BRESS, Circuit Judge

Appeal from the United States District Court for the Northern District of California, Jacqueline Scott Corley, Magistrate Judge, Presiding, D.C. No. 3:21-cv-00227-JSC Brock J. Specht (argued), Paul J. Lukas, and Grace I. Chanin, Nichols Kaster PLLP, Minneapolis, Minnesota; Matthew C. Helland and Daniel S. Brome, Nichols Kaster LLP, San Francisco, California; for Plaintiffs-Appellants. Mark C. Nielsen (argued), David N. Levine, Sarah M. Adams, Kara P. Wheatley, Paul J. Rinefierd, and Kalena R. Kettering, Groom Law Group Chartered, Washington, D.C.; Jeffrey S. Bosley and Andrew G. Row, Davis Wright Tremaine LLP, San Francisco, California; for Defendants-Appellees. Jamie Bowers (argued), Trial Attorney; Thomas Tso, Counsel for Appellate and Special Litigation; G. William Scott, Associate Solicitor for Plan Benefits Security; Seema Nanda, Solicitor of Labor; United States Department of Labor, Office of the Solicitor, Plan Benefits Security Division; Washington, D.C.; for Amicus Curiae Secretary of Labor. Meaghan VerGow and Alexander Reed, O'Melveny & Myers LLP, Washington, D.C.; Janet Galeria and Paul Lettow, United States Chamber Litigation Center, Washington, D.C.; for Amicus Curiae Chamber of Commerce of the United States of America. Before: Daniel A. Bress and Lawrence VanDyke, Circuit Judges, and Jane A. Restani, Judge.

OPINION

BRESS, Circuit Judge:

Participants in an ERISA welfare benefits plan sued the manager of a Multiple Employer Welfare Arrangement (MEWA) for alleged breach of fiduciary duty. The question is whether plaintiffs have Article III standing. We hold that under the facts alleged, they do not. We affirm the dismissal of plaintiffs' complaint.

I

A

Plaintiffs are current and former employees of RingCentral, a technology company. Plaintiffs participated in RingCentral's employee welfare benefits plan, which is governed by the Employee Retirement Income Security Act of 1974 (ERISA). 29 U.S.C. § 1001, et seq. RingCentral sponsored this plan to provide its employees with benefits such as medical, dental, and vision insurance.

From 2013 to 2019, the RingCentral plan participated in the "Tech Benefits Program" administered by defendants Sequoia Benefits and Insurance Services, LLC, and Gregory S. Golub. We will refer to the defendants collectively as "Sequoia." The Tech Benefits Program is a Multiple Employer Welfare Arrangement (MEWA), see 29 U.S.C. § 1002(40)(A), that pools assets from more than 180 employer-sponsored plans into a trust fund for the purpose of obtaining insurance benefits for employees at large-group rates that may otherwise be unattainable for individual employer plans.

RingCentral's ERISA plan was funded in part by contributions from RingCentral and in part by employee contributions. RingCentral determined which insurance options to make available to its employees and how much, if anything, employees were required to contribute for the different benefits. Under the Tech Benefit Program's governing documents, RingCentral had broad discretion to determine employee contributions. The program's terms did not set a formula for RingCentral to calculate employee contributions, and RingCentral's discretion was subject only to an obligation to cover at least 75% of the cost for employees who select single coverage and at least 50% for employees who select a group health plan. As alleged in plaintiffs' complaint, when asked how it determined the amount employees must contribute, RingCentral "did not identify a specific formula or set of factors" and instead said the decision was based on "various factors and discussion." The record reflects that in some instances, RingCentral paid all the premium contributions for certain benefit options, with participating employees paying nothing.

Under the Tech Benefits Program, Sequoia selected the insurance benefits that would be made available to employers, negotiated the cost of any given benefit with the insurance provider, and determined how much each employer plan must contribute to the Tech Benefits Program's trust fund in exchange for the plan participants' selected benefits. The insurance companies charged certain costs for the employee benefits, including general administrative fees and premiums based on the coverage provided. Sequoia paid those costs out of a trust account maintained in the name of the Tech Benefits Program. Participating employers like RingCentral funded this trust account with plan assets which, as we have noted, included some employee contributions. The document governing the Tech Benefits Program provided that "[n]o assets of the program will be used or diverted to purposes other than for the exclusive benefit of [participating employees] and for defraying the reasonable expenses of administering the program."

Under this arrangement, Sequoia effectively operated as an insurance broker between participating employer plans and insurance companies. As compensation for these services, the insurance companies paid Sequoia commissions. These commissions were set as a portion of the total fees and premiums paid to each insurance company. Sequoia's contracts with RingCentral did not specify the amount of such commissions—a figure that was instead negotiated by Sequoia and the insurance companies. But an agreement between Sequoia and RingCentral did acknowledge that Sequoia would receive commissions from insurers. In the case of the Tech Benefits Program's primary medical benefit provider, the commission was 6% of the total cost to the plans. These commissions paid to Sequoia were not taken directly from the assets of the Tech Benefits Program but were instead separately paid by the insurance companies to Sequoia after Sequoia used the program's assets to pay for plaintiffs' insurance benefits.

B

A few years after RingCentral began participating in the Tech Benefits Program, plaintiffs filed this putative class action on behalf of the RingCentral plan and other Tech Benefits Program participants. RingCentral is not a party to this case.

In their complaint, plaintiffs asserted that Sequoia owed fiduciary duties to the plan under ERISA because Sequoia allegedly exercised control over plan assets through its operation of the Tech Benefits Program. See 29 U.S.C. § 1002(21)(A) (defining who qualifies as a plan fiduciary). Plaintiffs did not allege that they were deprived of any health benefits, or that the health benefits they signed up for were not fully insured.

Instead, plaintiffs alleged that Sequoia violated its fiduciary duties in two ways: (1) by receiving and retaining commission payments from insurers, which plaintiffs regard as kickbacks; and (2) by negotiating allegedly excessive administrative fees with insurers, which led to higher commissions for Sequoia. To some extent, plaintiffs may believe it is improper for a company like Sequoia to receive commissions at all, or at least through the arrangement as designed here. In a way, their complaint thus purports to challenge a business model for providing employee benefits that we are told is common.

The district court dismissed plaintiffs' complaint for lack of Article III standing, concluding that plaintiffs had not alleged sufficient facts indicating that Sequoia's conduct led plaintiffs to pay higher contributions or to receive fewer benefits. Plaintiffs were given leave to amend, and they subsequently filed what is now the operative complaint.

In their amended complaint, plaintiffs alleged that Sequoia's supposed breach of fiduciary duty injured them by requiring plaintiffs to pay higher contributions toward their benefits and by allegedly interfering with plaintiffs' purported equitable ownership interest in the Tech Benefits Program trust fund. Plaintiffs contended that these alleged injuries could be redressed in either of two ways: (1) through direct disgorgement to plaintiffs of Sequoia's improper profits, using an equitable remedy such as a constructive trust; or (2) by forcing Sequoia to reimburse the RingCentral plan. In this latter circumstance, plaintiffs claim that the RingCentral plan would then likely refund the plaintiffs that portion of their contributions attributable to Sequoia's alleged misconduct.

The district court again dismissed for lack of standing, this time without leave to amend. The court found that plaintiffs had still provided "no allegations that support an inference that had Defendants not charged commissions to the insurers, or had they charged a lower commission, Plaintiffs would have contributed less toward their health benefits." And the court held that plaintiffs' theory of injury based on an equitable ownership interest in the program's assets was foreclosed by Thole v. U.S. Bank N.A., — U.S. —, 140 S. Ct. 1615, 1619, 207 L.Ed.2d 85 (2020). Plaintiffs thus had not established an injury in fact. And even if they had, plaintiffs had not shown that any injury they suffered would be redressed by a favorable decision.

Plaintiffs timely appealed. We review de novo the district court's dismissal of plaintiffs' complaint for lack of standing. Meland v. Weber, 2 F.4th 838, 843 (9th Cir. 2021). And we construe all material factual allegations in the complaint in plaintiffs' favor. Id.

II

Plaintiffs bear the burden of establishing each of the three "irreducible" elements of Article III standing. Lujan v. Defs. of Wildlife, 504 U.S. 555, 560, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992); Meland, 2 F.4th at 843 (quotation omitted). They must sufficiently allege "(i) that [they] suffered an injury in fact that is concrete, particularized, and actual or imminent; (ii) that the injury was likely caused by the defendant; and (iii) that the injury would likely be redressed by judicial relief." TransUnion LLC v. Ramirez, — U.S. —, 141 S. Ct. 2190, 2203, 210 L.Ed.2d 568 (2021) (citing Lujan, 504 U.S. at 560-61, 112 S.Ct. 2130). At the pleading stage, plaintiffs must " 'clearly . . . allege facts demonstrating' each element." Spokeo, Inc. v. Robins, 578 U.S. 330, 338, 136 S.Ct. 1540, 194 L.Ed.2d 635 (2016) (alteration in original) (quoting Warth v. Seldin, 422 U.S. 490, 518, 95 S.Ct. 2197, 45 L.Ed.2d 343 (1975)). This case is a putative class action, but "even named plaintiffs who represent a class 'must allege and show that they personally have been injured.' " Id. at 338 n.6, 136 S.Ct. 1540 (quoting Simon v. Eastern Ky. Welfare Rights Org., 426 U.S. 26, 40 n.20, 96 S.Ct. 1917, 48 L.Ed.2d 450 (1976)).

In many ERISA cases, the plaintiffs are suing the ERISA plan itself or their employer, parties who have a direct role in designing and administering the plan. This case is less typical because the plaintiffs are leapfrogging the RingCentral plan and seeking to recover directly from Sequoia, a management and insurance brokerage company that is a step removed from the contributions plaintiffs pay and the benefits they receive. As we will explain, this structural feature of this case contributes to plaintiffs' failure to sufficiently allege Article III standing.

A

We begin with the plaintiffs' first theory of injury, which is that Sequoia's actions allegedly caused plaintiffs to pay higher contributions for their insurance, and that eliminating Sequoia's commissions and reducing administrative fees would therefore have lowered plaintiffs' payments. This theory applied to the allegations in the complaint fails to satisfy the requirements for Article III standing.

"To establish injury in fact, a plaintiff must show that he or she suffered 'an invasion of a legally protected interest' that is 'concrete and particularized' and 'actual or imminent, not conjectural or hypothetical.' " Spokeo, 578 U.S. at 339, 136 S.Ct. 1540 (quoting Lujan, 504 U.S. at 560, 112 S.Ct. 2130). A "concrete" injury "must actually exist," and must be "real, and not abstract." Id. at 340, 136 S.Ct. 1540 (quotations omitted).

Even assuming plaintiffs' factual allegations are true, plaintiffs have not "clearly . . . allege[d] facts demonstrating" a concrete injury. Id. at 338, 136 S.Ct. 1540 (ellipsis in original) (quotation omitted). Plaintiffs' out-of-pocket-injury theory boils down to the following thesis: (1) Sequoia failed adequately to negotiate administrative fees and accepted improper commissions; (2) the RingCentral plan therefore had to pay higher total premiums than it would have absent the alleged misconduct; (3) and plaintiffs thus paid higher contributions, and would have paid lower contributions if Sequoia's allegedly wrongful conduct had never occurred.

The problem with plaintiffs' theory is that plaintiffs have not pleaded facts tending to show that Sequoia's alleged breach of fiduciary duty led to plaintiffs paying higher contributions. It is RingCentral, and not Sequoia, that sets plaintiffs' contribution amounts. It is likewise RingCentral that decides which coverage options to make available to its employees through the RingCentral plan. Plaintiffs have not alleged that RingCentral has changed or would change employee contribution rates based on Sequoia's alleged breaches of fiduciary duty, or that employee contribution rates are tied to overall premiums. Indeed, amicus the U.S. Department of Labor—which supports plaintiffs (though more heavily on their second theory of injury)—candidly acknowledges that "Plaintiffs do not expressly allege that they would pay lower contributions in the future if Defendants' commissions were eliminated."

Lacking express factual allegations, plaintiffs urge us to infer this key premise. But several of plaintiffs' allegations directly undermine their argument. Plaintiffs in their complaint allege that, under the RingCentral plan and Tech Benefits Program, RingCentral alone determined the share of employee contributions—if it required any contributions in the first place. In making such determinations, RingCentral had broad discretion, restrained only by a clause in the Tech Benefits Program agreement which required RingCentral to contribute at least 75% of the cost for employees who selected single insurance coverage or 50% of the cost for family coverage. During the relevant period, RingCentral also offered employees benefit options that required no employee contribution at all. Perhaps most significantly, plaintiffs allege in their complaint that when they asked RingCentral how employee contributions were determined, RingCentral "did not identify a specific formula or set of factors." Instead, RingCentral merely "cited 'various factors and discussion.' " These allegations underscore the role that RingCentral played in setting employee contributions and highlight plaintiffs' failure to plead facts indicating that RingCentral set those contributions based on overall premium costs.

Plaintiffs draw our attention to other allegations, which they claim support the inference they wish us to draw. Plaintiffs allege that RingCentral "contributes 80-90% of the cost for employee medical benefits." Plaintiffs further allege that Sequoia "advise[s] employers that a 'common strateg[y]' for determining the employee contribution is to require [employers] to pay 90% of the required contribution for [individual] coverage and 75% for family members." And plaintiffs allege that when the total insurance premium for one participant's vision plan decreased in 2019, her contribution decreased in a roughly proportionate manner, remaining at about 5% of the total premium.

But these allegations are general in nature and do not solve the variable of RingCentral's discretion in setting employee contribution rates. And even if these particular allegations are broadly consistent with plaintiffs' theory, "[w]here a complaint pleads facts that are merely consistent with a defendant's liability, it stops short of the line between possibility and plausibility of entitlement to relief." Ashcroft v. Iqbal, 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009) (quotations omitted). Given RingCentral's discretion in setting employee contributions, and the fact that, as alleged, RingCentral determined those contributions based not on a "specific formula or set of factors," but on "various factors and discussion"—which in some instances resulted in employees paying nothing—we lack a sufficient basis to draw the inference that plaintiffs seek. The facts as pleaded are thus insufficient to support plaintiffs' assertion that "employee contributions are calculated as a pro rata share of the total benefits cost." See Iqbal, 556 U.S. at 681, 129 S.Ct. 1937 ("[T]he allegations are conclusory and not entitled to be assumed true."). Plaintiffs' allegations do not demonstrate that they paid higher contributions because of Sequoia's allegedly wrongful conduct.

This deficiency in plaintiffs' allegations can also be understood as a failure to plead causation, the second element of Article III standing. To establish causation, plaintiffs must allege that their injuries are "fairly traceable" to Sequoia's conduct and "not the result of the independent action of some third party not before the court." Namisnak v. Uber Techs., Inc., 971 F.3d 1088, 1094 (9th Cir. 2020) (quoting Lujan, 504 U.S. at 560, 112 S.Ct. 2130). Plaintiffs must thus allege a "substantial probability" that Sequoia caused the harm they claim to have suffered. City of Oakland v. Oakland Raiders, 20 F.4th 441, 452-53 (9th Cir. 2021) (quotation omitted). Plaintiffs have not done so.

Even assuming Sequoia's alleged breach of fiduciary duty resulted in higher total insurance costs for the RingCentral plan (and that any increased costs were not caused by other factors), plaintiffs' chain of causation again encounters the same issue: RingCentral's discretion in setting employee contributions, and the lack of factual allegations tying RingCentral's employee contribution amounts to overall premium rates. Indeed, as discussed above, the few facts plaintiffs have alleged on this score contradict that inference. RingCentral's lack of set formula and its weighing of "various factors and discussion" in setting employee contributions cannot be squared with plaintiffs' mechanistic assertion that RingCentral sets contribution rates based on Sequoia's premium rates. RingCentral was always free to change the employee contribution rates (subject to the minimum limitations of the Tech Benefits Program); reducing Sequoia's commissions would not require RingCentral to change its plan design, and plaintiffs do not plead facts suggesting RingCentral would do so. Plaintiffs have not sufficiently alleged that any out-of-pocket financial injuries would be fairly traceable to Sequoia's alleged breach of fiduciary duty.

Finally, plaintiffs fail to plead redressability for their out-of-pocket-injury theory. To establish redressability, plaintiffs must allege that it is "likely, as opposed to merely speculative," that a favorable decision will redress their injuries. Skyline Wesleyan Church v. Cal. Dep't of Managed Health Care, 968 F.3d 738, 749 (9th Cir. 2020) (quoting Friends of the Earth, Inc. v. Laidlaw Env't Servs. (TOC), Inc., 528 U.S. 167, 181, 120 S.Ct. 693, 145 L.Ed.2d 610 (2000)); see also TransUnion, 141 S. Ct. at 2203. Plaintiffs argue that their injuries may be redressed in two ways: (1) by imposing "a constructive trust on [Sequoia's] ill-gotten profits for the purpose of distributing those funds to Plaintiffs"; and (2) by returning money to the RingCentral plan, which would in turn, plaintiffs maintain, apportion the recovery to the plan participants. Both theories are deficient as pleaded.

Plaintiffs' theory that they could personally recover funds directly from Sequoia lacks sufficient supporting allegations for much the same reason we have already explained. We will assume that, in this context, ERISA permits this form of constructive-trust relief (as opposed to recovery by the plan only). Even so, plaintiffs do not explain how a court could place Sequoia's "ill-gotten profits" directly into plaintiffs' pockets when plaintiffs have not alleged how a court could identify the discrete "profits" supposedly owed to them, given RingCentral's discretion in setting employee contribution amounts and the manner in which RingCentral exercised this discretion. Tellingly, the Department of Labor devotes almost no attention to this point, either.

Plaintiffs' second redressability theory—that awarding damages to the RingCentral plan would redress their injury—is more practicable, but it is foreclosed by our decision in Glanton ex rel. ALCOA Prescription Drug Plan v. AdvancePCS Inc., 465 F.3d 1123, 1125 (9th Cir. 2006). In Glanton, the plaintiffs claimed that a plan fiduciary charged their employers' health plans too much for prescription drugs, which allegedly required the plans to demand higher co-payments and contributions from participants. Id. Part of the Glanton plaintiffs' theory of redressability was that awarding damages to their employers' plans would redress their injuries. Id. But we held that the plaintiffs failed to plead redressability, explaining that "any one-time award to the plans for past overpayments [would not] inure to the benefit of participants." Id. This was so because the employers "would be free to reduce their contributions or cease funding the plans altogether until any such funds were exhausted." Id. We therefore held that "[t]here is no redressability, and thus no standing, where (as is the case here) any prospective benefits depend on an independent actor who retains broad and legitimate discretion the courts cannot presume either to control or to predict." Id. (quotations omitted). That same logic governs here.

Plaintiffs contend that Glanton is factually distinguishable on the theory that there, the plaintiffs did not make contributions to their plan and thus did not contribute to the disputed amounts that the plan had paid. But that is not what Glanton says. Glanton repeatedly referenced the fact that plaintiffs were alleging that defendants had "caused the plans to demand higher co-payments and contributions from participants." Id. (emphasis added); see also id. ("Plaintiffs claim that, if their suit is successful, the plans' drug costs will decrease, and that the plans might then reduce contributions or co-payments."); Appellant's Br., Glanton v. AdvancePCS Health, L.P., No. 04-15328, 2004 WL 1533744, at *25-26 (9th Cir. June 2, 2004) ("The more the Plan needs to be funded, the greater the direct employee contributions, whether in the form of co-payments, co-insurance, deductibles, or in the form of monthly contributions to the Plan." (emphasis added)).

As Glanton held, if RingCentral were to receive Sequoia's "ill-gotten" gains, RingCentral would "be free to reduce [its] contributions or cease funding the plan[ ] altogether until any such funds were exhausted." Id. By using the funds to pay for plaintiffs' health insurance, RingCentral would still be using them for plaintiffs' benefit. Glanton indicates that this would remain true even if RingCentral used the recovered funds to offset its own contributions while making no reduction to the contributions it required from plaintiffs. See id. There is thus no basis for plaintiffs' assertion that, if the RingCentral plan received money from Sequoia, the plan would "likely" remit that money to plaintiffs. Plaintiffs have identified nothing in the plan documents or in law that would require this or make it probable.

Plaintiffs rely instead on Department of Labor guidance that supposedly recommends such offsetting. See U.S. Dep't of Labor, Emp. Benefits Sec. Admin., Technical Release No. 2011-04, at 1-2 (Dec. 2, 2011). But, as the Department of Labor acknowledges, "the Technical Release is guidance for group health plans rebates pursuant to the Medical Loss Ratio Requirements of the Public Health Service Act." The Technical Release thus does not govern here. Neither plaintiffs nor the Department of Labor cite any authority, or anything in the governing plan documents, suggesting that RingCentral would violate its fiduciary duties under ERISA if it did not pass along to plaintiffs any money recovered from Sequoia.

For these reasons, plaintiffs' out-of-pocket-injury theory fails on each requirement of the Article III standing calculus.

B

Plaintiffs' second theory of injury is that they retained an "equitable" ownership interest in the Tech Benefits Program's trust fund. They rely on a series of cases and secondary sources for the proposition that "the harm of a trustee engaging in self-dealing with trust assets traditionally provide[s] a basis for a lawsuit on the part of the trust's beneficiaries." Plaintiffs contend that their equitable interest as beneficiaries of the Tech Benefits Program trust provides them standing to pursue relief such as surcharge or disgorgement, even if they suffered no tangible out-of-pocket loss.

This theory of standing runs aground under Thole v. U.S. Bank N.A., — U.S. —, 140 S. Ct. 1615, 207 L.Ed.2d 85 (2020). In Thole, the Supreme Court held that participants in a defined-benefit pension plan lacked standing to bring an ERISA claim alleging that the plan's fiduciaries had violated their duties of loyalty and prudence by poorly investing the plan's assets. Id. at 1618-19. Under the defined-benefit plan at issue in Thole, the plan participants received a fixed monthly payment, which did not fluctuate based on the value of the plan. Id. at 1618. The plaintiffs had therefore suffered no monetary injury because they had received all the monthly payments they were owed, and they were legally entitled to receive those same payments for the rest of their lives. Id. at 1622.

To get around this, the plaintiffs in Thole had pointed to trust law principles and contended that "an ERISA defined-benefit plan participant possesses an equitable or property interest in the plan." Id. at 1619. They asserted that "a plan fiduciary's breach of a trust-law duty of prudence or duty of loyalty itself harms ERISA defined-benefit plan participants, even if the participants themselves have not suffered (and will not suffer) any monetary losses." Id. at 1619. But the Supreme Court rejected this theory of standing, explaining that "plan participants possess no equitable or property interest in the plan." Id. at 1620 (first citing Hughes Aircraft Co. v. Jacobson, 525 U.S. 432, 439-41, 119 S.Ct. 755, 142 L.Ed.2d 881 (1999); and then citing LaRue v. DeWolff, Boberg & Assocs., Inc., 552 U.S. 248, 254-56, 128 S.Ct. 1020, 169 L.Ed.2d 847 (2008)).

Just like the plaintiffs here, the Thole plaintiffs had supported their theory of standing with an analogy to trust law. But the Supreme Court was not persuaded. The Court in Thole explained that the "basic flaw in the plaintiffs' trust-based theory of standing [was] that the participants in a defined-benefit plan are not similarly situated to the beneficiaries of a private trust or to the participants in a defined-contribution plan," such as a 401(k). Id. at 1619. "In the private trust context, the value of the trust property and the ultimate amount of money received by the beneficiaries will typically depend on how well the trust is managed, so every penny of gain or loss is at the beneficiaries' risk." Id. But in a defined-benefit plan, which "is more in the nature of a contract," the participants' benefits "will not change, regardless of how well or poorly the plan is managed." Id. at 1620. Moreover, the employer gets to keep any surplus and must make up for any shortfall. Id. The plaintiffs' trust law analogy therefore "d[id] not fit th[at] case and d[id] not support Article III standing for plaintiffs who allege mismanagement of a defined-benefit plan." Id.

Here, plaintiffs have not established that they have some equitable interest in plan funds that the Thole plaintiffs lacked, or that the comparison to trust law can have purchase here when it did not in Thole. Although the Tech Benefits Program is not a defined-benefit pension plan, it similarly provides a fixed set of benefits as promised in plan documents. Like the plan in Thole, the Tech Benefits Program differs from a private trust. The program is a large pool of money that is not divided into individual accounts. Plaintiffs do not own beneficial interests that increase or decrease depending on the management of trust assets. Once employer plans have paid into the program, the program provides a set level of agreed-upon benefits. See Smith v. Med. Benefit Adm'rs Grp., Inc., 639 F.3d 277, 283 (7th Cir. 2011) (describing a group health insurance plan as "the kind of defined benefit plan . . . which typically holds no assets in trust for any individual participant"). Unlike private trust beneficiaries, plaintiffs have not alleged that they are entitled to receive the funds held by the program. Instead, plaintiffs were contractually entitled to the insurance benefits that Sequoia agreed to purchase for them with the program's funds—benefits that plaintiffs have received.

Plaintiffs' attempts to avoid the import of Thole are unpersuasive. Plaintiffs (but not the Department of Labor) rely on the Fourth Circuit's decision in Peters v. Aetna, Inc., 2 F.4th 199 (4th Cir. 2021), to support their theory of injury premised on a purported equitable interest in the Tech Benefits Program funds. Peters considered the Article III standing of ERISA plan participants based in part on the participants' alleged equitable interest in their plans. Yet even though Thole was decided a year before Peters—and was flagged for the Peters court in supplemental authority letters while Peters was still pending—the Fourth Circuit in Peters did not address Thole. See Peters, 2 F.4th at 217-21. Peters also independently found Article III standing based on the plaintiffs' "financial" injury, Id. at 218-19, and plaintiffs here have not explained how that fact-specific holding is relevant to this case.

Plaintiffs also attempt to distinguish Thole on the facts, arguing that they have a concrete interest because Sequoia's alleged mismanagement increased their insurance costs. But in so arguing, plaintiffs effectively revert to their out-of-pocket-injury theory, which we have already held is insufficiently pleaded. In short, here, as in Thole, plaintiffs argue that they had an equitable interest in funds used for ERISA benefits. Here, as in Thole, plaintiffs argue that defendants breached their fiduciary duties of prudence and loyalty with respect to those funds. And here, as in Thole, plaintiffs have not alleged how this claimed breach concretely affected them.

* * *

We hold that plaintiffs have failed to plead the requirements for Article III standing. Like the district court, we therefore do not reach Sequoia's argument that it is not an ERISA fiduciary.

AFFIRMED.