IN RE FACEBOOK PPC ADVERTISING LITIGATION

---

---

ch-com-inc-v-los-angeles-cellular-tel-co#p18020 Cal. 4th 163, 180 (1999). Because the statute is written in the disjunctive, it applies separately to business practices that are (1) unlawful, (2) unfair, or (3) fraudulent. See Pastoria v. Nationwide Ins., 112 Cal. App. 4th 1490, 1496 (2003). See also Albillo v. Intermodal Container Services, Inc., 114 Cal. App. 4th 190, 206 (Cal. App. 2d Dist. 2003) (citations omitted) (To state a claim under the UCL, "[a] plaintiff must establish that the practice is either unlawful ( i.e., is forbidden by law), unfair ( i.e., harm to victim outweighs any benefit) or fraudulent ( i.e., is likely to deceive members of the public)."). Plaintiffs concede that they have not stated a claim under the UCL's unlawful prong. (Pls.' Opp'n at 7 n. 5.) However, they allege that Facebook has violated the UCL's unfair and fraud prongs by:

a) Representing that advertisers are billed based on the number of actual and valid clicks, yet charging and collecting . . . payments for all types of invalid clicks, including those referred to as click fraud[;]

b) Representing that [it] has adequate measures in place to ensure that it only reports and charges advertisers for valid clicks, when it does not;

c) Failing to disclose or adequately disclose that it does not have adequate measure in place to avoid reporting and charging for all types of invalid clicks; and

d) Failing to implement adequate measures to protect advertisers against charges for all types of invalid clicks, including those referred to as click fraud.

(SAC ¶ 99.) Plaintiffs' claim focuses on alleged misrepresentations and omissions made in Facebook's Help Center. Whether the allegations are framed under the UCL's fraud prong or its unfair prong appears to be unimportant, as the inquiries collapse together. If the statements in the Help Center are not likely to deceive, they cannot be "fraudulent" under the UCL, nor would they be "unfair," as non-deceptive statements would not appear to cause harm that would outweigh the benefit derived from the statements. The parties thus focus primarily on the fraud prong.

A. Whether the Court's previous rulings preclude the claim

Facebook argues that this Court's previous ruling precludes any UCL claim predicated on click fraud. In that ruling, the Court concluded that "it is difficult to see how Plaintiffs' allegations of `click fraud' can survive disclaimers that state unambiguously that Facebook will not be liable for `click fraud.'" In re Facebook, 2010 WL 3341062, at *8. However, the Court clearly was discussing third-party click fraud in connection with Plaintiffs' breach of contract claim, a theory of liability not asserted in the SAC. Facebook argues that because its representations in the Help Center do not render the disclaimers ambiguous, the representations cannot be fraudulent under the UCL. However, the inquiries are distinct. In connection with the breach of contract claim, the issue was whether "the language of a contract, in the light of all the circumstances, `is fairly susceptible of either one of the two interpretations contended for . . .'" Pacific Gas E. Co. v. G. W. Thomas Drayage etc. Co., 69 Cal. 2d 33, 39-40 (1968) (citation omitted and emphasis added). Here, while the Court must consider the unambiguous disclaimer as a relevant circumstance, the focus is on whether the language in the Help Center was "likely to deceive members of the public."

Facebook also cites the Court's earlier observation that one aspect of Plaintiffs' claim under the UCL's unfair prong "boils down to a dispute over contractual interpretation." Id. at *11. That observation concerned an allegation that nearly was identical to "allegation (d)," referenced above. Id.; (SAC ¶ 99). Plaintiffs framed this specific allegation under the UCL's unfair prong, contending that the harm caused by Facebook's failure to implement adequate measures to protect advertisers against charges for all types of invalid clicks was not outweighed by the benefits. The Court held that this aspect of the claim rose and fell with Plaintiffs' breach of contract claim because it alleged essentially that Facebook was liable merely for charging Plaintiffs for click fraud. However, at oral argument, Plaintiffs reframed their claim, contending that Facebook's failure to implement adequate measures to protect against click fraud was "fraudulent" under the UCL. They argue that while it may have disclaimed liability in the event that click fraud did occur, Facebook nonetheless represented in the Help Center that it would protect Plaintiffs from click fraud and thus is liable for failing to implement that protection. Under this theory, liability under the UCL, including liability premised on "allegation (d)," is distinct from the breach of contract claim, resting instead on an alleged misrepresentation outside of the contract.

The Court previously dismissed Plaintiffs' claims under the UCL's unfair and fraud prongs because Plaintiffs had not alleged reliance on the representations in the Help Center, as required by In re Tobacco II Cases, 46 Cal. 4th 298, 326 (2009). Id. The previous motions did not address the issue of whether the disclaimers in the contracts that shield Facebook from liability for charging for click fraud also shield it from liability for pre-contractual misrepresentations in the Help Center concerning its ability to prevent click fraud.

B. Whether compliance with the terms of an unambiguous contract precludes UCL liability for misrepresentations made prior to the contract's formation

The parties appear to agree that under certain circumstances, unambiguous contract language is insufficient to preclude UCL liability where one of the parties to the contract makes contradictory or misleading representations in order to obfuscate or obscure the actual terms of the contract. In In re First Alliance Mortgage Co., 471 F.3d 977 (9th Cir. 2006), the plaintiffs alleged that the defendant engaged in fraudulent practices to "induce" them into taking out subprime mortgage loans. While the annual percentage rate of the loan "was ultimately disclosed," "[l]oan officers were taught to deflect attention away from things that consumers might normally look at[,] and the loan sales presentation was conducted in such a way as to lead a consumer to disregard the high annual percentage rate. . . ." Id. at 985. The Ninth Circuit found that the "elaborate and detailed sales presentation . . . was unquestionably designed to obfuscate the true principal amount of the loan." Id. The defendant was found liable for fraud, notwithstanding that the terms of the mortgage eventually were disclosed. While In re First Alliance addressed common-law fraud rather than a claim under the UCL's fraud prong, the Court finds it instructive, as it illustrates that certain types of misrepresentations outside of a contract still may create liability even in the face of unambiguous contractual provisions. Accord, Janda v. T-Mobile, USA, Inc., No. C 05-03729 JSW, 2009 WL 667206, at *6-7 (N.D. Cal. March 13, 2009) (analyzing In re First Alliance during a discussion of whether advertising was misleading in the context of Cal. Bus. Prof. Code § 17500 despite clear contractual disclosures)

The California Supreme Court has held that subsequent to its amendment on November 2, 2004, the UCL "imposes an actual reliance requirement on plaintiffs prosecuting a private enforcement action under the . . . fraud prong." In re Tobacco, 46 Cal. 4th at 326. The disclaimer in this case is unambiguous with respect to third-party click fraud, and Plaintiffs must show that they were reasonable in relying on the representations in the Help Center even in light of those disclaimers. "[R]eliance is proved by showing that the defendant's misrepresentation or nondisclosure was `an immediate cause' of the plaintiff's injury-producing conduct. [Citation.] A plaintiff may establish that the defendant's misrepresentation is an `immediate cause' of the plaintiff's conduct by showing that in its absence the plaintiff `in all reasonable probability' would not have engaged in the injury-producing conduct." Id. (internal quotation marks and citations omitted).

While it was a pre-2004 case, the plaintiffs in In re First Alliance reasonably relied on the representations of the loan officers even in the face of clear contractual provisions because the misrepresentations were "unquestionably designed to obfuscate the true principal amount of the loan." 471 F.3d at 985. The issue thus is whether under the circumstances in this case, a reasonable jury could find that Plaintiffs were reasonable in relying on the statements in the Help Center notwithstanding an unambiguous disclaimer in the click-through agreement providing that Facebook would not be liable for click-fraud.

Plaintiffs allege that "in all reasonable probability" they would not have entered into contracts that disclaimed liability for third-party click fraud absent Facebook's assurances in the Help Center that they would be protected from such fraud. Facebook represented that it "ha[s] a variety of measures in place to ensure that [it] only report[s] and charge[s] advertisers for legitimate clicks, and not clicks that come from automated programs, or clicks that may be repetitive, abusive, or otherwise inauthentic." (SAC Ex. B (emphasis added).) Plaintiffs contend that click fraud necessarily involves clicks that are repetitive, abusive, inauthentic, and generated by an automated program. They argue that the statements in the Help Center thus were likely to cause a reasonable consumer to believe that Facebook employed a filtering system that was adequate to protect them against click fraud, making them more willing to accept the disclaimers.

1. Case law

In Chern v. Bank of America, 15 Cal. 3d 866 (1976), the defendant had represented prior to the formation of a contract that the interest rate would be calculated on a "per annum" basis. The California Supreme Court observed that the public likely would assume that a "per annum" rate would be based on a 365-day calendar year. Id. at 876. However, under the actual terms of the contract, the rate was calculated on the basis of a 360-day year. Id. Because the contract unambiguously provided for the shorter period, the Court affirmed summary judgment against the plaintiff on a breach of contract claim. Id. at 874-75. However, it concluded that the bank's practice still could be viewed as deceptive, noting that the "[d]efendant offer[ed] no sound reason why it could not quote an accurate annual rate during its initial communication with its customers, rather than withhold[ing] such information until the customer ha[d] entered its premises." Id. at 876. Because Chern was decided before In re Tobacco and the 2004 amendments to the UCL, Facebook contends that it no longer is good law.

However, in a case decided after In re Tobacco, the Ninth Circuit similarly reversed the dismissal of a claim under the UCL's fraud prong while at the same time affirming the dismissal of a breach of contract claim. Rubio v. Capital One Bank, 613 F.3d 1195, 1204 (9th Cir. 2010). In that case, the plaintiff received a solicitation from a credit card company, indicating that the card had a "fixed" annual percentage rate ("APR") of 6.99%. The plaintiff responded to the solicitation and opened a credit card account by consenting to the card holder agreement. The card holder agreement permitted the defendant to amend or change the APR unilaterally. When the defendant increased the APR, the plaintiff filed suit, alleging inter alia that the defendant had breached the contract and violated both the Truth in Lending Act ("TILA") and the UCL.

The Ninth Circuit concluded that the plaintiff had stated a claim under TILA — which requires "clear and conspicuous" disclosure of certain terms for credit cards — because "[i]n the minds of a significant number of consumers, the term `fixed' means `unchangeable.'" Id. at 1202. The court also concluded that the plaintiff had stated a claim under the UCL's unlawful and unfair prongs. Id. at 1204-05. At the same time, it affirmed the dismissal of the plaintiff's breach of contract claim because the agreement itself unambiguously gave the defendant discretion to increase the APR, and the solicitation — which arguably contained statements to the contrary — did not constitute an enforceable contract. Id. at 1205. Finally, the court held that the plaintiff had stated a claim under the UCL's fraud prong because a reasonable consumer was likely to be deceived by the solicitation's representation of a "fixed" rate when that rate was subject to change under the terms of the actual card holder agreement.

Facebook argues that Rubio is distinguishable because the plaintiff's TILA claim constituted a separate statutory violation. However, that violation served as the predicate only for the plaintiff's claims under the UCL's unlawful and unfair prongs. Rubio's holding with respect to the UCL's fraud prong did not depend on the TILA violation, and the court observed expressly that the UCL's fraud prong and TILA are governed by the same test: "[a]s in TILA, this prong of the UCL [the fraud prong] is `governed by the reasonable consumer test' [, in which a] plaintiff may demonstrate a violation by `showing that reasonable members of the public are likely to be deceived.'" Id. at 1204 (quoting Williams v. Gerber Prods. Co., 552 F.3d 934, 938 (9th Cir. 2008)) (alternations in the original omitted).

Facebook argues that the instant case is more similar to Spiegler v. Home Depot U.S.A., Inc., 552 F. Supp. 2d 1036 (C.D. Cal. 2008), in which the court dismissed a UCL claim in light of an unambiguous contract. The plaintiffs in that case contracted with the defendant for the resurfacing of their kitchen cabinets. The defendant dispatched a "Sales Professional" to the plaintiffs' homes to create an estimate for the work. The defendant also dispatched a "Measurement Technician" to confirm the measurements of the "Sales Professional" and the needs of the plaintiffs. The estimates of the "Sale Professional" tended to be higher than those based on the measurements of the "Measurement Technicians." However, the plaintiffs were charged for the higher estimate and were not informed of the lower estimate. They claimed that this practice was fraudulent under the UCL. Spiegler held that the plaintiffs had not stated a claim because the defendant had charged the plaintiffs the price that was specified by the contracts to which they had agreed. Importantly, the "plaintiffs concede[d] that [the] defendants did not make any representations regarding how they established the price to be charged." Id. at 1047.

Facebook also points briefly to a number of other cases. In Janda v. T-Mobile, USA, Inc., No. C 05-03729 JSW, 2009 WL 667206 (N.D. Cal. March 13, 2009), the court dismissed both the plaintiffs' breach of contract claim and their claim under the UCL's fraud prong. The plaintiffs alleged that the defendant's advertising, contracts, and billing practices were misleading and constituted a "bait and switch." Id. at *4. However, the court concluded that the defendant had made unambiguous disclosures that were not obfuscated by other representations. In Daugherty v. American Honda Motor Co., Inc., 144 Cal. App. 4th 824, 838 (Cal. App. 2d Dist. 2006), the court "[could not] agree that a failure to disclose a fact one has no affirmative duty to disclose is `likely to deceive' anyone within the meaning of the UCL." Finally, Facebook reminds the Court of the general rule that the UCL's fraud prong "does not give the courts a general license to review the fairness of contracts." Samura v. Kaiser Foundation Health Plan, Inc., 17 Cal. App. 4th 1284, 1299 n. 6 (Cal. App. 1st Dist. 1993).

2. Whether Plaintiffs' reliance was reasonable

As discussed above, Facebook represented in the Help Center that it "ha[s] a variety of measures in place to ensure that [it] only report[s] and charge[s] advertisers for legitimate clicks, and not clicks that come from automated programs, or clicks that may be repetitive, abusive, or otherwise inauthentic." (SAC Ex. B.) Plaintiffs contend that members of the public reasonably could interpret this statement to mean that Facebook employs some means of "pre-filtering or filtering software . . . used throughout the industry to limit, and in many cases eliminate, the charges for both invalid clicks and . . . fraudulent clicks." (SAC ¶ 44.)

To the extent that Plaintiffs allege that Facebook is subject to UCL liability merely because its filtering system is insufficient or ineffective, such allegations fail to state a claim. It is true that Facebook represented that it would "ensure" against charges for a variety of clicks, and "ensure" can be defined as "[t]o make certain . . . the attainment of (a result)." Oxford English Dictionary (2d ed 1989). Arguably, Facebook could have avoided any degree of uncertainty by avoiding the word "ensure," representing instead that it would "defend against," "protect from," or "make every effort to prevent" fraudulent clicks. However, given Plaintiffs' own explicit allegation that no existing filtering system is capable of detecting all fraudulent clicks, (SAC ¶ 44), a reasonable member of the public could not view Facebook's representation as a guarantee that its filtering system always will be effective.

To the extent that Plaintiffs could allege that Facebook failed to employ any "filtering" system to protect against click fraud or knew but did not disclose that its system was subject to regular and frequent failure in excess of flaws inherent in such systems, they might state a claim. Although Plaintiffs do not contend that the statements in the Help Center obfuscated the terms of the contract, the statements obviously represent to the public that Facebook employs a system that at least attempts to minimize the risk of click fraud. Under these circumstances, Facebook could be liable for misrepresentation, notwithstanding the contractual disclaimers.

With respect to defects in Facebook's filtering system, RootZoo alleges that it was charged for "clicks from small towns . . . that are not indicative of the population and that would almost be statistically impossible given RootZoo's prior tracking history with Facebook advertisements." (SAC ¶ 62.) RootZoo informed Facebook of the discrepancy, but Facebook denied that it had charged for invalid clicks. ( Id. at ¶ 64.) Price alleges that he used two separate tracking programs, Google Analytics and Statcounter.com, to discover that sixty-six percent of the clicks for which he was charged consistently were "invalid clicks that had never even occurred." ( Id. at ¶¶ 71, 73.) He informed Facebook of the discrepancy, and while Facebook did offer a refund, it "[did] not even come close to compensating Price for the monies that he has been wrongfully charged. . . ." ( Id. at ¶ 75.) Fox used Google Analytics to monitor its traffic and discovered that approximately eleven percent of the clicks for which Facebook charged involved visitors that spent an average time of one second on "the website." ( Id. at ¶ 77.) Facebook again failed to provide adequate refunds. ( Id. at ¶ 79.)

While the allegations in the SAC may be sufficient at the pleading stage to support a claim that Facebook's filtering system suffers from serious defects, Plaintiffs' current allegations, even viewed in the light most favorable to them, are insufficient to support a reasonable inference that Facebook knew of the problems at the time that it made its representations in the Help Center with respect to the filtering systems. Price and Fox alerted Facebook of their concerns only after they had entered into their respective contracts. Plaintiffs allege that Facebook acknowledged the click-fraud issue after a series of complaints was aired on WickedFire.com, (SAC ¶¶ 47, 51), but that alleged acknowledgment occurred on June 21, 2009, again after each of the Plaintiffs had entered into their respective contract. While RootZoo alleges that in June 2008, it informed Facebook of a "material discrepancy" and an "almost . . . statistically impossible" number clicks from small towns, (SAC ¶¶ 57-58), allegations that Facebook knew but did not disclose serious flaws in its system are allegations that sound in fraud. Accordingly, the claims must be plead with particularity, Ciba-Geigy, 317 F.3d at 1103-04, and RootZoo does not provide more than conclusory allegations with respect to the materiality of the discrepancy or the "almost" impossibility of the number of clicks.

V. DISPOSITION

The motion will be granted with respect to a claim under the UCL's fraud prong based on allegations that Facebook employs an insufficient click-filtering system. Leave to amend is granted in a manner consistent with the foregoing discussion. Any amended complaint shall be filed within thirty (30) days of the date of this order.

IT IS SO ORDERED.

DATED: 12/15/10