PRECEDENTIAL
UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT ______
No. 22-2552 ______
ROBERT MATOR; NANCY MATOR, individually and as representatives of a class of participants and beneficiaries in and on behalf of WESCO Distribution, Inc. Retirement Savings Plan, Appellants
v.
WESCO DISTRIBUTION, INC.; THE ADMINISTRATIVE AND INVESTMENT COMMITTEE FOR WESCO DISTRIBUTION INC RETIREMENT SAVINGS PLAN; JOHN AND JANE DOES 1-30 ______
On Appeal from the United States District Court for the Western District of Pennsylvania (D.C. No. 2-21-cv-00403) District Judge: Honorable Marilyn J. Horan ______
Argued on April 18, 2023 Before: HARDIMAN, PORTER and FISHER, Circuit Judges. (Filed: May 16, 2024)
Timothy L. Foster Paul R. Wood Franklin D. Azar & Associates 14426 East Evans Avenue Aurora, CO 80014
Mariah Heinzerling Beena M. McDonald Steven A. Schwartz [ARGUED] Chimicles Schwartz Kriner & Donaldson-Smith 361 W Lancaster Avenue One Haverford Centre Haverford, PA 19041 Counsel for Appellant
Deborah S. Davidson Morgan Lewis & Bockius 110 N Wacker Drive Suite 2800 Chicago, IL 60606
Michael E. Kenneally Matthew J. Sharbaugh [ARGUED] Morgan Lewis & Bockius 1111 Pennsylvania Avenue NW Suite 800 North Washington, DC 20004
Stephanie R. Reiss Morgan Lewis & Bockius
2 301 Grant Street One Oxford Centre, Suite 3200 Pittsburgh, PA 15219 Counsel for Appellee
Jordan Bock Goodwin Procter 100 Northern Avenue Boston, MA 02210
Jaime A. Santos Goodwin Procter 1900 N Street NW Washington, DC 20036
Jordan Von Bokern United States Chamber Litigation Center 1615 H Street NW Washington, DC 20062 Counsel for Amicus Appellee ______
OPINION OF THE COURT ______ FISHER, Circuit Judge. Plaintiffs Nancy Mator and Robert Mator participate in the Wesco Distribution, Inc. Retirement Savings Plan. On behalf of themselves and a class of participants and beneficiaries, the Mators sued the Plan, its fiduciaries, and Wesco Distribution, Inc. (collectively, “Wesco”). The Mators allege Wesco violated fiduciary duties imposed by the Employee Retirement Income Security Act of 1974 (ERISA) because it paid excessive recordkeeping fees and failed to
3 monitor the Plan. The District Court dismissed the complaint with prejudice. But under controlling law, the complaint properly states these claims. We will therefore vacate and remand.
At the motion-to-dismiss stage, the facts are limited to the allegations in the complaint. In re Asbestos Prods. Liab. Litig. (No. VI), 822 F.3d 125, 133 (3d Cir. 2016). Courts accept the factual allegations as true and view them in the light most favorable to the plaintiff. Doe v. Princeton Univ., 30 F.4th 335, 340 (3d Cir. 2022). Courts may also “consider documents integral to or explicitly relied upon in the complaint or any undisputedly authentic document that a defendant attaches as an exhibit to a motion to dismiss if the plaintiff’s claims are based on the document.” In re Asbestos Prods., 822 F.3d at 133 n.7 (internal quotation marks, citations, and emphasis omitted); see also Steinhardt Grp. Inc. v. Citicorp, 126 F.3d 144, 145 & n.1 (3d Cir. 1997) (considering contracts underlying the complaint, which were attached to the motion to dismiss). To its motion to dismiss, Wesco attached the Plan’s fee disclosures and excerpts of the Plan’s service agreements. It also attached several Form 5500s, which are reports that retirement plans must file annually with the federal government. The Mators did not take exception to Wesco’s attachments and, in fact, drew on them in amending their complaint. No one has disputed the documents’ authenticity and the Mators’ claims are based on them. The District Court therefore properly relied on these documents, and we may as well. See In re Asbestos Prods., 822 F.3d at 133 n.7. According to the documents, as well as the allegations in the complaint, the facts are as follows.
4 1. The Plan and its recordkeeping fees Wesco Distribution, Inc., a Pittsburgh company, maintains the Wesco Distribution, Inc. Retirement Savings Plan. The Plan is subject to ERISA. Wesco and its Administrative and Investment Committee are the Plan’s administrators, which makes them ERISA fiduciaries. See 29 U.S.C. § 1002(21)(A). At the end of 2020, the Plan had about $837 million in assets and nearly 8,300 participants. Among those participants are plaintiffs Robert Mator and Nancy Mator. The Plan is a defined contribution plan. Thus, it “promises the participant the value of [his or her] individual account at retirement.” Boley v. Universal Health Servs., Inc., 36 F.4th 124, 128 n.2 (3d Cir. 2022) (quoting LaRue v. DeWolff, Boberg & Assocs., 552 U.S. 248, 250 n.1 (2008)). That value is determined by how much was put into the account on the participant’s behalf, whether the chosen investments gain or lose value, and how much the account pays in fees. Id. Between 2015 and 2020, Wells Fargo was the Plan’s recordkeeper. Wells Fargo provided participants with internet access to their accounts, transaction processing, quarterly statements, communications including disclosures and newsletters, retirement education, telephone support, and a “brokerage window” that allowed participants to invest in stocks that were not part of the plan’s menu of options. App. 2137. Wells Fargo was paid for these recordkeeping services through direct and indirect fees. Direct fees are paid from a plan’s assets: “the fiduciary contracts with the recordkeeper to obtain services in exchange for a flat annual fee based upon the number of participants.” App. 2130. Indirect fees are paid by participants as a result of revenue-sharing agreements between recordkeepers and plan investments, such as mutual funds. “In
5 a revenue sharing arrangement, the mutual fund pays the plan’s recordkeeper putatively for providing recordkeeping and administrative services for the [mutual] fund.” App. 2131. Indirect fees paid through revenue sharing are based on the amount of assets in participants’ mutual fund accounts. The Mators allege the retirement plan services market is “highly competitive,” App. 2126, and large plans like Wesco’s “have the bargaining power to obtain the highest level of service and the lowest fees,” App. 2121. Such plans “possess tremendous economies of scale,” App. 2127, because “the marginal cost of adding an additional participant to a recordkeeping platform is relatively low” and recordkeeping services for any participant cost about the same regardless of the participant’s account balance, App. 2126. “Therefore, . . . [a]s the number of participants in the plan increases, the cost per[ ]participant to deliver the recordkeeping and administrative services decreases.” App. 2127. Given these dynamics, “a flat price per participant . . . ensures that the compensation [paid to a recordkeeper] is tied to the actual services provided and does not grow” just because participants have contributed more to their accounts or the market has gone up. App. 2130–31. Indirect fees, “if not closely monitored,” may become unreasonable because they “bear no relation to the actual cost to provide reasonable recordkeeping and administrative services.” App. 2133–34. The Mators allege the fiduciary’s standard of care for negotiating and monitoring recordkeeping fees is “well established . . . based on [Department of Labor] guidelines, case law, and best practices as shared by retirement plan professionals.” App. 2134. With regard to the substance of the standard, the Mators cite a consulting firm’s publication to allege prudent plan fiduciaries pay administrative fees on a per- participant basis, benchmark and negotiate fees every other
6 year, ensure that only contractually required fees are actually paid, and review services annually. And they allege a survey of retirement plans determined that plans with 5,000 to 10,000 participants (like the Plan) pay $40 to $60 in fees. The Mators further allege that “[r]ecordkeeping fees for large plans have declined significantly in recent years due to increased technological efficiency, competition, and increased attention to fees . . . .” App. 2135. Thus, “fees that may have been reasonable at one time may have become excessive.” Id. The Mators allege that to avoid overpaying, prudent fiduciaries regularly ask other recordkeepers to bid on services. Getting bids is easy, they allege, because plans need only provide a few basic facts: the number of participants and, possibly, the amount of assets. The Mators claim the competitive bidding process should be conducted “at least once every three to five years.” App. 2136. Yet Wesco failed to do so for over ten years. From 2015 through 2020, the Plan paid Wells Fargo direct fees of between $50 and $82 per participant—on average, $66. As opposed to a typical direct fee (a flat amount of dollars per participant), this was a “direct asset-based fee”— a percentage of each participant’s account balance paid directly from his or her account to Wells Fargo. App. 2138. Wells Fargo was also paid indirect fees (revenue sharing payments from Plan investments) of between $80 and $103—on average, $91. Thus, the total per-participant fees (direct plus indirect) ranged from $110 to $185—on average, $153–$154. This is something like two to four times the alleged $40 to $60 industry average as of 2018. Fees were not capped, nor was Wells Fargo “required to refund any excess amounts collected.” App. 2138–39. The Mators allege “the Plan should unquestionably have been able to obtain recordkeeping and administrative services [at] significantly lower rates” than it paid. App. 2141. And the Plan’s fees “were excessive relative
7 to the type and quality of the services received by the Plan when benchmarked against other similar-sized plans for the same or similar recordkeeping and administrative services.” App. 2137–38. “These excessive fees led to lower net returns” for Plan participants. App. 2138. The Mators provide a table in their complaint showing five other plans for which Wells Fargo provided recordkeeping services. In 2018, these plans had between 6,500 and 12,700 participants (compared to the Plan’s 8,600 participants) and $219 million to $1.1 billion in plan assets (compared to the Plan’s $671 million). The other plans allegedly paid $37 to $52 per participant—an average of $44—while the Plan paid $154. The Mators next provide a table showing eleven plans that received recordkeeping services from other providers such as Vanguard and Fidelity. In 2018, these plans had between 4,950 and 13,500 participants and plan assets between $221 million and $2.1 billion. The other plans allegedly paid from $31 to $53 per participant—an average of $42—compared to the Plan’s $154. But there is a caveat to these comparisons: the Mators do not have complete information. They calculated direct fees based on publicly available information, including the plans’ Form 5500s. The Form 5500—a joint creation of the IRS, the Department of Labor, and the Pension Benefit Guaranty Corporation—is the “Annual/Return Report of [an] Employee Benefit Plan.” 2018 Instructions for Form 5500;1 see also 29 U.S.C. §§ 1021, 1024(a) (requiring plans to file annual
1 https://www.dol.gov/sites/dolgov/files/EBSA/employ ers-and-advisers/plan-administration-and- compliance/reporting-and-filing/form-5500/2018- instructions.pdf [https://perma.cc/6RP3-6N5C].
8 reports). The Form 5500 requires a plan to list the amount of direct compensation it paid. But the Form 5500 asks only whether indirect compensation was paid—yes or no—rather than asking the dollar amount. To determine indirect fees, the Mators looked at the list of investments a plan offered its participants (such as money market funds and various types of mutual funds) and then sussed out the revenue sharing rates for those investments by checking a different part of the Form 5500 or using “publicly available” rates. App. 2142. While these allegations provide a somewhat detailed description of how the Mators calculated indirect fees, they do not allow us to replicate the calculations. For instance, the complaint does not say where the revenue-sharing rates were publicly available or what the rates were. Much of the dispute about the sufficiency of the complaint turns on whether it provides apples-to-apples comparisons between the Plan and other plans. The Mators beefed up successive versions of their complaint to show their comparisons are valid. For instance, they allege that either the Plan’s “direct fees alone” or its “indirect fees alone” were “unreasonable compared to the total fees . . . that other similar plans paid.” App. 2138, 2139. The implication is that the Plan overpaid, even if the comparisons are imperfect. To support the contention that the comparisons actually are sound, the Mators repeatedly allege that all defined contribution plans buy essentially the same bundle of recordkeeping services with no appreciable difference in quality. App. 2125 (“For large plans . . . , any minor variations in the way that these essential services are delivered have no material impact on the fees charged . . . .”); App. 2137 (recordkeeping services Wells Fargo provided to the Plan were “typical of the services provided to any large defined
9 contribution plan”); App. 2145 (“During the Class Period, Fidelity, Vanguard, T. Rowe Price, Voya and Transamerica all provided identical or similar services of the same quality to the comparator plans as those provided by Wells Fargo to the Plan.”). Finally, the Mators allege that in July 2020, the Plan switched to Fidelity for recordkeeping services and paid a much lower flat fee of $54 per participant. “Plaintiffs and other Plan participants received the same services from Fidelity that they had previously received from Wells Fargo” and “did not experience a reduction in the level or quality of retirement plan services . . . .” App. 2147–48. “This further confirms that the services offered by Fidelity (and other large recordkeepers) are substantially similar in all material respects to the services offered by Wells Fargo.” App. 2148. In sum, the Mators allege that as a result of Wesco’s imprudence, the Plan paid “four times the reasonable cost of recordkeeping,” which caused the participants to lose “millions of dollars in their retirement savings over the last six-plus years.” App. 2149.
2. Mutual fund share classes offered by the Plan The Mators allege the Plan imprudently offered participants expensive classes of mutual fund shares. Mutual funds offer multiple classes of shares that are identical in every way except cost. The more expensive retail-class shares “are targeted at smaller investors with less bargaining power, while lower-cost shares are targeted at institutional investors with more assets, generally $1 million or more, and therefore greater bargaining power.” App. 2151. The Mators allege that “[e]ven when a plan does not yet meet the investment minimum to qualify for the cheapest available share class, it is well-known among institutional investors that mutual fund companies will
10 typically waive those investment minimums for a large plan.” App. 2152. The Mators concede that “plans often select mutual fund share classes that include revenue sharing to pay for some or all of the plan administrative expenses.” Id. But, they allege, “prudent fiduciaries monitor the amount of revenue sharing” to make sure indirect fees do not become “unreasonably high.” Id. If they do, the plans switch to cheaper share classes or obtain rebates. The Mators say that for nineteen of the mutual funds Wesco offered participants, it chose expensive share classes that were subject to revenue sharing and therefore “had higher operating expenses than other available classes of the same mutual funds.” Id. The Mators allege the Plan was already paying too much in direct fees, so if the defendants had been monitoring properly, “they would have realized that it was not necessary to allow Wells Fargo to collect additional fees [indirectly] through revenue sharing.” App. 2153. Offering expensive share classes allegedly caused participants to pay “excess costs of between $700,000 and $900,000 per year.” App. 2155. And the fees allegedly paid for services that were not needed because Wells Fargo was already providing them. Finally, the Mators allege the higher-cost share class offerings show that Wesco did not have “a prudent process to evaluate, negotiate, and/or monitor . . . fees.” Id.
3. Failure to monitor The Mators allege Wesco had the duty to monitor those “responsible for overseeing retirement plan service fees for the Plan to ensure that they were adequately performing their fiduciary obligations.” App. 2161. They allege Wesco breached that duty by, among other things, failing to monitor the responsible individuals and the processes by which those individuals administered the Plan.
11 The Mators filed their complaint in March 2021. In Count I, they alleged Wesco breached its duty of prudence by causing the Plan to pay excessive recordkeeping fees and by offering mutual fund investment options in higher-cost share classes. In Count II, they alleged Wesco breached its duty to monitor the processes and people administering the Plan. Wesco moved to dismiss for failure to state a claim, and the District Court granted the motion without prejudice. The Court held that in order for the excessive fee allegations to “pass from ‘possible to the plausible,’ an apples-to-apples comparison is necessary.” App. 57. But the Mators “allege[d] no facts about the level of services provided to the Plan’s participants” and did “not allege the complete nature and scope of services provided by the alleged comparator plans.” App. 56. The share-class allegations were “conclusory.” App. 59. The Court held the failure-to-monitor claim was insufficient as a matter of law because it was derived from Count I, the excessive-fee and share-class claims. The Mators amended their complaint in October 2021, but the District Court again dismissed it. Their final amended complaint, filed in April 2022, met with the same fate—except this time it was dismissed with prejudice. The Court continued to view the successive iterations of the complaint as conclusory and insufficiently specific. The Mators appeal.
12 2
Congress enacted ERISA to protect “employees and their dependents” whose “well-being and security” was affected by “the lack of . . . adequate safeguards” for employee benefit plans. 29 U.S.C. § 1001(a). Congress also did not want “to create a system that is so complex that administrative costs, or litigation expenses, unduly discourage employers from offering welfare benefit plans in the first place.” Fifth Third Bancorp v. Dudenhoeffer, 573 U.S. 409, 425 (2014) (quoting Varity Corp. v. Howe, 516 U.S. 489, 497 (1996)). ERISA therefore “represents a careful balancing between ensuring fair and prompt enforcement of rights under a plan and the encouragement of the creation of such plans.” Id. at 424 (internal citation and quotation marks omitted); see also Sweda v. Univ. of Pa., 923 F.3d 320, 327 (3d Cir. 2019) (ERISA furthers the distinct goals of “safeguarding anticipated employee benefits” and “assuring a predictable set of liabilities” for employers) (quoting first Cutaiar v. Marshall, 590 F.2d 523, 529 (3d Cir. 1979), then Renfro v. Unisys Corp., 671 F.3d 314, 321 (3d Cir. 2011)). ERISA requires the appointment of “one or more named fiduciaries who . . . shall have authority to control and manage the operation and administration” of an employee benefit plan. 29 U.S.C. § 1102(a)(1). A fiduciary must administer the plan “solely in the interest of the participants and beneficiaries and
2 The District Court had jurisdiction under 28 U.S.C. § 1331 (actions arising under the laws of the United States) and 29 U.S.C. § 1132(e)(1) and (f) (ERISA civil enforcement actions). This Court has jurisdiction under 28 U.S.C. § 1291 (final orders of district courts). We review on a plenary basis an order granting a motion to dismiss for failure to state a claim. In re Asbestos Prods., 822 F.3d at 131.
13 . . . with the care, skill, prudence, and diligence under the circumstances then prevailing that a prudent [person] acting in a like capacity and familiar with such matters would use in the conduct of an enterprise of a like character and with like aims.” Id. § 1104(a)(1). “A fiduciary’s process must bear the marks of loyalty, skill, and diligence expected of an expert in the field. It is not enough to avoid misconduct, kickback schemes, and bad-faith dealings. The law expects more than good intentions.” Sweda, 923 F.3d at 329. ERISA entitles a participant in a defined contribution plan to “the value of his account unencumbered by any fiduciary impropriety.” Graden v. Conexant Sys. Inc., 496 F.3d 291, 297 (3d Cir. 2007). A participant may sue a fiduciary who breaches his duties, 29 U.S.C. § 1132(a)(2), and the fiduciary is “personally liable to make good to such plan any losses to the plan resulting from each such breach,” id. § 1109(a). The standard for pleading an ERISA claim is, of course, key to the resolution of this case. As in all civil cases, the familiar pleading standards of Federal Rules of Civil Procedure 8(a) and 12(b)(6) require an ERISA plaintiff to “allege enough facts to state a claim to relief that is plausible on its face.” Renfro, 671 F.3d at 320–21 (quoting Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27, 46 n.12 (2011)). When assessing the sufficiency of the complaint, we pay attention to “the context of [the] claim, including the underlying substantive law.” Id. This means we evaluate the allegations bearing in mind ERISA’s twin goals of protecting participants and encouraging plan creation through a predictable set of liabilities for employers. Sweda, 923 F.3d at 326. In addition to the legal context, the factual context is paramount. “[A]pplying the pleading standard discussed in Ashcroft v. Iqbal, 556 U.S. 662, 129 S. Ct. 1937, 173 L. Ed. 2d 868 (2009), and Bell Atlantic Corp. v. Twombly, 550 U.S. 544,
14 127 S. Ct. 1955, 167 L. Ed. 2d 929 (2007),” to ERISA fiduciary breach claims “will necessarily be context specific” because “‘the content of the duty of prudence turns on the circumstances . . . prevailing at the time the fiduciary acts.’” Hughes v. Nw. Univ., 595 U.S. 170, 177 (2022) (quoting Dudenhoeffer, 573 U.S. at 425). The circumstances confronting a fiduciary sometimes require “difficult tradeoffs,” so “courts must give due regard to the range of reasonable judgments a fiduciary may make based on her experience and expertise.” Id. To plead a breach of the duty of prudence under ERISA, then, a plaintiff must plausibly allege fiduciary decisions outside a range of reasonableness. Id. Consistent with these rules, a plaintiff alleging an ERISA fiduciary breach need not “rule out every possible lawful explanation for the conduct he challenges.” Sweda, 923 F.3d at 326 (quoting Braden v. Wal-Mart Stores, Inc., 588 F.3d 585, 597 (8th Cir. 2009)). In addition to the Eighth Circuit, the Second and Seventh Circuits agree on this point. See Hughes v. Nw. Univ., 63 F.4th 615, 629 (7th Cir. 2023); Sacerdote v. New York Univ., 9 F.4th 95, 108 (2d Cir. 2021). However, the Rules require dismissal when fiduciary defendants offer an alternative explanation for their conduct that is “obvious,” “natural,” or simply “more likely” than the plaintiff’s theory of
15 misconduct. See Twombly, 550 U.S. at 567–68 (“obvious,” “natural”); Iqbal, 556 U.S. at 680 (“more likely”).3 The elements of an ERISA breach of fiduciary duty claim are: “(1) a plan fiduciary (2) breaches an ERISA- imposed duty (3) causing a loss to the plan.” Sweda, 923 F.3d at 328 (quoting Leckey v. Stefano, 501 F.3d 212, 225–26 (3d Cir. 2007). The parties dispute only the second element. In the two counts of their complaint, the Mators allege Wesco (A) breached its duty of prudence by causing or permitting the Plan to pay excessive fees for recordkeeping services and offering retail-class shares of mutual funds, and (B) breached its duty to monitor the fiduciaries and the administration of the Plan.
1. Excessive recordkeeping fees To explain why the Mators’ complaint states a claim, we return to Sweda, where we reversed a dismissal. 923 F.3d at 320. Sweda alleged the following: the market for recordkeeping services is competitive; large plans offer economies of scale for recordkeeping; “paying for recordkeeping with asset-based revenue sharing,” though “not per se [a] violation of ERISA, . . . can lead to excessive fees if not monitored and capped”; and competitive bids should be
3 In Sweda, we held that “Twombly’s discussion of alleged misconduct that is ‘just as much in line with a wide swath of rational and competitive business strategy’ is specific to antitrust cases.” 923 F.3d at 326 (quoting Twombly, 550 U.S. at 554). The Supreme Court recently abrogated that specific portion of Sweda by reiterating that in ERISA cases, courts are to “apply[] the pleading standard discussed in [Iqbal] and [Twombly].” Hughes, 595 U.S. at 177.
16 obtained for recordkeeping services. Amended Complaint, Sweda v. Univ. of Pa., No. 16-4329, Dkt. No. 27, pp. 44–48 (E.D. Pa. Nov. 21, 2016). Sweda alleged the University of Pennsylvania retirement plan paid fees of about $220 to $250 per participant—but based on the plan’s features, the recordkeeping services provided, the number of participants, and market conditions, “experts in the recordkeeping industry” determined the plan should have paid no more than $35 per participant. Id. at 50–51. Sweda did not support that allegation with any comparisons to other plans. See id. Keeping in mind the importance of context and the regard due to the “range of reasonable judgments” a fiduciary might make when faced with “difficult tradeoffs,” Hughes, 595 U.S. at 177, we observe that the Mators’ allegations are more specific than Sweda’s and provide more context. As in Sweda, the Mators allege the Plan’s fees were several times larger than what similar plans paid; the Plan’s fiduciary did not negotiate a fee cap or solicit bids (although revenue sharing rates fell on a percentage basis); the asset-based fee structure caused the Plan’s fees to rise when there was no corresponding increase in services; and similarly situated fiduciaries requested proposals and negotiated with recordkeepers to keep fees reasonable. Unlike Sweda, the Mators provide context for these allegations by making additional allegations about the amount of the fees paid by comparator plans. These comparisons nudge their complaint across the line from conceivable to plausible. See Twombly, 550 U.S. at 570. Although reasonable fiduciaries need not prioritize cost minimization above all else, there is enough evidence alleged here of a failure to act prudently. The District Court held that the comparisons merely provided a “side by side list,” not an “apples to apples” comparison. App. 19 (citation omitted), 22. It reached this
17 conclusion for three main reasons: differences in services provided by the comparator plans’ recordkeepers, differences in plan sizes, and problems with how the Mators calculated the fees paid by the Plan and its comparators. But a close examination of the complaint and attached documents shows that the District Court’s criticisms do not scuttle the comparisons or the complaint. Services provided by the various recordkeepers. On the Form 5500, plans are asked to list codes showing what services were rendered by the provider. The District Court held that the complaint did not account for the differences in the recordkeeping services listed on the comparator plans’ Form 5500s. There are indeed variations between the service codes, but the differences are not fatal. From 2015 to 2019, the Plan obtained six services from recordkeeper Wells Fargo: “Recordkeeping and information management (computing, tabulating, data processing),” “Trustee (bank, trust company),” “Direct payment from the plan,” “Float revenue,” “Participant loan processing,” and “Recordkeeping fees.” App. 2260, 2303, 2346, 2387, 2427; see also 2018 Form 5500 Instructions p. 27 (providing list of service codes). In 2020, when Fidelity became the Plan’s recordkeeper, the Plan’s Form 5500 retained two of these service codes, “Participant loan processing” and “Recordkeeping fees,” and listed three new ones: “Sub-transfer agency fees,” “Account maintenance fees,” and “Securities brokerage commissions and fees.” App. 2697. The Form 5500 for one comparator, the Red Lobster plan, lists six service codes that exactly match the Plan’s six service codes from 2015 to 2019. But the other comparators’ codes differ from the Plan’s and from one another’s. In addition to the previously mentioned services, some comparators listed: “Contract administrator,” “Consulting,”
18 “Trustee (directed),” “Investment advisory (participants),” “Securities brokerage,” “Investment management fees paid indirectly by plan,” and “Sub-transfer agency fees.” App. 2495, 2535, 2574, 2612, 2649, 2748. The different service codes do not undermine the Mators’ comparisons because they apparently overlap. All of the plans list either “Recordkeeping fees,” “Recordkeeping and information management (computing, tabulating, data processing),” or both. These two codes seem similar. Other codes seem to overlap as well, such as “Securities brokerage” and “Securities brokerage commissions and fees.” “Consulting” could cover a wide variety of services and could intersect with other categories. And it is unclear why the code “Direct payment from the plan” exists at all, since every dollar reported on this part of Form 5500 is “direct compensation paid by the plan.” See, e.g., App. 2260. At this stage, the record does not reveal the codes’ precise meanings, nor whether all plans define the codes consistently. But given that all the plans received some portion of an overlapping constellation of recordkeeping services, the comparisons help nudge the Mators’ claims across the line from possible to plausible. For their part, the Mators argue that “due to the nature and competitiveness of the market at the top levels [of] recordkeeping service for large plans, it would be reasonable to infer that similar services” were provided. Appellants’ Br. 41–42. Their complaint emphasizes this point. It alleges, for instance, that “[f]or large plans with greater than 5,000 participants, like the Plan, any minor variations in the way that these essential services are delivered have no material impact on the fees charged,” which is demonstrated by the fact that “all service providers quot[e] fees on a per-participant basis without regard for any individual differences in services requested.” App. 2125. Assuming the truth of this allegation,
19 the information about comparator plans helps render the Mators’ claim plausible despite the differences in service codes.4 Plan sizes. Second, the District Court faulted the Mators for providing comparators that did “not match the Plan relative to number of participants or asset sizes, or the average of all other plans of similar sizes.” App. 21. We know of no authority on how close comparators must be in size, and the Mators’ choices are sound. The comparator plans ranged from 4,950 to 13,502 participants and had assets of $221 million to $2.1 billion. The Plan—with 8,600 participants and $671 million in assets—falls roughly in the middle of that range. If we were to limit comparators to the 7,000 to 10,000 participant range, that would leave six of the eleven comparators. If we were to limit comparators to the $350 million to $1 billion asset range, that would leave four of the eleven comparators. Four does not seem insufficient. See Johnson v. PNC Fin. Servs. Grp., Inc., 2022 WL 973581, at *1 (W.D. Pa. Mar. 31, 2022) (denying motion to dismiss complaint that made allegations about four comparator plans). Wesco says that if we consider only those comparator plans whose participants and assets are 75% to 125% of the Plan’s 2018 participants and assets, three comparators would
4 Amicus Chamber of Commerce argues that “[r]ecordkeeping services are highly customizable” and “myriad services are available at different fee levels,” usually with “extraordinarily complicated” fee arrangements. Amicus Br. 17–18. If the Chamber is correct, it would be virtually impossible for a plaintiff to identify plans that buy precisely the same bundle of services as a defendant plan. That would make it more important for courts not to disregard every less- than-identical comparator.
20 remain. Wesco does not offer any authority supporting a 75%- to-125% limitation. Regardless, even accepting Wesco’s argument, the three remaining comparators allegedly paid 2018 per-participant fees of $51, $31, and $44, compared to the Plan’s $154. That is a stark difference. Drawing bright lines would run counter to the required contextual analysis, so we do not adopt any rules limiting comparator size or requiring a specific number of comparators. Regardless, winnowing the comparators as Wesco proposes would not leave the complaint bereft of allegations that help make a fiduciary breach plausible. By rejecting the comparator plans on the basis of size, the District Court erroneously failed to view the allegations in the light most favorable to the Mators. Construed most favorably to them, the comparisons show the Plan paid well above what others did. Moreover, assuming the truth of the allegations that large plans have superior bargaining power and the cost per participant falls as participant numbers increase, the smaller comparators actually strengthen the complaint. If smaller plans can obtain lower fees than the Plan despite less bargaining power and higher per-participant costs, it is more likely the Plan’s fiduciaries breached their duty of prudence. Fee calculations. Third, the District Court faulted the Mators’ comparisons based on their fee calculations. To understand the complications in calculating recordkeeping fees, we begin with Form 5500, which asks for the amount of “direct compensation paid by the plan to service providers.” See, e.g., App. 2260. With this data, ascertaining direct compensation seems straightforward enough. But for indirect compensation, the form asks only whether such fees were paid, not their amount. The District Court observed that for some comparators—two, to be precise—the Mators allege that the
21 total fee amounts are the same as the direct fee amounts shown on the Form 5500s, even though the plans also reported they paid indirect fees. We can infer that by leaving out indirect fees, the Mators underreported the total fees paid by these two plans. But even if we disregard the two plans, we are left with fourteen comparators that paid significantly lower fees than the Plan. Like the District Court, Wesco finds fault with the Mators’ calculations of comparators’ fees, saying: “Plaintiffs . . . claim to be adding indirect fee amounts ‘using publicly available revenue sharing rates’,” but “[t]hey do not explain where or how those rates for other plans are ‘publicly available.’” Appellees’ Br. 40–41 (quoting App. 2142–43). In other words, Wesco argues the Mators need to explain precisely how they calculated the fees alleged in the complaint. Although plaintiffs may be well advised to do so, we have not required this in the past. The Sweda plaintiff did not say how she calculated Penn’s indirect fees—only that her allegations were based on Penn’s Form 5500s and “information from sources including industry experts.” Sweda, No. 16-4329 (E.D. Pa.), Amended Complaint, ECF No. 27, p. 49. Yet her allegations, in context, stated a claim. Sweda, 923 F.3d at 332. The same is true here. In sum, the Mators’ comparisons between the Plan and other plans, while not perfect, are sufficient to plausibly state a claim for breach of fiduciary duty. Wesco cites several cases in support of its argument for affirmance. We agree with our sister Circuits’ articulation of the relevant law in those cases. See Matousek v. MidAmerican Energy Co., 51 F.4th 274, 278–79 (8th Cir. 2022); Smith v. CommonSpirit Health, 37 F.4th 1160, 1164–65 (6th Cir. 2022). Contrary to Wesco’s arguments, however, the fact that Matousek and Smith affirmed dismissal of the respective
22 complaints has little bearing on our decision. The complaints in both cases fell short in ways the Mators’ does not. The Smith opinion mostly analyzes a fiduciary breach claim based on excessive investment fees—which is a type of claim the Mators do not allege. 37 F.4th at 1165–69.5 The portion of Smith dealing with what we are looking at here—excessive recordkeeping fees—held the complaint was properly dismissed because the plaintiff’s only comparators were averages from an industry publication. Id. at 1169. That “fail[ed] to give the kind of context that could move [the] claim from possibility to plausibility,” because the plaintiff did “not plead[] that the services that CommonSpirit’s fee covers are equivalent to those provided by the plans comprising the average in the industry publication that she cites.” Id. Matousek is similar. See 51 F.4th at 279–80 (“[T]he way to plausibly plead a claim of this type is to identify similar plans offering the same services for less,” but “[r]ather than point to the fees paid by other specific, comparably sized plans, the plaintiffs rel[ied] on industry-wide averages” without accounting for differences in the services purchased.). Here, as we have explained, the Mators give the context the Smith and Matousek plaintiffs omitted: they provide specific plan comparators, not just industry averages, and plausibly allege that the services purchased were sufficiently similar to render the comparisons valid.
5 Another of Wesco’s cases, Forman v. TriHealth, Inc., is distinguishable insofar as it discussed excessive investment fees and breaches of the duty of loyalty. 40 F.4th 443, 449–50 (6th Cir. 2022). When it came to the share class claim, which is similar to the Mators’ claim we discuss below, the Sixth Circuit concluded—as we do here—that the complaint was sufficient. Id. at 450–51; see also Part II.A.2., below.
23 Wesco contends the statute of limitations does not allow us to consider the Mators’ allegation that Wesco failed to engage in a bidding process every three to five years. Wesco’s argument goes as follows: the six-year ERISA statute of limitations permits us to look back only as far as 2015, and because less than five years elapsed between that point and the Plan’s 2020 change in recordkeepers, the Mators are unable to allege that Wesco failed to get bids for more than five years. The Mators reply that allegations in the complaint are not bounded by the statute of limitations. They agree they cannot collect damages for breaches before 2015, but argue it is proper for their allegations to encompass a longer time period to allow for the necessary context-specific analysis. Reply Br. 16. They are correct. Discovery may be permitted for “events that occurred before an applicable limitations period” if “the information sought is . . . relevant to issues in the case.” Oppenheimer Fund, Inc. v. Sanders, 437 U.S. 340, 352 (1978). Therefore, Wesco’s statute-of-limitations based argument is unpersuasive. We agree with the Seventh Circuit that “a failure to regularly solicit quotes or competitive bids from service providers” does not necessarily give rise to an imprudence claim. Albert v. Oshkosh Corp., 47 F.4th 570, 579 (7th Cir. 2022). Still, “fiduciaries who fail to monitor the reasonableness of plan fees and fail to take action to mitigate excessive fees . . . may violate their duty of prudence.” Hughes, 63 F.4th at 625–26 (emphasis added). Wesco argues the complaint was correctly dismissed because the Plan’s recordkeeping fees fell between 2015 and 2020, from fourteen to six basis points (that is, from 0.14% to 0.06%), and the Plan rebated some fees to participants. As the Supreme Court instructs, we must give “due regard to the range of reasonable judgments a fiduciary may make.” Hughes, 595 U.S. at 177. We agree that “[s]ometimes an alternative
24 explanation for an ERISA fiduciary’s conduct may be patently more reasonable and better supported by the [alleged] facts than any theory of fiduciary duty violation pleaded by a plaintiff,” and in those circumstances “courts should not hesitate to dismiss.” Hughes, 63 F.4th at 630. But Wesco’s alternative explanation is not more reasonable or better supported than the Mators’ theory of misconduct, given the magnitude of the differences in fees—about two to four times what comparable plans paid. Wesco also attacks the Mators’ calculations of the Plan’s fees. Wesco says that by adding direct and indirect fees to come up with the total fees, the Mators “essentially double- count” because “Wells Fargo did not receive any ‘direct’ fees from the Plan, as confirmed by the absence of any such fees in the participant disclosures.” Appellees’ Br. 42 n.6 (citing App. 2200–30). But the Plan’s Form 5500s seem to say otherwise: each year, the Plan reported paying hundreds of thousands of dollars to Wells Fargo as “direct compensation.” App. 2260, 2303, 2346, 2387, 2427. Wesco’s explanation is therefore not obvious, natural, or more likely than the allegations of misconduct. See Twombly, 550 U.S. at 567–68; Iqbal, 556 U.S. at 680. Wesco also contends that “[f]or price comparisons to raise an inference of an imprudent fiduciary process, there must also be nonconclusory, fact-based allegations that the cheaper services were as good as or better.” Appellees’ Br. 33– 34. There are. The Mators allege that when the Plan switched to Fidelity for recordkeeping services, lowering fees from about $154 to $54 per participant, there was no change in the kind or quality of recordkeeping services provided to participants. In other words, Fidelity’s services allegedly were as good as Wells Fargo’s, but at a fraction of the price.
25 Although the Mators are correct that their complaint states a claim, it is worth noting that two of their arguments are off base. First, they assert the District Court applied an incorrect dismissal standard. They point out that the Court, when dismissing the original complaint, quoted another district judge’s opinion that erroneously said “allegations must cross ‘the threshold from possible to probable.’” App. 55 (quoting Johnson v. PNC Fin. Servs. Grp., Inc., 2021 WL 3417843, at *4 (W.D. Pa. Aug. 3, 2021)) (emphasis added). The correct standard is less demanding: the allegations must move the claim “from conceivable to plausible.” Twombly, 550 U.S. at 570 (emphasis added). When dismissing the amended and second amended complaints, the District Court retained the flawed citation, thus replicating the mistake. But despite this error, the Court stated the correct “plausibility” standard numerous times and clearly applied it. App. 23; see also App. 17, 19, 20, 38, 39, 42, 48, 55. Second, the Mators go too far when they contend the District Court improperly weighed credibility and decided facts by pointing out problems with their calculations. Judges draw on “common sense” when “[d]etermining whether a complaint states a plausible claim for relief.” Iqbal, 556 U.S. at 679. And if allegations are based on incorrect arithmetic, common sense says they are not well-pled. The calculation problems are not fatal here because even taking those problems into account, there are enough comparators, and the comparators are sufficiently similar to the Plan, to state a claim. But in a different case, calculation errors could conceivably lead to dismissal of a complaint. When considering whether a plaintiff has stated a claim for breach of fiduciary duty under ERISA, “we employ a holistic approach” that takes into account all of the well-pled facts. Sweda, 923 F.3d at 331. “The complaint should not be
26 ‘parsed piece by piece to determine whether each allegation, in isolation, is plausible.’” Id. (quoting Braden, 588 F.3d at 594). Here, the District Court parsed the allegations, found some of them not well-pled, and dismissed. But the Court’s criticisms, although partly valid, only nibble around the edges of the complaint. Even allowing for these criticisms, what remains plausibly states a claim.
2. Retail-class mutual fund shares The District Court concluded the Mators did not state a claim for breach of fiduciary duty based on their allegation that the Plan offered retail-class shares of some mutual funds, rather than identical but cheaper institutional-class shares. Wesco argues we should affirm. It points to the allegation that more expensive share classes that include revenue sharing can help pay for a plan’s administrative expenses. It argues that the more expensive share classes the Plan chose—which did indeed pay revenue sharing to the recordkeeper—represented a choice to pay some fees indirectly rather than directly. Therefore, Wesco says, the share-class claim is “interrelated[]” with the excessive-fee claim: if the fees were not too high overall, then paying them partly through revenue sharing was not imprudent. Appellees’ Br. 45. Wesco is correct that, as the Mators have pled their fiduciary breach claim, the excessiveness of the recordkeeping fees and the impropriety of offering retail-class shares are intertwined. But, as explained above, the excessive-fee claim is adequately pled. Because the Mators plausibly allege the fees were too high overall, it is therefore also plausible that it was a fiduciary breach to cause participants to pay indirect fees by offering mutual fund shares subject to revenue sharing. The District Court dismissed partly based on its conclusion that the Mators did not “address[] whether the retail
27 share class may have offered other benefits.” App. 42. To the contrary, the Mators do so. They allege plans might choose share classes subject to revenue-sharing agreements in order “to pay for some or all of the plan administrative expenses.” App. 2152. But the Mators allege this benefit was not realized here: first, the direct fees alone were too high, so paying additional fees through revenue sharing was imprudent; and second, the direct fees were already paying for the needed administrative services, so there was no reason to pay for more services indirectly. Although the Mators have alleged a fiduciary breach based on the Plan’s offerings of retail-class mutual fund shares, they overstate the holdings of Tibble v. Edison International, 575 U.S. 523 (2015), and Hughes, 595 U.S. 170. In both cases, the Supreme Court reversed the dismissal of share-class claims. Tibble, 575 U.S. at 525–26, 530; Hughes, 595 U.S. at 176–77. But in neither case did the Court decide whether offering retail-class shares breaches a fiduciary’s duty. In Tibble, the Ninth Circuit had erroneously held some claims were time-barred. 575 U.S. at 530. In Hughes, the Seventh Circuit had erroneously held that offering an array of cheap and expensive investment options insulates a fiduciary from liability. 595 U.S. at 176. The Supreme Court rejected those bases for dismissal and remanded for further proceedings without expressing a “view on the scope of [defendants’] fiduciary duty.” Tibble, 575 U.S. at 531; Hughes, 595 U.S. at 177. We therefore decline to articulate a bright-line rule that a plan administrator breaches its fiduciary duty merely by offering retail-class investment shares.
Count II of the complaint alleges Wesco breached its fiduciary duty by failing to monitor those responsible for the
28 Plan and the processes by which it was administered. “[F]ailure to ‘monitor . . . investments and remove imprudent ones’ may constitute a [fiduciary] breach.” Sweda, 923 F.3d at 328 (quoting Tibble, 575 U.S. at 530). Whether a “monitoring claim survives depends on whether [the] underlying breach of fiduciary duty . . . claims survive.” In re Allergan ERISA Litig., 975 F.3d 348, 354 n.11 (3d Cir. 2020) (citation omitted). The District Court held the failure to monitor claim should be dismissed because the underlying fiduciary breach claim failed. Id. The parties agree the failure to monitor claim derives from the primary claim of breach of the duty of prudence. Having vacated the Court’s dismissal of the fiduciary breach claim, we will vacate its dismissal of the derivative monitoring claim as well.
For all these reasons, we will vacate and remand for further proceedings.