Andrei Fenner v. General Motors, LLC

• Court: Court of Appeals for the Sixth Circuit
• Decided: August 21, 2024
• Docket: 23-1697
• Status: Published

Opinion

RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 24a0190p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

┐ ANDREI FENNER et al., │ Plaintiffs, │ │ PHILLIP BURNS, et al. (23-1648); NANCY ANDERTON, │ et al. (23-1696); MIKE BULAON, et al. (23-1697); > Nos. 23-1648/1696/1697/1698 │ TAYLOR PANTEL, et al. (23-1698), │ Plaintiffs-Appellants, │ │ v. │ │ │ GENERAL MOTORS, LLC; ROBERT BOSCH GMBH; │ ROBERT BOSCH LLC, │ Defendants-Appellees. │ ┘

Appeal from the United States District Court for the Eastern District of Michigan at Bay City. No. 1:17-cv-11661—Thomas L. Ludington, District Judge.

Argued: May 9, 2024

Decided and Filed: August 21, 2024

Before: MOORE, KETHLEDGE, and BLOOMEKATZ, Circuit Judges.

_________________

COUNSEL

ARGUED: Steve W. Berman, HAGENS BERMAN SOBOL SHAPIRO, LLP, Seattle, Washington, for all Appellants. Jay P. Lefkowitz, KIRKLAND & ELLIS LLP, New York, New York, for Appellee General Motors. ON BRIEF: Steve W. Berman, HAGENS BERMAN SOBOL SHAPIRO, LLP, Seattle, Washington, E. Powell Miller, THE MILLER LAW FIRM, P.C., Rochester, Michigan, for the Burns Appellants. Jay P. Lefkowitz, KIRKLAND & ELLIS LLP, New York, New York, Renee D. Smith, Jeffrey S. Bramson, Cole T. Carter, KIRKLAND & ELLIS LLP, Chicago, Illinois, for Appellee General Motors. Eric D. Pearson, Charles W. Miller, HEYGOOD, ORR & PEARSON, Irving, Texas, for the Anderton, Bulaon, and Patel Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 2

Appellants. Jonathan S. Martel, ARNOLD & PORTER KAYE SCHOLER LLP, Washington, D.C., for Amici Curiae.

MOORE, J., delivered the opinion of the court in which BLOOMEKATZ, J., joined in full. KETHLEDGE, J. (pp. 25–29), delivered a separate opinion concurring in the judgment in part and dissenting in part. _________________

OPINION _________________

KAREN NELSON MOORE, Circuit Judge. Plaintiffs are a group of consumers who purchased or leased a model year 2011–2016 GM Silverado or Sierra 2500 or 3500. Plaintiffs allege that they selected and ultimately purchased or leased their vehicles, at least in part, because of the Duramax diesel engine and systems therein, as advertised and represented by GM. In advertisements for the subject vehicles, GM claimed that the vehicles ran “clean diesel,” had “low emissions,” had “a whopping 63%” reduction of “Nitrogen Oxide (NOx) emissions” when compared to previous models and turned “heavy diesel fuel into a fine mist.” GM omitted any reference to how—or when—its emissions system worked to accomplish these “clean diesel” imperatives. Contrary to GM’s advertisements, however, Plaintiffs allege that the subject vehicles actually emit NOx and other pollutants at levels many times higher than (i) their counterparts, (ii) what a reasonable consumer would expect, (iii) what GM advertised, (iv) the Environmental Protection Agency’s emissions standards, and (v) the levels set for the vehicles to obtain a certificate of compliance that allows them to be sold in the United States. On those bases, Plaintiffs brought this action against General Motors LLC, Robert Bosch GMBH, and Robert Bosch LLC, alleging violations of state consumer protection, fraud, and deceptive trade practices laws, as well as the Racketeer Influenced and Corrupt Organizations (RICO) Act. Defendants filed motions for summary judgment on all claims. The district court granted summary judgment, finding that (1) Plaintiffs’ state-law claims were preempted by the Clean Air Act, and (2) Plaintiffs did not have standing to bring a RICO action. Because Plaintiffs’ state- law claims are not impliedly preempted by the Clean Air Act, we REVERSE the district court’s grant of summary judgment on the state-law claims. Because Plaintiffs are indirect-purchasers Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 3

and thus do not have standing under RICO, however, we AFFIRM the district court’s grant of summary judgment on the RICO claims.

I. BACKGROUND

A. Factual Background

“Plaintiffs are a group of consumers who purchased or leased a model year 2011–2016 Chevrolet Silverado 2500HD or 3500HD, or a GMC Sierra 2500HD or 3500HD” (collectively “Duramax Trucks”). R. 444 (Summ. J. Order at 1) (Page ID #48701).1 Plaintiffs include both individuals and a putative class of consumers.2 Defendants are General Motors LLC (“GM”) and Robert Bosch GMBH and Robert Bosch LLC (collectively “Bosch”). GM manufactures the Duramax Trucks, whereas Bosch developed and manufactured engine components related to the emissions-control system for the Duramax Trucks. See R. 18 (First Am. Compl. ¶ 20) (Page ID #901).

The Duramax Trucks are equipped with Duramax diesel engines. See id. ¶ 108 (Page ID #962); D. 38 (Appellee Br. at 4). Whereas “gasoline engines require a spark from a spark plug to ignite fuel within the cylinders, diesel vehicles utilize a high level of compression to ignite the fuel.” R. 365-1 (Harrington Rep. at 118) (Page ID #21127). “This causes a more powerful compression of the pistons, which produces greater engine torque (that is, more power).” R. 18 (First Am. Compl. ¶ 56) (Page ID #937); see generally R. 365-1 (Harrington Rep. at 13–55) (Page ID #21022–64). In addition to having more power than gasoline engines, diesel engines “typically produce . . . more particulate matter (PM) and [oxides of nitrogen, also known as] NOx” than their gasoline counterparts. R. 365-1 (Harrington Rep. at 118–19) (Page ID #21127– 28).

1 As detailed below, this opinion addresses four consolidated cases. See infra Part I, Section B. Record citations refer to the lead case, Fenner et al. v. General Motors, LLC, et al., Case No. 23-1648, unless otherwise indicated. 2 Fenner et al. v. General Motors, LLC, et al., Case No. 23-1648 (“class action” or “Fenner class action”), was filed as a putative class action. The three individual cases at issue are Anderton et al. v. General Motors, LLC, et al., Case No. 23-1696 (“Anderton Case”), Bulaon et al. v. General Motors, LLC, et al., Case No. 23-1697 (“Bulaon Case”), and Pantel et al. v. General Motors, LLC, et al., Case No. 23-1698 (“Pantel Case”). Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 4

The Duramax Trucks utilize several mechanisms to reduce their NOx emissions. See R. 366-2 (Smithers Rep. at 6–14) (Page ID #21990–98). The Duramax Trucks also utilize auxiliary emission control devices (AECDs). See D. 38 (Appellee Br. at 6–7). “AECDs are a typical aspect of vehicle design used to modulate and control systems that impact vehicle emissions.” R. 365-1 (Harrington Rep. at 11) (Page ID #21020). AECDs “reduce[] the effectiveness of the emission control system,” in order to maintain other vehicle features, such as torque (i.e., power), or to “protect[] the vehicle against damage or accident.” 40 C.F.R. § 86.004-2.

As with all vehicles sold in the United States, before the Duramax Trucks could be introduced to the U.S. market, they were subject to extensive federal government regulations and testing pursuant to the Clean Air Act (CAA). See, e.g., 42 U.S.C. § 7525. A vehicle manufacturer must certify to the Environmental Protection Agency (EPA) that its vehicle meets federal emissions standards—and must obtain an EPA-issued certificate of conformity—before it can enter the U.S. market. Id. § 7525(a)(1).

In its application for a certificate of conformity, a manufacturer must disclose, describe, and justify to the EPA all AECDs that its vehicle utilizes. 40 C.F.R. § 86.1844-01(d)(11). The EPA then determines if the AECD has a legitimate purpose or if it is an illegal “defeat device,” i.e., a mechanism that unjustifiably turns emission controls down or off in certain circumstances. See id. § 86.004-2. The EPA is responsible for determining when an AECD is justified and thus permissible, or when it is unjustified, and thus an unlawful defeat device. Id. “GM made substantial and detailed disclosures of the [Duramax Trucks’] AECDs” in its certificate of conformity applications. R. 365-1 (Harrington Rep. at 11–12) (Page ID #21020–21). The Duramax Trucks all received EPA-issued certificates of conformity, indicating that the vehicles met EPA emissions standards. D. 38 (Appellee Br. at 7–8).

In marketing its Duramax Trucks, GM advertisements stated that the Duramax Trucks “turn heavy diesel fuel into a fine mist, burning cleaner and faster with lower emissions and greater power than the previous model,” R. 366-50 (Humphreys Rep. at 26) (Page ID #25538), and delivered “the latest emission control technology” that “reduc[ed] Nitrogen Oxide (NOx) emissions by a whopping 63%, when compared to” previous models, id. at App. D (Page ID #25594). GM advertisements also stated that the Duramax Trucks’ “[a]dvanced emission control Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 5

technology makes [it] one of the cleanest diesels in the segment.” Id. (Page ID #25593). GM omitted any reference to how—or when—the “[a]dvanced emission control technology” worked. Id. GM never informed consumers, for example, that the emission-control system that allowed for lower NOx emissions had “significantly reduce[d] . . . effectiveness . . . during real-world driving conditions.” R. 18 (First Am. Compl. ¶ 14) (Page ID #899). Along similar lines, GM failed to inform consumers that “high fuel economy, power, and durability” in real-world driving conditions were made possible only by utilizing AECDs to “reduc[e] emissions controls,” thus causing greater levels of NOx emissions. Id. ¶ 18 (Page ID #900); see R. 366-50 (Humphreys Rep. at App. B1, App. D) (Page ID #25562–64, 25593–98) (reviewing GM and Bosch advertisements and public communications about the Duramax Trucks). Stated otherwise, GM did not tell consumers that the emission control systems that it advertised would, under some conditions, shut off.

Consumers indicated that, based on these GM representations, they believed the Duramax Trucks to be “clean diesel” vehicles. See, e.g., R. 393-16 (Roberts Dep. at 195:19-24) (Page ID #38761). Consumers interpreted “clean diesel” to reflect several different things. Some consumers expected their “clean diesel” vehicles simply to comply with EPA emissions standards. See, e.g., R. 363-31 (Golden Dep. at 94–95) (Page ID #18928). Other consumers expected their “clean diesel” vehicles to have lower emissions than previous diesel models, see R. 393-16 (Roberts Dep. at 102:1-11, 174:2-8) (Page ID #38751, 38757); to have lower emissions than other vehicles on the market, see id. at 195:19-24 (Page ID #38761); R. 393-19 (Henderson Dep. at 79:11-15) (Page ID #38809); or to have lower emissions, “be cleaner[,] and run a lot better” than the older pickup trucks that they previously owned, see R. 393-19 (Henderson Dep. at 72–73) (Page ID #38806–07). Some consumers simply measured “clean diesel” visually, expecting their vehicles not to “blow the black smoke,” that diesel engines are sometimes known for. R. 393-12 (Mizell Dep. at 43:4-6) (Page ID #38675).

B. Procedural History

This case began in May 2017 when Plaintiffs Andrei Fenner and Joshua Herman filed a class action complaint against GM and Bosch. R. 1 (Compl.) (Page ID #1). Just one month later, in June 2017, eight additional Plaintiffs filed a complaint against the same Defendants. Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 6

See R. 16 (Consol. Order at 2) (Page ID #876). In July 2017, the cases were consolidated. Id. at 5 (Page ID #879). On August 4, 2017, Plaintiffs filed a First Amended and Consolidated Class Action Complaint. R. 18 (First Am. Compl.) (Page ID #884). The First Amended Complaint asserted claims under RICO and various state-law causes of action related to the emissions performance of Model Years 2011-2016 GM Silverado or Sierra 2500 or 3500. See generally id. Specifically, the complaint alleged:

In contrast to GM’s promises, . . . the Silverado and Sierra 2500 and 3500 models emit levels of NOx many times higher than (i) their gasoline counterparts, (ii) what a reasonable consumer would expect, (iii) what GM had advertised, (iv) the Environmental Protection Agency’s maximum standards, and (v) the levels set for the vehicles to obtain a certificate of compliance that allows them to be sold in the United States. Further, the vehicles’ promised power, fuel economy, and efficiency is obtained only by turning off or turning down emissions controls when the software in these vehicles senses they are not in an emissions testing environment.

Id. ¶ 2 (Page ID #892–93).

Both GM and Bosch moved to dismiss Plaintiffs’ claims, and, on February 20, 2018, the district court denied both motions to dismiss. R. 61 (Mot. to Dismiss Order) (Page ID #3473). Relevant here, the district court found that (1) Plaintiffs’ state-law claims were neither expressly nor impliedly preempted by the Clean Air Act, id. at 31, 33 (Page ID #3503, 3505), and (2) Plaintiffs sufficiently established standing under RICO, id. at 46 (Page ID #3518).

In May 2019, over 2700 plaintiffs (the “Anderton Plaintiffs”) filed twenty-six complaints against GM and Bosch. See R. 144 (Consol. Order at 7–8) (Page ID #5847–48). Like the consolidated class action, these complaints each “allege[d] violations of RICO and a multitude of state law fraud claims.” Id. In January 2020, the district court consolidated the Fenner class action and the Anderton Cases “for all purposes, except for trial.” Id. at 14 (Page ID #5854). In August 2020, these cases were consolidated with two additional sets of cases—the “Bulaon Cases” and the “Pantel Cases.” See R. 196 (Pantel Consol. Order at 12) (Page ID #11337); R. 198 (Bulaon Consol. Order at 12) (Page ID #11351).

On August 19, 2022, GM and Bosch filed motions for summary judgment in the consolidated cases. R. 365 (GM Mot. Summ. J.) (Page ID #20906); R. 373 (Bosch Mot. Summ. Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 7

J.) (Page ID #29812). As relevant here, Defendants argued that (1) the Clean Air Act preempted Plaintiffs’ state-law claims, and (2) Plaintiffs lacked standing to bring their RICO claims under the indirect-purchaser rule. See generally R. 365 (GM Mot. Summ. J.) (Page ID #20906); R. 373 (Bosch Mot. Summ. J.) (Page ID #29812).

While Defendants’ motions for summary judgment were pending, this court released In re Ford Motor Co. F-150 and Ranger Truck Fuel Economy Marketing and Sales Practices Litigation, 65 F.4th 851 (6th Cir.), cert. denied sub nom. Lloyd v. Ford Motor Co., 144 S. Ct. 332 (2023). Noting that the Ford case dismissed a “substantially similar claim as preempted by the Energy Policy and Conservation Act (EPCA),” the district court directed the parties “to show cause for why Plaintiffs’ state-law claims should not be dismissed.” R. 433 (Show Cause Order at 1) (Page ID #48077). On July 12, 2023, the district court granted summary judgment to Defendants, dismissing all claims “with prejudice because Plaintiffs’ state-law claims are impliedly preempted by the Clean Air Act, 42 U.S.C. § 7401 et seq., and Plaintiffs lack statutory standing for their RICO claim because they are indirect purchasers.” R. 444 (Summ. J. Order at 1) (Page ID #48701). This appeal followed.

II. ANALYSIS

A. Standard of Review

We review de novo a district court’s decision to grant summary judgment on the issues of preemption and standing. See State Farm Bank, FSB v. Reardon, 539 F.3d 336, 340 (6th Cir. 2008) (preemption); B & H Med., L.L.C. v. ABP Admin., Inc., 526 F.3d 257, 264 (6th Cir. 2008) (standing). We will affirm a grant of summary judgment only when “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). “In reviewing the district court’s decision to grant summary judgment, we must view all evidence in the light most favorable to the nonmoving party.” B & H Med., 526 F.3d at 264–65 (quoting Blair v. Henry Filters, Inc., 505 F.3d 517, 523 (6th Cir. 2007)). Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 8

B. State-Law Claims

Plaintiffs bring state-law claims under numerous varied state laws. These include claims under state consumer fraud laws, see, e.g., Ariz. Rev. Stat. § 44-1521 et seq.; Md. Code Ann., Com. Law § 13-101 et seq., and state unfair or deceptive trade practices laws, see, e.g., Nev. Rev. Stat. § 598.0903 et seq.; Minn. Stat. § 325D.43-48 et seq. The state-law claims also include common law claims sounding in breach of contract and fraudulent concealment. See, e.g., R. 18 (First Am. Compl. ¶ 325–49) (Page ID #1047–56). On appeal, Defendants argue that all of Plaintiffs’ state-law claims are preempted by the CAA.

1. Preemption Doctrine

The Supremacy Clause of the United States Constitution provides in relevant part that federal law “shall be the supreme Law of the Land . . . any Thing in the Constitution or Laws of any State to the Contrary notwithstanding.” U.S. Const. art. VI, cl. 2. Accordingly, “[w]hen state and federal laws clash, federal law reigns supreme and state law is preempted.” McDaniel v. Upsher-Smith Lab’ys, Inc., 893 F.3d 941, 944 (6th Cir. 2018).

State laws may be preempted both expressly and impliedly. Id. In the absence of express preemption, as is the case here, “federal law may impliedly preempt state law to the extent the two laws conflict.” Id. That said, “[i]mplied preemption analysis does not justify a ‘freewheeling judicial inquiry into whether a state statute is in tension with federal objectives’; such an endeavor ‘would undercut the principle that it is Congress rather than the courts that pre- empts state law.’” Chamber of Com. of U.S. v. Whiting, 563 U.S. 582, 607 (2011) (quoting Gade v. Nat’l Solid Wastes Mgmt. Ass’n, 505 U.S. 88, 111 (1992) (Kennedy, J., concurring in part and concurring in judgment)).

One form of implied preemption is conflict preemption.3 A state law is “conflict” preempted only “to the extent it actually conflicts with federal law, that is, when it is impossible to comply with both state and federal law, or where the state law stands as an obstacle to the accomplishment of the full purposes and objectives of Congress.” Silkwood v. Kerr-McGee

3 “Field” preemption—another form of implied preemption—is not at issue in the case at bar. Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 9

Corp., 464 U.S. 238, 248 (1984) (citations omitted). Here, Defendants do not argue that it is impossible to comply with both the state and federal laws. Instead, Defendants argue only that the state claims are obstacle preempted, i.e., they are an obstacle to achieving Congress’s full purposes and objectives. There is a “high threshold” for such an argument. Whiting, 563 U.S. at 607 (quoting Gade, 505 U.S. at 110).

Congress’s purpose “‘is the ultimate touchstone’ in every pre-emption case.” Medtronic, Inc. v. Lohr, 518 U.S. 470, 485 (1996) (quoting Retail Clerks v. Schermerhorn, 375 U.S. 96, 103 (1963)). “[B]ecause the States are independent sovereigns in our federal system, we have long presumed that Congress does not cavalierly pre-empt state-law causes of action.” Id. This is especially true in the case of conflict preemption. See Wimbush v. Wyeth, 619 F.3d 632, 643 (6th Cir. 2010) (“Conflict preemption analysis ‘should be narrow and precise, to prevent the diminution of the role Congress reserved to the States while at the same time preserving the federal role.’” (quoting Downhour v. Somani, 85 F.3d 261, 266 (6th Cir. 1996) (internal quotation marks omitted)). “When Congress acts to preempt state law—especially in areas of longstanding state concern—it treads on the states’ customary prerogatives in ways that risk upsetting the traditional federal-state balance of authority.” Merrick v. Diageo Americas Supply, Inc., 805 F.3d 685, 694 (6th Cir. 2015). On that basis, when considering a preemption argument, we apply a presumption against preemption. See Medtronic, 518 U.S. at 485.

The presumption against preemption “operates with special force in cases ‘in which Congress has legislated . . . in a field which the States have traditionally occupied.’” Merrick, 805 F.3d at 694 (quoting Medtronic, 518 U.S. at 485). States have traditionally occupied the field of consumer protection. See, e.g., Fla. Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 144 (1963) (“States have always possessed a legitimate interest in ‘the protection of [their] people against fraud and deception in the sale of food products . . .’” (quoting Plumley v. Massachusetts, 155 U.S. 461, 472 (1894)) (alteration in original)); Durnford v. MusclePharm Corp., 907 F.3d 595, 601 (9th Cir. 2018) (“Consumer protection falls well within that category.”). States have, likewise, traditionally occupied the field of environmental regulation. Merrick, 805 F.3d at 694 (citing Huron Portland Cement Co. v. Detroit, 362 U.S. 440, 442 (1960)). Crucially, this presumption “accounts for the historic presence of state law but does not Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 10

rely on the absence of federal regulation.” Wyeth v. Levine, 555 U.S. 555, 566 n.3 (2009). Stated otherwise, the existence of federal regulations does not undermine the presumption against preemption. See id.

2. In re Ford

Defendants’ preemption argument starts with In re Ford. 65 F.4th 851. The instant case, however, meaningfully differs from Ford. In Ford, a group of consumers brought claims “alleging that defendant Ford Motor Company intentionally submitted false fuel economy testing figures for certain vehicles to the U.S. Environmental Protection Agency (EPA).” Id. at 854. When the EPA “published its fuel-economy estimates for those vehicles . . . Ford used these figures in its advertisements,” attributing the figures to the EPA. Id. at 857. The Ford plaintiffs’ claims asserted that the fraud on the EPA “led the agency to provide an inaccurate fuel economy estimate to consumers, which induced consumers (including plaintiffs) to buy” the relevant Ford vehicles. Id. at 854. In Ford, we held that the EPCA impliedly preempted plaintiffs’ fraud-on- the-agency claims. Id. at 866–67.

In making that determination in Ford, we first held that “no presumption against preemption exists” in fraud-on-the-agency cases. Id. at 863 n.7. We explained that “unlike in other circumstances where states have traditionally regulated conduct,” a presumption against preemption was not appropriate because “state law has not traditionally regulated fraud against a federal agency; that relationship is ‘inherently’ federal because it owes its very existence to federal law.” Id. (quoting Buckman Co. v. Plaintiffs’ Legal Comm., 531 U.S. 341, 347–48 (2001)). We then listed four conflicts between the Ford plaintiffs’ state-law claims and the EPCA that warranted implied preemption: “First, because the EPA accepted Ford’s testing information and published its estimate based on that information, plaintiffs’ claims essentially challenge the EPA’s figures.” Id. at 863. Because the plaintiffs’ claims asserted fraud on the EPA itself, “a jury would have to decide whether Ford’s testing figures are correct or fraudulent,” thus “inescapably and impermissibly put[ting] a jury into the EPA’s regulatory shoes.” Id. “Second, allowing juries to second-guess the EPA’s fuel economy figures would permit them to rebalance the EPA’s objectives,” which would “‘disrupt the expert balancing underlying the federal scheme.’” Id. (quoting Farina v. Nokia, Inc., 625 F.3d 97, 126 (3d Cir. Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 11

2010)). Third, the EPA is itself empowered to monitor, police, punish, and deter fraud against the agency; allowing private plaintiffs to do so would usurp EPA power. Id. And fourth, state- law fraud-on-the-agency claims would “skew the disclosures that manufacturers need to make to the EPA,” encouraging overproduction and thus “burden[ing] the agency’s approval process and obstruct[ing] its goal[s].” Id. at 864. On these four bases, we held that the Ford plaintiffs’ claims were conflict-preempted by the EPCA. Id.

Defendants in this case argue that all four conflicts between the state-law claims and the EPCA that required preemption in Ford are present in this case. D. 38 (Appellee Br. at 29). Before analyzing each of the four conflicts, it is crucial to note a preliminary distinction between Ford and the present case: In Ford, there was no presumption against preemption, 65 F.4th at 863 n.7, whereas here the presumption against preemption is particularly strong, see Merrick, 805 F.3d at 694 (explaining that the presumption against preemption “operates with special force in cases ‘in which Congress has legislated . . . in a field which the States have traditionally occupied” (quoting Medtronic, 518 U.S. at 485)); Paul, 373 U.S. at 144; Durnford, 907 F.3d at 601 (holding that states have traditionally occupied the field of consumer protection). Causes of action for fraudulent omissions to consumers “cannot reasonably be characterized as a state’s attempt to police fraud against the [EPA].” Desiano v. Warner-Lambert & Co., 467 F.3d 85, 94 (2d Cir. 2006). In contrast to a pure fraud-on-the-agency claim, the present consumer protection claims include theories of liability based exclusively on fraudulent omissions to consumers, not the EPA. See infra p. 13–14 (discussing theories of liability). Such focused consumer- protection theories warrant a strong presumption against preemption. That presumption necessarily affects the applicability of Ford’s analysis to this case. See Desiano, 467 F.3d at 94 (“[T]he existence of the presumption in the instant case requires an altogether different analysis from that made in Buckman.”). With that presumption in mind, we consider in turn each of the four Ford justifications for preemption.

a. Ford’s First Reason: An Impermissible Challenge to EPA Decisions

Defendants first argue that, like the state-law claims in Ford, “Plaintiffs’ fraud-based claims represent impermissible legal challenges to the EPA’s decision to approve GM’s testing data and AECD disclosures.” D. 38 (Appellee Br. at 29). As stated above, Plaintiffs’ claims rest Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 12

on five possible theories of liability: GM violated state consumer fraud and deception laws when the Duramax Trucks “emit[ted] levels of NOx many times higher than (i) their gasoline counterparts, (ii) what a reasonable consumer would expect, (iii) what GM had advertised, (iv) the Environmental Protection Agency’s maximum standards, and (v) the levels set for the vehicles to obtain a certificate of compliance that allows them to be sold in the United States.” R. 18 (First Am. Compl. ¶ 2) (Page ID #892). The fourth and fifth theories of liability would require a showing that, contrary to the EPA’s decision, the Duramax Trucks failed to meet EPA standards. These two theories of liability depend entirely upon a CAA violation. Like the Ford plaintiffs’ claims, these two theories are thus “impermissible legal challenges to the EPA’s decision” and, under Ford, are preempted. D. 38 (Appellee Br. at 29); see Ford, 65 F.4th at 863.

Contrary to Defendants’ position, however, the remaining three theories of liability—that GM violated state law through its misleading omissions regarding its emissions as compared to: (i) gasoline counterparts, (ii) a reasonable consumer’s expectations, and (iii) what GM advertised—do not implicate or challenge the EPA’s determinations. Cf. Loreto v. Procter & Gamble Co., 515 F. App’x 576, 579–80 (6th Cir. 2013) (holding that, though one theory of liability is impliedly preempted, plaintiffs’ second theory of liability was not preempted). Plaintiffs claim that GM fraudulently omitted emissions information in its advertisements and communication to the public. See R. 18 (First Am. Compl. ¶ 87–106) (Page ID #952–61) (listing alleged fraudulent omissions in GM’s advertising materials). These theories of liability rest on traditional state tort law and exist independently of EPA decisions.

The evidence in support of these remaining theories of liability exists independently of EPA standards. When Duramax Truck consumers heard “clean diesel,” they expected their Duramax Truck to have lower emissions than previous diesel models. See R. 393-16 (Roberts Dep. at 102:1-11, 174:2-8) (Page ID #38751, 38757). Based on GM advertisements and communications to consumers, consumers likewise expected that the Duramax Trucks would have lower emissions than other vehicles on the market. See id. at 195:19-24 (Page ID #38761); R. 393-19 (Henderson Dep. at 79:11-15) (Page ID #38809). Some consumers measured “clean diesel” visually, expecting their vehicles not to “blow the black smoke,” that diesel engines can be known for, R. 393-12 (Mizell Dep. at 43:4-6) (Page ID #38675), whereas others understood Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 13

GM’s communications to mean that the Duramax Trucks would have lower emissions, “be cleaner[,] and run a lot better” than the older pickup trucks that they previously owned, R. 393- 19 (Henderson Dep. at 72–73) (Page ID #38806–07).

Of note, the Ford plaintiffs also alleged fraud on consumers in addition to their fraud-on- the-agency claims. In contrast to the instant case, however, Ford’s alleged fraudulent statements to consumers “relied solely on the EPA estimates,” and lacked “any further disclosures about a vehicle’s supposed real-world fuel economy,” not directly tied to the EPA-published numbers. Ford, 65 F.4th at 866. We found it “crucial[]” that “the regulatory scheme governing fuel economy standards requires the EPA to approve those figures and publish them as its own.” Id. at 865; see also id. at 856, 862 (repeatedly noting that the EPA adopts and publishes those figures as its own). “[B]ecause the EPA accepted Ford’s testing information and published its estimate” as its own, “plaintiffs’ claims essentially challenge the EPA’s figures” themselves. Id. at 863. Likewise, Ford was required to inform consumers of the EPA-approved estimate through a specified label on the car and was warned by the Federal Trade Commission that using any other figure in its advertising could be deceptive. Id. at 856–57. Then Ford used those figures in its ads, explicitly stating that they were “EPA-estimated”—because they were. Id. at 857, 866. The Ford plaintiffs’ consumer fraud claims “exist[ed] solely because of the EPCA’s requirements” and were thus “a by-product of alleged fraud committed on the EPA.” Id. at 866. On that basis, we held that the Ford consumer fraud claims were preempted. Id. at 866–67.

The Supreme Court, like this court in Ford, has also distinguished between claims based on government-controlled information and claims based on privately controlled information. In Ford, the basis of the consumer fraud claims were EPA estimates that Ford used in its communications to consumers. Id. at 866. The EPA published and controlled those estimates. Id. In contrast, in Wyeth v. Levine, the Supreme Court held that state-law claims about a medication label were not preempted by Food and Drug Administration (FDA) regulations in part because “the manufacturer[, not the FDA,] bears responsibility for the content of its label at all times.” 555 U.S. at 570–71. Because the private manufacturer had the ultimate control over Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 14

the content of its label, state-law claims based on that content were not preempted.4 Id. at 570– 73. The Ford court specifically distinguished Levine on the basis that, unlike the regulation in Levine, the EPCA “regulatory scheme . . . requires the EPA to approve those figures and publish them as its own.” Ford, 65 F.4th at 865–66 (citing Levine, 555 U.S. at 568–73).

On this question, the instant case mirrors Levine, not Ford. Like Levine, and unlike Ford, GM “was responsible for the contents of the [advertisements about clean diesel] and could alter [them] unilaterally without agency approval.” Ford, 65 F.4th at 866 (citing Levine, 555 U.S. at 568–73). Because GM “bears responsibility for the content of its [advertisements] at all times,” and the advertisements are the basis of the state-law claims, the state-law claims are not preempted. Levine, 555 U.S. at 570–71. Moreover, unlike Ford, GM’s advertisements did not utilize any EPA-owned figures and, in fact, made significant disclosures about real-world emissions unconnected to EPA determinations. See Ford, 65 F.4th at 866; R. 18 (First Am. Compl. ¶ 87–106) (Page ID #952–61). The Plaintiffs here, accordingly, do not challenge EPA figures or even representations that derive from those figures.

Defendants also argue that, like the Ford claims, “Plaintiffs’ claims of consumer fraud necessarily depend on . . . alleged violations of the federal emissions standards because . . . ‘Plaintiffs have not identified an [alternative] emissions benchmark.’” D. 38 (Appellee Br. at 30) (quoting R. 444 (Summ. J. Order at 10) (Page ID #48710)). As noted above, however, Plaintiffs’ three remaining theories of liability rest on (i) comparative gas vehicles, (ii) consumer expectations, and (iii) GM’s own statements in advertisements. R. 18 (First Am. Compl. ¶ 2) (Page ID #892). These theories of liability may be supported by evidence of, for example, consumers expecting their Duramax Trucks to have lower emissions than previous diesel models, see R. 393-16 (Roberts Dep. at 102:1-11, 174:2-8) (Page ID #38751, 38757), or lower emissions than other vehicles on the market, see id. at 195:19-24 (Page ID #38761); R. 393-19 (Henderson Dep. at 79:11-15) (Page ID #38809). These plausible alternative emissions

4 The Levine Court’s discussion of this issue arose in the context of impossibility preemption. See 555 U.S. at 570–71. The Levine Court, however, held that the state-law claims were neither impossibility nor obstacle preempted. Id. at 581. Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 15

benchmarks do not depend on federal emissions standards. Ford’s first reason for finding preemption accordingly does not apply.

b. Ford’s Second Reason: Undermines EPA Policy Considerations

Ford’s second reason for finding preemption—that allowing juries to second guess EPA decisions undermines EPA balancing and policy considerations—likewise does not apply to the case at bar. Contrary to Ford, Plaintiffs in this case “can prevail without showing that the subject vehicles violate EPA regulations.” R. 61 (Mot. to Dismiss Order at 28) (Page ID #3500). Compare id. with Ford, 65 F.4th at 865 (“To demonstrate that Ford committed fraud, plaintiffs would need to show that Ford failed to follow the EPA-proscribed testing procedures or its obligation to report truthful information to the EPA.”). The “gravamen” of Plaintiffs’ state-law claims “is that they purchased a vehicle which polluted at levels far greater than a reasonable consumer would expect. In other words, the vehicle operated differently than they expected.” R. 61 (Mot. to Dismiss Order at 28) (Page ID #3500); see also supra Part I, Section A (discussing evidence of consumer expectations).

Plaintiffs’ state-law claims are entirely premised on fraudulent omissions to consumers; “these allegations do not require proof of noncompliance with EPA regulations. In fact, it is conceivably possible that Defendants could simultaneously comply with EPA regulations while still concealing material information from consumers.” R. 61 (Mot. to Dismiss Order at 29) (Page ID #3501). In Ford, the fraud-on-the-agency claim meant that, “[t]o evaluate their claims, a jury would have to decide whether Ford’s testing figures are correct or fraudulent.” Ford, 65 F.4th at 863. Here, a jury would not be required to look at any EPA-approved figures or determine whether the EPA was correct in accepting GM’s submissions. Rather, a jury would need to look at GM advertisements and the expectations of a reasonable consumer alone. The EPA’s balancing and policy considerations would, accordingly, remain in the EPA’s hands.

c. Ford’s Third Reason: Usurps EPA Authority to Police Fraud

Ford’s third reason for finding preemption—that state-law claims usurp the EPA’s authority to police fraud, id.—does not apply to the present case. In Ford, the court explained that “[t]he EPA has several statutory and regulatory ways to police suspected fraud and monitor Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 16

compliance with its testing procedures.” Id. Relying on Buckman, this court explained that it was this “explicit authority [that] was a foundational reason Buckman determined the claims at issue were preempted.” Id. (citing Buckman, 531 U.S. at 350). Unlike in Ford and Buckman, the EPA here has no explicit—nor even implicit—authority to police the suspected fraud at issue, i.e., fraud on consumers.

Defendants argue that that the EPA has “authority to punish and deter fraud in the emissions-testing process” because “the EPA has the power to ‘investigate if any “defeat device” lurks beneath the deck’ of a vehicle.” D. 38 (Appellee Br. at 32) (quoting R. 444 (Summ. J. Order at 7) (Page ID #48707)). The EPA can also approve permissible AECDs or reject them as so-called defeat devices. See 40 C.F.R. § 86.004-2. GM argues that, if an AECD is approved, any claim premised on the use of that AECD thus usurps EPA authority. D. 38 (Appellee Br. at 32). Plaintiffs’ claims, however, are based on fraudulent omissions to consumers about how the Duramax Trucks’ emissions systems run. Though Plaintiffs discuss defeat devices, their remaining theories of liability do not depend on the use (or misuse) of a “defeat device,” nor on any fraud “in the emissions-testing process.” Id.; see also R. 439 (Pls. Show Cause Resp. at 19) (Page ID #48312) (“It may be that GM also installed defeat devices in the vehicles to aid in their deception. It may be that these defeat devices also enabled Defendants to commit fraud on the EPA. But GM’s fraud on consumers is not based on any fraud on the EPA or based on repeating any finding made by the EPA, and Plaintiffs can prove their allegations without proving that GM defrauded the EPA.”). Discussion of defeat devices does not necessitate that Plaintiffs’ theories of liability depend on the CAA. See Loreto, 515 F. App’x at 580 (holding that even though “the complaint does include extensive reference” to an agency determination, the “claim does not depend upon this determination and would logically exist even in its absence”). Instead, Plaintiffs’ claim is that GM omitted information about the truck’s emissions during regular use in its representations to consumers, regardless of whether the AECD’s use was approved. Stated otherwise, even if an AECD is approved, GM cannot mislead consumers by omitting information about how the (permissible) AECD works. Because Plaintiffs’ remaining theories of liability do not implicate fraud in the emissions-testing process, but rather implicate only fraud in communications to consumers, and because policing fraud on consumers is not within the EPA’s scope of authority, Plaintiffs’ state-law claims do not usurp the EPA’s role. Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 17

d. Ford’s Fourth Reason: Skews Disclosures to the EPA

Ford’s fourth and final reason for finding preemption—that “state-law claims would skew the disclosures that manufacturers need to make to the EPA,” Ford, 65 F.4th at 864—is likewise not applicable to this case. In Ford, we reasoned that, because the EPA was responsible for determining whether a car manufacturer’s “documentation is sufficient,” a state-law claim that depended on fraud-on-the-EPA would allow “a jury [to] find this documentation inadequate even if the EPA had previously determined otherwise.” Id. Such a finding would then incentivize manufacturers to over-document and “submit a deluge of information that the [EPA] neither wants nor needs, resulting in additional burdens on the” agency. Id. (quoting Buckman, 531 U.S. at 351). Defendants argue that these concerns apply equally here.

Contrary to Defendants’ argument, however, and as discussed above, Plaintiffs’ remaining theories of liability do not depend on any fraud on the EPA. The documentation that GM provided to the EPA thus has no bearing on whether GM provided sufficient information to consumers. Because the information provided to the EPA would not serve as a defense to these state-law consumer fraud or fraudulent concealment claims, manufacturers would have no incentive to over-disclose information to the EPA.

Without any presumption against preemption, the Ford court identified four reasons that the Ford plaintiffs’ state-law claims were preempted by the EPCA. Contrary to that case, none of the four Ford reasons apply to the case at bar. Ford, accordingly, does not control whether the CAA preempts the state-law claims in this case.

3. Buckman and Progeny

Given that Ford does not mandate preemption in this case, the question of whether the CAA impliedly preempts Plaintiffs’ state-law claims remains unanswered. As we properly noted in Ford, Buckman Co. v. Plaintiffs’ Legal Committee is the central case in this area. 531 U.S. 341. In Buckman, plaintiffs claimed injuries resulting from a medical device that had been reviewed and approved by the FDA. See id. at 343. The Buckman plaintiffs claimed that the defendant company had defrauded the FDA in order to get approval of the device. Id. The Supreme Court began its analysis by stating that no presumption against preemption would apply Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 18

because “[p]olicing fraud against federal agencies is hardly ‘a field which the States have traditionally occupied.’” Id. at 347 (quoting Rice v. Santa Fe Elevator Corp., 331 U.S. 218, 230 (1947)). In the absence of any presumption against preemption, the Court, distinguishing these “fraud-on-the-FDA claims” from claims sounding in “traditional state tort law,” held that plaintiffs’ claims were preempted. Id. at 351–52. The Court explained that, “were plaintiffs to maintain their fraud-on-the-agency claims here, they would not be relying on traditional state tort law which had predated the federal enactments in question. On the contrary, the existence of these federal enactments is a critical element in their case.” Id. at 353.

Under Buckman, the critical distinctions between preempted fraud-on-the-agency claims and not-preempted traditional state tort law claims are, accordingly, (1) whether the source of the allegedly violated duty stems from federal law or regulation, as opposed to preexisting state tort law, and (2) whether “proof of fraud against the [agency] is alone sufficient to impose liability.” Desiano, 467 F.3d at 94–95; see also Garcia v. Wyeth-Ayerst Lab’ys, 385 F.3d 961, 966 (6th Cir. 2004) (“Buckman teaches that state tort remedies requiring proof of fraud committed against the FDA are foreclosed since federal law preempts such claims.”). In Fulgenzi v. PLIVA, Inc., we likewise explained that if claims “‘exist solely by virtue of’ the regulatory scheme, they are preempted,” but if a claim is based on an “independent, pre-existing state law cause[] of action that parallel[s] federal safety requirements, the suit is not preempted.” 711 F.3d 578, 586 (6th Cir. 2013) (quoting Buckman, 531 U.S. at 353).

Unlike in Buckman, the source of the duty in this case is state consumer-protection law; it is not federal regulation. If the EPA did not exist or if there were no federal emissions regulations, Plaintiffs’ claims that GM fraudulently omitted facts about its emissions systems in advertisements and communications to consumers would live on. Under Buckman, the instant state-law claims are accordingly non-preempted “traditional state tort law” claims, rather than the preempted “fraud-on-the-agency” claims. Buckman, 531 U.S. at 351–52. Additionally, the consumer protection state-law claims in this case do not require proof of fraud against the EPA to demonstrate liability. See R. 61 (Mot. to Dismiss Order at 28) (Page ID #3500). Plaintiffs’ state-law claims are entirely premised on fraudulent omissions to consumers; “these allegations do not require proof of noncompliance with EPA regulations. In fact, it is conceivably possible Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 19

that Defendants could simultaneously comply with EPA regulations while still concealing material information from consumers.” Id. at 29 (Page ID #3501). Under Buckman, Plaintiffs’ state-law claims are thus not preempted.

In Loreto v. Procter & Gamble Co., plaintiffs claimed that Procter & Gamble violated state consumer-protection laws when it falsely marketed an FDA-approved drug as alleviating certain cold symptoms. 515 F. App’x at 577–78. Plaintiffs asserted claims under state consumer-protection laws based on two theories. Id. at 579. The first theory—that “Procter & Gamble omitted telling consumers that its products were ‘illegal’”—was preempted by federal law because it “depends entirely upon an FDCA violation.” Id. In contrast, the second theory— “that Procter & Gamble violated state law when it represented to the public that taking Vitamin C can blunt the effects of a cold, a statement plaintiffs contend is false or misleading”—was not preempted, because it “relies solely on traditional state tort law predating the FDCA, and would exist in the absence of the [federal] Act.” Id. at 579–80 (citing Buckman, 531 U.S. at 353). Even though the Loreto plaintiffs’ complaint, including the second, non-preempted theory, “include[d] extensive reference” to FDA regulations, “plaintiffs’ claim does not depend upon [any FDA determination or regulation] and would logically exist even in its absence.” Id. at 580. On that basis, this second theory of liability was not preempted.

Like the Loreto plaintiffs, Plaintiffs here bring state-law claims under several theories of liability. And like the Loreto plaintiffs’ first theory of liability, Plaintiffs’ assertions that GM omitted in its communications with consumers that the Duramax Trucks violate EPA regulations are preempted. See supra Part II, Section B(2) (discussing Plaintiffs’ five theories of liability). That said, Plaintiffs’ assertions that GM violated state consumer protection laws through its omissions and misrepresentations regarding its Duramax Trucks, which had emissions that far exceeded (i) gasoline counterparts, (ii) reasonable consumer expectations, and (iii) GM’s own advertisements, are analogous to the Loreto plaintiffs’ non-preempted, second theory of liability. These theories of liability are based on GM’s statements and omissions to the public. These theories assert that, based on GM’s omissions to consumers, consumers reasonably expected their Duramax Trucks to have lower emissions than previous diesel models, other vehicles on the market, and their own older trucks. See, e.g., see R. 393-16 (Roberts Dep. at 102:1-11, 174:2-8, Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 20

195:19-24) (Page ID #38751, 38757, 38761); R. 393-19 (Henderson Dep. at 79:11-15) (Page ID #38809). These theories assert that the Duramax Trucks ultimately failed to meet these expectations, and that GM fraudulently omitted information to induce these consumer beliefs and consequent purchase. This theory “relies solely on traditional state tort law predating the [CAA], and would exist in the absence of the [federal] Act.” Loreto, 515 F. App’x at 580 (citing Buckman, 531 U.S. at 353). Without the CAA, Plaintiffs could nonetheless bring a claim that GM fraudulently omitted relevant information about its emission system in its communications to consumers. These theories of liability benchmark against other vehicles; these claims exist independently of federal regulation.

The instant case is controlled by Buckman and informed by Loreto. Unlike the claims in Ford, these state-law claims exist independently of federal law and any EPA findings, and do not depend on proving fraud on the EPA or a violation of federal law. These state-law claims are, accordingly, not preempted by the CAA.

4. Savings Clause & Citizen-Suit Provision

Plaintiffs argue in the alternative that, even if their claims require proving violations of federal emissions standards, the claims can nonetheless move forward—and are not preempted— thanks to the CAA’s citizen-suit provision and savings clause. See D. 30 (Appellant Br. at 39– 40). The CAA’s citizen-suit provision authorizes private civil actions “against any person . . . in violation of . . . an emission standard or limitation under this chapter.” 42 U.S.C. § 7604(a). The provision also gives district courts the jurisdiction “to enforce such an emission standard.” Id. The CAA’s savings clause states that “[n]othing in this section shall restrict any right which any person (or class of persons) may have under any statute or common law to seek enforcement of any emission standard or limitation or to seek any other relief (including relief against the Administrator or a State agency).” Id. § 7604(e). Defendants argue that “the normal principles of both express preemption and implied preemption still apply, notwithstanding the existence of the savings clause in the citizen-suit provision.” D. 38 (Appellee Br. at 42).

We have twice addressed these provisions. In Her Majesty the Queen in Right of the Province of Ontario v. City of Detroit, we held that “the plain language of the CAA’s savings Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 21

clause . . . clearly indicates that Congress did not wish to abolish state control.” 874 F.2d 332, 342–43 (6th Cir. 1989). In so finding, we held that the CAA did not “entirely preempt[] the field in question.” Id. at 342. In Her Majesty the Queen, we did not, however, address conflict preemption in particular. In Merrick v. Diageo Americas Supply, Inc., on the other hand, property owners brought nuisance claims under state common law against a company emitting ethanol within the state. 805 F.3d 685. We held that these particular state common law claims were not conflict preempted by the CAA. Id. at 695. In making that determination, we considered all evidence of Congress’s intent, including the CAA’s citizen-suit provision and savings clause, “[o]ther parts of the text of the Clean Air Act,” legislative history, and case law. Id. at 691–95. Though the citizen-suit provision was highly relevant to the conflict-preemption analysis, it did not stand alone. Stated otherwise, the CAA’s citizen-suit provision and savings clause were merely evidence that implied conflict preemption was not appropriate. Despite that strong evidence, we nonetheless engaged in a normal conflict-preemption analysis. Id.

The CAA’s citizen-suit provision and savings clause are direct evidence of Congress’s intent not to preempt state-law causes of action. Such direct evidence is central to our preemption analysis. See Medtronic, 518 U.S. at 485. These provisions, however, do not stand alone to overcome any and all preemption arguments. See Merrick, 805 F.3d at 691–95; Int’l Paper Co. v. Ouellette, 479 U.S. 481, 493 (1987) (holding that the analogous citizen-suit and savings clause provisions of the Clean Water Act do “not purport to preclude pre-emption of state law by other provisions of the Act”); Buckman, 531 U.S. at 352 (holding “that neither an express pre-emption provision nor a saving clause ‘bar[s] the ordinary working of conflict pre- emption principles’” (quoting Geier v. Am. Honda Motor Co., 529 U.S. 861, 869 (2000)) (alteration in original)). Instead, the CAA’s citizen-suit provision and savings clause buttress our holding that Plaintiffs’ state-law claims are not preempted.

5. California State-Law Claims

Plaintiffs argue that, were we to hold that their state-law claims are generally preempted, their state-law claims arising out of California state law would nonetheless remain. See D. 30 (Appellant Br. at 46). The CAA expressly waives preemption of California emissions Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 22

standards.5 See 42 U.S.C. § 7543(b)(1); California v. EPA, 940 F.3d 1342, 1345 (D.C. Cir. 2019). Under the CAA’s citizen-suit provision, as noted above, citizens have authority to bring civil actions “against any person . . . in violation of . . . an emission standard or limitation under this chapter.” 42 U.S.C. § 7604(a). Because California’s emissions standards fall under this chapter of the CAA, citizens have authority to bring civil actions against any person in violation of California emissions standards. See id.; id. § 7543(b)(1). Because we hold that Plaintiffs’ state-law claims are not preempted by the CAA, we need not analyze California state-law claims standing alone.

C. RICO Claims

In a discussion totaling a mere two paragraphs, the district court found that Plaintiffs did not have standing to bring their RICO claims under the indirect-purchaser rule. R. 444 (Summ. J. Order at 16–17) (Page ID #48716–17). The district court found that, because (1) “neither GM nor Bosch ever charged Plaintiffs a dime,” and (2) “Plaintiffs are trying to recover ‘pass-through’ overcharges,” Plaintiffs therefore lack statutory standing under RICO. Id.

Plaintiffs argue that their RICO claims are not barred by the indirect-purchaser rule because “[t]hat rule applies only if an indirect purchaser asserts a pass-on theory of liability, but Plaintiffs make no such claim.” D. 30 (Appellant Br. at 47). Stated otherwise, Plaintiffs argue that “mere proof that a plaintiff is an ‘indirect purchaser’ is insufficient to establish that the plaintiff lacks standing. Proof that the plaintiff seeks passed-on damages is also required, but Plaintiffs here do not seek such damages and, therefore, have RICO standing.” Id. at 57–58. Defendants, on the other hand, argue that the indirect-purchaser rule is “a bright-line rule that authorizes suits by direct purchasers but bars suits by indirect purchasers.” D. 38 (Appellee Br. at 47) (quoting Apple, Inc v. Pepper, 587 U.S. 273, 279 (2019)). Because “the only question is whether the defendant directly sold to the plaintiff, which did not occur here,” Defendants argue that the indirect-purchaser rule bars Plaintiffs’ RICO claims. Id. at 49.

5 The CAA’s express preemption waiver has been expanded to cover California emissions standards as well as any other state’s standards that are “identical to the California standards.” 42 U.S.C. § 7507. References to California state-law claims thus encompass state-law claims arising from other states with “identical” standards. Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 23

The indirect-purchaser rule was initially developed in the anti-trust realm but applies to civil RICO claims with equal force. See Trollinger v. Tyson Foods, Inc., 370 F.3d 602, 612–14 (6th Cir. 2004). Under the indirect-purchaser rule, “indirect purchasers who are two or more steps removed from the violator in a distribution chain may not sue.” Pepper, 587 U.S. at 279. The indirect-purchaser rule is “a bright-line rule that authorizes suits by direct purchasers but bars suits by indirect purchasers.” Id. Stated otherwise, and as applicable to this case, “consumers at the bottom of a vertical distribution chain” do not have standing under RICO “to sue manufacturers at the top of the chain.” Id. at 281.

Plaintiffs are correct that the indirect-purchaser rule typically applies to indirect- purchasers who are seeking pass-through overcharges. See, e.g. Trollinger, 370 F.3d 602. Plaintiffs point to several cases in which this court and others have held that the indirect- purchaser rule bars claims by indirect-purchasers seeking pass-through charges. See D. 30 (Appellant Br. at 47–57). That indirect-purchasers’ claims for pass-through charges are covered by the indirect-purchaser rule, however, does not mean that other claims by indirect-purchasers are necessarily outside the indirect-purchaser rule. Plaintiffs point to no authority that suggests that the indirect-purchaser bright-line rule applies only to indirect-purchasers seeking pass- through overcharges. Supreme Court precedent is clear: “indirect purchasers who are two or more steps removed from the [RICO] violator in a distribution chain may not sue” under RICO. Pepper, 587 U.S. at 279. Nowhere in that bright-line rule are pass-through charges mentioned.

The Supreme Court in Illinois Brick Co. v. Illinois, 431 U.S. 720, 730–32 (1977), listed three reasons for adopting the indirect-purchaser rule: “(1) facilitating more effective enforcement of antitrust laws; (2) avoiding complicated damages calculations; and (3) eliminating duplicative damages against antitrust defendants.” Pepper, 587 U.S. at 285. The Plaintiffs here further argue that none of these “policy considerations that animated the Illinois Brick decision come into play here.” D. 45 (Reply Br. at 26). Even if Plaintiffs are correct that these reasons for the indirect-purchaser rule do not apply in the current case, however, “the bright-line rule of Illinois Brick means that there is no reason to ask whether the rationales of Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 24

Illinois Brick ‘apply with equal force’ in every individual case.”6 Pepper, 587 U.S. at 285 (quoting Kansas v. UtiliCorp United, Inc., 497 U.S. 199, 216 (1990)). The bright-line rule applies, and we will “not engage in ‘an unwarranted and counterproductive exercise to litigate a series of exceptions.’” Id. (quoting UtiliCorp United, 497 U.S. at 217).

III. CONCLUSION

Because Plaintiffs’ state-law claims are not impliedly preempted by the Clean Air Act, we REVERSE the district court’s grant of summary judgment on the state-law claims. Because Plaintiffs are indirect-purchasers and thus do not have standing under RICO, however, we AFFIRM the district court’s grant of summary judgment on the RICO claims.

6 Though we must be faithful to the Supreme Court’s admonition against finding exceptions to the indirect- purchaser rule based on the rationales of the rule, Plaintiffs’ point is well taken. In Pepper, the Supreme Court explained that, “if manufacturer A sells to retailer B, and retailer B sells to consumer C, then C may not sue A.” 587 U.S. at 280. Under this bright-line rule, it matters not whether retailer B is actually injured by manufacturer A’s unlawful conduct; even if consumer C is the only injured party, “C may not sue A.” Id. We would be remiss not to note the consequences of such a bright-line rule that mixes law and economics but does not necessarily reflect economic reality today. Under this rule, major manufacturers can insulate themselves from all antitrust and RICO liability, simply by selling their products through intermediaries. Because the business and success of intermediary car dealerships is dependent on car manufacturers, for example, consumers cannot rely on the intermediary car dealerships to vindicate their interests. As this case illustrates, car dealerships are not suing car manufacturers for RICO violations; doing so is against the intermediary’s interest. As a result, average car consumers have no recourse and major manufacturers are insulated from any liability. At bottom, rules must be supported by sensible principles. In a world with interdependent vertical economic structures, a bright-line indirect-purchaser rule does not facilitate effective enforcement of RICO and antitrust laws nor eliminate duplicative damages—this rule, instead, immunizes major manufacturers from any RICO or antitrust liability and hurts consumers. See id. at 285. A bright-line indirect-purchaser rule is, accordingly, unsupported by sensible principles. Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 25

________________________________________________________________________

CONCURRING IN THE JUDGMENT IN PART AND DISSENTING IN PART ________________________________________________________________________

KETHLEDGE, Circuit Judge, concurring in the judgment in part and dissenting in part. Purchasers of full-size diesel pickup trucks typically lack any walking-around sense of what the nitrogen-oxide levels of their trucks’ emissions should be. Regulating those levels, rather, falls squarely within the EPA’s expertise. Unsurprisingly, then, the named plaintiffs in this case— purchasers and lessors of 2011-16 GM “Duramax” diesel pickup trucks—uniformly testified during their depositions that they understood GM’s advertising about “clean diesel” to mean that the trucks’ emissions would comply with applicable federal regulations. Indeed none of them identified any other benchmark by which they could possibly assess the cleanliness of their trucks’ emissions.1 Nor did the plaintiffs’ emissions expert, Juston Smithers, identify any benchmark other than the EPA’s own standards: his report is instead a brief as to why, in his view, the EPA was wrong to conclude that the Duramax trucks met them. For good reasons, then, the district court held that this case was preempted on the ground that it sought to relitigate the EPA’s certification that these trucks complied with the EPA’s emissions standards. Yet our court now remands this case so that a lay jury can second-guess the EPA’s certification. And in doing so, the majority limits to its facts our recent decision in In re Ford Motor Co. F-150 and

1 For example, when the three plaintiffs whose testimony is cited by the majority were asked what “clean diesel” or “low emissions” meant to them, one said that it “meets government standards” and nothing else, Roberts Dep. 95:14-20; another said it “would meet, you know, all of the EPA standards that it was supposed to meet,” Mizell Dep. 52:9-25; and the third explained that the trucks would have “the DEF tank and some emissions [controls] on it,” Henderson Dep. 73:1-2, 76:13-21. See also, e.g., Biggs Dep. 122:16-25 (explaining that, as a consumer, he “assume[d]” the emissions would “meet the standards”); Reichert Dep. 111:15-19 (when asked what GM lied about, he said GM “lied about [] meeting the government reg”); McLean Dep. 84:12-17 (when asked if he had any expectation about his truck’s emissions, he said only that “it was supposed to be up to the standards of what the government puts out”); Burns Dep. 61:8-14 (agreed that his truck would provide the performance he expected if it “turned out” to be “EPA-compliant”); Golden Dep. 94:16-24 (explaining his only expectation as to emissions was that “it should meet all federal and state guidelines”); Iosue Dep. 81:1-5, 105:12-15 (saying he purchased the truck “thinking that they conformed to the emissions standards set by the government” and “[t]hat was it”); Crunkleton Dep. 179:14-17 (explaining that the term “clean diesel” meant to him that “you don’t see the black smoke and it’s meeting or exceeding the standard”); Prichard Dep. 101:22-102:5 (agreed that his only expectation as to emissions was that “it did comply” with EPA standards); Raleigh Dep. 61:25-62:7 (when asked “what kinds of emissions expectations” he had, he said: “That it would meet government standards.”). Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 26

Ranger Truck Fuel Economy Marketing and Sales Practices Litigation (Ford F-150), 65 F.4th 851 (6th Cir. 2023). I would affirm the district court’s judgment across the board.

The question presented in this appeal is whether the plaintiffs’ claims are subject to conflict preemption on the ground that they present “an obstacle to the accomplishment of the full purposes and objectives” of federal law. Id. at 860 (citation omitted). Specifically, the question here—as in Ford—is whether the plaintiffs’ “claims conflict with the EPA’s testing and fraud-policing authority[,]” in this case as to vehicle emissions. Id.

As relevant here, the Clean Air Act requires the EPA to set and enforce federal standards for vehicle emissions. See 42 U.S.C. §§ 7521(a)-(b), 7525; 40 C.F.R. §§ 86.127-12, 86.135.12. Motor vehicles generally cannot be sold in the United States without a “certification of conformity,” which is the EPA’s certification that a vehicle complies with all federal emissions standards. See 42 U.S.C. § 7521(a); 40 C.F.R. § 86.1848-01(e). As part of the certification process, manufacturers must disclose whether a vehicle has any “auxiliary emission control devices,” (AECDs), which for various reasons can increase vehicle emissions under certain operating circumstances. 40 C.F.R. § 86.1844-01(d)(11). When a vehicle is equipped with AECDs, the manufacturer must provide the EPA with “a justification for each AECD, the parameters they sense and control, a detailed justification of each AECD that results in a reduction in effectiveness of the emission control system, and a rationale for why it is not a defeat device.” Id. A “defeat device,” in turn, is an AECD that unjustifiably “reduces the effectiveness of the emissions control system under conditions which may reasonably be expected to be encountered in normal vehicle operation and use[.]” 40 C.F.R. § 86.004-2. What a “defeat device” defeats, therefore, is the proper functioning of a vehicle’s emissions system, as determined by the EPA; and the EPA may not issue a certificate of compliance as to any vehicle equipped with one. 40 C.F.R. § 86.1809-01(a).

GM went through this whole regulatory process for the Duramax trucks. In its application for a certificate of compliance, GM disclosed 309 pages of information about the AECDs utilized in these trucks—including 18 pages of explanation about precisely the “online- dosing” system that the plaintiffs complain about now. GM engineers also met with EPA officials in 2009 and 2010 to answer their questions about the online-dosing system. And in Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 27

May 2010, the EPA issued a certificate of compliance for the emissions systems of the Duramax trucks at issue here.

Yet, after all that disclosure on GM’s part and review on the EPA’s, the plaintiffs’ core factual allegation in this case—the through-thread of all their claims and “theories”—is that the online-dosing system in Duramax trucks was a “defeat device” nonetheless. See, e.g., Pl. Br. at 13; Smithers Rep. at 2. Every one of the plaintiffs’ theories—including the three the majority revives today—all include as a premise that the online-dosing system unjustifiably increases the trucks’ emissions during “real-world” driving conditions. The plaintiffs complain about no other AECD on these pickup trucks; and they do complain, over and over again, that the “online dosing in Duramax trucks caused excessive NOx emissions during real-world driving[.]” Pl. Br. at 9. But that is precisely the definition of a defeat device, 42 C.F.R. § 86.004-2—which would preclude the EPA from certifying these trucks as compliant with its emissions standards. That the EPA certified these trucks as compliant necessarily means that it concluded the online-dosing system was not a defeat device. And—in a rational legal order—that was a judgment for the EPA to make before GM sold the vehicles, not for a lay jury to make afterward.

Our decision in Ford confirms as much. There too the issue concerned the real-world performance of pickup trucks—specifically the gas mileage of Ford F-150 and Ranger pickups, which (again as here) the plaintiffs said was worse than advertised. Under the regime there, after extensive testing and disclosure, the EPA had adopted Ford’s estimates of the pickups’ gas mileage as its own. The plaintiffs claimed those estimates were wrong, which if true would cause an actual pocketbook injury. (The injury here, by any practical measure, is entirely fictive.) Yet we held that the plaintiffs’ claims were preempted, because they “inescapably and impermissibly put[] a jury into the EPA’s regulatory shoes.” 65 F.4th at 863. Thus, we said, “even though the EPA exercised its statutory duty and found Ford’s testing to be acceptable, a jury would still make its own determination, thus conflicting with the EPA’s authority to set its own fuel-economy figures.” Id. The same is true here as to the EPA’s determination that the online-dosing system was not a defeat device.

Relatedly, in Ford, we said that “allowing juries to second-guess the EPA’s fuel economy figures would permit them to rebalance the EPA’s objectives.” Id. The same is true here: Nos. 23-1648/1696/1697/1698 Fenner, et al. v. General Motors, LLC, et al. Page 28

AECDs by definition involve a tradeoff between increased emissions, on the one hand, and the “justification for each AECD” and the “rationale for why it is not a defeat device[,]” on the other. 40 C.F.R. § 86.1844-01(d)(11). This case invites lay jurors to strike that balance differently than the EPA did as to the online-dosing system in Duramax trucks—which “would disrupt the expert balancing underlying the federal scheme.” Ford, 65 F.4th at 863.

That is reason enough to show that the plaintiffs’ claims—all of which rest on the premise that the online-dosing system is a defeat device—are preempted. True, in Ford, the EPA adopted Ford’s mileage numbers as the agency’s own, whereas here the agency reached a negative conclusion that the online-dosing system was not a defeat device. But that should make no difference as to preemption: in both cases the claims would have a jury revisit the EPA’s regulatory judgment.

Nor should it matter that three of the plaintiffs’ “theories” of liability for the online- dosing system are that the trucks’ “emissions exceeded (i) gasoline counterparts, (ii) a reasonable consumer’s expectations, and (iii) what GM advertised[.]” Maj. Op. at 12. Those theories are just so much semantics: for the factual reason why the plaintiffs say the trucks’ emissions exceeded these different benchmarks, such as they are, is that the online-dosing system caused excessive NOx emissions during real-world driving—which again is to say that it was a defeat device. See, e.g., Smithers Rep. p. 2 (“The defeat device here is a technically unjustified use of a strategy called ‘online dosing.’”). And the EPA concluded that the system was not a defeat device. Moreover, as noted above, the benchmark for the named plaintiffs and their expert alike—as to whether these trucks’ emissions were “clean”—was whether they complied with federal emissions standards. See supra, n. 1; Smithers Report pp. 1-2. (Also, the idea that any purchaser of a full-size pickup would expect a diesel truck to have cleaner emissions than a gasoline one is implausible on its face.) Labels aside, in substance this case would relitigate the EPA’s regulatory judgment. The district court was right to grant summary judgment to the defendants.

Making complex technical judgments about vehicle emissions is precisely the kind of thing that expert agencies are good at—and lay jurors are not. We lose our legal bearings and misapply binding precedent in concluding otherwise here. I respectfully dissent.