Snow v. Align Technology, Inc.

• Court: District Court, N.D. California
• Decided: November 29, 2023
• Docket: 3:21-cv-03269
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA

SIMON AND SIMON, PC, et al., Case No. 20-cv-03754-VC Plaintiffs, v. Re: Dkt. No. 316 ALIGN TECHNOLOGY, INC., Defendant.

MISTY SNOW, et al., Case No. 21-cv-03269-VC Plaintiffs, v. Re: Dkt. No. 389 ALIGN TECHNOLOGY, INC., Defendant. ORDER GRANTING IN PART AND DENYING IN PART THE MOTIONS FOR CLASS CERTIFICATION; DENYING MOTIONS TO EXCLUDE DR. SINGER AND DR. VOGT

The motion for class certification is granted in part and denied in part. The motions to exclude the plaintiffs’ experts are denied. This ruling assumes the reader is familiar with the facts, the applicable legal standards, and the arguments made by the parties. Align makes three categories of arguments in opposition to class certification in both actions. First, it argues that the alleged exclusive dealing agreements—the Fusion program, the Advantage program, and the Dental Services Organization contracts—are inappropriately incorporated into the expert models that purport to show common antitrust injury and damages. Second, it argues that other flaws with the expert models render them unreliable and the expert testimony excludable, or, at least, render them so flawed that the questions they purport to answer have not been shown to be capable of class-wide resolution. Third, it argues that various issues with the assorted named plaintiffs render them unable to satisfy the typicality or adequacy requirements of Rule 23. None of these arguments defeats class certification, with one exception: neither of the direct purchaser plaintiffs purchased a scanner within the class period, and thus there is no typical scanner purchaser plaintiff. Certification of a scanner purchaser class is therefore denied. Class certification is granted for the class of direct purchasers of aligners, the injunctive relief class of indirect purchasers of aligners, and the state-law damages classes of indirect purchasers of aligners. I At the motion to dismiss stage, the plaintiffs in these related cases were allowed to proceed on two different theories. First, that the termination of interoperability was a refusal to deal that amounts to a Section 2 violation. Call this the TOI-only theory. Simon and Simon, PC v. Align Technology, Inc., 533 F.Supp.3d 904, 913–15 (N.D. Cal. April 2021).1 Second, that various exclusive dealing agreements can be understood in combination with the termination of interoperability to be an even greater Section 2 violation. Call this the combined theory. Id. at 918.The Court clearly held that the alleged exclusive dealing agreements did not state a separate or independent Section 2 claim—they were deemed actionable only in combination with the termination of interoperability. Id. The experts properly incorporated the exclusive dealing agreements into their common impact and common damages models.2 Both experts developed their own criteria for determining

1 The Section 2 ruling from the direct purchaser action was also applied in the indirect purchaser action. Snow v. Align Technology, Inc., No. 21-cv-03269-VC, 2022 WL 468703, at *1 (N.D. Cal. Feb. 16, 2022). 2 The parties dispute whether the contracts in question can be properly characterized as exclusionary. This ruling adopts the terminology “exclusionary agreements” for ease of reference to this category of alleged conduct—it does not express an opinion on the underlying question of whether the agreements were in fact exclusionary. whether a particular contract had the potential to cause anticompetitive effect as part of the overall scheme. Align argues that these criteria were inappropriate because the experts did not determine that each exclusionary agreement that they considered was itself illegal under Section 2.E.g., Simon & Simon, Dkt. No. 323 at 8-9, 22-24 (critiquing the plaintiffs’ expert for not performing a discount attribution test on the bundled discounts); Snow, Dkt. No. 422 at 23-26 (critiquing the plaintiffs’ expert for not focusing on the short-term nature or easy terminability of certain discount programs). But both experts offered opinions about the exclusionary agreements that were consistent with the theory permitted by this Court’s prior order: they assessed the anticompetitive effect of and damages from the exclusionary agreements in combination with the TOI as part of a broader Section 2 claim. In the direct purchaser action, Dr. Singer performed two regressions. First, the TOI-only model, which found that aligner prices would have been 7.1 percent lower absent the TOI. Second, the combined model, which found that aligner prices would have been 10 percent lower absent the TOI and certain exclusionary agreements. To be more precise, the combined model involved two regression models: an exclusivity-share analysis (which assessed how the TOI and the exclusionary agreements affected Align’s exclusive share of the aligner market), and then a price analysis (which assessed how the exclusivity share affected the resulting prices). At no point did Singer merely assume that the exclusionary agreements were either illegal or anticompetitive—instead, his combined model measures their anticompetitive effect, which is what makes them cognizable as part of the broader Section 2 claim. And the TOI-only model is offered as a backup, as one way of measuring antitrust impact and damages in the event the jury disagrees with Singer and concludes that only the termination of interoperability (and not the exclusionary agreements) had anticompetitive effect. Singer’s models thus map directly onto the two theories of liability. In the indirect purchaser action, Dr. Vogt performed a benchmark analysis and a passthrough analysis. The benchmark compared a similar market—the market for dental implants—to the aligner market, and based on that analysis, he concluded that prices in the aligner market should have fallen by more than 10% but instead rose by 5% after key patents expired. The passthrough demonstrated that the increase in prices paid by direct purchasers was passed on to individual consumers at a consistent rate. The key thing to understand about Vogt’s analysis is that the benchmark compares the real world to a world without any of Align’s alleged anticompetitive conduct. Thus, if the jury were to reject Vogt’s conclusion that the exclusive dealing agreements had anticompetitive effect, then the benchmark would still function to capture the anticompetitive effect of the remaining conduct at issue—it would just indicate that the remaining conduct had greater impact, since the exclusionary agreements had none. Vogt’s models also map on to the two theories of liability. Align’s main argument against class certification is that the common impact and common damages models are incapable of differentiating legal and illegal conduct and thus cannot support class certification. This is based on Comcast Corp. v. Behrend, 569 U.S. 27 (2013). In that case, the Supreme Court held that class certification was inappropriate where the district court had only permitted one out of four antitrust theories to proceed, but the common damages model included all four theories and could not differentiate between them. Id. 30–32. In this case, the Comcast-style argument just begs the question. Align says that the models cannot include the exclusionary agreements because those agreements were lawful, and that the agreements were lawful because they were not antitrust violations in isolation—but the motion to dismiss ruling says that is not the way to look at it. Rather, as explained in that ruling, to the extent the exclusionary agreements had anticompetitive impact, they can be considered actionable as part of the broader antitrust scheme, even if they could not, on their own, be deemed a Section 2 violation. So at the class certification stage, liability and damages are common questions capable of class-wide resolution; those questions include whether the TOI caused antitrust injury, what damages are traceable to that conduct, whether the TOI in combination with exclusionary agreements caused antitrust injury, and what damages are traceable to that conduct. None of those questions require a determination as to whether individual exclusionary agreements (or particularly types of exclusionary agreements) would be Section 2 violations when viewed in isolation. At summary judgment, Align is free to argue that—as a legal matter—the exclusionary agreements may not be considered as part of a broader Section 2 claim even if they have some anticompetitive effect as part of an overall scheme. That argument seems dubious, but if the Court were to accept it at summary judgment, then the implications for the ability of the plaintiffs to proceed can be addressed at that time. But that possibility does not change the reality that common questions predominate now.3 Relatedly, at trial, Align may try to convince the jury that the exclusionary agreements did not have antitrust impact and thus should not contribute to liability. But that also does not negate predominance. That goes to whether the expert opinions will ultimately convince the jury, not whether the questions the opinions seek to answer can be answered on a class-wide basis. And anyway, both experts have a viable method of calculating damages from just the termination of interoperability if the jury were to conclude that the exclusionary agreements lacked anticompetitive effect. The plaintiffs in both actions have thus met their burden under Comcast: they have shown that their models match their liability theories. See Singh v. Google LLC, No. 16-cv-03734-BLF, 2022 WL 94985, at *13 (N.D. Cal. Jan. 10, 2022) (“Courts are required to ‘conduct a rigorous analysis to ensure at the class certification stage that any models supporting a plaintiff’s damage case is consistent with its liability case.’” (quoting Comcast, 569 U.S. at 35)); Vaquero v. Ashley

3 In the unlikely event that the Court eliminates the exclusionary agreements as a legal matter, while allowing that they may have some anticompetitive effect, the direct purchaser action would still be amenable to class treatment. Singer has the combined model that enables him to “turn off” the exclusionary agreements and measure the TOI alone, thus controlling for any impact of the exclusionary agreements. (At the hearing, he satisfied the Court that this would merely be a mechanical calculation drawn from models already vetted.) The indirect purchaser action might face a greater challenge. As noted, Vogt’s model compares the real world to a hypothetical world with none of the anticompetitive conduct, and so it cannot differentiate the effect of the TOI from the effect of the exclusionary agreements. But for now, the main point is that this scenario is highly speculative. Sometimes a case will take an unlikely legal turn that requires decertification of a class, but, at least on this record, it would not be appropriate to deny certification of an indirect purchaser class at the front end based on speculation about a legal ruling that seems unlikely to come. Furniture Industries, Inc., 824 F.3d 1150, 1154 (9th Cir. 2016) (“We have interpreted Comcast to mean that ‘plaintiffs must be able to show that their damages stemmed from the defendant’s actions that created the legal liability.’” (quoting Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979, 987–88 (9th Cir. 2015)). The plaintiffs are not required by Comcast to show that their models match other liability theories that the defendant may argue for at summary judgment. II At the class certification stage, the district court “must make a rigorous assessment of the available evidence and the method or methods by which plaintiffs propose to use the class-wide evidence to prove the common question in one stroke.” Olean Wholesale Grocery Cooperative, Inc. v. Bumble Bee Foods LLC, 31 F.4th 651, 666 (9th Cir. 2022) (quotations and brackets omitted). “Put another way, the inquiry must be to determine if the proffered expert testimony has the requisite integrity to demonstrate class-wide” effect. In re Optical Disk Drive Antitrust Litigation, No. 3:10-MD-2143 RS, 2016 WL 467444, at *6 (N.D. Cal. Oct. 18, 2019). But that inquiry is “limited to resolving whether the evidence establishes that a common question is capable of class-wide resolution, not whether the evidence in fact establishes that plaintiffs would win at trial.” Olean, 31 F.4th at 667 (emphasis in original). And “a district court cannot decline certification merely because it considers plaintiff’ evidence relating to the common question to be unpersuasive and unlikely to succeed in carrying the plaintiffs’ burden of proof on that issue.” Id. After examining the expert reports, considering the arguments of both parties, and conducting a two-day evidentiary hearing at which both Singer and Vogt testified for several hours, the Court concludes that the expert opinions in both actions are reliable and capable of providing class-wide answers to the questions central to the plaintiffs’ claims. Align has offered reasons that the models are flawed, and “those arguments may appropriately be considered by the trier of fact,” but they “do not rise to a level that would require denial of certification.” Optical Disk Drive, 2016 WL 467444, at *4. The primary objection to Singer’s analysis is that he measures the effect of the termination of interoperability one year after it occurred. Align argues that Singer assumes the one-year gap only because the price effects do not show up immediately or even three months after. Moreover, Align argues that if it had market dominance, then it would have had the “ability and incentive to immediately raise prices,” and so the impact should have occurred sooner. Simon & Simon, Dkt. No. 323 at 26. But Singer explained—in his initial report, in his rebuttal report, and at the hearing—that he did not merely assume that there would be a one-year lag. Rather, he hypothesized the one-year lag, then tested it. And he gave cogent reasons for his hypothesis: First, that Align feared market backlash over the termination of interoperability and wanted to create distance between that policy change and the price increase to avoid more ill will. Second, that Align needed time to monitor the effect of the termination of interoperability to ensure that it could support the price increase. Third, that Align changed its list prices on set, infrequent intervals. And fourth, that Align needed to adjust its discount programs to raise prices, and a key change happened about a year after the termination of interoperability. Moreover, Singer cited literature about lagged price effects. Simon & Simon, Dkt. No. 337 at 20. Singer’s justifications may not carry the day—the jury may be persuaded that the price effect showing up later makes the model less convincing or that Singer’s analysis was too results-oriented to be credible. But the methodology is sufficiently reliable to support a classwide showing of antitrust impact and damages. Align also argues that Singer’s regression analysis fails to control for key variables. In its opposition, Align asserts that Singer’s model “does not even control for inflation . . . or any other market-wide factors that could affect prices.” Simon & Simon, Dkt. No. 323 at 27. But Singer did include controls meant to capture inflation and other market-wide factors: the linear time trend and the county income variable. In the papers, Align did not identify any other factors that Singer missed. During cross-examination, Align’s counsel identified another possible issue with one of Singer’s models: the TOI-only model does not control for the potential effect of the exclusionary agreements. It’s true that normally it’s important to control for any variable that could have a significant impact on the thing being measured; and Singer’s combined model supports a conclusion that the exclusionary agreements did have anticompetitive impact. In response to this line of attorney questioning, Singer explained that his combined model can be used to assess just the effects of termination of interoperability, controlling for the exclusionary agreements. Although Singer did not include those calculations in his expert report, he explained how to produce that result with only mechanical changes to the same combined model that has been vetted by the parties and the Court. The TOI-only model is just an alternative—if the jury disagrees with Singer’s conclusion that the exclusionary agreement had anticompetitive effect, then the TOI-only model remains a reliable way of measuring the effects of the termination of operability alone (because you don’t need to control for a variable that you already know has no impact). Perhaps the failure to control for the effect of the exclusionary agreements in the TOI- only model would be an issue if the Court were to rule that those agreements may not be considered even if a jury could conclude that they had anticompetitive effect. But as discussed in the preceding section, that scenario is highly speculative, and not a basis for excluding the TOI- only model at the front end. Similarly, perhaps the failure to control for the effect of the exclusionary agreements would be an issue if a jury found that some of those agreements had anticompetitive effect while others did not, but again that is highly speculative. In any event, this issue was not so much as mentioned in Align’s briefs in opposition to class certification or in support of its motion to exclude Singer.4 Because of that, the issue is severely underdeveloped. At the pretrial conference (assuming the case gets past summary judgment), the Court would consider a renewed motion to exclude Singer's TOI-only model based on the failure to control for the effects of the exclusionary agreements in the event this issue were somehow relevant at that stage.5

4 There is brief mention of the issue in paragraph 227 of one of Align’s many expert reports, but that is obviously not an adequate presentation. 5 Align’s other objections to Singer’s analysis are even less convincing. Some examples: Align argues that one model produced certain nonsensical results, such as suggesting that higher costs lead to lower prices for the scanner. But the rebuttal report explains that unexpected relationships among control variables is not an issue for a regression if the key variable is unaffected. Align The primary objection to Vogt’s analysis is that the dental implant market was an improper benchmark. Vogt explained that he was looking for a comparator market in which key patents expired and competitive entry followed. Moreover, these are similar products with similar customization, and several of the same firms entered both the dental implant market and aligner market as competitors. Additionally, in 2018, Align itself received a competitive intelligence report that compared the impact of expiring patents in the two markets, and that report is part of the record. Align raises several issues with the benchmark, some of which may be compelling to the jury, but none of which invalidates the comparison. See In re Blood Reagents Antitrust Litigation, MDL No. 09-2081, 2015 WL 6123211, at *22 (E.D. Pa. Oct. 19, 2015) (concluding that “a proposed yardstick must be rejected as inadmissible where the expert testimony is so deficient that the comparison is manifestly unreliable” which “generally occurs when an expert has failed to perform any substantive analysis of those factors most relevant to comparability”); id. at 22 n.18 (noting that “many courts [ ] have held that the question whether plaintiffs have met their burden of proving comparability should be left to the trier of fact to resolve because comparability challenges generally involve weighing facts”). A few of Align’s benchmark points warrant further discussion. For example, Align argues that the dental implant market had multiple high-end and low-end brands and so is an improper comparator for the prices of a single, well-known product (Invisalign) from a premium brand. The plaintiffs respond in a couple ways. First, they note that Vogt relied on analysis from another expert for his conclusions about the impact of brand recognition in both markets—he did not ignore it. Second, they note that Vogt’s point is that the patent expiration led to the

argues that the combined model lacks sufficient factual foundation because Singer deemed certain agreements exclusionary based on an exclusivity provision without checking whether Align enforced that provision. But, as Singer explained in his report with citations to economic literature, provisions like that can have effect even if unenforced. Singer adequately articulated rational criteria that he used for ruling in certain agreements and ruling out others. Align argues that Singer’s combined model is unreliable because it was contrary to law. But this is just the flawed Comcast argument in Daubert clothes. Align may make any and all of these arguments to try to convince the jury that Singer is wrong. But none of them render his opinion so flawed as to be unreliable. emergence of a value segment in the implant market because of the lack of a comprehensive anticompetitive scheme to suppress it, and thus that the absence of that value segment is a result of the challenged conduct. Align also argues that because Vogt was unaware of and did not consider certain anticompetitive conduct in the dental implants market, his benchmark is flawed. But, as Vogt observed, the asserted anticompetitive conduct in the dental implants market was far more limited than the conduct at issue in this action: there was no comprehensive scheme. He expressed skepticism about the conduct in the dental implant market having had any material impact on prices—and Align certainly hasn’t shown or even really argued that the price impact would have been so substantial as to affect the integrity of the comparison. But perhaps more importantly, even if the comparator market did have some anticompetitive conduct that kept prices artificially high, that would only mean that Vogt’s benchmark analysis underestimated the impact of Align’s anticompetitive conduct. Ultimately, as Vogt explained at the hearing, the benchmark provides the basis for a legitimate estimate. This is unlike In re Live Concert Antirust Litigation, 863 F. Supp. 2d 966, 974–76 (C.D. Cal. 2012), in which a district court ruled that a yardstick analysis was unreliable because it compared average rock concert prices in Los Angeles and Denver without giving any real consideration to the quality or popularity of artists promoted during the relevant time periods. In that case, the defendants’ expert also performed an analysis that demonstrated the predictiveness of artist popularity on ticket prices. That is a glaring flaw, unlike the more nitpicky differences identified in this case. Dkt. No. 422 at 27. Moreover, this is the only case that Align cites in which a district court excluded a benchmark analysis due to dissimilarities between the markets. The weight of authority seems to favor allowing the jury to determine whether dissimilarities are salient. Align can argue that all or some of the difference that Vogt found and attributed to anticompetitive conduct can actually be attributed to Align’s marketing skill or product quality. Align can also try to argue that the anticompetitive conduct in the dental implant market muddies the waters in a way that makes the benchmark less persuasive. But the benchmark is still a reliable way of measuring class-wide impact and providing a basis for damages. Align also takes issue with Vogt’s passthrough analysis. But the plaintiffs have shown that the passthrough analysis is consistent with qualitative evidence, is founded on substantial data, and was formed using reliable methods. Vogt has adequately explained his use of averages and his decision not to look at a single dental practice. Vogt also performed an analysis on a data set with product differentiation that supported his findings from the other data. The plaintiffs have provided a reliable opinion that there was a consistent passthrough of prices from direct to indirect purchasers, and thus have shown that indirect purchaser damages are susceptible to class-wide proof. III The class of scanner purchasers cannot be certified because there is no named plaintiff who purchased a scanner within the class period. The plaintiffs argue that this is not an issue: even if the named plaintiffs didn’t purchase scanners, they purchased aligners; and antitrust classes often include purchasers of multiple products that were all affected by the same anticompetitive scheme. See, e.g., In re Static Random Access (SRAM) Antitrust Litigation, No. C 07-01819 CW, 2008 WL 4447592, at *3 (N.D. Cal. Sept. 29, 2008) (allowing certification of a class of fast SRAM and slow SRAM purchasers challenging a price-fixing scheme where the named plaintiff only purchased fast SRAM). Determining whether named plaintiffs have typical claims in antitrust class actions is a case-by-case inquiry, and it may sometimes be a close question. See In re Graphics Processing Units Antitrust Litigation, 253 F.R.D. 478, 484–89 (N.D. Cal. July 2008) (surveying different approaches). But a plaintiff alleging an antitrust violation based on aligner purchases is too different from a plaintiff alleging an antitrust violation based on scanner purchases. Those plaintiffs suffered different injuries, and they must prove different things to recover. A plaintiff who purchased aligners must prove monopoly power, anticompetitive conduct, and anticompetitive impact in the aligner market, but that plaintiff is not necessarily required to prove any of those things with respect to the scanner market. So this is not a case where the named plaintiffs, as aligner purchasers, would be unable to “prove their own claims without proving those of the class” of scanner purchasers. Pecover v. Electronic Arts Inc., No. C 08-2820 VRW, 2010 WL 8742757, at *12 (N.D. Cal. Dec. 21, 2010). It’s one thing to certify a class where the named plaintiff might need to prove an additional element to recover as compared to absent class members. See id. at *13. But the reverse situation is highly problematic from a class certification standpoint. Nor is it enough that the expert reports are already commissioned and cover both scanner and aligner purchasers: the fact that plaintiffs’ counsel has developed evidence that can establish claims beyond those asserted by the plaintiff representatives does not make the plaintiffs any more representative. In fact, Singer’s expert report confirms the problem; it has different, dedicated sections for dominance and injury in the two markets. And even if the markets and products were affected by the same scheme, they were not necessarily affected the same way. Ultimately, the scanner and aligner products are different enough that certification of a class including scanner purchasers without any scanner purchaser representative is inappropriate under Rule 23. At the hearing regarding typicality, the Court inquired whether denial of class certification for the scanner purchasers should be without prejudice, which would allow the plaintiffs to seek leave to add a named plaintiff who purchased a scanner during the class period and then file a renewed motion for certification of that class. As the Court explained, this would require the trial date for the aligner class to be pushed back substantially, because it would be unfair to Align to require trial (not to mention adjudication of summary judgment motions) to occur on separate tracks for the two portions of the class. Plaintiffs’ counsel responded that, under those circumstances, denial of class certification for the scanner purchasers should be without leave to refile. Counsel asserted that it is in the best interest of the aligner class (whose members overlap substantially with the proposed scanner class) to proceed to a prompt trial. The Court is okay with allowing counsel to make that decision because there is no prejudice to the absent scanner purchasers: There is nothing preventing a future lawsuit by scanner purchasers, in which class certification could well be appropriate. There just will not be class certification in this action (at least, assuming no other unexpected delays in the litigation schedule). The other differences between the named plaintiffs and the potential class members that Align identifies are not material to the case. See In re JUUL Labs, Inc., Marketing Sales Practices and Products Liability Litigation, 609 F. Supp. 3d 942, 957-58 (N.D. Cal. June 2022). In the direct purchaser action, Align offers a list of characteristics of the named plaintiff dental practices that may differ from other class members (e.g., they do not treat patients with braces, they refer some patents to orthodontists). Align then asserts that the plaintiffs have not carried their burden of showing typicality because the plaintiffs have not determined whether these traits are pervasive among the class members. But these differences are not relevant to the claims in this class action. The direct purchaser named plaintiffs will have to prove the same things as the class members that they seek to represent. There is no reason to believe that the identified differences would skew the named plaintiffs’ interests or incentives relative to the rest of the class. The same is true of the traits of the indirect purchaser named plaintiffs that Align pulls out (e.g., still owing some money for an aligner purchase, having a conviction for underage alcohol possession, being employed by a court reporting company that has apparently contracted with class counsel in the past). Cummings v. Connell, 316 F.3d 886, 896 (9th Cir. 2003) (noting that the Ninth Circuit “does not favor denial of class certification on the basis of speculative conflicts”). Similarly, there is no issue with plaintiff Vo representing insurers in the nationwide injunctive relief class: her claims are sufficiently like theirs for the purposes of seeking an end to the anticompetitive conduct. Additionally, there is still no standing issue for plaintiff Vo. She has said that she would like to purchase Invisalign for her child in the future when financially ready and has taken steps toward doing so. See Snow, Dkt. No. 448 at 26 n.77 (collecting record citations). The motion to dismiss ruling held that this was sufficient imminence of injury for standing to seek injunctive relief. Snow v. Align Technology, Inc., No. 21-CV-03269-VC, 2022 WL 468703, at *1 (N.D. Cal. Feb. 16, 2022). Nothing in the record has altered that conclusion. Finally, the plaintiffs have shown that the named plaintiffs are sufficiently familiar with and involved in the action to be adequate class representatives. Snow, Dkt. No. 337 at 29-35 (Appendix A). A few gaps in knowledge or memory do not change that. Snow, Dkt. No. 422 at 38. Thus, the plaintiffs have demonstrated both adequacy and typicality for the remaining classes. IV In light of the foregoing, the following classes are certified: e Direct Purchasers Class. All persons or entities in the United States that purchased Invisalign Aligners directly from Defendant Align Technology, Inc. during the period beginning January 1, 2019 through March 31, 2023 (the “Class Period”). Excluded from the Class are: (1) Defendant, its subsidiaries, affiliate entities, and employees; and (2) all federal or state government entities or agencies. e Nationwide Injunctive Relief Indirect Purchaser Class. All persons or entities in the United States that purchase, pay, and/or provide reimbursement for some or all of the purchase price for Invisalign aligners acquired for personal use, until such time as the anticompetitive conduct alleged herein has ceased (the “Class Period”). Excluded from the Class are: (1) Defendant, its subsidiaries, affiliate entities, and employees; and (2) all federal or state government entities or agencies. e State Indirect Purchaser Classes. All persons that purchased, paid and/or provided reimbursement for some or all of the purchase price for Invisalign Treatments that used any of the following types of Invisalign products—“Comprehensive”, “Moderate”, “GO”, “Teen”, “Assist Product Tracking”, “Lite”, “Express Ten”, “Multi-stage Comprehensive”, or “Assist” treatment—while residing in a relevant state, acquired for personal use during the period beginning July 1, 2018 until such time as the anticompetitive conduct alleged herein has ceased (the “Class Period”). The relevant states are: Arizona, California, Maryland, Massachusetts, Michigan, Minnesota, Nebraska, Nevada, North Carolina, and Oregon. Excluded from the Class are: (1) insurance providers; (2) Defendant, its subsidiaries, affiliate entities, and employees; and (3) all federal or state government entities or agencies. IT IS SO ORDERED. Dated: November 29, 2023 Ze VINCE CHHABRIA United States District Judge