ORDER DENYING MOTION TO EXPAND CLASS CERTIFICATION BEYOND CALIFORNIA CONSUMERS

RICHARD SEEBORG, District Judge.

I. INTRODUCTION

Plaintiff Kathie Sonner represents a class of California consumers who purchased Joint Juice, a drinkable glucosamine hydrochloride and chondroitin sulfate supplement. She has sought to expand those classes to include members: (1) who purchased Joint Juice in all fifty states or, in the alternative, (2) who purchased the product in ten specific states. Questions remained about whether the nationwide or ten-state class could proceed where the claims are based on California law, and therefore the parties were directed to file briefs addressing that issue.

The responses submitted by the parties exposed the material conflicts between California's Unfair Competition Law ("UCL"), Cal. Bus. & Prof. Code § 17200, and the Consumer Legal Remedies Act ("CLRA"), Cal. Civ. Code § 1770, on the one hand, and the laws of other effected states. Premier demonstrated there are at least some states with laws that materially conflict with the UCL and CLRA and, on balance, the interests of the states where the advertising occurred outweigh those of California. Sonner concedes the existence of true conflicts between the law of some states and California law, thereby making a nationwide class improper. Similarly, Premier has demonstrated there are material conflicts between the laws of those ten states in the alternative sub-group, and that those states have more interest in applying their laws to the advertisements at issue than does California.

Sonner now offers a new proposal: to carve out those states whose laws materially conflict with the UCL or CLRA. The trouble, however, is that the material conflicts that Premier has identified conflict with the UCL, but not the CLRA, or vice versa. Sonner has not offered a proposal that resolves all conflicts cleanly. Nor has Premier had the opportunity to challenge Sonner's new proposal. Accordingly, because Sonner has not demonstrated that California law should apply to the two classes originally proposed, she will represent only a class of California consumers who have purchased Joint Juice since March 1, 2009. Her request to represent a larger class of consumers therefore must be denied.

II. PROCEDURAL HISTORY1

Plaintiff Kathie Sonner seeks to represent consumers in all fifty states, or, in the alternative, consumers in ten specific states in this class action premised on California consumer-protection statutes. Sonner satisfied her burden to prove that the consumer claims she advances are amenable to classwide adjudication and that common questions predominate. Accordingly, at a minimum, a class of California consumers who have purchased Joint Juice since March 1, 2009, was certified. Difficult questions remained, however, about whether, under California's government-interests test, California law could apply nationwide or to the proposed ten-state subclass in light of Mazza v. American Honda Motor Co., 666 F.3d 581 (9th Cir. 2012). To assist with these issues, the parties submitted additional briefs addressing the following two questions:

III. LEGAL STANDARD

"A federal court sitting in diversity must look to the forum state's choice of law rules to determine the controlling substantive law." Mazza, 666 F.3d at 589 (internal quotation marks omitted). Thus, in this case, California's choice-of-law rules apply. Application of California law to class members from other jurisdictions is permissible only if Sonner shows "doing so comports with both (1) due process, and (2) California's choice of law rules." In re Optical Disk Drive Antitrust Litig., No. 3:10-MD-2143 RS, 2016 WL 467444, at *12 (N.D. Cal. Feb. 8, 2016) (citing Mazza, 666 F.3d at 589). Sonner has already demonstrated application of California law to California-based Premier will not offend due process, and thus the only remaining question is whether California's choice of law rules are satisfied here. Premier bears the burden to show "`foreign law, rather than California law, should apply to class claims.'" Mazza, 666 F.3d at 590 (quoting Wash. Mut. Bank, F.A. v. Superior Court, 24 Cal.4th 906, 921 (2001)).

There are typically three steps to decide which state's law applies. The first task is to determine whether the laws of the affected jurisdictions are "the same or different." Id. at 590 (quoting McCann v. Foster Wheeler LLC, 48 Cal.4th 68, 81-82 (2010)). If the laws are different, the second step requires an examination of "each jurisdiction's interest in the application of its own law" to determine whether a true conflict exists. Id. If it does, then the final step involves analyzing "which state's interest would be more impaired if its policy were subordinated to the law of the other state." Id.

IV. DISCUSSION

A. The Impact of Mazza

The language used by the Mazza court speaks broadly, suggesting nationwide consumer class actions are never permissible because some states require proof of reliance or scienter, or have different methods for compensating victims of consumer fraud. See Mazza, 666 F.3d at 592-93 ("Each of our states has an interest in balancing the range of products and prices offered to consumers with the legal protections afforded to them. Each of our states also has an interest in being able to assure individuals and commercial entities operating within its territory that applicable limitations on liability set forth in the jurisdiction's law will be available. . . ." (internal quotation marks omitted)). Yet, the court was careful to limit its holding to "the facts and circumstances of this case," id. at 594, and thus district courts routinely require fact- and case-specific analysis to determine whether to apply California law to absent class members in other jurisdictions. See, e.g., Forcellati v. Hyland's, Inc., 876 F.Supp.2d 1155, 1160-61 (C.D. Cal. 2012) ("Defendants can only meet [their] burden by engaging in an analytically rigorous discussion of each prong of California's `government interests' test based on the facts and circumstances of this case and this Plaintiff's allegations." (emphasis in original)); In re Clorox Consumer Litig., 894 F.Supp.2d 1224, 1237 (N.D. Cal. 2012) ("Since the parties have yet to develop a factual record, it is unclear whether applying different state consumer protection statutes could have a material impact on the viability of Plaintiffs' claims."); Allen v. Hyland's Inc., 300 F.R.D. 643, 658 (C.D. Cal. 2014); Bruno v. Eckhart Corp., 280 F.R.D. 540, 550 (C.D. Cal. 2012) ("Defendants cannot profitably rely on the work of a different party in a different case with different facts—or on the Ninth Circuit finding error in a district court rejecting an argument Defendants did not themselves present to this Court—to correct their failure" to show "the law of other states conflicted with California law as applied to this particular case." (emphasis in original)).

Premier correctly notes that when district courts have permitted nationwide class actions involving claims for violations of consumer protection laws, the defendants failed to offer any analysis of states' laws or case-specific discussion whatsoever. See Allen, 300 F.R.D. at 643; Forcellati, 2014 WL 1410264, at *3; Bruno, 280 F.R.D. at 549-50. Of course, defendants need do more than simply submit a table of state law to show the UCL and CLRA are inapplicable nationwide; they must demonstrate the elements at issue are actually in dispute and amount to true conflicts.

This fact-specific approach comports with the California Supreme Court's decision to abandon its prior categorical rule "that in tort actions the law of the place of the wrong was the applicable law in a California forum regardless of the issues before the court." Hurtado v. Superior Court, 11 Cal.3d 574, 579 (1974). Instead, the court chose to adopt the more flexible government-interests test. Thus, under that test, "[t]he fact that two states are involved does not itself indicate there is a `conflict of laws' or `choice of law' problem. There is obviously no problem where the laws of the two states are identical." Id. at 580. In other words, the analysis begins by identifying whether the statutes are worded differently, and, if so, whether the differences rise to the level of substantive requirements. "The key step in this [choice of law] process is delineating the issue to be decided." Id. at 584.

B. Nationwide Class

Neither party seriously disputes the fact the consumer protection statutes of all fifty states vary. Instead, they clash about whether the conflicts identified—reliance, scienter, and remedies—are true or false conflicts. Sonner contends most of the conflicts identified are of no consequence. To some extent, her assessment is accurate. She acknowledges, however, that at least some conflicts cannot be reconciled, rendering inappropriate the application of California law nationwide.

The issue of reliance, for example, makes a nationwide class difficult to maintain. The CLRA requires proof of "damage," Cal. Civ. Code § 1780(a), and thus named and unnamed plaintiffs must prove the alleged deceptive conduct caused that damage. In re Steroid Hormone Prod. Cases, 181 Cal.App.4th 145, 156 (2010). In the context of claims asserting alleged misrepresentations, a showing that the alleged misrepresentation was material suffices to establish damage. Id. at 156-57. In contrast, the UCL requires class representatives to prove reliance as part of the statute's standing requirement, but does not require absent class members to do so. See In re Tobacco II Cases, 46 Cal.4th 298, 324-26 (2009). A plaintiff satisfies the burden of proving reliance by showing the alleged "misrepresentation was an immediate cause of the injury-producing conduct"—not that it was the only cause—or that the misrepresentation was material. Id. at 326-27. When a misrepresentation is material, a presumption of reliance arises. Id. at 327. Sonner agrees that most states, like the CLRA, require proof of reliance. Thus, most states' laws materially conflict with the UCL. Idaho and North Dakota, however, do not mandate such proof, and therefore conflict with the CLRA, but not the UCL.

Sonner argues nonetheless that any conflict as to the level of proof of reliance is not meaningful here as that element is not seriously in dispute. To the contrary, on summary judgment, she successfully demonstrated that a dispute of fact exists about whether Premier's alleged misrepresentations were material (thereby giving rise to a presumption of reliance), and whether consumers actually relied on the advertising. Those questions are not beyond dispute, and therefore a true conflict exists.

Scienter is also a stumbling block for Sonner's attempt to apply California law nationwide. South Dakota and Pennsylvania would require a plaintiff to prove Premier intended to deceive consumers. See Coleman v. Commonwealth Land Title Ins. Co., 684 F.Supp.2d 595, 619 (E.D. Pa. 2010) (citing Weinberg v. Sun Co., 565 Pa. 612, 777 A.2d 442, 445-46 (2001); Santana Products, Inc. v. Bobrick Washroom Equip., Inc., 401 F.3d 123, 136 (3d Cir. 2005) ("To bring a claim of fraud under the UTPCPL, Pennsylvania state court precedent requires Plaintiffs to meet the elements of common law fraud. . . . Under Pennsylvania law, common law fraud requires: (1) a misrepresentation, (2) material to the transaction, (3) made falsely, (4) with the intent of misleading another to rely on it, (5) justifiable reliance resulted, and (6) injury was proximately caused by the reliance."); S.D. Codified Laws § 20-10-1 ("One who willfully deceives another, with intent to induce him to alter his position to his injury or risk, is liable for any damage which he thereby suffers.").

Sonner's response to this conflict is to assert that Premier's intent to deceive consumers is not seriously in dispute. While the Ninth Circuit has not offered specific guidance about whether a judge or jury resolves factual disputes necessary to choose which state law should apply, the better practice would be for courts to resolve those factual disputes at the choice-of-law stage. See generally In re Facebook Biometric Info. Privacy Litig., No. 15-CV-03747-JD, 2016 WL 2593853, at *2-4 (N.D. Cal. May 5, 2016). Premier hotly contests that it and its employees intended to deceive the public and has proffered evidence demonstrating how seriously that question is in dispute. Accordingly, it is not proper to conclude as a matter of law that Premier intended to deceive consumers, and therefore the CLRA and ULC materially conflict with the statutes of South Dakota and Pennsylvania.

In the face of these true conflicts, Sonner does not attempt to argue that California's interest in the application of its law outweighs the other states' interests in the same.2 Instead, she proposes carving out these four states from the class or proceeding with two different classes: one class that includes consumers in states whose laws do not conflict with the UCL, and a second class of consumers from states whose statutes do not conflict with the CLRA. There are two problems with this proposal. First, this approach requires clearly delineating which states have material conflicts with the UCL and which have conflicts with the CLRA, and there are notable differences between the UCL and the CLRA. As the proponent of a nationwide (or 46-state) class, Sonner bears the burden to demonstrate the solution she proposes is proper. Vinole v. Countrywide Home Loans, Inc., 571 F.3d 935, 944 n.9 (9th Cir. 2009) ("The party seeking certification bears the burden of demonstrating that he has met the requirements of Rule 23(b)."). She has not, in the end, persuasively demonstrated that the solution she proposes is workable or would adequately resolve these conflicts.

Second, Premier has not had the opportunity to respond to this proposal. At the outset, Sonner offered three proposed class definitions: nationwide, ten-state, or California. Premier responded to those three proposals with specific, targeted analysis. To change the definitions of the proposed class at the last minute would be unfair under these circumstances, when Premier has already carried its burden to show that the government-interests test does not favor application of California law nationwide or, as explained below, to the claims of consumers in the specific ten states Sonner identifies. Accordingly, a nationwide class will not be certified.

C. Ten-State Class

Sonner's second proposal is to certify a class of consumers from the following ten states: California, Florida, Illinois, Massachusetts, Michigan, Minnesota, Missouri, New Jersey, New York, and Washington. Again, the central dispute is whether the conflicts identified are real and material. As explained below, Premier has demonstrated the existence of material conflicts and that each individual state's interest in the application of its laws outweighs California's interest in applying its law to certain members among the proposed ten-state group.

1. Reliance

Premier contends Minnesota's consumer protection statute conflicts with the UCL and the CLRA because it (1) requires proof of reliance, and (2) demands an individualized examination of evidence rebutting the presumption of reliance. As explained above, the CLRA requires proof of reliance, whereas the UCL does not require such proof for absent class members. Clearly, the Minnesota law materially conflicts with the UCL because it requires proof of reliance. The more difficult question is whether it conflicts with the CLRA.

Minnesota permits defendants to "negat[e] a plaintiff's direct or circumstantial showing of causation and reliance." In re St. Jude Med., Inc., 522 F.3d 836, 840 (8th Cir. 2008). The extent to which Minnesota's statute actually differs from California law remains unclear. At least one federal district court has concluded the reliance requirements under California and Minnesota law are similar. Khoday v. Symantec Corp., No. CIV. 11-180 JRT/TNL, 2014 WL 1281600, at *29-31 (D. Minn. Mar. 13, 2014) ("[S]imilar to California, Minnesota courts have found that class members' awareness of misrepresentations provides a sufficient causal nexus between a violation of the consumer protection statutes and damages in the form of restitution."). Indeed, class actions under the CLRA have failed where the evidence showed the defendant's message was not uniform, consumers received information about the product from myriad sources, and some consumers would use the product despite the risks. See In re Vioxx Class Cases, 180 Cal.App.4th 116, 133-34 (2009).

All in all, doubts remain about whether Premier has identified a true conflict. In this circumstance, one of the guiding principles of federalism breaks the tie: "each State may make its own reasoned judgment about what conduct is permitted or proscribed within its borders." State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 422 (2003). Thus, each state must decide for itself how to resolve these unanswered questions about the scope and evidentiary requirements needed to establish reliance. Until state courts clarify otherwise, this potential difference between the Minnesota and California statutes presumably is material. As discussed above, Minnesota's interest in regulating transactions occurring within its borders outweighs California's interest in the same. Given the fact that California's interest in transactions in Minnesota is attenuated, the law of Minnesota should apply. See Mazza, 666 F.3d at 594.

2. Scienter

In Mazza, the Ninth Circuit observed, "[i]n cases where a defendant acted without scienter, a scienter requirement will spell the difference between the success and failure of a claim." 666 F.3d at 591. Premier asserts numerous states require proof of scienter that materially differs from those at issue under the CLRA. In particular, Premier contends the following states included in Sonner's proposed ten-state class require proof of various types of scienter: Michigan (intent to deceive) and Illinois and Minnesota (intent to induce reliance).

Sonner has shown that scienter is not an impediment to proceeding with a class of consumers from the ten states she has identified under the CLRA, which offers numerous ways to prove an unfair business practice some of which have a scienter requirement in place, such as "intent not to sell [goods] as advertised." Cal. Civ. Code § 1770(a)(9). This intent requirement aligns with the intent requirements of Michigan. Compare Mich. Comp. Laws Ann. § 445.903(1)(e) ("Representing that goods or services are of a particular standard, quality, or grade . . . if they are not"), with Cal. Civ. Code § 1770(a)(7) ("Representing that goods . . . are of a particular standard, quality, or grade . . . if they are not."); see also Brownlow v. McCall Enters., Inc., No. 325843, 2016 WL 1576919 (Mich. Ct. App. Apr. 19, 2016) (rejecting the contention that plaintiffs had to prove the defendant had actual knowledge of the misrepresentation or reckless disregard for the truth when asserting goods were advertised as having qualities they did not).

The trouble for Sonner is that, in contrast to the CLRA, "[t]he UCL does not impose a scienter requirement." Lazebnik v. Apple, Inc., No. 5:13-CV-04145-EJD, 2014 WL 4275008, at *7 (N.D. Cal. Aug. 29, 2014); see also Cortez v. Purolator Air Filtration Prods. Co., 23 Cal.4th 163, 181 (2000) ("[A] plaintiff need not show that a UCL defendant intended to injure anyone through its unfair or unlawful conduct."). The fact Illinois, Michigan, and Minnesota impose any intent requirement at all therefore renders Michigan's law materially in conflict with the UCL. As explained above, California's interest in the application of its law to transactions occurring in Michigan is attenuated, and therefore Michigan's law should govern these potential claims.

3. Remedies Available

The remedies the CLRA and UCL offer vary considerably. The UCL offers limited remedies—restitution and injunctive relief. In re Vioxx Cases, 180 Cal.App.4th 116, 131-32 (2009), whereas the CLRA offers plaintiffs a chance to obtain actual and punitive damages, restitution, injunctive relief, and mandatory attorney fees for a prevailing plaintiff. Cal. Civ. Code §§ 1780(a), (e). Premier has identified differences in the way all fifty states have chosen to compensate victims of misleading advertising. Some states, like California, award punitive damages, while some do not. Other states agree with the policies underlying awards of punitive damages, but choose instead to authorize treble damages. Moreover, some states take different approaches to the calculation of actual damages. All of these differences, Premier argues, are material and preclude application of California law to the proposed ten-state group. Sonner lumps the two statutes together without differentiating the two, an unsatisfactory approach—particularly in light of the variations between remedies offered under the ten state statutes.

a. Punitive Damages

Premier points out that Florida and Michigan do not affirmatively permit plaintiffs to recover punitive damages; each of these states remains silent on the issue. See Fla. Stat. Ann. § 501.211(2); Mich. Comp. Laws Ann. § 445.911; Minn. Stat. §8:31(3a). State and federal courts interpreting these statutes have read this silence to mean punitive damages are unavailable under these statutes. Nowicki-Hockey v. Bank of Am., N.A., 593 F. App'x 420, 421 (6th Cir. 2014) ("[W]e note that the MCPA lacks express language permitting exemplary damages. . . . The absence of such language in the authorizing statute—or clear legislative intent—precludes an award of exemplary damages under Michigan law." (citation omitted)); LaFerney v. Scott Smith Oldsmobile, Inc., 410 So.2d 534, 535-36 (Fla. Dist. Ct. App. 1982) ("[T]he court reversed an attorney fee award under [Fla. Stat. Ann. §] 501.2105 because it included time spent on a punitive damage claim, which is outside the scope of Chapter 501."); In re Lutheran Bhd. Variable Ins. Prods. Co. Sales Practices Litig., No. 99-MD-1309(PAM/JGL), 2004 WL 909741, at *7 (D. Minn. Apr. 28, 2004) ("Because the statute that gives rise to Plaintiffs' cause of action contains no authorization for punitive damages, the Court finds that punitive damages are not available in actions based on the CFA.").

Sonner apparently concedes the statutes that do not permit punitive-damages awards materially conflict with the CLRA, but insists California's interests outweigh those of Florida, Michigan, and Minnesota. To support this contention, she relies on the fact California believes strongly in awarding punitive damages for violations of the CLRA as a mechanism to remedy a public wrong. Indeed, at least one district court concluded "punitive damages are . . . a `fundamental' part of the [CLRA's] statutory scheme." Walter v. Hughes Commc'ns, Inc., 682 F.Supp.2d 1031, 1041 (N.D. Cal. 2010). In contrast, Sonner contends, the legislatures of Florida, Michigan, and Minnesota are agnostic about the value of punitive damages, which she views as tantamount to indifference.

"Punitive damages may properly be imposed to further a State's legitimate interests in punishing unlawful conduct and deterring its repetition." BMW of N. Am., Inc. v. Gore, 517 U.S. 559, 568 (1996). Our federal system allows states to reach different conclusions about whether to authorize punitive damages and when to permit awards of such damages. See id. at 568-72 (describing the various consumer protection statutes as "a patchwork of rules representing the diverse policy judgments of lawmakers in 50 States."). The silence of Florida, Michigan, and Minnesota should not be taken as an expression of indifference. In many respects, their silence speaks volumes: they have chosen not to adopt California's preferred mechanism for protecting consumers. States need not affirmatively state they decline to expose business to the threat of punitive damages for it to be so. Moreover, states may choose to limit corporations' liability in an effort to "attract[] foreign businesses, with resulting increase in commerce and jobs" by limiting liability and punishment. Mazza, 666 F.3d at 592. Thus, the interests of those states that choose not to permit punitive damages overshadow California's separate interest.

b. Treble Damages

Premier further notes Massachusetts and Washington have chosen to make available treble damages instead of punitive damages in the appropriate circumstances.3 New Jersey has chosen to make treble-damages awards mandatory. See N.J. Stat. Ann. § 56:8-19. Sonner suggests there is no true conflict between punitive damages in cases of oppression,4 fraud, or malice,5 Trammell v. W. Union Tel. Co., 57 Cal.App.3d 538, 557 (1976), and treble damages for willful or knowing violations because both systems are designed to punish more culpable actors and to deter future misconduct, see Kraft Power Corp. v. Merrill, 981 N.E.2d 671, 685-86 (Mass. 2013) ("Like punitive damages in tort actions, multiple damages . . . can no longer achieve the goals of punishing a defendant or deterring him from future misconduct when the wrongdoer has died. . . .").

Despite some similarities, treble damages differ from punitive damages in one significant way: there is a set maximum amount a plaintiff may receive. In Washington, for example, the Consumer Protection Act caps treble damages at $25,000. Wash. Rev. Code §19.86.090. Washington's decision to limit treble damages reflects a policy decision to limit the risk of advertising and selling in an effort to encourage businesses to offer products within its borders. Thus, as in Mazza, application of California's law would impair the policy preferences of the states where the advertising occurred. See 666 F.3d at 594 (concluding the state where the communication of advertisements took place have an interest in the application of their laws to transactions between their citizens and businesses operating in the state). Here, that would mean the interests of Massachusetts, New Jersey, and Washington outweigh those of California.6

c. Actual Damages vs. Restitution7

The final material differences at issue concern restitution and actual damages and the different approaches to calculating actual damages. The UCL permits plaintiffs to recover only restitution—the "difference between what was paid and what a reasonable consumer would have paid at the time of purchase without the fraudulent or omitted information." Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979, 989 (9th Cir. 2015).8 In contrast, Washington does not; it awards only actual damages under their consumer protection laws. See Paris v. Steinberg & Steinberg, 828 F.Supp.2d 1212, 1217 (W.D. Wash. 2011) (restitution is permitted only in actions the attorney general initiates) (citing Wash. Rev. Code § 19.86.080; State v. Ralph Williams' N. W. Chrysler Plymouth, Inc., 510 P.2d 233, 241-42 (Wash. 1973)).

Washington's preference for actual damages creates a conflict with the UCL, but not the CLRA, which permits recovery of actual damages. In her motion for class certification, however, Sonner focused on whether restitution could be determined on a classwide basis and did not offer argument or evidence with respect to actual damages. Accordingly, there is no basis to conclude she seeks actual damages under the CLRA, and the conflict remains.9

In sum, Premier has carried its burden to show that California's interest in applying its law to the consumers in the specific ten states is weak. Sonner's proposed ten-state class is therefore not appropriate under the facts and circumstances of this case.

IV. CONCLUSION

The facts and circumstances presented in this case compel the conclusion that neither a nationwide class nor a ten-state class is appropriate. Accordingly, the class will remain limited to California consumers.

IT IS SO ORDERED.