THE UCL-NOW A MONEY BACK GUARANTEE?

By Michele Floyd¹

I. INTRODUCTION

The Ninth Circuit issued its opinion in Pulaski & Middleman, LLC v. Google, Inc.² on September 21, 2015. After adopting an expansive definition ofrestitution under the California Unfair Competition Law ("UCL"), Pulaski reversed the district court’s denial of certification on the ground that calculating restitution under the UCL posed no individualized questions. Underlying its reversal of the district court’s decision were three findings: (1) California law creates a "conclusive presumption" that a consumer is entitled to restitution once liability is established under the UCL thus making individualized proof of entitlement to restitution unnecessary; (2) the value received post-purchase is irrelevant to a restitution calculation because the proper measure is the diference between the amount the class member paid and the amount the class member would have paid had he or she known the truth; and, in any event, (3) individualized damage calculations cannot defeat certiication in the Ninth Circuit, even after Comcast Corp. v. Behrend.³

One week later, the California Court of Appeal affirmed the final judgment in In re Tobacco Cases II⁴ and affirmed the trial court’s refusal to award any restitution under the UCL. There, the court conirmed that UCL restitution must account for post-purchase value received, applying the definition formulated in In re Vioxx Class Cases.⁵ Tobacco II further entrenched the developing weight of authority holding that a full refund is not available under the UCL except in the rare circumstance where the product at issue has no intrinsic value.

Pulaski is inconsistent with Tobacco II in at least two respects. First, as Tobacco II illustrates, California law does not create a conclusive presumption that a restitution award is proper once liability is established. Second, the Pulaski court applied the incorrect deinition of restitution under the UCL—a definition that allowed it to sidestep Comcast. While it is true that there is no California Supreme Court authority conclusively deining restitution for purposes of the UCL, Pulaski is a clear departure from the body of law developed by the Courts of Appeal. Pulaski is also inconsistent with Comcast. Its bright-line refusal to consider the impact of damage calculations on the Rule 23(b) analysis flies in the face of Comcast’s now clear holding that plaintifs must come forward at the certiication stage with a damages methodology capable of class-wide application.

[Page 294]

This article will analyze the Pulaski opinion in light of existing California law defining restitution for purposes of the UCL and then look at the viability of the Ninth Circuit’s ultimate conclusion in light of Comcast. We begin with a brief discussion of Pulaski, Tobacco II and Vioxx. We then turn to an examination of the evolution of the definition of UCL restitution. Then we will analyze the Pulaski holding in light of Comcast, reaching the conclusion that it was only the incorrect definition of restitution that allowed the Court to sidestep Comcast.

II. SUMMARY OF PULASKI: UCL RESTITUTION CAN’T DEFEAT PREDOMINANCE

Pulaski involved a challenge to Google’s AdWords click advertising program. Specifically, plaintiffs, who were advertising agencies and businesses, alleged that Google uniformly failed to disclose in its AdWords documentation that it would place ads on low quality web pages such as parked domains (an undeveloped page consisting of nothing but links to ads) and error pages (reached when a user mistypes a domain or URL). At certification, plaintiffs proposed three alternative methods for calculating the amount of restitution owed to class members: (1) using Google’s "Smart Pricing" algorithm, which would approximate the difference between what plaintiffs actually paid and what they would have paid had Google disclosed the ad placement; (2) a "Content Pricing" approach, which plaintiffs described as factoring in the lower bidding that would have occurred had advertisers been allowed to bid separately for parked domain and error page placement; and (3) a full refund.⁶ The District Court, relying on the definition of restitution set forth in Vioxx, denied certification on the ground that determining not only who was entitled to restitution but in what amount, were necessarily individualized, and complicated, inquiries; the Court would essentially have to assess the intrinsic value of the advertising received, which would vary among class members.⁷ Accordingly, the need to make individualized calculations defeated predominance under Rule 23.

The Ninth Circuit reversed, finding that the restitution calculations were irrelevant to the Rule 23 analysis. The Court’s conclusion was premised on its findings that: (1) California has created a "conclusive presumption" of entitlement to restitution once liability under the UCL is established; and (2) liability under the UCL does not require individualized proof of deception, reliance or injury.⁸ Therefore, the Court concluded, the amount of restitution is determined at the time of purchase and the proper measure is the difference between what the class members paid and what they would have paid absent the misrepresentation or omission.

Having adopted an expansive definition of restitution, the Court turned to predominance under Rule 23(b)(3), making it clear that Comcast did not change Ninth Circuit law holding that varying amounts of damage cannot defeat predominance.⁹ It then expanded that familiar, and somewhat innocuous, proposition, however, to damage calculations, holding that the need to individually calculate the amount of each class member’s damage award can never defeat predominance. In the case before it, however, the statement was of little influence because pursuant to the Court’s view of UCL restitution, individual calculations were unnecessary anyway. The Court formulated a measure of restitution that focused only on the point of purchase—the difference between the price paid and the price the plaintiff would have paid had she known the truth about the alleged misrepresented or omitted information. Under the Court’s deinition, then, post-purchase value, while undoubtedly individualized, was irrelevant to the calculation.¹⁰ Taken to its logical extension, the Court’s deinition of restitution would allow a full refund if the consumer would not have purchased the product at all.

[Page 295]

Ultimately, however, the Court did not endorse a full refund. Rather, it found that using the Smart Pricing algorithm would result in a nonspeculative approximation of the proper amount of restitution.¹¹ The algorithm essentially used a ratio from Google’s data that adjusted the ad price for web page quality. Thus, it was both targeted to remedying the alleged harm (because it measured what the advertiser should have paid at the onset rather than accounting for what occurred post-purchase) and did not turn on individual circumstances.¹² Because damage calculations posed no impediment to certification, the Court reversed the District Court.

III. Summary of Tobacco II: Restitution is Individualized

The recent Tobacco II opinion was the result of an appeal after inal judgment. The facts of Tobacco II are familiar: plaintiffs claimed that defendants falsely advertised Marlboro Lights as being less harmful than other "full-flavored" cigarettes. The trial court found that defendant’s advertising was likely to mislead the general public but denied plaintiffs’ request for restitution on the ground that plaintiffs’ evidence of the difference between the price they paid for the cigarettes and the value they actually received was "incompetent and inadmissible."¹³

Plaintiffs made two contentions on appeal. First, they argued that the Vioxx measure of restitution (the difference between the amount paid and the actual value received) was not the only measure permitted under California law.¹⁴ Second, they argued that California law did not require them to show any loss attributable to the false advertising because the court had the authority to order a full refund under the UCL exclusively for the purpose of deterrence.¹⁵

The Court of Appeal rejected both arguments and conirmed that Vioxx accurately represents California law. The Court further made it clear that under Vioxx, a full refund would be appropriate under the UCL only when the product in question has no intrinsic value. In so holding (and rejecting plaintiffs’ arguments), the Court made three notable observations. First, it noted that Sections 17203 (remedies) and 17204 (standing) have distinct requirements.¹⁶ The Court discussed Kwikset Corp. v. Super. Ct.,¹⁷ noting that it established the standard only for standing, a standard that was inapplicable to calculating restitution once liability was established. Second, the Court noted that there is no "entitlement" to restitution under section 17203.¹⁸ Rather, injunctive relief is the primary remedy under the UCL and restitution is only ancillary.¹⁹ Last, while section 17203 may be broad enough to allow recovery without proof of individual injury under certain circumstances, it is not so broad as to allow an actual monetary award with no evidence of the actual value of the product. On this last point, the Court confirmed that courts have no statutory authority to award restitution under the UCL as a deterrent.²⁰ Accordingly, a restitution award must account for the value each class member actually received in order to avoid a windfall. In short, restitution is an individualized calculation that cannot be avoided by awarding a full refund unless the product at issue has no intrinsic value.

[Page 296]

Ultimately, the Court rejected plaintiffs’ proposed methodology for calculating restitution. The Court also found plaintiffs’ evidence to be inconsistent with a restitution award, specifically:

• Testimony from plaintiffs indicating that Marlboro Lights provided a reasonable value for the price paid notwithstanding the misrepresentations;
• Testimony from some class members establishing that they purchased cigarettes for reasons wholly unrelated to the health misrepresentations, such as taste and smoothness; and
• Testimony from some class members showing that they continued to buy the cigarettes for many years after they learned the truth.²¹

Accordingly, the Court awarded no restitution despite its liability finding.

IV. SUMMARY OF VIOXX: RESTITUTION REQUIRES SUBSTANTIAL EVIDENCE

Since Vioxx is credited with formulating the definition of restitution for purposes of the UCL, it warrants a brief discussion. Vioxx was a products liability action in which plaintiffs sued Merck for allegedly misrepresenting the safety and efficacy of the pain reliever Vioxx. Specifically, plaintiffs alleged that Merck represented that Vioxx was safe when it was actually less safe and no more effective than its generic competitors. Thus, plaintiffs alleged that they overpaid for Vioxx. On their motion for class certification, plaintiffs argued that they could establish the amount of restitution due to each class member by using the price of a generic drug, naproxen, as a comparator for calculating the amount of their overpayment (the difference between the price of Vioxx and the price of naproxen).

[Page 297]

Merck made a compelling argument that naproxen was an inappropriate comparator because not allVioxx patients could safely switch to it. It supported its argument by submitting evidence showing a myriad of reasons why naproxen was inappropriate. For example:

• Plaintiffs did not contend that they would have purchased naproxen absent the misrepresentations and omissions, just that they would not have purchasedVioxx;
• Because naproxen was safer than Vioxx from a gastrointestinal perspective but not a cardiovascular perspective, only patients with gastrointestinal disorders might take naproxen rather than Vioxx;
• Many individuals would take Vioxx regardless of its risks because of its efficacy;
• Few patients switched to naproxen when Vioxx was taken off the market.²²

The Court agreed with Merck and rejected naproxen as a valid comparator. On the issue of UCL restitution, the Court confirmed that restitution under the UCL is the difference between what the plaintiffs paid and the value of what they received.²³ Accordingly, plaintiffs must submit substantial evidence establishing the amount of restitution (i.e., evidence of the actual value they received).²⁴ Where, as there, the calculus was necessarily subjective, certification was inappropriate because the appropriate methodology was not subject to class-wide proof.²⁵ The Court denied certification on the ground that restitution could not be calculated on a class-wide basis.

V. THE PROBLEM(S) WITH PULASKI

A. The Pulaski Definition of Restitution is Inconsistent with Vioxx

In light of Vioxx and Tobacco II, it is clear that the Pulaski court misstepped. First, its expansive definition ofrestitution is not supportable. Tobacco II confirmed that Vioxx sets forth the correct definition—the difference between the amount paid and the value received.²⁶ True, Pulaski did not have the benefit of Tobacco II at the time it issued its holding, but it did have Vioxx and a number of lower court opinions (both federal and state) confirming that the calculation required deductions for value received.²⁷ Nonetheless, Pulaski relied on Kwikset, which discussed standing, in formulating the definition of restitution as: "what the purchaser would have paid at the time of purchase had the purchaser received all the information" with the focus being "on the value of the service at the time of purchase."²⁸

[Page 298]

Pulaski’s definition is broad enough to encompass a full refund under circumstances where the plaintiff claims he or she would not have purchased the product at all but for the misrepresentation (or had she been aware of the omitted information). The problem here is that the UCL, however, is not exclusively deterrent. Accordingly, as Tobacco II explained, California law does not support a full refund under the UCL unless the product has no intrinsic value:

We conclude that as a matter of law, restitution is not available here for the exclusive purpose of deterrence. Under plaintiff’s theory, a full refund would be available on any product, even costly items such as cars, yachts, and planes, based on UCL violations that had little or no impact on value. That, of course, is not the law. "Section 17203 makes injunctive relief ‘the primary form of relief available under the UCL,’ while restitution is merely ‘ancillary.’" Without any showing of loss to plaintiffs, there can be no restoration of money "which may have been acquired by means of such unfair competition’"
While a full refund may be proper when a product confers no benefit on consumers, such is not the scenario here. Plaintiffs do not dispute the court’s finding that they obtained value from Marlboro Lights apart from the deceptive advertising. Indeed, it appears inherently implausible to show that a class of smokers received no value from a particular type of cigarette.²⁹

There are circumstances where a full refund can be appropriate, and Ortega v. Natural Balance, Inc.³⁰provides an example. In Ortega, the plaintiffs challenged label representations indicating that the dietary supplements at issue had aphrodisiac properties, which, of course, was false. As it turned out, the evidence established that the supplements provided no health benefits at all, represented or not, and, in fact, contained illegal substances. Under those circumstances, the court determined that the Vioxx measure of restitution was actually zero because the product had no intrinsic value.³¹ Thus, plaintiffs were entitled to a full refund.

So how did Pulaski stray so far afield in measuring restitution? Instead of relying on Vioxx and its progeny, it turned instead to standing cases and the Restatement of Restitution; based on them, the Court formulated a punitive definition: "[r]estitution has two purposes: ‘to restore the defrauded party . . . and to deny the fraudulent party any benefits, whether or not foreseeable, which derive from his wrongful act.’"³² Pulaski relied on Nelson v. Serwold, as an example of a punitive application, but Nelson was a securities fraud case and it did not involve violations of the UCL or California False Advertising Law ("FAL"). Thus, Nelson was defining restitution for entirely different purposes.

[Page 299]

As referenced above, Pulaski also relied on Kwikset and Colgan v. Leatherman Tool Group, Inc.³³ While those are big names in UCL jurisprudence, neither were applicable because they addressed standing and liability standards under the UCL, not restitution calculations for purposes of class certification. Moreover, neither relied on the Restatement definition that Pulaski used. While Colgan made a reference to the Restatement, that part of the opinion was explaining simply that the amount of a restitution award has to be supported by evidence.³⁴ Thus, Colgan does not support Pulaski’s adoption of the Restatement definition for purposes of calculating restitution under the UCL.

Pulaski similarly finds no support for the Restatement definition in Kwikset. In fact, Pulaski’s reliance on Kwikset clearly conflated UCL standing requirements with the standards applicable to restitution calculations.³⁵ Pointing to language in Kwikset, Pulaski concluded that every consumer who is misled by a misrepresentation suffers the same harm—they paid more for the product than they otherwise would have.³⁶ Thus, the Court concluded that the measure of restitution is the same for all class members. Kwikset, however, does not support Pulaski’s conclusion because Kwikset was addressing injury sufficient to support UCL standing requirements—it was not calculating the amount of restitution or analyzing the predominance requirement for class certification:

For each consumer who relies on the truth and accuracy of a label and is deceived by misrepresentations into making a purchase, the economic harm is the same: the consumer has purchased a product that he or she paid more for than he or she otherwise might have been willing to pay if the product had been labeled accurately. This economic harm—the loss of real dollars from a consumer>s pocket—is the same whether or not a court might objectively view the products as functionally equivalent. A counterfeit Rolex might be proven to tell the time as accurately as a genuine Rolex and in other ways be functionally equivalent, but we do not doubt the consumer (as well as the company that was deprived of a sale) has been economically harmed by the substitution in a manner sufficient to create standing to sue. Two wines might to almost any palate taste indistinguishable— but to serious oenophiles, the difference between one year and the next, between grapes from one valley and another nearby, might be sufficient to carry with it real economic differences in how much they would pay. Nonkosher meat might taste and in every respect be nutritionally identical to kosher meat, but to an observant Jew who keeps kosher, the former would be worthless.³⁷

The above examples illustrate Pulaski’s flaw. The difference between the price paid and what a consumer would have been willing to pay had the product been accurately labeled demonstrates the existence of some injury for purposes of standing. But, calculating the amount of that injury for purposes of a monetary award is an entirely different story.

[Page 300]

Kwikset said nothing about placing a value on that amount for purposes of an award, nor did it excuse the plaintiff from ultimately quantifying the loss at the certification stage by resorting to admissible, nonspeculative proof of value. In fact, even in addressing standing only, Kwikset undermined Pulaski because it specifically called out the distinction between standing and calculating restitution:

As we recently have noted, however, the standards for establishing standing under section 17204 and eligibility for restitution under section 17203 are wholly distinct. For the drafters of Proposition 64, which amended both sections 17203 and 17204, to make standing under section 17204 expressly dependent on eligibility for restitution under section 17203 would have been easy enough, but nothing in the text or history of Proposition 64 suggests this was intended. We thus rejected in Clayworth the argument that if the plaintiffs could demonstrate no compensable losses or entitlement to restitution under section 17203, they would lack standing under section 17204. As we explained, "this argument conflates the issue of standing with the issue of the remedies to which a party may be entitled. That a party may ultimately be unable to prove a right to damages (or, here, restitution) does not demonstrate that it lacks standing to argue for its entitlement to them."³⁸

Colgan,also cited by Pulaski, actually confronted the distinction between liability and calculating restitution when that case reached the remedy phase.³⁹ The result there was no restitution award. Colgan was a challenge to an allegedly false "Made in the USA" label used in connection with marketing tools. The trial court found that the tools were not "Made in the USA" and thus found liability, but had difficulty quantifying the restitution. Accordingly, it awarded a flat amount of25% of the gross sales receipts. Appellate court reversed noting that the amount was not supported by substantial evidence.⁴⁰ Because the plaintiffs’ expert could not quantify the value of the "Made in the USA" label, plaintiffs failed in their proof.⁴¹ Importantly, this did not leave the plaintiffs without remedy for the violation because the primary relief under the UCL is an injunction—in this case a labeling change. This outcome was entirely consistent with Tobacco II.

Thus, Pulaski’s definition of restitution for purposes of remedy finds no support in California law. It has long been held that the UCL has no punitive component.⁴² Vioxx supplied a definition of restitution for purposes of the UCL. And, Tobacco II confirmed it, noting "'[c]ourts ordering restitution under the UCL are not concerned with restoring the violator to the status quo ante. The focus instead is on the victim.’ Obviously, restitution without proof of any actual loss to any plaintiff cannot be characterized as restitutionary."⁴³

[Page 301]

B. CALIFORNIA LAW DOES NOT CREATE A "CONCLUSIVE PRESUMPTION" OF RESTITUTION

The second failing of Pulaski is its conclusion that California law imposes a "conclusive presumption" of restitution once liability under the UCL is established.⁴⁴ This conclusion is flatly contrary to Colgan and Tobacco II, both of which ended up awarding no restitution at all based on plaintiffs’ failure of proof. Pulaski arrived at its contrary conclusion by relying on Stearns. The problem with its analysis, however, is that Pulaski’s jump from liability standards under the UCL to standards applicable to quantifying restitution is legally unsupportable. In fact, Pulaski cites no authority to support its conclusion that entitlement to restitution never requires individualized inquiries because Stearns did not address restitution at all. Stearns addressed liability and standing standards after the first Tobacco II decision and, in fact, the footnote that Pulaski relied on was part of a discussion of Article III standing. Accordingly, Stearns does not support Pulaski.

C. DAMAGE CALCULATIONS DO MATTER: PULASKI’S RELIANCE ON YOKOYAMA IS INCONSISTENT WITH COMCAST

Relying on Yokoyama v. Midland National Life Insurance Co., Pulaski concluded that damage calculations can never defeat predominance in the Ninth Circuit.⁴⁵ Yokoyama, decided well before Comcast, holds simply that "damage calculations alone cannot defeat certification."⁴⁶ Comcast, on the other hand, holds that plaintiffs’ damage model must be capable of measuring damage resulting from the theory of liability advanced.⁴⁷ Underlying the Comcast opinion is another widely accepted rule: the inability to submit a class-wide methodology for establishing damages (or restitution in the context of the UCL) can defeat predominance.⁴⁸ In short, "[n]o damages model, no predominance, no class certification."⁴⁹ Despite the Ninth Circuit’s attempt to state otherwise, Yokoyama is not inconsistent with Comcast because at bottom, it really stands for the same proposition. While the opinion speaks in terms of damage "calculations" when it says "damage calculations alone cannot defeat certification," the Court was clearly speaking in terms of one calculation that might yield varying amounts for different class members rather than a different methodology applicable to individual class members:

[D]amage calculations alone cannot defeat certification. We have said that ‘[t]he amount of damages is invariably an individual question and does not defeat class action treatment.’⁵⁰

[Page 302]

Yokoyama said nothing about the necessity of differing proof to establish entitlement to an award. It spoke only to a situation where the plaintiffs have established that the class is entitled to recover. There, the fact that a uniform calculation may yield differing amounts for individual members (even if the amount is zero for some) usually does not impede certification. Where, however, class-wide proof of entitlement is not possible, the issue is different and predominance is usually lacking.

The distinction is subtle and often conflated, but the facts of Yokoyama are helpful. There, the plaintiff claimed that uniformly distributed written information about variable annuity products were deceptive. Ironically, the issue in Yokoyama did not involve damage calculations but rather whether individual reliance was an element of liability under the Hawaii consumer protection statute.⁵¹ With respect to damages, the Ninth Circuit concluded that because the lawsuit was premised only on uniform written statements (and not what any individual sales agent said) that all class members allegedly saw, entitlement to damage could be determined on a class-wide basis. It further dismissed the district court’s finding that individualized inquiries might be necessary to calculate the amount of damage because calculations alone do not defeat certification. But, Yokoyama was a case where it appeared that the same general formula could be applied to all class members even though the inputs were variable and the calculations might yield different amounts for different class members. And, importantly, the variables were not so individualized that they would swamp common proof, and the formula was "mechanical." Despite the fact that it is widely cited, particularly as of late, Yokoyama really does not support the broad proposition that damage calculations can never defeat certification in the Ninth Circuit.

Blackie v. Barrack, also oft-cited (including by Yokoyama), likewise illustrates the point.⁵² Blackie was a securities fraud case. In recognizing that individual damage calculations usually do not defeat certification, the Court noted that before it was a case where "the amount of price inflation during the period can be charted and the process of computing individual damages will be virtually a mechanical task."⁵³ Again, Blackie was a case where one formula could be applied class-wide even though its application might result in different amounts for different class members.

Leyva v. Medline Industries, Inc.,⁵⁴ the last of the Ninth Circuit damage trifecta (and cited by Pulaski), is also consistent with Comcast. Levya was a wage and hour case where plaintiffs proposed a "mechanical calculation" for restitution. Thus, the Court found that the damage calculations were not so individualized as to swamp common proof.⁵⁵

There are plenty of examples, however, of cases where either entitlement to recover or the damage calculation itself is so individualized an inquiry that it swamps common proof and thus does defeat certification. In Blades v. Monsanto Co., for example, the Eighth Circuit denied certification because the damages there were so individualized that attempting to figure out who was entitled to recover, and in what amount, would swamp common proof.⁵⁶ The Sixth Circuit came to a similar conclusion in Rodney v. Northwest Airlines, Inc. and denied certification based on the individualized nature of the damage calculations presented there.⁵⁷ In Heerwagen v. Clear Channel Communications, the Second Circuit denied certification because the plaintiffs failed to submit a viable, class-wide methodology for proving damages.⁵⁸

[Page 303]

VI. CONCLUSION: RECONCILING PULASKI,TOBACCOIIAND COMCAST

Pulaski,Tobacco II and Comcast may be reconcilable by limiting Pulaski to its facts. Pulaski was an omission case based on documentation that uniformly omitted information about ad placement on parked domains and error pages. Accordingly, the line of California cases holding that there must be some showing of reliance on misrepresentations in order to justify restitution did not come into play because all plaintiffs were subjected to the same omissions and the materiality of those omissions was not challenged. Thus, the appropriate measure of damage may have been calculable on a class-wide basis because Google actually had a methodology in place (the algorithm) that could calculate the value of the ads considering their placement. The Court itself seemed to suggest such was the case by using Google’s algorithm, which resulted in a mechanical, formulaic approach consistent with Rule 23(b) and, incidentally, Comcast. But, this damage model, while uniform, still provided a windfall to those advertisers whose sales increased or who received some other benefit from the ads. Thus, this damage model is not supported by Vioxx and Tobacco II.

At bottom, though, California law does not support the reasoning underlying Pulaski’s restitution definition and federal law does not support the manner in which it applied Yokoyama in light of Comcast. The defendant in Pulaski petitioned for certiorari, so it remains to be seen how the Ninth Circuit’s positions will hold up. For now, however, there is at least an argument (albeit not a legally sound one) that the UCL can provide a money-back guarantee on a class-wide basis in Ninth Circuit courts.

[Page 304]

——–

Notes:

^1. Michele Floyd is a partner at Sacks Ricketts & Case, a women-owned elite litigation boutique that focuses on competitive, consumer and employment class action defense across a number of industries, including video games, computer hardware and software, consumer retail sales, franchise, food/restaurant, manufacturing, and technology. Michele’s practice focuses on antitrust and consumer class action defense, primarily for technology, online and retail clients.

^2. 802 F.3d 979 (9th Cir. 2015), petition for cert. filed, Mar. 1, 2016 (No. 15-1101).

^3. 133 S. Ct. 1426 (2013) (reversing the Third Circuit’s decision and holding that there is a duty to conduct a "rigorous analysis" of whether the Rule 23 requirements have been met even if that means considering the merits of the claim at class certiication).

^4. 192 Cal. Rptr. 3d 881 (Cal. Ct. App. 2015).

^5. 103 Cal. Rptr. 3d 83 (Cal. Ct. App. 2009).

^6. Pulaski, 802 F.3d at 983.

^7. Id. at 982.

^8. Pulaski, 802 F.3d at 986 (stating "’that when a defendant puts out tainted bait and a person sees it and bites, the defendant has caused an injury; restitution is the remedy’") (citing Stearns v. Ticketmaster LLC, 655 F.3d 1013, 1021 n.13 (9th Cir. 2013)).

^9. The Court relied on Yokoyama v. Midland Nat’l Life Ins. Co., 594 F.3d 1087 (9th Cir. 2010) (holding that damage calculations alone could not defeat class certification). This proposition is not new in the Ninth Circuit. See Leyva v. Medline Indus., Inc., 716 F.3d 510 (9th Cir. 2013).

^10. Pulaski, 802 F.3d at 989-990 (citing Kwikset Corp. v. Super. Ct., 246 P.3d 877, 890 (Cal. 2011)).

^11. Pulaski, 802 F.3d at 989 (noting California law requires only that some reasonable basis for computation of damages be used and damages may be computed even if the result reached is an approximation) (citing Marsu, B.V. v. Walt Disney Co., 185 F.3d 932, 938-39 (9th Cir. 1999)).

^12. Pulaski, 802 F.3d at 990.

^13. Tobacco II, 192 Cal. Rptr. 3d at 886.

^14. Id. at 887.

^15. Id.

^16. Tobacco II, 192 Cal. Rptr. 3d at 898 (citing Kwikset Corp. v. Super. Ct., 246 P.3d 877, 894 (Cal. 2011)).

^17. 246 P.3d 877 (Cal. 2011). Pulaski relied on Kwikset as well, but as support for its point of purchase measure of restitution.

^18. Tobacco II, 192 Cal. Rptr. 3d at 898 (citing Kwikset, 246 P.3d at 894).

^19. Id. at 891.

^20. Id. at 894.

^21. Id. at 889.

^22. In re Vioxx Class Cases, 103 Cal. Rptr. 3d 83, 91-92 (Cal. Ct. App. 2009).

^23. Vioxx, 103 Cal. Rptr. 3d at 96 (citing Cortez v. Purolator Air Filtration Products Co., 99 P.2d 706, 713 (Cal. 2000)).

^24. Vioxx, 103 Cal. Rptr. 3d at 100-101 (citing Colgan v. Leatherman Tool Group, Inc., 38 Cal. Rptr. 3d 36, 61 (Cal. Ct. App. 2006)).

^25. Vioxx, 103 Cal. Rptr. 3d at 101 (holding that the issue of a proper comparator was a patient-specific issue incorporating the patient’s medical history, treatment need and drug interactions).

^26. Tobacco II, 192 Cal. Rptr. 3d 881, 895 (citing Zhang v. Superior Court, 304 P.3d 163 (Cal. 2013) and Dunkin v. Boskey, 98 Cal. Rptr. 2d 44 (Cal. Ct. App. 2000) (party seeking restitution must return value received)).

^27. See, e.g., In re Pom Wonderful LLC Mktg and Sales Practices Litig., No. ML 10-02199 DDP, 2014 WL 1225184, at *3 (C. D. Cal. Mar. 25 2014) (citing Vioxx); Ortega v. Natural Balance, Inc., 300 F.R.D. 422, 429 (C. D. Cal. 2014) (same).

^28. Pulaski, 802 F.3d at 989.

^29. Tobacco II, 192 Cal. Rptr. 3d 881, 900-901 (citing Clayworth v. Pfizer, Inc., 233 P.3d 1066, 1088 (Cal. 2010)) (emphasis in original).

^30. 300 F.R.D. 422 (C. D. Cal. 2014).

^31. Id. at 430 (holding full refund may be available where dietary supplement had no value apart from the misrepresented health benefit).

^32. Pulaski, 802 F.3d at 988 (citing Nelson v. Serwold, 687 F.2d 278, 281 (9th Cir. 1982) (citing the Restatement of Restitution)).

^33. 38 Cal. Rptr. 3d 36 (Cal. Ct. App. 2006).

^34. Id. at 62. ("The Restatement contemplates restitution supported by evidence.").

^35. Pulaski, 802 F.3d at 989.

^36. This price premium measure has been rejected by lower courts. See, e.g., Pom Wonderful, 2014 WL 1225184, at *5.

^37. Kwikset, 246 P.3d at 890 (emphasis added).

^38. Id. at 894-895 (citing Clayworth v. Pfizer, Inc., 233 P.3d 1066, 1087 (Cal. 2010)).

^39. Colgan v. Leatherman Tool Group, Inc., 38 Cal. Rptr. 3d 36 (Cal. Ct. App. 2006).

^40. Id. at 63.

^41. Id.

^42. Korea Supply Co. v. Lockheed Martin Co. made it clear that restitution under the UCL is essentially limited to an out-of-pocket loss. 63 P.3d 937, 946-947 (Cal. 2003). The remedial sections of the statutory scheme do not provide for the recovery of compensatory or punitive damages. Korea Supply held clearly that disgorgement of profits was not "restitutionary" and therefore not recoverable under the UCL. Id. at 949. Madrid v. Perot Sys. Corp. later extended Korea Supply to class actions. 30 Cal. Rptr. 3d 210, 221-224 (Cal. Ct. App. 2005).

^43. Tobacco II, 192 Cal. Rptr. 3d at 900 (citing Madrid, 30 Cal. Rptr. 3d 210 at 221).

^44. Pulaski, 802 F.3d at 986 (citing Stearns v. Ticketmaster Corp., 655 F.3d 1013, 1021 n.13 (9th Cir. 2011)).

^45. Id. at 988 ("Yokoyama remains the law of this court, even after Comcast").

^46. Yokoyama v. Midland Nat’l Life Ins. Co., 594 F.3d 1087, 1094 (9th Cir. 2010) (emphasis added).

^47. Comcast, 133 S. Ct. at 1433.

^48. See In re NJOY, Inc. Consumer Class Action Litig., 120 F. Supp.3d 1050, 1117 (C. D. Cal. 2015) (citing Comcast, 133 S. Ct. at 1433).

^49. In re Rail Freight Surcharge Antitrust Litig., 725 F.3d 244, 253 (D.C. Cir. 2013).

^50. Yokoyama, 594 F.3d at 1094 (citing Blackie v. Barrack, 524 F.2d 891, 905 (9th Cir. 1975)) (emphasis added).

^51. Yokoyama, 594 F.3d at 1092.

^52. 524 F.2d 891 (9th Cir. 1975).

^53. Id. at 905.

^54. 716 F.3d 510 (9th Cir. 2013).

^55. Id. at 514 (noting "Medline’s computerized payroll and time-keeping database would enable the court to accurately calculate damages and related penalties for each claim.").

^56. 400 F.3d 562 (8th Cir. 2005) (denying certification on ground that individualized damage calculations would defeat predominance).

^57. 146 Fed. App’x. 783 (6th Cir. 2005) (rejecting expert’s proposed methodology for common proof of damages and denying certification). Lower courts within the Ninth Circuit have come to the same conclusion following Comcast. The Central District of California noted that "Rule 23(b)(3) is satisfied only if plaintiffs can show that damages are capable of measurement on a classwide basis." In re NJOY, Inc. Consumer Class Action Litig., 120 F. Supp. 3d at 1117 (citing Comcast, 133 S. Ct. at 1433); In re Con Agra Foods, Inc., 90 F. Supp. 3d at 1022-1023 (C.D. Cal. 2015) (citing Comcast, 133 S. Ct. at 1433). McVicar v. Goodman Global, Inc., recently reached the same conclusion: "At the class certification stage, Plaintiffs must present a theory that can measure, on a class-wide basis, damages attributable to Plaintiffs’ theory of liability." McVicar, No. SA CV 13-1223-DOC, 2015 WL 4945730, at *14 (C.D. Cal. Aug. 20, 2015) (citing Comcast, 133 S. Ct. at 1433) (emphasis added).

^58. 435 F.3d 219, 235 (2d. Cir. 2006).