Glenn TIBBLE; William Bauer; William Izral; Henry Runowiecki; Frederick Suhadolc; Hugh Tinman, Jr., as representatives of a class of similarly situated persons, and on behalf of the Plan, Plaintiffs–Appellants, v. EDISON INTERNATIONAL; the Edison International Benefits Committee, FKA The Southern California Edison Benefits Committee; Edison International Trust Investment Committee; Secretary of the Edison International Benefits Committee; Southern California Edison's Vice President of Human Resources; Manager of Southern California Edison's HR Service Center, Defendants–Appellees.
Glenn Tibble; William Bauer; William Izral; Henry Runowieck I; Frederick Suhadolc; Hugh Tinma N, Jr., as representatives of a class of similarly situated persons, and on behalf of the Plan, Plaintiffs–Appellees, v. Edison International; The Southern California Edison Benefits Committee, incorrectly named The Edison International Benefits Committee; Edison International Trust Investment Committee; Secretary of the Southern California Edison CompanyBenefits Committee, incorrectly named Secretary of the Edison International Benefits Committee; Southern California Edison's Vice President of Human Resources; Manager of Southern California Edison's Hr Service Center, Defendants–Appellants.
Current and former beneficiaries sued their employer's benefit plan administrator under the Employee Retirement Income Security Act charging that their pension plan had been managed imprudently and in a self-interested fashion. We must decide, among other issues, whether the Act's limitations period or its safe harbor provision are obstacles to their suit.
Edison International is a holding company for various electric utilities and other energy interests including Southern California Edison Company and the Edison Mission Group (collectively “Edison”), which itself consists of the Chicago-based Midwest Generation. Like most employer-organizations offering pensions today, Edison sponsors a 401(k) retirement plan for its workforce. During litigation, the total valuation of the “Edison 401(k) Savings Plan” was $3.8 billion, and it served approximately 20,000 employee-beneficiaries across the entire Edison International workforce. Unlike the guaranteed benefit pension plans of yesteryear, this kind of defined-contribution plan entitles retirees only to the value of their own individual investment accounts. See 29 U.S.C. § 1002(34). That value is a function of the inputs, here a portion of the employee's salary and a partial match by Edison, as well as of the market performance of the investments selected.
To assist their decision making, Edison employees are provided a menu of possible investment options. Originally they had six choices. In response to a study and union negotiations, in 1999 the Plan grew to contain ten institutional or commingled pools, forty mutual fund-type investments, and an indirect investment in Edison stock known as a unitized fund. The mutual funds were similar to those offered to the general investing public, so-called retail-class mutual funds, which had higher administrative fees than alternatives available only to institutional investors. The addition of a wider array of mutual funds also introduced a practice known as revenue sharing into the mix. Under this, certain mutual funds collected fees out of fund assets and disbursed them to the Plan's service provider. Edison, in turn, received a credit on its invoices from that provider.
Past and present Midwest Generation employees Glenn Tibbie, William Bauer, William Izral, Henry Runowiecki, Frederick Suhadolc, and Hugh Tinman, Jr. (“beneficiaries”) sued under the Employee Retirement Income Security Act of 1974 (ERISA), 29 U.S.C. § 1001, etseq., which governs the 401(k) Plan, and obtained certification as a class action representing the whole of Edison's eligible workforce.1 Beneficiaries objected to the inclusion of the retail-class mutual funds, specifically claiming that their inclusion had been imprudent, and that the practice of revenue sharing had violated both the Plan document and a conflict-of-interest provision. Beneficiaries also claimed that offering a unitized stock fund, money market-style investments, and mutual funds, had been imprudent.
The district court granted summary judgment to Edison on virtually all these claims. See Tibbie v. Edison Int'l, 639 F.Supp.2d 1074 (C.D.Cal.2009). The court also determined that ERISA's limitations period barred recovery for claims arising out of investments included in the Plan more than six years before beneficiaries had initiated suit. Id. at 1086; see 29 U.S.C. § 1113(1)(A).
Remaining for trial after these rulings was beneficiaries' claim that the inclusion of specific retail-class mutual funds had been imprudent. Without retreating from an earlier decision—at summary judgment—that retail mutual funds were not categorically imprudent, the court agreed with beneficiaries that Edison had been imprudent in failing to investigate the possibility of institutional-class alternatives. See Tibbie v. Edison Int'l, No. CV 07–5359, 2010 WL 2757153, at *30 (CD.Cal. July 8, 2010). It awarded damages of $370,000.
Beneficiaries timely appeal the district court's partial grant of summary judgment to Edison.2 Edison timely cross appeals, chiefly contesting the post-trial judgment.
Beneficiaries' first contention on appeal is that the district court incorrectly applied ERISA's six-year limitations period to bar certain of its claims. Edison argues for application of the shorter three-year period. We reject both parties' approaches to timeliness.
For claims of fiduciary breach, ERISA § 413 provides that no action may be commenced “after the earlier of”:
(1) six years after (A) the date of the last action which constituted a part of the breach or violation, or (B) in the case of an omission the latest date on which the fiduciary could have cured the breach or violation, or
(2) three years after the earliest date on which the plaintiff had actual knowledge of the breach or violation;
except that in the case of fraud or concealment, such action maybe commenced not later than six years after the date of discovery of such breach or violation.
29 U.S.C. § 1113.
Beneficiaries argue that the court erred by measuring the timeliness under ERISA § 413(1) for claims alleging imprudence in plan design from when the decision to include those investments in the Plan was initially made. They are joined in this contention by the United States Department of Labor (“DOL”). Because fiduciary duties are ongoing, and because section 413(1)(A) speaks of the “last action” that constitutes the breach, these claims are said to be timely for as long as the underlying investments remain in the plan. Essentially, they argue that we should either equitably engraft onto, or discern from the text of section 413 a “continuing violation theory.”
Beneficiaries' argument, though, would make hash out of ERISA's limitation period and lead to an unworkable result. We have previously declined to read the section 413(2) actual-knowledge provision as permitting the maintenance of the status-quo, absent a new breach, to restart the limitations period under the banner of a “continuing violation.” Phillips v. Alaska Hotel & Rest. Emps. Pension Fund, 944 F.2d 509, 520 (9th Cir.1991). In Phillips, the controlling opinion did not reach whether the same was true for section 413(1)(A). 944 F.2d at 520–21. Today we hold that the act of designating an investment for inclusion starts the six-year period under section 413(1)(A) for claims asserting imprudence in the design of the plan menu.
Preliminarily, we observe that in the case of omissions the statute already embodies what the beneficiaries urge for the last action. Section 413(1)(B) ties the limitations period to “the latest date on which the fiduciary could have cured the breach or violation.” Importing the concept into (1)(A), then, would render (1)(B) surplusage. This must be avoided when, as here, distinct meanings can be discerned from statutory parts. See Freeman v. Quicken Loans, Inc., ––– U.S. ––––, ––––, 132 S.Ct. 2034, 2043, 182 L.Ed.2d 955 (2012).
Second, beneficiaries' logic “confuse[s] the failure to remedy the alleged breach of an obligation, with the commission of an alleged second breach, which, as an overt act of its own recommences the limitations period.” Phillips, 944 F.2d at 523 (O'Scannlain, J., concurring). Characterizing the mere continued offering of a plan option, without more, as a subsequent breach would render section 413(1)(A) “meaningless and [could even] expose present Plan fiduciaries to liability for decisions made by their predecessors—decisions which may have been made decades before and as to which institutional memory may no longer exist.” David v. Alphin, 817 F.Supp.2d 764, 777 (W.D.N.C.2011), aff'd, 704 F.3d 327, 342–43 (4th Cir.2013). We decline to proceed down that path. As with the application of any statute of limitations, we recognize that injustices can be imagined, but section 413(1) “suggests a judgment by Congress that when six years has passed after a breach or violation, and no fraud or concealment occurs, the value of repose will trump other interests, such as a plaintiff's right to seek a remedy.” Larson v. Northrop Corp., 21 F.3d 1164, 1172 (D.C.Cir.1994).
Finally, we are unpersuaded by DOL's suggestion that our holding will give ERISA fiduciaries carte blanche to leave imprudent plan menus in place. The district court allowed beneficiaries to put on evidence that significant changes in conditions occurred within the limitations period that should have prompted “a full due diligence review of the funds, equivalent to the diligence review Defendants conduct when adding new funds to the Plan.” These particular beneficiaries could not establish changed circumstances engendering a new breach, but the district court was entirely correct to have entertained that possibility. See, e.g., Quan v. Computer Scis. Corp., 623 F.3d 870, 878–79 (9th Cir.2010) (explaining that “fiduciaries are required to act ‘prudently’ when determining whether or not to invest, or continue to invest”). The potential for future beneficiaries to succeed in making that showing illustrates why our interpretation of section 413(1)(A) will not alter the duty of fiduciaries to exercise prudence on an ongoing basis.
For its part, Edison contends that beneficiaries had actual knowledge of conduct concerning retail-class mutual funds, triggering ERISA § 413(2), more than three years before August 16, 2007, when the complaint was filed.3
In order to apply ERISA's limitation periods, the court “must first isolate and define the underlying violation.” Ziegler v. Conn. Gen. Life Ins. Co., 916 F.2d 548, 550–51 (9th Cir.1990). Here, as we explore in greater detail below,4 the crux of beneficiaries' successful theory of liability at trial was that alternatives to retail shares had not been investigated—not simply that their inclusion had been imprudent. Second, specific to section 413(2), the court must inquire as to when the plaintiffs had actual knowledge of that violation or breach. Id. at 552. Edison points to Summary Plan Descriptions provided to all participants, as well as to mutual fund prospectuses furnished to investors, claiming that these materials made the inclusion of retail shares known. Similar information was also furnished to the unions during negotiations.
But as the nature of the breach makes apparent, Edison is citing evidence of the wrong type of knowledge. When beneficiaries claim “the fiduciary made an imprudent investment, actual knowledge of the breach [will] usually require some knowledge of how the fiduciary selected the investment.” Brown v. Am. Life Holdings, Inc., 190 F.3d 856, 859 (8th Cir.1999). For example, in Waller v. Blue Cross of California, we explained that the three-year ERISA limitations period did not run from the time when the plaintiffs had purchased the subject annuities because their theory of breach was that the fiduciaries had “unlawfully employ[ed] an infirm bidding process” to acquire such annuities. 32F.3d 1337, 1339, 1341 (9th Cir.1994); see also Frommert v. Conkright, 433 F.3d 254, 272 (2d Cir.2006)(“The flaw with the district court's conclusion [under section 413(2) ] is that the plaintiffs' claim for breach of fiduciary duty is not premised solely on the defendants' adoption of the phantom account; rather, it is based on allegations that the defendants made ongoing misrepresentations about the origins of the phantom account in an effort to justify its usage.”).
Therefore, because these beneficiaries' trial claims hinged on infirmities in the selection process for investments, we hold that mere notification that retail funds were in the Plan menu falls short of providing “actual knowledge of the breach or violation.” § 413(2).
On its cross appeal, Edison claims that beneficiaries' entire case is proscribed by ERISA § 404(c), a safe harbor that can apply to a pension plan that “provides for individual accounts and permits a participant or beneficiary to exercise control over the assets in his account.” 29 U.S.C. § 1104(c)(1)(A).
As the Edison 401(k) is clearly such a plan we consider the terms of section 404(c). It provides that:
[N]o person who is otherwise a fiduciary shall be liable under this part for any loss, or by reason of any breach, which results from such participant's or beneficiary's exercise of control.
Id. § 1104(c)(1)(A)(ii).
Edison reads this statutory language as insulating it from all of beneficiaries' claims because each challenged investment was a product of a “participant's or beneficiary's exercise of control,” by virtue of his selection of it from the Plan menu. Disagreeing, the DOL directs us to its previously announced interpretations. In a 1992 regulation it stated that in order to fall within section 404's ambit, the breach or loss would need to be the “direct and necessary result” of the action by the beneficiary. 29 C.F.R. § 2550.404c–1(d)(2). A preamble that went through the notice-and-comment process and appeared in the agency's final rule, stated that “the act of limiting or designating investment options which are intended to constitute all or part of the investment universe of an ERISA section 404(c) plan is a fiduciary function which ․ is not a direct or necessary result of any participant direction.” 57 Fed.Reg. 46,922, 46,924 n.27 (Oct. 13, 1992).
To “reiterate its long held position,” 73 Fed.Reg. 43,014, 43,018 (July 23, 2008), DOL recently codified this guidance in the body of a new regulation so that it now appears in the Code of Federal Regulations, rather than in the preamble to a rule.5 See 75 Fed.Reg. 64,910, 64,946 (Oct. 20, 2010) (codified at 29 C.F.R. pt. 2550) (Section 404(c) “does not serve to relieve a fiduciary from its duty to prudently select and monitor any service provider or designated investment alternative offered under the plan”). This amended regulation, however, was not in effect during the time period at issue in this case.6 Our inquiry therefore centers on what appeared in the 1992 final rule.
As to these earlier materials, the parties and amici join issue on the status this court should accord them. Beneficiaries and DOL argue that they are entitled to the robust sort of administrative-law deference dictated by Chevron, U.S.A., Inc. v. Natural Resource Defense Council, Inc., 467 U.S. 837, 842–43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). Edison claims that a preamble is not the type of material to which courts properly defer. In any event, the California Employment Law Council, as amicus for Edison, argues that DOL's interpretation is an impermissible construction of the statute. See id. (“If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.”). Both Edison and the Employment Council rely on a divided opinion from the Fifth Circuit, and on an older case from the Third Circuit in which the alleged violations preceded the effective date of even the 1992 rule. See Langbecker v. Elec. Data Sys. Corp., 476 F.3d 299, 310–12 (5th Cir.2007); In re Unisys Sav. Plan Litig., 74 F.3d 420, 444–48 & n. 21 (3d Cir.1996).
Several other circuits, by contrast, have accepted the position advocated by DOL. See, e.g., Pfeil v. State St. Bank & Trust Co., 671 F.3d 585, 599–600 (6th Cir.2012)(favoring DOL's position in its “amicus curiae brief in this appeal and with the preamble to the regulations implementing the safe harbor”), cert. denied, ––– U.S. ––––, 133 S.Ct. 758, 184 L.Ed.2d 519 (2012); Howell v. Motorola, Inc., 633 F.3d 552, 567 (7th Cir.2011)(similar); DiFelice v. U.S. Airways, Inc., 497 F.3d 410, 418 n. 3 (4th Cir.2007)(implicitly deferring to the 1992 rulemaking).
The Chevron framework can apply only if two initial conditions are met: (1) Congress has delegated the power to that agency to pronounce rules that carry the force of law and (2) the interpretation for which deference is sought was rendered pursuant to that authority. Price v. Stevedoring Servs. of Am., Inc., 697 F.3d 820, 833 (9th Cir.2012)(en banc). That was the teaching of United States v. Mead Corp., 533 U.S. 218,226–27(2001).
Congress gave the Secretary of Labor authority to promulgate binding regulations interpreting Title I of ERISA, which includes section 404(c). 29 U.S.C. § 1135. It also empowered the Secretary to bring civil enforcement actions. Id. § 1132(a)(2). These charges plainly satisfy the first requirement under Mead. See, e.g ., Gonzales v. Oregon, 546 U.S. 243, 258, 126 S.Ct. 904, 163 L.Ed.2d 748 (2006) (explaining that “[i]n many cases authority is clear because the statute gives an agency broad power to enforce” its provisions). As for Mead 's second consideration, we do not view the fact that the interpretation appears in a final rule's preamble as disqualifying it from Chevron deference. Edison cites nothing authoritative for cabining that doctrine to materials destined for the pages of the Code of Federal Regulations. Though not a necessary condition, a notice-and-comment rule is virtually assured eligibility for Chevron deference. See, e.g., Mead, 533 U.S. at 230–31; Renee v. Duncan, 686 F.3d 1002, 1011 (9th Cir.2012). Additionally, other factors significant to whether deference is owed are present here. DOL has expressed its position for two decades, ERISA is “an enormously complex and detailed statute,” Conkright v. Frommert, 559 U.S. 506, 130 S.Ct. 1640, 1644, 176 L.Ed.2d 469 (2010), and this question is of central import to its administration. See Barnhart v. Walton, 535 U.S. 212, 222, 122 S.Ct. 1265, 152 L.Ed.2d 330 (2002).7
Because the 1992 interpretation clears the Mead threshold, we proceed to the well-trod Chevron inquiry.8 This calls on the court to examine the plain meaning of the text and apply other relevant canons of statutory interpretation to ascertain whether Congress had a fixed “intention on the precise question at issue” that the agency must abide. Wilderness Soc'y v. U.S. Fish & Wildlife Serv., 353 F.3d 1051, 1060 (9th Cir.2003)(en banc).
If so, “that intention is the law and must be given effect.” Id. If not, the court defers to the agency, provided that its interpretation is not “arbitrary, capricious, or manifestly contrary to the statute.” Id. at 1059; see also Nat'l Cable & Telecomm. Ass'n v. Brand X Internet Servs., 545 U.S. 967, 980, 125 S.Ct. 2688, 162 L.Ed.2d 820 (2005)(explaining that “ambiguities in statutes within an agency's jurisdiction to administer are delegations of authority to the agency to fill the statutory gap in reasonable fashion”). These inquiries can be pursued in two steps, or all at once. Compare Wilderness Soc'y, 353 F.3d at 1059, with Entergy Corp. v. Riverkeeper, Inc., 556 U.S. 208, 218 n. 4, 129 S.Ct. 1498, 173 L.Ed.2d 369 (2009) (embracing single-step analysis because “if Congress has directly spoken to an issue then any agency interpretation contradicting what Congress has said would be unreasonable”).
In Langbecker, the Fifth Circuit concluded that the DOL's interpretation of section 404(c) could not receive Chevron deference “because it contradicts the governing statutory language.” 476 F.3d at 311. Respectfully, we disagree. Section 404(c) speaks of “any breach, which results from” a participant's exercise of control. “Result from” means “[t]o arise as a consequence, effect, or outcome of some action.” Oxford English Dictionary (3d ed.2010); see Wilderness Soc'y, 353 F.3d at 1060 (“[A] fundamental canon of construction provides that unless otherwise defined, words will be interpreted as taking their ordinary, contemporary, common meaning.” (internal quotation marks ommitted)).
Thus as cogently explained by DOL in its brief, “the selection of the particular funds to include and retain as investment options in a retirement plan is the responsibility of the plan's fiduciaries, and logically precedes (and thus cannot ‘result[ ] from’) a participant's decision to invest in any particular option.” As previously noted, the DOL expressed the same position in a notice-and-comment rule—albeit less succinctly. The preamble to the 1992 final rule states
that the act of limiting or designating investment options which are intended to constitute all or part of the investment universe of an ERISA 404(c) plan is a fiduciary function which, whether achieved through fiduciary designation or express plan language, is not a direct or necessary result of any participant direction of such plan. Thus, for example, in the case of look-through investment vehicles, the plan fiduciary has a fiduciary obligation to prudently select such vehicles, as well as a residual fiduciary obligation to periodically evaluate the performance of such vehicles to determine, based on that evaluation, whether the vehicles should continue to be available as participant investment options. Similar fiduciary obligations would exist in the case of an investment universe consisting of investment alternatives which are not look-through investment vehicles but which are specifically designated by plan fiduciaries.
57 Fed.Reg. at 46,924 n.27 (emphasis added). Although this rule invokes the regulatory terms “direct and necessary,” 29 C.F.R. § 2550.404c–1(d)(2), the agency's ability to make the same point in its amicus brief and in the new 2010 rule without that terminology suggests that this gloss may not be essential. See Langbecker, 476 F.3d at 311. In our view, though, this does not diminish the validity of its interpretation.
In an opinion that has been read by some to support the no-deference view, the Third Circuit keyed in on the fact that section 404(c) also speaks of “any loss” resulting from a participant's control. In re Unisys, 74 F.3d at 445.9 For a 401(k) (or for any defined-contribution plan for that matter), it is admittedly the case that monetary damage flowing from a fiduciary's imprudent design of the investment menu passes through the participant, as intermediary. But is it proper to conclude that those losses, in the language of section 404(c), “result from” the participant's choice? This might seem an odd question given that, literally speaking, there can be no loss without the participant selecting an investment.
But, “[i]njuries have countless causes, and not all should give rise to legal liability.” CSX Tramp., Inc. v. McBride, ––– U.S. ––––, ––––, 131 S.Ct. 2630, 2637, 180 L.Ed.2d 637 (2011). Undoubtedly, in these situations, a fiduciary's decision to include an investment option on the plan menu also is a cause of any participant's loss. Confronted with this difficulty, DOL has effectively imported the tort-law notion of proximate cause to conclude that the most salient cause (as between the two) is the fiduciary's imprudence. See id. (“What we ․ mean by the word proximate, one noted jurist has explained, is simply this: Because of convenience, of public policy, of a rough sense of justice, the law arbitrarily declines to trace a series of events beyond a certain point.”)(omission in original)(internal quotation marks and alteration omitted).
We deem this “a reasonable interpretation of the statute.” Entergy Corp., 556 U.S. at 218. ERISA “allocates liability for plan-related misdeeds in reasonable proportion to the respective actors' power to control and prevent the misdeeds.” Mertens v. Hewitt Assocs., 508 U.S. 248, 262, 113 S.Ct. 2063, 124 L.Ed.2d 161 (1993). As compared to the beneficiary, the fiduciary is better situated to prevent the losses that would stem from the inclusion of unsound investment options. It can design a prudent menu of options. Second, Chevron deference is meant to foster “coherent and uniform construction of federal law.” Orthopaedic Hosp. v. Belshe, 103 F.3d 1491, 1495 (9th Cir.1997). Our acknowledgment of the flexibility inherent in the phrase “result from” promotes this, because DOL adopts a similar interpretation with regard to breaches that-unlike claims of imprudent plan design—do chronologically follow a participant's decision. Concluding that “a fiduciary is relieved of responsibility only for the direct and necessary consequences of a participant's exercise of control,” 57 Fed.Reg. at 46,924, DOL takes the position that errors in carrying out the investment elections of a beneficiary give rise to liability notwithstanding that any associated loss technically also “results from such participant's or beneficiary's exercise of control.” 29 U.S.C. § 1104(c)(1)(A)(ii). These are just the sort of “difficult policy choices that agencies are better equipped to make than courts.” Brand X, 545 U.S. at 980.
We also reject the argument raised by Edison and the Employment Law Council that DOL's interpretation renders section 404(c) a meaningless provision. When certain conditions are complied with,10 the provision safeguards fiduciaries from being liable for participants' substantive investment decisions. 57 Fed.Reg. at 46,924. “The purpose of section 404(c) is to relieve the fiduciary of responsibility for choices made by someone beyond its control.” Howell, 633 F.3d at 567. These include matters such as, hypothetically, “the participant's decision to invest 40% of her assets in Fund A and 60% in Fund B, rather than splitting assets somehow among four different funds, [or] emphasizing A rather than B .” Id.
It is, indeed, the contrary view pressed by Edison that would render parts of the ERISA statute a nullity by making it nearly impossible for defined-contribution-plan beneficiaries to vindicate fiduciary imprudence. Cf. LaRue v. DeWolff, Boberg & Assocs., Inc., 552 U.S. 248, 256, 128 S.Ct. 1020, 169 L.Ed.2d 847 (2008) (citing the DOL's regulations implementing section 404(c) in rejecting the converse interpretation); see also Langbecker, 476 F.3d at 321 (Reavley, J., dissenting)(“All commentators recognize that § 404(c) does not shift liability for a plan fiduciary's duty to ensure that each investment option is and continues to be a prudent one.”).
Because DOL's interpretation of how the safe harbor functions is consistent with the statutory language, we conclude that the district court properly decided that section 404(c) did not preclude merits consideration of beneficiaries' claims. See Tibbie, 639 F.Supp.2d at 1121.
Edison on its cross appeal raises another argument that could waylay our analysis of beneficiaries' substantive claims on their appeal. It contends that the district court improperly certified beneficiaries' case as a class action under Federal Rule of Civil Procedure 23.
Rule 23 sets out four prerequisites in subsection (a). A class must be “so numerous that joinder of all members is impracticable,” (a)(1), there must be “questions of law or fact common to the class,” (a)(2), “the claims or defenses of the representative parties” must be “typical of the claims or defenses of the class,” (a)(3), and those representatives must “fairly and adequately protect the interests of the class,” (a)(4). Classes must also comply with “at least one of the requirements of Rule 23(b).” Zinser v. Accufix Research Inst, Inc., 253 F.3d 1180, 1186 (9th Cir.2001).
For the first time on its cross appeal and relying on out-of-circuit authority, Edison argues that this class action was improperly certified because the claims of the representative plaintiffs are not typical to the claims of the class at large. See Spano v. Boeing Co., 633 F.3d 574, 586 (7th Cir.2011)(expounding on Rule 23(a)(3)'s “typicality requirement”). In Spano, the court stated that “it seems that a class representative in a defined-contribution case would at a minimum need to have invested in the same funds as the class members.” Id. Seizing on this statement, Edison contends that one of the three funds successfully litigated at trial was not held by any of the six named plaintiffs.11 This violates Rule 23(a)(3), it claims, and requires that we reverse the class certification order.
Beneficiaries correctly argue that arguments not raised in the district court ordinarily will not be considered on appeal. Dream Palace v. Cnty. of Maricopa, 384 F.3d 990, 1005 (9th Cir.2004). “This rule serves to ensure that legal arguments are considered with the benefit of a fully developed factual record, offers appellate courts the benefit of the district court's prior analysis, and prevents parties from sandbagging their opponents with new arguments on appeal.” Id. In contrast to this typicality argument, Edison's only Rule 23(a) arguments below were (i) a lack of commonality because the then-live misrepresentation claims would require individualized proof of reliance and (ii) a failure of adequacy. Edison concedes that it framed its argument strictly “as an adequacy issue below” but claims that because this inquiry can overlap with the typicality analysis, its presentation in the lower court suffices.
While we have indulged some liberality as to whether a particular Rule 23(a) subdivision has been pressed,12 the presentation must have been “raised sufficiently for the trial court to rule on it.” In re Mercury Interactive Corp. Sec. Litig., 618 F.3d 988, 992 (9th Cir.2010). Here, the district court found that “[d]efendants [did] not challenge whether the claims of the individual plaintiffs are typical to the class.” As to adequacy, Edison's critique below centered on a “contention that the named plaintiffs [were] nothing more than ‘window dressing or puppets for class counsel” ’ in that they were not knowledgeable about their legal claims—a far cry from its appellate contention about these beneficiaries' investments.13 In light of the failure to present the issue to the district court, we expressly reserve the question of whether the Ninth Circuit should adopt a rule akin to that articulated in Spano, or whether the circumstances of that case would be distinguishable from ours.14
We now turn to the merits of the main appeal. Beneficiaries argue that the district court erred in granting summary judgment to Edison on their claim that revenue sharing between mutual funds and the administrative service provider violated the Plan's governing document, as well as was a conflict of interest.
Because ERISA requires fiduciaries to discharge their duties “in accordance with the documents and instruments governing the plan,” 29 U.S.C. § 1104(a)(1)(D), violations of the written plan have been recognized as a basis for liability. See, e.g., Cal. Ironworkers Field Pension Trust v. Loomis Sayles & Co., 259 F.3d 1036, 1042 (9th Cir.2001).15
Since 1997, Plan section 19.02 has stated: “The cost of the administration of the Plan will be paid by the Company.” Edison contracted with Hewitt Associates, LLC, for a variety of services, including the drafting of Plan updates and regulatory reports. Hewitt also maintained the system by which beneficiaries designate their contribution amounts and make their investment elections. The addition of a large menu of mutual funds in 1999 made the Plan more expensive to administer, so Edison availed itself of a practice known in the industry as revenue sharing. Under this arrangement, mutual funds transfer a portion of their fees to the Plan's service provider, Hewitt. That revenue reimburses Hewitt for its recordkeeping and other costs. In turn, Edison receives a credit on its bills from Hewitt.
Beneficiaries, while conceding this new practice of revenue sharing was disclosed during the negotiations to expand the Plan offerings, argue that the arrangement violated the language of the Plan because it allowed Edison to escape from part of the obligation to pay. With a December 26, 2006 amendment this Plan language was revised to state that “[t]he cost of administration of the Plan, net of any adjustments by service providers, will be paid by the Company.” (emphasis added). The parties agree that under the new language these offsets are perfectly appropriate. The issue that arises, however, is whether the district court correctly determined that no triable issue existed over whether the pre-amendment version of section 19.02 allowed offsets. See Fed.R.Civ.P. 56(a). At bottom, this is a simple interpretive matter, but like most issues arising under ERISA there are complications.
In addition to the pension plan at issue in this case, ERISA also governs “employee welfare benefit” plans such as those for health or disability. See 29 U.S.C. § 1002(1)-(2). “[T]he validity of a claim to benefits under an ERISA plan is likely to turn on the interpretation of terms in the plan at issue.” Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115, 109 S.Ct. 948, 103 L.Ed.2d 80 (1989). The Supreme Court has handed down a trio of opinions explaining the framework for review when those disputes reach the judiciary. See Conkright, 130 S.Ct. at 1646 (discussing the Court's two prior precedents, Firestone and Metropolitan Life Insurance Co. v. Glenn ). The proper standard of review hinges, in part, on what the plan instrument says about interpretation. When the plan is silent, judges review its terms de novo. But, when the plan grants interpretive authority to its administrator, as is usually the case, a deferential abuse of discretion standard applies to the administrator's determinations.
The Edison Plan has a provision that speaks to interpretation; it vests the company's Benefits Committee with the “full discretion to construe and interpret [its] terms and provisions.” See, e.g., Sandy v. Reliance Std. Life Ins. Co., 222 F.3d 1202, 1206–07 & n. 6 (9th Cir.2000). The Plan even purports to make interpretations by the Committee “final and binding on all parties.” Taking stock of these principles, the district court applied the abuse of discretion standard and then concluded that Edison's view that the language did not foreclose revenue sharing had been reasonable.
Yet, as we noted at the outset, the Supreme Court expounded these interpretive principles in the context of “ § 1132(a)(1)(B) actions challenging denials of benefits.” Firestone, 489 U.S. at 108. At least one court has held that in cases implicating ERISA § 404 fiduciary duties, the standard fleshed out in Firestone, Glenn, and Conkright is not applicable. See John Blair Commc'ns, Inc. Profit Sharing Plan v. Telemundo Grp., Inc. Profit Sharing Plan, 26 F.3d 360, 369–70 (2d Cir.1994). Other courts of appeals have declined to follow suit. See Hunter v. Caliber Sys., Inc., 220 F.3d 702, 711–12 (6th Cir.2000); Moench v. Robertson, 62 F.3d 553, 565 (3d Cir.1995)(expressly disagreeing with John Blair ). We agree with the Third and Sixth Circuits.
At least three considerations prompt us to hold that the usual abuse of discretion standard applies to cases such as this. First, there are disquieting parallels between the John Blair exception and that same circuit's “one-strike-and-you're-out” approach to conflicts-of-interest, which the Supreme Court repudiated in 2010. See Conkright, 130 S.Ct. at 1646–47. In so doing, the Court explained that Firestone “set out a broad standard of deference without any suggestion that the standard was susceptible to ad hoc exceptions like the one adopted by the Court of Appeals.” Id.
Second, though mindful that the Firestone case expressed “no view as to the appropriate standard of review for actions under other remedial provisions of ERISA,” we conclude that the principles underlying that 1989 decision, as well as subsequent guidance on the matter, leave little doubt that its teaching governs ERISA globally. 489 U.S. at 108. After uttering that caveat, Firestone, in nearly the next breath, announces that its holding does not stem from an interpretive gloss on the welfare-benefits provision, or from any section of ERISA for that matter. Id. at 109 (“ERISA does not set out the appropriate standard of review for actions under § 1132(a)(1)(B) challenging benefit eligibility”). Instead, because “ERISA abounds with the language and terminology of trust law” and because of legislative history to that effect, that body of law—not a discrete provision—dictated “the appropriate standard of review.” Id. at 110–11 (“Trust principles make a deferential standard of review appropriate when a trustee exercises discretionary powers”).16
This precise insight led the Third Circuit to reject John Blair. See Moench, 62 F.3d at 565 (“[W]e believe that after Firestone, trust law should guide the standard of review over claims ․ not only under section 1132(a)(1)(B) but also over claims filed pursuant to 29 U.S.C. § 1132(a)(2) based on violations of the fiduciary duties set forth in section 1104(a).”). Further evidence that the principles underlying the trilogy of benefits cases extend here is that “common law trust principles animate the fiduciary responsibility provisions of ERISA.” Acosta v. Pac. Enters., 950 F.2d 611, 618 (9th Cir.1991); see also Cent. States, Se. & Sw. Areas Pension Fund v. Central Transp., Inc., 472 U.S. 559, 570–71, 105 S.Ct. 2833, 86 L.Ed.2d 447 (1985)(identifying the statutorily prescribed duties of loyalty and of prudence as imported from trust law).
Third, we observe that applying deference across the board, “by permitting an employer to grant primary interpretive authority over an ERISA plan to the plan administrator,” has the added virtue of “preserv[ing] the ‘careful balancing’ on which ERISA is based.” Conkright, 130 S.Ct. at 1649. In particular, it helps keep administrative and litigation expenses under control, which could otherwise “discourage employers from offering [ERISA] plans in the first place.” Id. (alteration in original).
In addition to these primary considerations, there are shortcomings with the John Blair decision itself. In it, the Second Circuit appealed to the notion that fiduciary duty and conflict-of-interest suits, i.e., under ERISA § 406, arise when the plan administrator has fallen prey to invalid considerations—matters other than the well-being of beneficiaries. See John Blair, 26 F.3d at 369. Yet when the Supreme Court had the opportunity to craft a new review standard for plan administrators adjudicating claims under a financial conflict of interest, it saw “no reason to forsake Firestone 's reliance upon trust law.” Metro. Life Ins. Co. v. Glenn, 554 U.S. 105, 116, 128 S.Ct. 2343, 171 L.Ed.2d 299 (2008); see also Conkright, 130 S.Ct. at 1647 (explaining that “we held in Glenn [that] a systemic conflict of interest does not strip a plan administrator of deference”). Furthermore, John Blair drew its insight about the need to limit Firestone from a then-decade-old Third Circuit opinion that the Third Circuit came to read differently. Compare John Blair, 26 F.3d at 369 (discussing Struble v. N.J. Brewery Emps. Welfare Trust Fund, 732 F.2d 325, 333–34 (3d Cir.1984)), with Moench, 62 F.3d at 565.
Both the affirmative case for the abuse of discretion standard and difficulties with John Blair impel us to apply Firestone, and so we do.
ERISA administrators abuse their discretion if they act without explanation or “construe provisions of the plan in a way that conflicts with the plain language of the plan.” Day v. AT & T Disability Income Plan, 698 F.3d 1091, 1096 (9th Cir.2012). We are instructed not to disturb those interpretations if they are reasonable. See Conkright, 130 S.Ct. at 1651.
To start with, we discern no explicit conflict with the plain language of the Plan. See Day, 698 F.3d at 1096. Section 19.02 required the company to pay the costs, and Edison did. Although beneficiaries argue that the “costs” are the expenses associated with Hewitt before the offsets, the more natural reading is that “costs” simply are whatever bills Hewitt presented Edison with. Under this commonsense reading, the Plan merely assigned Edison an affirmative obligation to pay. It did not, as beneficiaries would have it, prohibit “Hewitt's recordkeeping services from being paid by a third party such as mutual funds.” That kind of interpretation, nonsensically, would also imply that if Hewitt had simply lowered its prices (maybe due to efficiency or market pressure) Edison would be somehow shirking its obligation under Plan § 19.02.
Beyond the text, in conducting abuse of discretion review, courts consider “various [other] criteria for determining the reasonableness of a fiduciary's discretionary decision.” Booth v. Wal–Mart Stores, Inc. Assocs. Health & Welfare Plan, 201 F.3d 335, 342 (4th Cir.2000). Viewing the matter in terms of those considerations further establishes the soundness of Edison's position. Its view is most “consistent with the goals of the plan,” as it facilitated the expansion of the Plan's mutual fund offerings. Id. We also note that section 19.02 has been applied consistently over time. Undisputed evidence showed that the union negotiators and Edison had “extensive discussions with regard to how revenue sharing from the mutual funds would be used.” Also, between 1999 when the process started, and 2006 when the language was modified, on at least seventeen occasions participants were specifically advised that mutual funds were being used to reduce the cost of retaining Hewitt. For example, one Summary Plan Description in evidence said: “the fees received by Edison's 401(k) plan recordkeeper are used to reduce the recordkeeping and communication expenses of the plan paid by the company.” Another consideration under the abuse of discretion standard is “whether the challenged interpretation is at odds with the procedural and substantive requirements of ERISA itself.” de Nobel v. Vitro Corp., 885 F.2d 1180, 1188 (4th Cir.1989)(citing Blau v. Del Monte Corp., 748 F.2d 1348, 1353 (9th Cir.1984)). Although we explain the reasoning behind this observation next, we are satisfied that revenue sharing as carried out by Edison does not violate ERISA.
Beneficiaries alternatively argue that the statute's conflicts provision, ERISA § 406(b)(3), prohibits the practice of revenue sharing. ERISA § 406 is similar to a duty-of-loyalty provision. See Mass. Mat. Life Ins. Co. v. Russell, 473 U.S. 134, 143 n. 10, 105 S.Ct. 3085, 87 L.Ed.2d 96 (1985). It prohibits the type of business deals “likely to injure the pension plan.” Wright v. Or. Metallurgical Corp., 360 F.3d 1090, 1100 (9th Cir.2004).
ERISA § 406(b)(3) provides that:
A fiduciary with respect to a plan shall not receive any consideration for his own personal account from any party dealing with such plan in connection with a transaction involving the assets of the plan.
29 U.S.C. § 1106(b)(3). Beneficiaries' claim is that Edison's revenue sharing arrangement violated this provision because Edison received “consideration” in the form of discounts for administrative expenses from Hewitt, which was a “party dealing with” the Plan. The DOL, though, has issued several non binding advisory opinions staking out the position that a fiduciary does not violate section 406(b)(3) so long as “the decision to invest in such funds is made by a fiduciary who is independent” of the fiduciary receiving the fee. DOL Advisory Op.2003–09A, 2003 WL 21514170 (June 25, 2003); see also DOL Advisory Op. 97–15A, 1997 WL 277980 (May 22, 1997) (fiduciary that “does not exercise any authority or control” to cause the suspect investment is not liable).
Relying on these concepts, the district court granted summary judgment to Edison. To do so, it conceived of “Edison,” not as a unified corporate entity, but in terms of its constituent parts. In brief, the “fiduciaries” named in the Plan include the Southern California Edison Benefits Committee and its members, as well as the Edison International Trust Investment Committee and its members. The “Plan Sponsor” is Southern California Edison, while its Benefits Committee is designated under ERISA as the “Plan Administrator.” See 29 U.S.C. § 1002(16)(A)(i), (B).17 Edison International's CEO appoints the Investment Committee and Southern California Edison's CEO handles appointments to the Benefits Committee.
In light of this diffusion of responsibility, the district court observed that, as the sole contracting party with Hewitt, only the subsidiary Southern California Edison had received the credit from administrative expenses. It then noted that it was the Investment Committee of the parent company, Edison International, which had selected the mutual funds that featured revenue sharing. From this, the court drew the conclusion that a different fiduciary had received the “consideration” than the fiduciary which had (in the DOL's parlance) exercised “authority or control” over the offending investment. Therefore, the mutual fund revenue sharing had not violated section 406(b)(3).
As amicus curiae, the DOL vigorously objects to the lower court's parsing of Edison International this way, and objects to what it considers an overly broad reading of its advisory opinions. DOL maintains that permitting “fiduciaries to make plan asset investment decisions that result in the company on which they serve as directors and officers receiving an economic benefit from a third party is precisely the kind of transaction—rife with the potential for abuse—that Congress intended to prohibit in section 406(b)(3).” In response, Edison argues that the separate legal identities of the committees and companies are meaningful, and calls to our attention the district court's finding that beneficiaries had not marshaled evidence that justified disregarding their putative separateness.
We review the district court's entry of summary judgment de novo, and we are empowered to affirm on any basis the record will support. See Gordon v. Virtumundo, Inc., 575 F.3d 1040, 1047 (9th Cir.2009). In light of that, we reserve for another case whether the lower court's control determinations are defensible and, instead, proceed to consider the basis for affirmance expressly advocated by the DOL.
The DOL directs our attention to its regulatory interpretation at 29 C.F.R. § 2550.408b–2(e)(3), which states that “[i]f a fiduciary provides services to a plan without the receipt of compensation or other consideration (other than reimbursement of direct expenses properly and actually incurred in the performance of such services ․ ), the provision of such services does not, in and of itself, constitute an act described in section 406(b) of the Act.” Assuming that the Edison Plan permitted revenue sharing (as we concluded above), then as DOL explains, the discounts on its invoices from Hewitt “would not constitute the receipt of any ‘consideration” ’ by Edison “within the meaning of the section 406(b)(3) prohibition.” In further support, the agency cites one of its opinion letters that permitted, under the authority of section 2550.408b–2(e), a fiduciary to receive reimbursement from an unrelated mutual fund of direct expenses for which the plan would otherwise be liable. See DOL Advisory Op. 97–19A, 1997 WL 540069 (Aug. 28, 1997).
The district court intimated that our Patelco Credit Union v. Sahni decision might be to the contrary. 262 F.3d 897 (9th Cir.2001). It is not, although we do not fault the district court for its misconception. It did not have the advantage, afforded us, of DOL's participation in tackling these regulatory intricacies. In Patelco, the fiduciary had wrongfully deposited ERISA Plan assets—two checks payable to the company—into his own account. Id. at 903, 908. This straightforwardly constituted “consideration for his own personal account” from a “party dealing with [the] plan,” in violation of ERISA § 406(b)(3). Id. at 909–10. Confronted with that scenario, we vindicated DOL's pronouncement that when a fiduciary self-deals in violation of ERISA § 406(b), the “reasonable compensation exception” found in section 408(b)(2) cannot be used as a shield from liability. Id. at 910–11; see also Dupree v. Prudential Ins. Co. of Am., No. 99–8337, 2007 WL 2263892, at *42 (S.D.Fla. Aug.7, 2007) (explaining this).18
By contrast in our case, section 2550.408b–2(e)(3), as it is “routinely interpreted by the DOL,” exempts revenue sharing payments from the very definition of consideration. Dupree, 2007 WL 2263892, at *42. The Department's position is that rather than constituting “consideration,” “such payments may be considered ‘reimbursement’ within the meaning of regulation section 2550.408b–2(e).” DOL Advisory Op. 97–19A.19 That means it is not a section 406(b)(3) violation at all.
Aside from citing Patelco as the lower court understood it, beneficiaries' only response is, in effect, that we ought to read DOL's regulations and opinion letters differently than DOL has counseled in its amicus brief. We decline to do so. Notably, courts are instructed to “defer to an agency's interpretation of its own regulation, advanced in a legal brief unless that interpretation is ‘plainly erroneous or inconsistent with the regulation.” ’ Chase Bank USA, N.A. v. McCoy, ––– U.S. ––––, ––––, 131 S.Ct. 871, 880, 178 L.Ed.2d 716 (2011) (discussing Auer deference). We mention this not because we resolve whether this view is permissible either under ERISA or the regulation, but simply to explain why beneficiaries have not convinced us to reject DOL's interpretation in this case.
Beneficiaries next claim that Edison violated its duty of prudence under ERISA by including several investment vehicles in the Plan menu: (i) mutual funds, (ii) a short-term investment fund akin to a money market, and (iii) a unitized fund for employees' investment in Edison stock.
ERISA demands that fiduciaries act with the type of “care, skill, prudence, and diligence under the circumstances” not of a lay person, but of one experienced and knowledgeable with these matters. 29 U.S.C. § 1104(a)(1)(B). Fiduciaries also must act exclusively in the interest of beneficiaries. Id. § 1104(a)(1). These obligations are more exacting than those associated with the business judgment rule so familiar to corporate practitioners, Howard v. Shay, 100 F.3d 1484, 1489 (9th Cir.1996), a standard under which courts eschew any evaluation of “substantive due care.” Brehm v. Eisner, 746 A.2d 244, 264 (Del.2000), cited in Pac. Nw. Generating Coop. v. Bonneville Power Admin., 596 F.3d 1065, 1077 (9th Cir.2010). To enforce this duty of prudence, we consider the merits of the transaction and “the thoroughness of the investigation into the merits of the transaction.” Howard, 100 F.3d at 1488 (emphasis added). Courts are in broad accord that engaging consultants, even well-qualified and impartial ones, will not alone satisfy the duty of prudence. See George v. Kraft Foods Global, Inc., 641 F.3d 786, 799–800 (7th Cir.2011)(collecting cases from the Second, Fifth, Seventh, and Ninth Circuits).
Under the common law of trusts, which helps inform ERISA, a fiduciary “is duty-bound ‘to make such investments and only such investments as a prudent [person] would make of his own property having in view the preservation of the [Plan] and the amount and regularity of the income to be derived.” ’ In re Unisys., 74 F.3d at 434 (quoting Restatement (Second) of Trusts § 227 (1959)) (first alternation in original).
A mutual fund is a pool of assets, chiefly a portfolio of securities bought with the capital contributions of the fund's shareholders. Jones v. Harris Assocs. L.P., 559 U.S. 335, 130 S.Ct. 1418, 1422, 176 L.Ed.2d 265 (2010). Joined by the AARP as an amicus, beneficiaries seek a ruling that including mutual funds of the sort available to the investing public at large (“retail” or “brand-name” funds) is categorically imprudent. Their position is that under ERISA, fiduciaries must offer institutional investment alternatives such as “commingled pools” or “separate accounts.”
Mutual funds, however, have a variety of unique regulatory and transparency features that make it an apples-to-oranges comparison to judge them against AARP and beneficiaries' suggested options. As Chief Judge Easterbook, writing for the Seventh Circuit, has usefully summarized:
A pension plan that directs participants into privately held trusts or commingled pools (the sort of vehicles that insurance companies use for assets under their management) lacks the mark-to-market benchmark provided by a retail mutual fund. It can be hard to tell whether a closed fund is doing well or poorly, or whether its expenses are excessive in relation to the benefits they provide. It can be hard to value the vehicle's assets (often real estate rather than stock or bonds) when someone wants to withdraw money, and any error in valuation can hurt other investors.
Loomis v. Exelon Corp., 658 F.3d 667, 671–72 (7th Cir.2011). As beneficiaries admit in their briefing, brand-name mutual funds are generally easy to track via newspaper or internet sources. This, in fact, was a stated goal of the report issued by the Joint Study Group of human resource managers and employee union representatives empaneled to expand the Plan menu. Relatedly, as other courts have recognized, non-mutual fund alternatives such as commingled pools are not subject to the same “reporting, governance, and transparency requirements” as mutual funds, which are governed by the Securities Act of 1933 and the Investment Company Act of 1940. See Renfro v. Unisys Corp., 671 F.3d 314, 318 (3d Cir.2011); Harris Assocs., 130 S.Ct. at 1422.
Further, the undisputed evidence was that during collective bargaining the union requested “forty name-brand retail mutual funds for inclusion in the Plan.” While conceding this, the beneficiaries claim that the union did not know what was in its members' best interest. Because participant choice is the centerpiece of what ERISA envisions for defined-contribution plans, these sorts of paternalistic arguments have had little traction in the courts. See, e.g., Loomis, 658 F.3d at 673; Renfro, 671 F.3d at 327–28 (observing that imprudence is less plausible “in light of an ERISA defined-contribution 401(k) plan having a reasonable range of investment options with a variety of risk profiles and fee rates”).
Also before us under the mutual fund umbrella is beneficiaries' claim that the particular mutual funds Edison selected charged excessive fees, which rendered their inclusion imprudent. Part of this challenge is a broadside against retail-class mutual funds, which do generally have higher expense ratios than their institutional-class counterparts. As the district court explained in its post-trial findings of fact, this is because with institutional-class mutual funds “the amount of assets invested is far greater than [that associated with] the typical individual investor.” The Seventh Circuit has repeatedly rejected the argument that a fiduciary “should have offered only ‘wholesale’ or ‘institutional’ funds.” See Loomis, 658 F.3d at 671; Hecker, 556 F.3d at 586 (“[N]othing in ERISA requires [a] fiduciary to scour the market to find and offer the cheapest possible fund (which might, of course, be plagued by other problems) .”). We agree. There are simply too many relevant considerations for a fiduciary, for that type of bright-line approach to prudence to be tenable. Cf. Braden v. Wal–Mart Stores, Inc., 588 F.3d 585, 596 (8th Cir.2009)(acknowledging that a fiduciary might “have chosen funds with higher fees for any number of reasons, including potential for higher return, lower financial risk, more services offered, or greater management flexibility”).
Nor is the particular expense ratio range out of the ordinary enough to make the funds imprudent. In Hecker, the court upheld the dismissal of a similar excessive fee claim where the range of expenses varied from .07 to 1% across a pool of twenty mutual funds. 556 F.3d at 586. Here, the summary-judgment facts showed that the expense ratio varied from .03 to 2%, and there were roughly forty mutual funds to choose from.
Before we leave the topic of mutual funds we find it necessary to make one last observation. Much time at oral argument and ink in the briefs were devoted to debating the question of whether the revenue sharing typically associated with mutual funds adversely impacts plan beneficiaries. Today we have held that the practice here did not violate the terms of the Edison Plan or violate ERISA § 406(b)(3).
Mutual funds generate this revenue by charging what is known as a Rule 12b–1 fee to all investors participating in the fund.20 Edison takes the position that because that fee applies to Plan beneficiaries and all other fund investors alike, the allocation of a portion of that total 12b–1 fee to Hewitt is irrelevant. As it put the matter at oral argument: “the mutual fund advisor can do whatever it wants with the fees; sometimes they share costs with service providers who assist them in providing service and sometimes they don't.” This benign-effect, of course, assumes that the “cost” of revenue sharing is not driving up the fund's total 12b–1 fee and, in turn, its overall expense ratio. It also assumes that fiduciaries are not being driven to select funds because they offer them the financial benefit of revenue sharing. The former was not explored in this case and the evidence did not bear out the latter,21 but we do not wish to be understood as ruling out the possibility that liability might—on a different record—attach on either of these bases.
The next contention can be addressed briefly. Beneficiaries argue that it was imprudent for Edison to include a short-term investment fund (or “STIF”) rather than a stable value fund. Both types of investments are conservative in that they emphasize capital preservation rather than the maximization of returns. A stable value fund generally consists of short-to-medium duration bonds paired with insurance contracts that guard against interest rate volatility, and the record here indicates that beneficiaries are correct that they typically outperform money market funds. A STIF is similar to a traditional money market fund, which invests in what might be loosely termed “money,” instruments such as “short-term securities of the United States Government or its agencies, bank certificates of deposit, and commercial paper.” Harris Assocs., 130 S.Ct. at 1426 n. 6. The regulatory regime is different for the two instruments however: registered money markets must comply with the Investment Company Act, whereas banking regulations set the rules of the road for STIFs.
When applying the prudence rule in section 1104(a)(1)(B), “the primary question is whether the fiduciaries, at the time they engaged in the challenged transactions, employed the appropriate methods to investigate the merits of the investment and to structure the investment.” Cal. Ironworkers, 259 F.3d at 1043 (internal quotation marks omitted). Thus, fatal to beneficiaries is uncontroverted evidence that there were discussions about the pros and cons of a stable-value alternative. Furthermore, an investment staffer testified at his deposition that in 1999 his team determined that a short-duration bond fund already on the menu filled the same investment niche as would have a stable value fund.
Beneficiaries also charge that the inclusion of the unitized stock investment was imprudent, despite it being an industry standard for large 401(k)'s. Their main contention is that during the class period a roughly 77% gain in Edison's stock price yielded Plan investors only around a 67% return. But hindsight is the wrong metric for evaluating fiduciary duty. See Roth v. Sawyer–Cleator Lumber Co., 16 F.3d 915, 918 (8th Cir.1994); DiFelice, 497 F.3d at 424.
This dilution, or “investment drag,” that occurs when stock prices rise as compared to a direct stock investment is a well-recognized characteristic of unitized funds. The reason they are called “unitized” is that participants own units of a fund that invests primarily in company stock, but also in “cash and other similar highly liquid investments.” George, 641 F.3d at 792. These non-stock portions of the unitized fund generate lower rates of return than does the stock. Why use the device then? The advantages are twofold. The cash-buffer gives investors increased liquidity. See id. at 793 (explaining that money can be dispersed without delay because sales of units are paid out from the cash). Also, “in a market in which the relevant stock is declining, the presence of cash in the fund would be a good thing” because it functions as a hedge. Id.
Citing George, beneficiaries correctly note that, there, the court withheld summary judgment because there was a genuine issue of material fact as to whether the fiduciary had considered “implementing changes to the [fund] in order to reduce or eliminate investment and transactional drag.” Id. at 796 n. 8. Yet, by contrast, the district court here found vigilance on the part of the Edison Investment Committee to minimize this phenomenon. “For example, in July 2004, the issue of how much cash should be held in the Edison Stock Fund was raised.” Because active trading had decreased, the decision was made to reduce the cash target. See Taylor v. United Techs. Corp., No. 3:06–CV–1494, 2009 WL 535779, at *9 (D.Conn.2009)(“The evidence indicates that UTC's evaluation of the merits of retaining cash to provide transactional liquidity satisfies the prudent person standard.”). Because the choice to include unitization was objectively reasonable as well as informed, and because the evidence establishes that Edison oversaw the fund as conditions changed, we agree that summary judgment was proper.
Continuing with our application of the prudence standard, we confront the final issue in the case: Edison's argument on cross appeal that the district court erred in concluding—after a three-day bench trial and months of post-trial evidence and briefing—that the company had been imprudent in deciding to include retail-class shares of three specific mutual funds in the Plan menu.22 The basis of liability was not the mere inclusion of retail-class shares, as the court had rejected that claim on summary judgment. Instead, beneficiaries prevailed on a theory that Edison has failed to investigate the possibility of institutional-share class alternatives.
In reviewing a judgment after a bench trial, we evaluate the district court's factual findings “for clear error and its legal conclusions de novo.” Lee v. W. Coast Life Ins. Co., 688 F.3d 1004, 1009 (9th Cir.2012).
Here, the lower court's unchallenged findings are that during the relevant time period (i) all three funds offered institutional options in which the Edison 401(k) Savings Plan almost certainly could have participated,23 (ii) those options were in the range of 24 to 40 basis points cheaper than the retail class options the Plan did include, and—crucially—(iii) between the class profiles, there were no salient differences in the investment quality or management.
Since at least 1999, Edison has contracted with Hewitt Financial Services (“HFS”)24 for investment consulting advice. It argued below, and re-urges here, that it reasonably depended on HFS for advice about which mutual fund share classes should be selected for the Plan.
HFS frequently engages with the Investment Committee staff at Edison to help design and manage the Plan menu. It applies the investment staff's criteria: (1) fund stability/management, (2) diversification, (3) performance relative to benchmarks, (4) expense ratio relative to the peer group, and (5) the accessibility of public information on the fund. HFS then approaches the Committee with options and discusses their respective merit with its members. And to keep Edison abreast of developments, it provides the Committee with monthly, quarterly, and annual investment reports. We offer this background to illustrate a point, which, though it should be unmistakable, seems to have eluded Edison in its briefing. HFS is its consultant, not the fiduciary. “As Judge Friendly has explained, independent expert advice is not a ‘whitewash.” ’ Shay, 100 F.3d at 1489 (quoting Donovan v. Bierwirth, 680 F.2d 263, 272 (2d Cir.1982)). Our Shay factors recognize this by not simply requiring that the fiduciary (1) probe the expert's qualifications, and (2) furnish the expert with reliable and complete information, but also requiring it to “(3) make certain that reliance on the expert's advice is reasonably justified under the circumstances.” Id.25
Applying Shay, the district court found that Edison failed to satisfy element (3)—reasonable reliance. We agree. Just as fiduciaries cannot blindly rely on counsel, Donovan v. Mazzola, 716 F.2d 1226, 1234 (9th Cir.1983), or on credit rating agencies, Bussian, 223 F.3d at 301, a firm in Edison's position cannot reflexively and uncritically adopt investment recommendations. See In re Unisys, 74 F.3d at 435–36 (“[W]e believe that ERISA's duty to investigate requires fiduciaries to review the data a consultant gathers, to assess its significance and to supplement it where necessary.”); Shay, 100 F.3d at 1490 (fiduciaries should “make an honest, objective effort” to grapple with the advice given and, if need be, “question the methods and assumptions that do not make sense”). The trial evidence—from both beneficiaries' and Edison's own experts—shows that an experienced investor would have reviewed all available share classes and the relative costs of each when selecting a mutual fund. The district court found an utter absence of evidence that Edison considered the possibility of institutional classes for the funds litigated—a startling fact considering that supposedly the “expense ratio” was a core investment criterion.
However, because the “goal is not to duplicate the expert's analysis,” had Edison made a showing that HFS engaged in a prudent process in considering share classes this might have been a different case. Bussian, 223 F.3d at 301. But despite having ample opportunities, Edison “did not present evidence of: the specific recommendations HFS made to the Investments Staff regarding those funds, what the scope of HFS's review was, whether HFS considered both the retail and institutional share classes” or what questions or “steps the Investments Staff [pursued] to evaluate HFS' recommendations.”
On this record we have little difficulty agreeing with the district court that Edison did not exercise the “care, skill, prudence, and diligence under the circumstances” that ERISA demands in the selection of these retail mutual funds. 29 U.S.C. § 1104(a)(1)(B). Its cross appeal thus fails.
For the foregoing reasons, the judgment of the district court is AFFIRMED. The parties shall bear their own costs on appeal.
O'SCANNLAIN, Circuit Judge: