Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 1 of 23

THE HONORABLE JOHN C. COUGHENOUR
UNITED STATES DISTRICT COURT
WESTERN DISTRICT OF WASHINGTON
AT SEATTLE

1
2
3

FEDERAL TRADE COMMISSION,

4

Plaintiff,

5
v.

6

AMAZON.COM, INC.,

7

Defendant.

8

CASE NO. C14-1038-JCC
REDACTED ORDER GRANTING
AMAZON’S MOTION FOR
PARTIAL SUMMARY JUDGMENT
AND GRANTING THE FTC’S
MOTION FOR SUMMARY
JUDGMENT

9
This matter comes before the Court on Amazon’s Motion for Partial Summary Judgment

10
11
12
13
14

(Dkt. Nos. 92 and 93), the FTC’s Response (Dkt. Nos. 156 and 158), Amazon’s Reply in Support
(Dkt. Nos. 188 and 189), the FTC’s Motion for Summary Judgment (Dkt. Nos. 109 and 138),
Amazon’s Opposition (Dkt. Nos. 167 and 179), and the FTC’s Reply (Dkt. Nos. 182 and 184).
The Court held oral argument on the FTC’s Motion on Thursday, April 21, 2016.
Having thoroughly considered the parties’ briefing, argument, and the relevant record, the

15
16
17
18
19
20
21
22
23
24
25
26

Court hereby GRANTS Amazon’s Motion with respect to injunctive relief, and GRANTS the
FTC’s Motion with respect to liability. The Court ORDERS further briefing with respect to the
appropriate remedy.

I.

BACKGROUND

A.

In-App Purchases
Defendant Amazon.com, Inc. (“Amazon”) operates an Appstore in which customers can

view and download apps to use on Android mobile devices or Kindle Fire tablets. (Dkt. No. 1 at
3.) Apps take many forms, but include functions that allow users to read books, play games,
stream movies, check weather, and organize files. (Dkt. No. 1 at 3; Dkt. No. 15 at 2.) Apps may
be free or come at a cost to download and install. (Dkt. No. 15 at 3.) Certain user activities
within some apps also come with monetary charges, starting at $0.99 and ranging up to $99.99.
REDACTED ORDER GRANTING AMAZON’S
MOTION FOR PARTIAL SUMMARY
JUDGMENT AND GRANTING THE FTC’S
MOTION FOR SUMMARY JUDGMENT
PAGE - 1

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 2 of 23

1 (Id.) These charges are known as “in-app purchases.” (Id.) Amazon started charging customers
2 for “in-app purchases” or “IAPs” in November 2011. (Id.)
3

While the app developers set the price for apps and in-app purchases, Amazon retains

4 30% of the revenue from every in-app sale. (Dkt. No. 1 at 4; Dkt. No. 15 at 3.)
5

Many apps geared towards children, and likely to be used by children, offer in-app

6 purchases. (Dkt. No. 15 at 2–3, 7.) For example, a child may be prompted to use or acquire
7 seemingly-fictitious currency, including a “boatload of doughnuts, a can of stars, and bars of
8 gold,” but in reality the child is making an in-app purchase using real money. (Dkt. No. 100 at
9 11.) In fact, in developing its Kindle Fire tablet, Amazon identified “soccer parents” as a key
10 target customer base, referring to them as “low-hanging fruit.” (Dkt. No. 121 at 8; see also Dkt.
11 No. 122 at 3.)
12

Moreover, the evidence demonstrates that Amazon was aware that many customers did

13 not understand in-app purchases when they were first implemented. In a confidential document
14 regarding Amazon’s marketing plan for launching in-app purchases, the company acknowledged
15 that “‘IAP’ isn’t a concept widely known by customers.” (Dkt. No. 120 at 5.) And, despite its
16 assertion that “[c]ustomers are not looking for apps based on how much they cost,” the company
17 was aware that customers’ top searches in selecting apps indicate that customers were seeking
18 free apps to use. (Id. at n. 2; five of the top searches included the word “free.”) Amazon was
19 aware that in many instances, the person initiating the in-app purchase was a child: in a
20 document discussing company strategy to promote increases in in-app purchasing, Amazon
21 acknowledged “the disconnect between the account owner (e.g., parent) and the app user (e.g.,
22 child).” (Dkt. No. 118 at 7.) On February 11, 2011, Amazon Appstore Director Aaron Rubenson
23 forwarded an article discussing scrutiny that Apple faced for enabling children to accrue in-app
24 purchases to Johanna May, who was working on designing the IAP user experience. (Dkt. No.
25 127 at 158.) Rubenson said to May, “We need to think this through for our IAP implementation.”
26 (Dkt. No. 115 at 10.)

PAGE - 2

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 3 of 23

1

When Amazon’s Appstore first implemented in-app purchases in November 2011, the

2 default setting did not require account holder approval, by entry of a password or any other
3 means, prior to completion of an in-app purchase. (Dkt. No. 127 at 160, 180.) Only if a customer
4 had previously enabled parental controls would the IAP require entry of a pin code or password.
5 (Dkt. No. 127 at 161.) As one Amazon employee explained, “If a parent gives the device to a
6 child and the child is playing the game and they – they just decide, you know, outside of the
7 parent’s view or whatever, to purchase, they could do that.” (Dkt. No. 181 at 304.)
8

In the Amazon Appstore, individual apps were displayed using what is called a “detail

9 page,” containing information relating to the app. (Dkt. No. 15 at ¶ 14; Dkt. No. 1 at ¶ 14.) When
10 in-app purchases were launched in November 2011, the detail page looked like this:
11
12
13
14
15
16
17
18
19
20
21

(See Dkt. No. 100 at 3.) A button below the name of the app in the upper-left corner

22 included the price of the app download: either “FREE” or a set dollar amount. (Id.) The price
23 button itself did not acknowledge the existence of in-app charges. (Id.) Underneath the header,
24 “Description” was a long note providing more information about the app. While the “Note”
25 included information about the presence of in-app purchases, a user often had to scroll down
26 “below the fold” to read it:

PAGE - 3

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 4 of 23

1
2
3
4
5
6
7
8
9
10
11
(Id. at 4–5.) Within the Note is language, about two-thirds of the way towards the bottom,

12
13
14
15
16
17
18
19
20
21
22
23
24
25

beginning, “PLEASE NOTE.” Prior to October 2012, the insertion of this language, warning
users that the app enables users to “purchase digital content using actual money,” was inserted
manually and in some instances omitted altogether. (Dkt. No. 127 at 299–300.) After making an
in-app purchase, Amazon customers were always notified via an e-mail. (See, e.g., Dkt. No. 181
at 305.)
B.

Fallout from In-App Purchases Made By Children
Amazon has received many complaints from adults who were surprised to find

themselves charged for in-app purchases made by children. By December 2011, Aaron Rubenson
referred to the amount of customer complaints as “near house on fire.” (Dkt. No. 115 at 19.)
Rubenson also referred to “accidental purchasing by kids” as one of two issues the company
needed to solve. (Id.) Rubenson additionally stated that “we’re clearly causing problems for a
large percentage of our customers.” (Id. at 14.) In the first two months of the Amazon Appstore
offering IAPs, Amazon received 11,110 customer service contacts requesting refunds for IAPs.

26

PAGE - 4

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 5 of 23

1 (Dkt. No. 122 at 16.) The “primary root of the contacts was accidental purchases due to parent
2 control issues.” (Id. at 14.)
3

In March 2012, Amazon introduced a password prompt feature for in-app charges of $20

4 or more. (See Dkt. No. 1 at ¶ 20, Dkt. No. 15 at ¶ 20.) This initial step did not include charges
5 below $20 or charges that, in combination, exceeded $20. (Dkt. No. 127 at 169–170; Dkt. No.
6 115 at 13.)
7

In August 2012, the FTC notified Amazon that it was investigating its in-app billing

8 practices. (Dkt. No. 158 at 7; Dkt. No. 215 at 3; Dkt. No. 216 at 2.) In February 2013, in
9 response to this investigation and high refund rates, Amazon began to require password prompts
10 more frequently, though not consistently. (Dkt. No. 15 at ¶ 21.) Passwords became required
11 when: a purchase over $20 was initiated, a second IAP purchase attempt was made within five
12 minutes of a first, and when parental controls were enabled. (Dkt. No. 124 at 15.) And once a
13 password was entered, in-app purchases were often then authorized for the next sixty minutes.
14 (Dkt. No. 15 at ¶ 21.) The fact that one password entry authorized in-app purchases for an
15 additional window of time was not explained during the process. (Dkt. No. 127 at 117.) When
16 asked in a Rule 30(b)(6) deposition whether the rules regarding when password prompts
17 appeared was likely “transparent to consumers,” Aaron Rubenson said, “they are not.” (Dkt. No.
18 127 at 126.)
19

In October 2012, Amazon released software entitled Kindle FreeTime, which allowed

20 parents to control tablet usage by children in a variety of ways. (Dkt. No. 95 at 7.) Within Kindle
21 FreeTime, in-app purchasing was disabled. (Id.) FreeTime was available on Kindle Fire tablets
22 sold after September 2012. (Id.)
23

In March 2013, Amazon developed a list of “High-Risk Apps,” those that had a refund

24 rate of greater than 15 percent of its revenue, and those that were targeted at children. (Dkt. No.
25 127 at 114–115.) In-app purchases in High-Risk Apps then required entry of a password. (Dkt.
26 No. 124 at 15.)

PAGE - 5

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 6 of 23

1

In May 2013, Amazon added a password requirement for all first-time in-app purchases

2 on Kindle Fire tablets. (Dkt. No. 95 at 8.) The prompt is pictured below:
3
4
5
6
7
8
9
10
11
12

(Dkt. No. 97 at 103.) The prompt refers to authorization of a singular “in-app purchase.”

13

In June 2013, Amazon changed the configuration of the AppStore so that the words “In-

14 App Purchasing” would appear on an app’s description page:
15
16
17
18
19
20
21
22
23
24
25
26

PAGE - 6

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 7 of 23

1

(Dkt. No. 200 at 6.) The words, “In-App Purchasing” are smaller than the remainder of

2 the text on the screen and in the same font and color. Notably, though the words “In-App
3 Purchasing” represented a clickable hyperlink which users could click on to learn more about in4 app charges, the lettering is not presented in such a way as to make that obvious, such as setting
5 the words in a different color or underlining them.
6

In June, after the FTC announced a settlement with Apple regarding its in-app purchases,

7 Amazon refined its password prompt for first-time in-app purchases to be consistent with the
8 terms of the Apple settlement. (Dkt. No. 95 at 9.) This new prompt enabled customers to select
9 whether they would like to require a password for future IAPs:
10
11
12
13
14
15
16
17
18
19
20
21
22
23

To date, Kindle devices of the “First Generation,” for which software updates are no

24 longer available, enable customers to make in-app purchases of $1 or less without authorization
25 via entry of a password. (Dkt. No. 160 at 4–6.)
26 C.

The Total Damage

PAGE - 7

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 8 of 23

1

The FTC has calculated its estimate of consumer damages due to unauthorized billing on

2 in-app purchases. As discussed below, Amazon objects to the inclusion of the declarations of
3 Julie Miller, the FTC employee who summarized and assessed the damages data as untimely
4 and/or speculative. Amazon’s objections are unwarranted.
5

To reach its estimate, the FTC summarized transactional and aggregate data produced by

6 Amazon pertaining to “High-Risk” Apps. In so doing, The FTC omitted apps with the category
7 “Casino,” and added other apps that “would have qualified as High-Risk” before April 2013
8 (“High-Risk Non-Casino apps”). (Dkt. No. 110 at 2.) Julie Miller, a lead FTC data analyst,
9 calculated the total in-app purchase revenue and refund amounts for seven different categories:
10 (1) orders of $20 or more in High-Risk Non-Casino apps from the earliest date available to
11 March 25, 2012, 1 (2) orders of $19.99 and below in High-Risk Non-Casino apps from the
12 earliest date available to February 5, 2013, (3) orders of $19.99 and below in High-Risk Non13 Casino apps from February 6–April 30, 2013 excluding those on the “Otter” device, (4) orders of
14 $19.99 and below in High-Risk Non-Casino apps from May 1–July 30, 2013 excluding those on
15 the Otter device, (5) orders of $19.98 and below in High-Risk Non-Casino apps from July 31,
16 2013–June 3, 2014 excluding those on the Otter device, (6) orders of $19.99 and below in High17 Risk Non-Casino apps from February 6–October 9, 2013 on the Otter device, and (7) orders of
18 $0.99 and below in High-Risk Non-Casino apps from October 10, 2013 to the latest date
19 available on the Otter device. (Id.) These categories were selected in order to omit authorized
20 charges. This calculation gave Ms. Miller a total of charges made without authorization by
21 password. Ms. Miller calculated $86,575,321.38 in revenue and also found that $10,060,646.48
22 was provided in refunds. (Dkt. No. 110 at 3.)
23

Ms. Miller then calculated an “unauthorized charge rate,” the rate at which users failed to

24 properly enter a password in initiating an in-app purchase as a percentage of the overall total of
25
26

1

As discussed above, Amazon began to require a password entry for all purchases of $20 or more in
March 2012. (Dkt. No. 1 at ¶ 20, Dkt. No. 15 at ¶ 20.)

PAGE - 8

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 9 of 23

1 password entry attempts. (Dkt. No. 110 at 4.) The FTC argues that “[t]he rate at which users
2 faced with such [password entry] prompts in a kids’ app failed to enter the correct password and
3 abandoned the charge attempt . . . is a reasonable proxy for the rate at which children would
4 incur an in-app charge without consent when a password entry was not required.” (Dkt. No. 182
5 at 18.) Ms. Miller identified the unauthorized charge rate as 42.03%. (Id.)
6

By applying the unauthorized charge rate to the total un-refunded revenue, and revising

7 in light of charges that were later deemed to have been authorized, the FTC estimates the total
8 customer injury as $26,242,025.30. (Dkt. No. 183 at 2.)
9 D.
10

The Present Litigation
The Federal Trade Commission (“FTC”) brings suit against Amazon, alleging that the

11 billing of parents and other account holders for in-app purchases incurred by children “without
12 having obtained the account holders’ express informed consent” is unlawful under Section 5 of
13 the FTC Act, 15 U.S.C. § 45(n). (Dkt. No. 1 at 11.) The FTC argues, on summary judgment, that
14 there is no remaining genuine dispute of material fact as to whether Amazon’s billing practices
15 for in-app purchases violated the FTC Act and seeks monetary and injunctive relief. (Dkt. No.
16 109.) Amazon also moves for partial summary judgment, solely over the FTC’s request for
17 injunctive relief. (Dkt. Nos. 92 and 93.)
18

II.

DISCUSSION

19
A.

Standard of Review

20
Pursuant to Rule 56 of the Federal Rules of Civil Procedure, “[t]he court shall grant
21
summary judgment if the movant shows that there is no genuine dispute as to any material fact
22
and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). In making such
23
a determination, the Court must view the facts and justifiable inferences to be drawn therefrom in
24
the light most favorable to the nonmoving party. Anderson v. Liberty Lobby, Inc., 477 U.S. 242,
25
255 (1986). Once a motion for summary judgment is properly made and supported, the opposing
26

PAGE - 9

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 10 of 23

1 party “must come forward with ‘specific facts showing that there is a genuine issue for trial.’”
2 Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587 (1986) (quoting Fed. R.
3 Civ. P. 56(e)). Material facts are those that may affect the outcome of the case, and a dispute
4 about a material fact is genuine if there is sufficient evidence for a reasonable jury to return a
5 verdict for the non-moving party. Anderson, 477 U.S. at 248–49. Ultimately, summary judgment
6 is appropriate against a party who “fails to make a showing sufficient to establish the existence
7 of an element essential to that party’s case, and on which that party will bear the burden of proof
8 at trial.” Celotex Corp. v. Catrett, 477 U.S. 317, 324 (1986).
9

Contrary to Amazon’s assertion, the material facts on record are not in meaningful

10 dispute. Amazon’s dispute pertains to whether the facts in the present case constitute an unfair
11 practice under Section 5 of the FTC Act.
12

B.

13

The Court first turns to Amazon’s Motion for Partial Summary Judgment. In its

Injunctive Relief

14 Complaint, the FTC seeks permanent injunctive relief, asserting that, absent such relief, Amazon
15 “is likely to continue to injure consumers, reap unjust enrichment, and harm the public interest.”
16 (Dkt. No. 1 at 11.) Such injunctive relief is authorized under Section 13(b) of the FTC Act. 15
17 U.S.C. § 53(b) (“in proper cases the Commission may seek, and after proper proof, the court may
18 issue, a permanent injunction.”). The injunction sought would subject Amazon to government
19 oversight for twenty years. (Dkt. No. 188 at 4.)
20

A court may permanently enjoin defendants from violating the FTC Act where there

21 exists “some cognizable danger of recurring violation.” FTC v. Gill, 71 F. Supp. 2d 1030, 1047
22 (C.D. Cal. 1999), aff’d, 265 F.3d 944 (9th Cir. 2001) (citing United States v. W.T. Grant, 345
23 U.S. 629, 633 (1953)). “The determination that such danger exists must be based on appropriate
24 findings supported by the record.” United States v. Laerdal Mfg. Corp., 73 F.3d 852, 854 (9th
25 Cir. 1995) (internal citation omitted).
26

Past violations of the FTC Act do not justify the imposition of a permanent injunction.

PAGE - 10

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 11 of 23

1 United States v. ACB Sales & Serv., Inc., 683 F. Supp. 734, 741 (D. Ariz. 1987) (citing FTC v.
2 Evans Products Co., 775 F.2d 1084, 1087 (9th Cir.1985)). However, the Court may consider past
3 conduct in determining the likelihood of a future, recurring violation. FTC v. Sharp, 782 F. Supp.
4 1445, 1454 (D. Nev. 1991) (citing C.FTC v. Co Petro Marketing, 502 F.Supp. 806, 818
5 (C.D.Cal.1980), aff’d, 680 F.2d 573 (9th Cir.1981)). “In drawing the inference from past
6 violations that future violations may occur, the Court should look at the ‘totality of
7 circumstances, and factors suggesting that the infraction might not have been an isolated
8 occurrence are always relevant.’” Id.
9

The FTC correctly asserts that “[m]ere voluntary cessation of allegedly illegal conduct

10 does not moot a case . . .” FTC v. Affordable Media, LLC, 179 F.3d 1228, 1238 (9th Cir. 1999)
11 (quoting United States v. Concentrated Phosphate Export Ass’n, Inc., 393 U.S. 199, 203 (1968)).
12 However, the Court’s inquiry upon Amazon’s motion for partial summary judgment hinges upon
13 whether the FTC has established, with evidence, a cognizable danger of a recurring violation: not
14 whether Amazon has met a mootness burden. See, e.g., Sheely v. MRI Radiology Network, P.A.,
15 505 F.3d 1173, 1182 n. 10 (11th Cir. 2007) (“To be sure, the analysis of whether a case is moot
16 overlaps with the analysis of whether a permanent injunction is appropriate on the merits because
17 both are concerned with the likelihood of future unlawful conduct. But the two inquiries are
18 strikingly different.”).
19

While permanent injunctions are often awarded in cases where liability under the FTC

20 Act is determined, Amazon correctly distinguishes those cases from the facts of this case. For
21 example, in FTC v. Gill, the Ninth Circuit upheld a permanent injunction where defendants
22 engaged in continuous, fraudulent practices and were deemed likely to reoffend based on the
23 “systemic nature” of their misrepresentations. 265 F.3d 944, 950, 957 (9th Cir. 2001). Other
24 cases in which a permanent injunction has been entered involved deceptive, ongoing practices.
25 See FTC v. Loewen, 2013 WL 5816420, at *7 (W.D. Wash. Oct. 29, 2013); FTC v. Commerce
26 Planet, Inc., 878 F. Supp. 2d 1048, 1086 (C.D. Cal 2012), aff’d in part and reversed on other

PAGE - 11

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 12 of 23

1 grounds, 815 F.3d 593 (9th Cir. 2016); FTC v. Inc21.com Corp., 745 F. Supp. 2d 975, 1010
2 (N.D. Cal. 2010).
3

The only potential ongoing violation of the FTC Act alleged is the fact that Amazon

4 customers are still billed for in-app purchases under $1 without authorization on First Generation
5 Kindle devices. The First Generation Kindle has not been sold since August 2012. (Dkt. No. 94
6 at 7.) The First Generation Kindle no longer receives software updates. (Id.) And the total in-app
7 purchases for 99-cent items on the First Generation Kindle are, all told, very low. For each
8 month since August 2015, the total refunds for such purchases have been less than $1,000. (Dkt.
9 No. 168 at 28.) Similarly, the total number of users making in-app purchases for 99-cents on
10 First Generation Kindles has been steadily declining. (Id.)
11

While unauthorized billing of customers, even for small purchases, constitutes part of the

12 substantial harm that Amazon caused customers, and for which monetary damages should be
13 assessed, the Court does not find this to represent a cognizable danger of a recurring violation.
14 Accordingly, the Court finds that injunctive relief is not warranted here, and GRANTS
15 Amazon’s Motion for Partial Summary Judgment (Dkt. Nos. 92 and 93.)
16

C.

Liability Under the FTC Act

17

The Court now turns to the FTC’s Motion for Summary Judgment and Amazon’s

18 opposition. The FTC presents strong support for its entitlement to judgment as a matter of law. In
19 opposition, Amazon argues that (1) the FTC is applying the wrong legal test for unfair business
20 practices, (2) the witness employed by the FTC to calculate money damages was not timely
21 disclosed and therefore her testimony should not be admitted or considered, and (3) its business
22 practices around in-app purchases did not violate Section 5.
23

1.

Proper Test for Liability Under the FTC Act

24

Section 5 of the FTC Act prohibits “unfair or deceptive acts or practices in or affecting

25 commerce.” 15 U.S.C. § 45(a)(1). Acts or practices are considered unfair if (1) they cause or are
26 likely to cause substantial injury to consumers, (2) the injury is not reasonably avoidable by

PAGE - 12

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 13 of 23

1 consumers, and (3) the injury is not outweighed by any countervailing benefits to consumers or
2 competition. 15 U.S.C. § 45(n).
3

Amazon classifies the three-part test for unfair business practices as a “necessary but not

4 sufficient prerequisite[] for unfairness.” (Dkt. No. 179 at 10.) However, Amazon does not point
5 the Court towards any concrete additional factors to incorporate into its review for unfairness.
6 Rather, in its effort to urge the Court to adopt a more searching inquiry under 15 U.S.C. § 45(n),
7 Amazon cites to a Third Circuit case in which the defendant suggested additional requirements
8 for a practice to be deemed “unfair,” such as unscrupulous or unethical behavior, and the court
9 declined to adopt those additional requirements. FTC v. Wyndham, 799 F.3d 235, 244 (3rd Cir.
10 2015).
11

While accusing the FTC of adopting an expanded version of the FTC Act, it is Amazon

12 that advocates for a new, more searching, definition of “unfair” practices. (See Dkt. No. 179 at
13 1.) The three-part test for whether a practice is “unfair” under the FTC Act, found in the statute
14 itself, is followed without embellishment by courts in this Circuit. See FTC v. Neovi, Inc., 604
15 F.3d 1150, 1153 (9th Cir. 2010); Davis v. HSBC Bank Nevada, N.A., 691 F.3d 1152, 1169 (9th
16 Cir. 2012). The Court follows suit in its analysis below.
17

2.

18

As another preliminary matter, Amazon objects to the testimony of Ms. Miller, who

Ms. Miller’s Declaration/Testimony

19 calculated the damages estimates, claiming that (1) she was not timely disclosed as a witness and
20 (2) that her statements are based on inadmissible conjecture. (Dkt. No. 179 at 16–21.)
21
22

a.

Timeliness of Disclosure

First, Amazon argues that Ms. Miller was disclosed as a witness after the discovery cut-

23 off. (Dkt. No. 179 at 17; Dkt. No. 172 at 2.) The calculated damages estimate was disclosed to
24 Amazon on January 7, 2016. (Dkt. No. 172 at 2, 26–27.) The discovery cut-off date was
25 originally September 28, 2015 (Dkt. No. 23), but after considering several discovery disputes,
26

PAGE - 13

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 14 of 23

1 the Court extended it to December 23, 2015. 2 (Dkt. No. 77.)
2

However, the FTC has made its intentions to seek monetary relief known since the

3 beginning of this case. In September 2014, in its initial disclosures, the FTC identified monetary
4 relief as “all charges consumers incurred in connection with [Amazon’s] unfair billing practices,
5 less any such charges that [Amazon] has already refunded.” (Dkt. No. 172 at 33.) On November
6 6, 2014, the FTC requested, via interrogatories, the data upon which the damages calculations
7 would be based. (Id.) Amazon objected and did not provide the information. (Id.) This request
8 was eventually the subject of a motion to compel (Dkt. No. 24) which the Court granted (Dkt.
9 No. 48.) On August 3, 2015, the Court ordered Amazon to provide the charge data by August 18,
10 2015. (Dkt. No. 48 at 5.) Amazon did not provide this data until nearly two months later on
11 October 14, 2015. (Dkt. No. 172 at 34.)
12

A party must make its initial disclosures “based on the information then reasonably

13 available to it,” including those pertaining to the calculation of relief. Fed. R. Civ. P. 26(a)(1)(E).
14 Moreover, “a party is not expected to provide a calculation of damages which depends on
15 information in the possession of another party or person.” United States ex rel. Parikh v.
16 Premera Blue Cross, 2006 WL 2927699, at *1 (W.D. Wash. Oct. 11, 2006) (quoting Fed. R. Civ.
17 P. 26(a) advisory committee’s note). Not only did the FTC make its intent to calculate damages
18 roughly using Ms. Miller’s methods well known, but it was Amazon’s own delay in turning over
19 the data that contributed to the timing of the disclosure of a more specific damages estimate. In
20 short, the Court does not exclude Ms. Miller’s declaration for a lack of timeliness. The two-week
21 “delay” in disclosure of the FTC’s damages estimate was substantially justified.
22

Finally, the FTC classifies Ms. Miller’s declaration as a summary of data provided by

23 Amazon, permissible under Fed. R. Evid. 1006. (Dkt. No. 184 at 20.) This is a fair assessment,
24 given that Ms. Miller does not offer additional facts or opinion evidence but simply ran
25
26

2

While the Court extended the discovery cutoff pursuant to several motions, it did not specify that the
extension of the cut-off was limited in any way.

PAGE - 14

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 15 of 23

1 calculations from data that Amazon provided. (See Dkt. Nos. 99, 157, 159, 183.)
2

b.

Ms. Miller’s Methods

3

Next, Amazon argues that Ms. Miller’s estimate is so “fundamentally flawed” as to not

4 be able to support a finding of substantial injury. (Dkt. No. 179 at 18.) In so arguing, Amazon
5 primarily takes issue with Ms. Miller’s calculation of an “Unauthorized charge rate.” (Id.) In
6 dividing the number of password entry “failures” and dividing that by the total number of
7 password prompts presented, the FTC argues that it identified a “reasonable proxy for the rate at
8 which children would incur an in-app charge without consent . . when password entry was not
9 required.” (Dkt. No. 184 at 18.) Amazon asserts that this rate calculation “assumes that every
10 single password failure was an attempt by a child that would otherwise have been a completed
11 in-app purchase.” (Dkt. No. 179 at 18.) This point is well taken: many password “failures” could
12 have occurred because the user got distracted, changed his or her mind, or simply could not
13 remember their password. However, it is reasonable to assume that of the group of users faced
14 with a password prompt who ultimately failed to provide a password, many were children who,
15 absent a password prompt, would have gone on to complete an in-app purchase.
16

The FTC’s burden is to “reasonably approximate” monetary relief. FTC v. Inc12.com

17 Corp, 475 F. App’x 106, 110 (9th Cir. 2012); see also FTC v. Febre, 128 F.3d 530, 535 (7th Cir.
18 1997) (“The Commission must show that its calculations reasonably approximated the amount of
19 customers’ net losses, and then the burden shifts to the defendants to show that those figures
20 were inaccurate.”) (citing SEC v. Lorin, 76 F.3d 458, 462 (2nd Cir.1996); United States
21 Department of Housing & Urban Development v. Cost Control Marketing & Sales Mgmt. of
22 Virginia, Inc., 64 F.3d 920, 927 (4th Cir.1995); FTC v. Direct Mktg. Concepts, Inc., 648 F. Supp.
23 2d 202,. 214 (D. Mass. 2009), aff’d, 624 F.3d 1 (1st Cir. 2010); FTC v. Verity Intern., Ltd., 443
24 F.3d 48, 69 (2nd Cir. 2006).
25

Considering that the FTC took reasonable measures to determine the rate of unauthorized

26 charges and applied that rate to actual charges that occurred without a password prompt, this is a

PAGE - 15

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 16 of 23

1 reasonable approximation based on the available data. However, the Court concludes that the
2 unauthorized charge rate of 42% is too inflated to be used in determining final money damages.
3 As discussed below, further briefing will assist the Court in determining a more appropriate
4 money damages award to enter in this case.
5

3.

Liability Under 15 U.S.C. § 45(n)

6

Turning now to the three-part test for an “unfair practice” under 15 U.S.C. § 45(n), the

7 Court finds that all three portions of the test are satisfied by the FTC’s evidence. The Court
8 reviews Amazon’s billing practices around in-app purchases to determine whether (1) they
9 caused (or were likely to cause) substantial injury to consumers, (2) that injury was not
10 reasonably avoidable by consumers, and (3) the injury was not outweighed by any countervailing
11 benefits to consumers or competition. 15 U.S.C. § 45(n).
12

a.

13

An act or practice may cause substantial injury either by doing “small harm to a large

Substantial Injury

14 number of people, or if it raises a significant risk of concrete harm.” FTC v. Neovi, Inc., 604 F.3d
15 1150, 1157 (9th Cir. 2010). Consumer injury can occur in “a variety of ways.” Id. at 1156. While
16 courts should look to any deception on the part of businesses, “the absence of deceit is not
17 dispositive.” Id. Nor is actual knowledge on the part of the consumer a requirement to establish
18 substantial harm. Id. Injury can be shown where consumers are “injured by a practice for which
19 they did not bargain.” Neovi, 604 F.3d at 1157.
20

Courts have repeatedly held that billing customers without permission causes injury for

21 the purposes of asserting a claim under Section 5 of the FTC Act. See, e.g., Neovi, 604 F.3d. at
22 1153; FTC v. Ideal Fin. Solutions, Inc., 2014 WL 2565688, at *5 (D. Nev. June 5, 2014); FTC v.
23 Commerce Planet, Inc., 878 F. Supp. 2d 1048, 1078 (C.D. Cal. 2012); FTC v. Inc21.com Corp.,
24 745 F. Supp. 2d 975, 1004 (N.D. Cal. 2010), aff’d, 475 F. App’x 106 (9th Cir. 2012); FTC v.
25 Crescent Publ’g Group, Inc., 129 F. Supp. 2d 311, 322 (S.D.N.Y. 2001); FTC v. J.K.
26 Publications, Inc., 99 F. Supp. 2d 1176, 1191–1192 (C.D. Cal. 2000); FTC v. Willms, 2011 WL

PAGE - 16

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 17 of 23

1 4103542, at *9 (W.D. Wash. Sept. 13, 2011); FTC v. Kennedy, 574 F. Supp. 2d 714, 719–720
2 (S.D. Tex. 2008).
3

The millions of dollars billed to Amazon customers without a mechanism for consent, the

4 thousands of customers complaining about unauthorized charges, and the time spent seeking
5 refunds for those charges, all demonstrate substantial injury. Amazon argues that because of its
6 liberal practices around providing refunds, its customers were not injured. Amazon’s argument
7 conflates complaints with the total universe of injury. However, given the design of the Appstore
8 and procedures around in-app purchases, it is reasonable to conclude that many customers were
9 never aware that they had made an in-app purchase. Moreover, regardless of its reputation for
10 customer service, it is Amazon’s stated policy that in-app purchases are final and non11 refundable, likely discouraging much of its customer base from attempting to seek refunds in the
12 first place. (See Dkt. No. 127 at 275.) (“Yeah, that’s the – that’s our official policy, is digital
13 content’s not refundable.”)
14

Finally, the time spent pursuing those refunds constitutes additional injury to Amazon’s

15 customers. Neovi, 598 F. Supp. 2d 1104 at 1115 (noting that “harm need not be monetary to
16 qualify as injury” and finding that time consumers spent contesting unauthorized checks and
17 “attempting to get their money back” contributed to substantial injury).
18

b.

Reasonably Avoidable

19

An injury is reasonably avoidable under Section 5 of the FTC Act if the consumer could

20 have made a “free and informed choice” to avoid it. Neovi, 604 F.3d at 1158. Amazon contends
21 that its customers could have mitigated damages either before an in-app purchase through
22 parental controls, or afterwards by pursuing a refund.
23

An injury is reasonably avoidable if consumers “have reason to anticipate the impending

24 harm and the means to avoid it,” or if consumers are aware of, and are reasonably capable of
25 pursuing potential avenues toward mitigating the injury after the fact. Davis v. HSBC Bank
26 Nevada, N.A., 691 F.3d 1152, 1168–69 (9th Cir. 2012) (citing Orkin Exterminating Co., Inc. v.

PAGE - 17

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 18 of 23

1 FTC, 849 F.2d 1354, 1365–66 (11th Cir.1988); Neovi, 604 F.3d at 1158).
2

Amazon argues that the harm to consumers was “reasonably avoidable” because (1) upon

3 first linking their mobile device to their Amazon account a password entry was required, (2) the
4 Appstore “Terms of Use” mentioned in-app purchasing, (3) in-app purchasing is mentioned on
5 the app-description pages, (4) parental controls could always be activated, and (5) complaining
6 customers often received refunds for their in-app purchases.
7

Amazon argues that a reasonable consumer would preemptively “identify and sidestep a

8 potential injury.” (Dkt. No. 179 at 29.) However, evidence establishes that, when in-app
9 purchases were introduced, most customers were unaware of their existence. (Dkt. No. 120 at 5.)
10 Accordingly, it is unreasonable to expect customers to be familiar with the potential to accrue in11 app purchases while using apps labeled as “FREE.” Many of Amazon’s arguments improperly
12 assume a familiarity with in-app purchases on the part of consumers. For example, Amazon cites
13 to a case determining that a “reasonable Amazon customer is accustomed to online shopping,”
14 but online shopping and spending real currency while obtaining virtual items in a game are
15 completely different user activities. (Dkt. No. 179 at 29) (citing Multi Time Mach., Inc. v.
16 Amazon.com, Inc., 804 F.3d 930, 939 (9th Cir. 2015)). In other words, while entering a password
17 linking her Amazon account to a new device, a reasonable consumer unaware of the possibility
18 of in-app purchases would not assume she was authorizing unforeseen charges. Nor would a
19 reasonable consumer have seen any connection between IAPs and parental controls until
20 Amazon changed its Appstore interface in June 2013.
21

Amazon directs the Court to Davis v. HSBC Bank Nevada, N.A., in which the Ninth

22 Circuit held that a consumer’s injury was “certainly avoidable” because the advertisement for the
23 credit card he applied for contained the disclaimer, “other restrictions may apply,” and such a
24 disclaimer “would have motivated a reasonable consumer to consult the terms and conditions.”
25 691 F.3d 1152, 1169 (9th Cir. 2012). Additionally, the Davis Court pointed to the credit card
26 application process in which a “boldface and oversized font” alerted the consumer to the credit

PAGE - 18

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 19 of 23

1 card’s terms and conditions and instructed him to “read the notice below carefully.” Id. The text
2 alerting the customer to read the terms and conditions of the card was twice the size as the
3 remaining text on the page. Id. at 1158. Finally, the consumer in Davis had to affirmatively
4 check a box agreeing to the terms and conditions, and certifying that he read them. Id. at 1169.
5

The notice that Amazon customers were given regarding in-app purchases is distinct from

6 the notices in Davis. From 2011 until June 2013, the only warning about in-app purchases that
7 customers would see during the download process on the Appstore was towards the bottom of a
8 long “description” note that a user would have to scroll down to see, in the same size and color
9 font as the rest of the text:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

Moreover, until Amazon began to introduce password prompts for in-app purchases in
March 2012, no affirmative assent to the charges was required, like the check-box in Davis. And
the password prompts introduced, asking customers to “confirm in-app purchase,” did not make
it clear that a single password entry served to authorize multiple IAPs within a certain timeframe.
Even in June 2013 when the Appstore was reconfigured, the notice that an app included “in-app
purchases” was not conspicuous: the text was in a smaller sized font of the same color as the rest

PAGE - 19

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 20 of 23

1 of the text on the page, and though it was a hyperlink to more information, this was not obvious
2 from the way the text was presented:
3
4
5
6
7
8
9

Finally, Amazon’s argument that customers could reasonably avoid their injury by

10 seeking a refund fails for four reasons. First, the refund rate of 17%, while high by Amazon’s
11 standards (see Dkt. No. 127 at 168), does not account for the entirety of customer injury. (See
12 also id. at 122, referring to 13% as a “high refund rate”). For example, the majority of in-app
13 purchases by customers were for $2.99 or less, and so many reasonable customers would not
14 have thought it worth the time to seek a refund for a relatively minor charge. (Dkt. No. 118 at 23;
15 Dkt. No. 127 at 257.)
16

Second, it was Amazon’s stated policy not to provide refunds for in-app purchases and so

17 many customers would have reasonably believed such recourse was not available to them. (Dkt.
18 No. 114 at 11.) Nothing on Amazon’s website states that in-app charges are refundable. (Dkt.
19 No. 102 at 54.) The confirmation email Amazon sends to consumers following an in-app charge
20 is consistent with this policy: it does not provide any information about whether refunds for in21 app charges are available or how to obtain one. (Dkt. No. 127 at 138; Dkt. No. 125 at 41–43.)
22

Third, the time customers spent seeking refunds actually constitutes additional injury to

23 them. Neovi, 598 F. Supp. 2d 1104 at 1115 (noting that “harm need not be monetary to qualify as
24 injury” and finding that time consumers spent contesting unauthorized checks and “attempting to
25 get their money back” contributed to substantial injury). Fourth, 1,573 known customers who
26 sought refunds did not receive them. (Dkt. No. 167 at 21.)

PAGE - 20

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 21 of 23

1

c.

Countervailing Benefits

2

Finally, Amazon’s billing practices around in-app purchases did not benefit consumers or

3 competition. Amazon’s argument to the contrary is twofold: (1) “consumers prefer a seamless,
4 efficient mobile experience,” essentially, that failing to require a password was a benefit, and (2)
5 that the general interest in innovation constitutes a benefit to consumers. (Dkt. No. 179 at 37–
6 39.)
7

First, even accepting as true the notion that consumers prefer a seamless and efficient

8 experience, the “benefit” of ensuring a streamlined experience is not incompatible with the
9 practice of affirmatively seeking a customer’s authorized consent to a charge. In fact, a clear and
10 conspicuous disclaimer regarding in-app purchases and request for authorization on the front-end
11 of a customer’s process could actually prove to better inform customers about their risk of
12 accruing in-app purchases and be more seamless than the somewhat unpredictable password
13 prompt formulas rolled out by Amazon. Moreover, as the FTC points out, Amazon has not
14 provided evidence of any customers who reported being upset or harmed by the existence of a
15 password prompt.
16

Amazon’s second argument about stifling innovation is too vague, and unsupported by

17 any evidence, to create a genuine issue with respect to this cost-benefit analysis.
18

The cost–benefit prong of the unfairness test is “easily satisfied” where, as here, “a

19 practice produces clear adverse consequences for consumers that are not accompanied by an
20 increase in services or benefits to consumers or by benefits to competition.” FTC v. J.K. Publ’ns,
21 Inc., 99 F. Supp. 2d 1176, 1201 (C.D. Cal. 2000) (internal citation omitted); see also In re In-app
22 Purchase Litig., 855 F. Supp. 2d 1030, 1041 (N.D. Cal. 2012).
23

In summary, the FTC has met its burden to demonstrate Amazon’s liability under Section

24 5 of the FTC Act. In this respect, its motion for summary judgment is GRANTED.
25

D.

26

While, as discussed above, the general methods used by the FTC to reasonably

PAGE - 21

Damages

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 22 of 23

1 approximate the damages to consumers by unauthorized in-app charges serve as a fair starting
2 place, the Court finds that the unauthorized charge rate of 42% is too high. The Court has
3 received Amazon’s “Adjustments to the FTC’s Estimates of Injury and Monetary Relief” (Dkt.
4 No. 221 at 2) and invites further briefing on the issue of the scope of appropriate monetary relief.
5

The Court determines that the scope of Amazon’s unfair billing practices pertains to all

6 in-app charges made by account users without express, informed authorization. The Court
7 concludes that Amazon’s injurious practices lasted up until users were clearly informed both
8 about the existence of in-app purchases and the scope of their consent by virtue of the revised in9 app purchase prompt on June 3, 2014. The Court finds that continuous unauthorized purchases
10 on First Generation devices, not provided the benefit of updated password prompts, constitutes
11 additional relevant injury.
12

The parties are hereby ORDERED to submit briefing with respect to the appropriate

13 methodologies and final amount of monetary relief as follows:
14

1.

Defendant Amazon is to submit a more detailed brief explaining all
of the methodologies for which it advocates in its Offer of Proof
and attached declaration, including (1) which apps to include in
determining the scope of injury, (2) the basis for excluding
purchases made by customers who had previously received
refunds, and (3) the basis for excluding purchases as described in
paragraphs (f) and (g) of Jonathan Werner’s declaration. (Dkt. No.
220-1.) This brief, not to exceed 16 pages, will be due to the Court
on or before May 27, 2016.

2.

Plaintiff FTC shall be afforded an opportunity to respond to
Amazon’s brief, in a response not to exceed 16 pages, due to the
Court on or before June 17, 2016.

3.

The parties are to notify the Court if they wish to hold oral
argument on this issue, and if so, include proposed time limits for
the hearing.

15
16
17
18
19
20
21
22
23
24
25
26

III.

CONCLUSION
For the foregoing reasons, Amazon’s motion for partial summary judgment (Dkt. Nos. 92

and 93) is GRANTED and injunctive relief is denied. However, with respect to liability, the

PAGE - 22

 Case 2:14-cv-01038-JCC Document 225 Filed 04/26/16 Page 23 of 23

1 FTC’s Motion for Summary Judgment (Dkt. Nos. 109 and 138) is GRANTED. Judgment is
2 hereby entered in the FTC’s favor. Following consideration of additional briefing and possibly
3 oral argument, the Court will make a determination as to the appropriate remedy. Amazon’s
4 currently-pending Motions in Limine (Dkt. No. 209) are hereby TERMINATED as moot.
5

DATED this 26th day of April 2016.

6

A

7
8
9

John C. Coughenour
UNITED STATES DISTRICT JUDGE

10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

PAGE - 23