation was low, and that the settlement was
collusive and of little value.
Allison and Frye also argued that another, almost-
identical class action lawsuit better demonstrated the value
of the class members’ claims: Candelore v. Tinder, Inc.,
228 Cal. Rptr. 3d 336 (2018), review denied (May 9, 2018).1
In Candelore, as here, the class action plaintiff alleged that
Tinder violated the Unruh Act by charging customers over
29 more than it charged younger customers for the same
service. Id. at 339. While Tinder initially moved successfully
to dismiss the complaint for failure to state a claim, id.
at 340, on appeal, Candelore secured a ruling that his
allegations did state a claim for age discrimination under the
Unruh Act—and that, if his allegations were true, Tinder’s
age-based distinction would not be justified by public policy
as a matter of law. See id. at 350. In the course of its decision,
the Candelore court “recognize[d] . . . that past cases,” like
Javorsky v. Western Athletic Clubs, Inc., 195 Cal. Rptr. 3d
706 (2015), “have embraced the notion that age may serve
as a reasonable proxy for income in upholding age-based
discounts against Unruh Act claims.” 228 Cal. Rptr. 3d
at 344. But those cases, the court reasoned, are “inconsistent
1 Allison’s and Frye’s attorneys also represented lead plaintiff Allan
Candelore in this lawsuit.

-- 9 of 25 --

10 ALLISON V . TINDER
with the ‘individual nature’ of the right secured by the
[Unruh] Act, which protects individuals from unequal
treatment based on generalizations about ‘a group’ to which
they belong.” Id. at 347.
Allison and Frye argued that Candelore showed the
relative weakness of Kim as a class representative. Unlike
Kim, Candelore was not compelled to arbitration, because
when Candelore signed up for Tinder Plus in March 2015,
the sign-up wrap agreement was not yet in place. Allison and
Frye also stressed that the appellate victory in Candelore
showed that the class members had meritorious Unruh Act
claims and significant leverage.
Preliminary Approval
Despite Allison’s and Frye’s objections, the district court
granted preliminary approval of the settlement agreement. In
its preliminary approval order, the district court recognized
that the settlement agreement contained a “clear sailing”
provision—that is, an agreement by Tinder not to oppose
plaintiff’s attorneys’ application for $1.2 million in
attorneys’ fees—and noted that such provisions “are
considered troubling and a possible sign of collusion.”
However, the court reasoned that the provision was unlikely
to indicate collusion here because “the attorneys’ fees
provision was not negotiated until after the parties had
agreed to the other settlement terms,” a neutral mediator
oversaw the fee negotiation, and the fee award
“represent[ed] . . . approximately 5 percent of the total
estimated value of the Settlement.” Id. Nevertheless, the
court promised that it would “closely scrutinize the
attorneys’ fees requested at the Final Fairness Hearing” to
confirm that collusion was unlikely. Id.

-- 10 of 25 --

ALLISON V. TINDER 11
Final Approval
At the final approval hearing, the district court
determined that the proposed settlement was “fair,
reasonable, and adequate” under Federal Rule of Civil
Procedure 23(e). The district court’s analysis focused on the
perceived weakness of Kim’s case, the supposed benefits the
settlement would provide to class members, and the low
number of class-member objections.
In the district court’s estimation, Kim’s case was weak.
Kim’s claim had been compelled to arbitration, and the court
found that “over 95 percent of the Class Members” could be
bound by similar arbitration agreements, which could
present a bar to ultimate class certification. The court also
determined that, while Candelore could help the class
members survive a motion to dismiss, “the [unfavorable]
holding of Javorsky [is] more compelling and more in line
with the weight of authority” on summary judgment or at
trial. The court also noted that other, as-of-yet unraised
defenses could end the class’s quest for relief, including
arguments related to choice-of-law and contractual limits on
liability.
At the same time, the court concluded that the agreement
“will provide direct and meaningful benefits to the
Settlement Class” to “a total value of $24 million.” This
value included $12 million worth of Super Likes disbursed
automatically, up to $6 million in “potential cash or cash-
equivalent benefits” distributed through the claims process,
and $6 million in injunctive relief. In regard to the latter, the
court noted that “Class Counsel estimates that the injunctive
relief negotiated on behalf of the Class and the public has a
value of at least $6 million,” but the court did not otherwise
evaluate why this was a reasonable value.

-- 11 of 25 --

12 ALLISON V . TINDER
Finally, the district court also found that the “reaction of
class members” favored the settlement, because of about
240,000 class members, only six had objected to the
settlement. The court found Allison’s and Frye’s allegations
of collusion unpersuasive, because the settlement offered a
significant benefit of $12 million in Super Likes to class
members automatically, those class members who preferred
a cash benefit could submit a claim form, the parties had
delayed negotiating attorneys’ fees in mediation until after
they had reached agreement on substantive settlement terms,
and the attorneys’ fees sought were proportionate to a
$24 million settlement value.
Turning to attorneys’ fees, per se, the court analyzed the
proposed attorneys’ fee award under the “percentage-of-
fund” and “lodestar” methods and concluded that the fee
award was reasonable under both. The court also found the
incentive award to Kim was reasonable because of “the time
and effort Plaintiff has devoted to this case” and because the
$5,000 amount was “presumptively reasonable.” (quoting
Faigman v. AT & T Mobility LLC, 2011 WL 672648, at *5
(N.D. Cal. Feb. 16, 2011)).
The district court then granted final approval of the class
action settlement and awarded $1.2 million in attorneys’ fees
and a $5,000 incentive payment to Kim.
DISCUSSION
In general, we review for abuse of discretion both a
district court’s grant of approval of a pre-certification class
action settlement and the court’s calculation of attorneys’
fees. Rodriguez v. W. Publ’g Corp., 563 F.3d 948, 963 (9th
Cir. 2009); In re Bluetooth Headset Prods. Liab., 654 F.3d
935, 940 (9th Cir. 2011). The factual findings underlying the

-- 12 of 25 --

ALLISON V. TINDER 13
fee award are reviewed for clear error. Bluetooth, 654 F.3d
at 940.
However, settlements that occur before class
certification are subject to “a high procedural standard.”
Allen v. Bedolla, 787 F.3d 1218, 1223 (9th Cir. 2015). The
district court must act as a fiduciary, protecting the interests
of absent class members by scrutinizing the settlement’s
fairness in light of well-established factors. Id. Accordingly,
“the district court must show it has explored
comprehensively all factors, and must give a reasoned
response to all non-frivolous objections.” Dennis v. Kellogg
Co., 697 F.3d 858, 864 (9th Cir. 2012) (internal citations and
quotation marks omitted). Reversal is warranted when the
settlement terms “contain convincing indications” that the
class representative and class counsel’s self-interest won out
over the class’s interest, and “the district court was wrong in
concluding otherwise.” Staton v. Boeing Co., 327 F.3d 938,
960 (9th Cir. 2003).
I
We turn first to the approval of the settlement overall.
Rule 23(e) authorizes district courts to approve class action
settlements when they are “fair, reasonable, and adequate.”
See Fed. R. Civ. P. 23(e)(2). In this Circuit, a district court
examining whether a proposed settlement comports with
Rule 23(e)(2) is guided by the eight “Churchill factors,” viz.,
“(1) the strength of the plaintiff’s case; (2) the risk, expense,
complexity, and likely duration of further litigation; (3) the
risk of maintaining class action status throughout the trial;
(4) the amount offered in settlement; (5) the extent of
discovery completed and the stage of the proceedings;
(6) the experience and views of counsel; (7) the presence of
a governmental participant; and (8) the reaction of the class
members of the proposed settlement.” In re Bluetooth

-- 13 of 25 --

14 ALLISON V . TINDER
Headset Prods. Liab., 654 F.3d at 946 (citing Churchill Vill.
v. Gen. Elec., 361 F.3d 566 (9th Cir. 2004)); see also Hanlon
v. Chrysler Corp., 150 F.3d 1011, 1026 (9th Cir. 1998). Only
when the district court “explore[s] these factors
comprehensively” can the settlement award “survive
appellate review.” See In re Mego Fin. Corp. Sec. Litig.,
213 F.3d 454, 458 (9th Cir. 2000).
However, “consideration of these eight Churchill factors
alone is not enough to survive appellate review.” In re
Bluetooth Headset Prod. Liab. Litig., 654 F.3d at 946; see
also Briseño v. Henderson, 998 F.3d 1014, 1025–26 (9th Cir.
2021) (holding that the revised Rule 23(e) requires courts “to
go beyond our precedent” by applying the heightened
scrutiny set forth in Bluetooth to all class action settlements).
Rule 23(e)(2) also requires the court to consider “the terms
of any proposed award of attorney’s fees” and scrutinize the
settlement for evidence of collusion or conflicts of interest
before approving the settlement as fair. Briseño, 998 F.3d at
1024–25.
As noted, the district court’s determinations in these
respects are reviewed on appeal for abuse of discretion. But
even applying that deferential standard of review, we find
that the district court so underrated the strength of the
plaintiff’s case, so overstated the settlement value, and so
overlooked the suggestions of collusion present as to
collectively constitute an abuse of discretion.
First, the district court discounted the strength and value
of the class members’ claims because the court “f[ound] the
holding of Javorsky to be more compelling and more in line
with the weight of authority than the holding in Candelore.”
In so doing, the district court ignored the fact that the
settlement class members are also putative members of the
class in Candelore—and the settlement agreement therefore

-- 14 of 25 --

ALLISON V. TINDER 15
releases claims where Candelore is the law of the case,
regardless of whether the district court finds the opinion
persuasive.
The district court also substantially overstated the
settlement’s worth. Tinder’s agreement to eliminate age-
based pricing going forward applies only to new California-
based subscribers—a group that, by definition, does not
include the class members. Yet, the court accepted class
counsel’s unsupported representation that the injunctive
relief was worth $6 million to the class. We see no basis for
this conclusion.
Furthermore, the universal participation component, an
award of 50 Super Likes, is available only to class members
who maintain or reactivate their Tinder accounts. In
December 2018, approximately 44% of the class did not
have a Tinder account. For many obvious reasons, these
class members might not want to resume their relationship
with Tinder to receive the class benefit, e.g., because in the
intervening years, they had entered into a satisfactory
romantic relationship, or because their experience of age
discrimination soured them on Tinder entirely, or whatever.
Further still, because those class members who still have
Tinder Plus or Tinder Gold accounts already receive
150 Super Likes per month, the marginal value of 50 more
is low.
Further still, the district court grossly overstated the
value of the claims that Tinder would actually pay as being
$6 million. This was based on the extremely doubtful
assumption that all members of the class would not only file
a claim but also elect the $25 cash alternative. In reality,
based on the actual claims rate at the time of final approval
of 0.745%, Tinder stood to pay less than $45,000.

-- 15 of 25 --

16 ALLISON V . TINDER
Third, as for possible collusion, the district court did not
adequately scrutinize the combination of a clear-sailing
provision and an attorneys’ fee award that outstripped the
likely financial benefit to the class. Because these early, pre-
certification settlements are so open to abuse and so little
subject to scrutiny at the time by the district court, the court
is required to search for “subtle signs” that plaintiff’s
counsel has subordinated class relief to self-interest. In re
Bluetooth Headset Prods. Liab., 654 F.3d at 947; see also
Briseño, 998 F.3d at 1024–25 (explaining that although class
certification “does not cleanse all sins, especially when it
involves potential collusion over divvying up funds between
class counsel and the class,” “[t]he potential for collusion
reaches its apex pre-class certification”). Signs of collusion
can include: (1) a handsome fee award despite little to no
monetary distribution for the class, (2) a “clear sailing”
provision under which defendant agrees not to object to the
attorneys’ fees sought, and (3) an agreement that fees not
awarded will revert to the defendant, not the class fund.
Allen, 787 F.3d at 1224. The presence of these three signs is
not a death knell—but when they exist, “they require[] the
district court to examine them, . . . develop the record to
support its final approval decision,” and thereby “assure
itself that the fees awarded in the agreement were not
unreasonably high.” Id. (quoting In re Bluetooth Headset
Prods. Liab., 654 F.3d at 947).
The settlement agreement here contained a clear-sailing
provision. This Court has frequently stressed that “‘clear
sailing’ agreements on attorneys’ fees are important warning
signs of collusion.” Roes, 1–2 v. SFBSC Mgmt., LLC,
944 F.3d 1035, 1051 (9th Cir. 2019) (quoting Lane v.
Facebook, 696 F.3d 811, 832 (9th Cir. 2012)). When a
district court encounters such a provision, it must “peer into
the provision and scrutinize closely the relationship between

-- 16 of 25 --

ALLISON V. TINDER 17
attorneys’ fees and benefit to the class,” even when the
settlement has been negotiated “with a neutral mediator
before turning to fees.” In re Bluetooth Headset Prods.
Liab., 654 F.3d at 948.
Here, however, the district court did not develop the
record to support its conclusion that the attorneys’ fees were
proportionate to the value to the class. Instead, the court, as
noted above, calculated the settlement value based on a
totally unrealistic claims rate and an unexplained injunctive
relief valuation, and then found the attorneys’ fee award
proportional to that inflated settlement value.
The district court gave “deference to the mediation
proceedings and the judgment of the parties regarding the
reasonableness of fees,” but in so doing abdicated its
independent duty to see whether these actually excessive
attorneys’ fees evidenced collusion in the settlement. See
Allen, 787 F.3d at 1224 (quoting In re Bluetooth Headset
Prods. Liab., 654 F.3d at 947).
II
Independent of the district court’s abuse of discretion in
its overall evaluation of the settlement, the approval of the
attorneys’ fees was itself an abuse of discretion. The district
court approved a $1.2 million attorneys’ fee award, because
“the $1,200,000 in attorneys’ fees sought equates to
approximately 5 percent of the estimated $24,000,000 in
settlement benefit provided to the Class.” But, as already
noted, the $24,000,000 figure was greatly overstated.
This Circuit permits two methods of calculating
attorneys’ fee awards in class actions: (1) the “lodestar”
method and (2) the “percentage-of-recovery” method. In re
Hyundai & Kia Fuel Econ. Litig., 926 F.3d 539, 570 (9th

-- 17 of 25 --

18 ALLISON V . TINDER
Cir. 2019) (en banc). Under the lodestar method, the court
multiplies the number of hours the prevailing party
reasonably spent on litigation by a reasonable hourly rate to
determine a presumptively reasonable fee award. See
Yamada v. Nobel Biocare Holding AG, 825 F.3d 536, 546
(9th Cir. 2016). The court may then “adjust” the award “by
an appropriate positive or negative multiplier reflecting . . .
the quality of representation, the benefit obtained for the
class, the complexity and novelty of the issues presented,
and the risk of nonpayment.” In re Bluetooth Headset Prods.
Liab. Litig., 654 F.3d at 941–42 (quoting Hanlon, 150 F.3d
at 1029). Benefit to the class is the “[f]oremost”
consideration. Id. at 942. This method is especially
appropriate in class actions “where the relief sought—and
obtained—is . . . primarily injunctive.” Id. at 941.
The percentage-of-recovery approach may be used
“where the defendants provide monetary compensation to
the plaintiffs” and class benefit is easy to quantify. See In re
Hyundai, 926 F.3d at 570. Under this method, “the court
simply awards the attorneys a percentage of the fund
sufficient to provide class counsel with a reasonable fee.”
Hanlon, 150 F.3d at 1029. Injunctive relief is inherently
difficult to monetize. Yamada, 825 F.3d at 547. Thus, a
district court must exercise caution when using the value of
injunctive relief to determine proportional attorneys’ fees
and should generally avoid valuing hard-to-measure
injunctive relief altogether, “because of the danger that
parties will overestimate the value of injunctive relief in
order to inflate fees.” SFBSC Mgmt., LLC, 944 F.3d at 1055.
When a settlement includes injunctive relief among other
forms of relief, the court should (1) “explain[] why the value
of the injunctive relief’s benefits to individual class members
was readily quantifiable and worth [the estimated value]” or
(2) “exclude[] the injunctive relief from the valuation of the

-- 18 of 25 --

ALLISON V. TINDER 19
settlement and explain[] why attorneys’ fees . . . were
justified.” Id. at 1056.
As already noted, the settlement agreement between Kim
and Tinder provides injunctive relief that eliminates age-
based pricing in California for new subscribers only, none of
whom are members of the class. It does so by raising the
price for new subscribers under 29 to match the price
previously charged to those 29 and over, while allowing
existing under-29 subscribers to be grandfathered in at the
lower price. So it is hard to see how the court could credit
the parties’ assertion that this benefit was worth $6 million
to the class, when it appears to be a benefit of zero. Instead,
the district court justified its determination that injunctive
relief was worth approximately $6 million without reference
to anything other than class counsel’s estimation.
Furthermore, the court should have excluded the
injunctive relief estimate in its calculation of reasonable
attorneys’ fees under the percentage-of-recovery method.
See Staton, 327 F.3d at 946 (“[P]arties ordinarily may not
include an estimated value of undifferentiated injunctive
relief in the amount of an actual or putative common fund
for purposes of determining an award of attorneys’ fees.”).
As for the parties’ suggestion that the award that Tinder
would pay in cash under the settlement should be valued at
$6 million, this again was totally without substance. When
assessing whether the fee award is disproportionate to the
class benefit the district court should have considered the
amount of anticipated monetary relief based on the timely
submitted claims already made. See Allen, 787 F.3d at 1224
n.4. Instead, the court assumed a 100% claims rate when less
than 1% of class members had submitted claims by the date
of final approval. In dollar terms, this meant that the actual

-- 19 of 25 --

20 ALLISON V . TINDER
amount that Tinder would pay in cash was not $6 million,
but something closer to $45,000.
By adopting without any scrutiny the purported value of
the injunctive relief and failing to consider the likely claims
rate, the district court shirked its independent duty to assess
the value of the settlement. See, e.g., SFBSC Mgmt., LLC,
944 F.3d at 1052–55 (reversing where the district court
included an unjustified valuation of injunctive relief in a
lodestar cross-check and did not “explain why the [claims-
made relief] should nevertheless be valued at its $1 million
maximum”).
CONCLUSION
For the foregoing reasons, we reverse the district court’s
approval of the settlement, finding that the district court did
not subject the settlement agreement to a “a higher standard
of fairness and a more probing inquiry than may normally be
required under Rule 23(e).” Dennis, 697 F.3d at 864 (internal
quotation marks omitted). Further, the district court abused
its discretion by awarding attorneys’ fees with reference to
an unsupported estimation of the value of injunctive relief
and a wholly inflated assessment of benefit to the class.
Accordingly, we vacate the judgment and remand the case
to the district court for further proceedings consistent with
this Opinion.
REVERSED AND REMANDED.

-- 20 of 25 --

ALLISON V. TINDER 21
CALLAHAN, Circuit Judge, dissenting:
While I agree with the majority that the district court’s
$24 million valuation of the settlement agreement is to some
degree overinflated, I respectfully dissent because the
district court nevertheless reasonably evaluated the
settlement class’s relatively weak claims. As the settlement
provides fair, reasonable, and adequate value for the release
of the class’s claims, even when substantially discounted to
reflect the problems identified by the majority, I would
affirm on the ground that the district court did not abuse its
discretion in approving the settlement.
The majority’s criticism of the district court’s evaluation
of the class’s claims is based entirely on an incomplete
reading of the district court’s application of Candelore v.
Tinder, Inc., 19 Cal. App. 5th 1138 (2018). Contrary to the
majority’s accusation, the district court did not “ignore[]”
Candelore’s impact. While the district court noted that it
found the rationale of Candelore less “compelling” than
other conflicting California court decisions, it did not
dismiss Candelore as irrelevant. Indeed, the district court
emphasized the mediator’s awareness of the decision and the
parties’ extensive discussion of its impact at mediation as
evidence that its impact was properly considered by the
parties in reaching their agreement.
The district court also accurately noted that Candelore
merely held that the plaintiff’s claim under the Unruh Civil
Rights Act could survive a demurrer; the case did not
establish that Candelore was necessarily entitled to relief.
The district court further reasonably considered that the
objectors had not identified any cases in which damages had
ever been awarded for comparable Unruh Act violations.
And as Tinder notes, Candelore leaves room for the
company to defend the lawfulness of its age-based pricing

-- 21 of 25 --

22 ALLISON V . TINDER
tiers in a subsequent summary judgment motion if it can
establish that its discounts for those under 30 years of age
“are independently justified by compelling ‘social policy
considerations as evidenced by legislative enactments.’”
Candelore, 19 Cal. App. 5th at 1149 (quoting Koire v. Metro
Car Wash, 40 Cal. 3d 24, 38 (1985)). While Candelore may
be the law of the case for the settlement class’s claims, the
district court acted reasonably and within its discretion in
determining that Candelore does not compel the conclusion
that those claims are particularly valuable.
Critically, the majority also fails to address other barriers
to the class’s recovery that the district court identified and
relied upon in discounting the value of the class’s claims.
For example, Tinder has required its users to agree to terms
of service which provide that Texas law governs disputes
regarding the company’s services since at least July 31,
2015. If that provision is enforceable, the Unruh Act would
not even apply to individuals who used Tinder’s services
during most of the class period, which began on March 2,
2015. The terms of service also include limitations on
Tinder’s liability and provisions compelling arbitration. The
impact of the arbitration provision in particular has already
proven to be a significant obstacle. The district court
determined that Kim (the named plaintiff here) was
compelled to arbitrate her claims against Tinder, a ruling that
is not currently before us but would have to be overturned
before Kim could even continue litigating the merits of her
claim. The district court reasonably determined that the
significant risks of failure on the merits, coupled with the
prospect of lengthy and costly litigation, lowered the value
of the class claims at issue.
There are, admittedly, good reasons to believe that the
district court’s $24 million settlement valuation was too

-- 22 of 25 --

ALLISON V. TINDER 23
high. But even disregarding the value the district court
attributed to the injunctive relief ($6 million), the value of
the cash benefits for class members who submitted a claim
($6 million), and the value of the automatic benefits (in the
form of free “Super Likes”) provided to inactive Tinder
users (approximately $6 million), the settlement still
provides approximately $6 million worth of automatic
benefits to current users. The significant obstacles the
district court identified that the class would have to
overcome to achieve certification and ultimately succeed on
the merits support the conclusion that the district court did
not abuse its discretion in approving the settlement, even at
this substantially reduced value.
Nor did the district court abuse its discretion in
determining that the $1.2 million in attorneys’ fees provided
for by the settlement agreement was proportionate to the
value received by the class (again, even if that value were to
be discounted to approximately $6 million). This is a far cry
from the situation recently presented in Briseño v.
Henderson, 998 F.3d 1014 (9th Cir. 2021), which held that
a nearly $7 million attorneys’ fee award was
disproportionate to the less than $1 million in value provided
to the settlement class. Id. at 1026. Here, the proposed fee
award represents only about 20 percent of the discounted
value of the settlement. See In re Bluetooth Headset Prods.
Liab. Litig., 654 F.3d 935, 942 (9th Cir. 2011) (noting that
“courts typically calculate 25% of the fund as the
‘benchmark’ for a reasonable fee award” in cases where
settlement “produces a common fund for the benefit of the
entire class”). Because the fees were proportionate to even
the discounted recovery, I would hold that the district court
did not abuse its discretion in analyzing the fairness of this
pre-certification settlement.

-- 23 of 25 --

24 ALLISON V . TINDER
Finally, I disagree with Part II of the majority’s opinion
discussing whether the district court’s award of attorneys’
fees was an abuse of discretion. This section is entirely
superfluous given the majority’s conclusion that the
approval of the settlement agreement should be vacated—in
the absence of an approved settlement agreement, class
counsel is obviously not entitled to the fees provided for by
that agreement.
Further, in determining whether a proposed fee award is
appropriate, a district court has discretion to apply “either
the lodestar method or the percentage-of-recovery method.”
In re Hyundai & Kia Fuel Econ. Litig., 926 F.3d 539, 570
(9th Cir. 2019) (en banc). Here, the district court held that
the fees were independently reasonable under both methods.
Even if the district court’s percentage-of-recovery analysis
were flawed,2 the objectors have not specifically challenged
any aspect of the district court’s lodestar analysis. The only
comment in the objectors’ opening brief on this point is a
footnote stating that they “raised other concerns with the fee
calculation below, including the inflated amount of time that
was submitted in light of the minimal amount of work
performed.” The objectors do not articulate what those
concerns were, explain how the lodestar calculation used
inflated amounts of time, or otherwise press this argument
on appeal. Accordingly, they have waived any challenge to
the district court’s lodestar calculations. Greenwood v.
F.A.A., 28 F.3d 971, 978 (9th Cir. 1994) (holding that an
2 It is not clear that the district court necessarily erred in this
analysis. Applying the percentage-of-recovery method, the fees were
likely in the reasonable range of about 20 percent of the settlement value
even if the settlement value is discounted to about $6 million to address
the problems identified in the majority’s opinion. See In re Bluetooth
Headset Prods. Liab. Litig., 654 F.3d at 942 (identifying “25% of the
fund as the ‘benchmark’ for a reasonable fee award”).

-- 24 of 25 --

ALLISON V. TINDER 25
argument not “specifically and distinctly” argued in the
appellant’s opening brief was waived); Hilao v. Est. of
Marcos, 103 F.3d 767, 778 n.4 (9th Cir. 1996) (“The
summary mention of an issue in a footnote, without
reasoning in support of the appellant’s argument, is
insufficient to raise the issue on appeal.”). If the panel had
affirmed the approval of the settlement and it became
necessary to reach this issue, I would uphold the award based
on the district court’s unchallenged lodestar calculations.
I fear that the majority’s embrace of the objectors’
exaggerated view of the class’s claims will ultimately result
in the class receiving far less than it would under the district
court’s reasonable judgment. For these reasons, I
respectfully dissent.

-- 25 of 25 --