GRIFFITH, Circuit Judge:

In 2005, the Washington, D.C. law firm of Feder Semo closed its doors and terminated its retirement plan. Appellant Denise Clark was an attorney at the law firm for almost a decade and participated in the plan. Unfortunately, when the plan was terminated, there were not enough assets to satisfy all of its obligations. Dissatisfied with the amount of money that came her way, Clark sued, alleging that decisions made by Joseph Semo and Howard Bard (the law firm's directors who administered the retirement plan) breached their fiduciary duties under the Employee Retirement Income Security Act of 1974 (ERISA). The district court rejected all of Clark's claims, and we affirm its judgment and reasoning. We think, however, that two issues merit further discussion.

I

There was enough money in the retirement plan at termination for Semo and Bard to distribute $229,949 to firm founder Gerald Feder. Clark argues this violated § 401(a)(4) of the Internal Revenue Code, which prohibits payments that favor highly compensated employees. The district court properly concluded that there is no cause of action under ERISA for a breach of § 401(a)(4), relying upon decisions of other circuits.1 But neither the district court nor any of those decisions addressed the particular statutory argument advanced by Clark. We write to explain its flaws.

Section 401(a)(4) provides that retirement plans may lose their tax-favored status if "the contributions or benefits provided under the plan ... discriminate in favor of highly compensated employees." 26 U.S.C. § 401(a)(4). It may well be that the distribution to Feder was discriminatory, but Clark doesn't seek to disqualify the plan; she seeks relief under ERISA. And here we must be cautious because the Supreme Court has repeatedly warned courts against permitting suits to proceed under ERISA based on novel causes of action not expressly authorized by the text of the statute. See Great-West Life & Annuity Ins. Co. v. Knudson, 534 U.S. 204, 209, 122 S.Ct. 708, 151 L.Ed.2d 635 (2002) ("ERISA is a comprehensive ... [statute that is] the product of a decade of congressional study of the Nation's private employee benefit system," and courts should avoid "extending remedies not specifically authorized by its text." (internal quotation marks omitted)); see also Harris Trust & Sav. Bank v. Salomon Smith Barney Inc., 530 U.S. 238, 246-47, 120 S.Ct. 2180, 147 L.Ed.2d 187 (2000).

Clark suggests that express authorization for her claim is found in 29 U.S.C. § 1344, a provision of ERISA that sets forth general rules governing the allocation of the assets of a retirement plan upon termination. She points to a portion of § 1344 that authorizes the Secretary of the Treasury to step in and override an application of those general rules that would violate § 401(a)(4).2 According to Clark, this authority for the Secretary to intervene into the workings of a plan also imposes upon a fiduciary the duty to avoid the discriminatory distributions barred by § 401(a)(4). But Clark never tells us how authority for the Secretary to intervene becomes the source of a duty for a plan fiduciary. She does not because she cannot. Section 1344 authorizes the Secretary of the Treasury to take action to prevent a plan from losing tax benefits, but says nothing at all about what a fiduciary may or may not do about distributions at termination. As Clark vaguely suggests, general principles of fiduciary law imported into ERISA may set bounds on the distributions Semo and Bard authorized, but Clark's argument is based upon § 401(a)(4), which is not the source of any such limits. Section 1344's reference to § 401(a)(4) stands in contrast to other ERISA provisions that use unequivocal language to describe the duties of plan fiduciaries. See, e.g., 29 U.S.C. § 1106(a)(1) ("A fiduciary with respect to a plan shall not cause the plan to engage in a transaction ... [that] constitutes a direct or indirect ... sale or exchange, or leasing, of any property between the plan and a party in interest...."); id. § 1106(b) ("A fiduciary with respect to a plan shall not... deal with the assets of the plan in his own interest or for his own account...."); id. § 1104(a)(1)(B) ("[A] fiduciary shall discharge his duties with respect to a plan... by diversifying the investments of the plan so as to minimize the risk of large losses....").

Furthermore, the terms of § 1344 operate only "[i]f the Secretary of the Treasury determines that" applying its allocation rules unfairly favors the highly compensated. 29 U.S.C. § 1344(b)(5). Clark suggests the Secretary made that determination when he mandated in a treasury regulation that retirement plans must comply with § 401(a)(4). See Treas. Reg. § 1.401(a)(4)-5(b)(2) (retirement plans must include a provision limiting distributions upon termination to "a benefit that is nondiscriminatory under section 401(a)(4)"). But surely this is not the type of particularized determination contemplated by § 1344. That determination comes only in the wake of a finding by the Secretary that the application of the allocation rules to the distribution of the assets of a specific retirement plan will violate the rule against discrimination. Nothing like that has happened here.

II

In calculating Clark's distribution, Semo and Bard placed her in a group of employees whose share was based on the firm's annual contribution to the retirement plan of 10% of their salary. Clark objected and asked that she be reassigned to the group whose share was based on the firm's annual contribution of 20% of their salary. Relying upon the advice of the plan's lawyer, William Anspach, Semo and Bard denied her request. Clark argued before the district court that Bard and Semo were not entitled to rely on that advice because it was based on a mistake of fact that they would have discovered had they undertaken an independent investigation. The district court properly concluded that relying on the advice of counsel was justified under the circumstances, but cited no authority in support. We write to clarify when ERISA permits plan fiduciaries to act in reliance on the advice of counsel.

Prior to ERISA's passage, retirement plans were governed in large part by the common law of trusts. See Varity Corp. v. Howe, 516 U.S. 489, 496, 116 S.Ct. 1065, 134 L.Ed.2d 130 (1996). A fundamental principle of that law holds trustees to the standard of conduct of an objectively prudent person. See id.; Fink v. Nat'l Sav. & Trust Co., 772 F.2d 951, 955 (D.C.Cir.1985); RESTATEMENT (THIRD) OF TRUSTS § 77 & cmt. a (2005). Over time, a body of case law developed that fleshed out the meaning of that standard. In ERISA, Congress provided that a plan fiduciary must act "with the care, skill, prudence, and diligence under the circumstances then prevailing that a prudent man acting in a like capacity and familiar with such matters would use." 29 U.S.C. § 1104(a)(1)(B). Doing so, ERISA adopted much of what the common law had, over time, come to require of fiduciaries. As the Supreme Court described it, "rather than explicitly enumerating all of the powers and duties of trustees and other fiduciaries, Congress invoked the common law of trusts to define the general scope of their authority and responsibility." Cent. States, Se. & Sw. Areas Pension Fund v. Cent. Transp., Inc., 472 U.S. 559, 570, 105 S.Ct. 2833, 86 L.Ed.2d 447 (1985); see also Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 110, 109 S.Ct. 948, 103 L.Ed.2d 80 (1989) ("ERISA abounds with the language and terminology of trust law. ERISA's legislative history confirms that the Act's fiduciary responsibility provisions codif[y] and mak[e] applicable to [ERISA] fiduciaries certain principles developed in the evolution of the law of trusts." (alterations in original) (citations omitted) (internal quotation marks omitted)).

Even so, the Supreme Court has cautioned that although trust law principles developed at common law are a good "starting point" for determining a fiduciary's duties under ERISA, Congress may not have adopted them all. See Varity Corp., 516 U.S. at 497, 116 S.Ct. 1065; see also Harris Trust & Sav. Bank, 530 U.S. at 250, 120 S.Ct. 2180. Courts must therefore be on the lookout for instances in which ERISA departs from the common law, sometimes requiring more, other times requiring less, of fiduciaries. See Varity Corp., 516 U.S. at 497, 116 S.Ct. 1065.

In determining the "starting point," the Supreme Court has relied on sources such as the Restatement of Trusts, see, e.g., Cent. States, 472 U.S. at 570 n. 11, 105 S.Ct. 2833; see also Eddy v. Colonial Life Ins. Co. of Am., 919 F.2d 747, 750 (D.C.Cir.1990), and well-known treatises on the law of trusts, including that of Professor Bogert, see, e.g., Varity Corp., 516 U.S. at 498, 116 S.Ct. 1065. Following the Supreme Court's example, our review of those sources shows that it is a principle firmly rooted and founded in the common law of trusts that a fiduciary may rely on the advice of counsel when reasonably justified under the circumstances.3 The propriety of that reliance must be judged based on the circumstances at the time of the challenged decision.4 The fundamental question is always whether a prudent trustee in those particular circumstances would have acted in reliance on counsel's advice. Of course, reliance would be improper if there were significant reasons to doubt the course counsel suggested.5

Because nothing in ERISA suggests that Congress displaced this common law principle, we conclude that ERISA's adoption of the common law's standard of fiduciary care in § 1104(a)(1)(B) permits prudent fiduciaries making important decisions to rely on the advice of counsel in appropriate circumstances. We join the other circuits that have indicated that ERISA permits such reliance.6

Following a six-day bench trial, the district court concluded that Semo and Bard had rightfully relied upon the view of Anspach that Clark had been properly placed in the 10% group. Our review of such a fact-intensive, case-specific determination is necessarily deferential. See Salve Regina Coll. v. Russell, 499 U.S. 225, 233, 111 S.Ct. 1217, 113 L.Ed.2d 190 (1991) (explaining that "probing appellate scrutiny" of a case-specific determination is unlikely to add "to the clarity of legal doctrine"). Ample evidence supported the district court's conclusion.

Prior to advising Semo and Bard about Clark's request, Anspach consulted what he believed to be the relevant documents. Based on his review, he concluded that Clark and Bard should be assigned to the same group. Both had started work at the firm around the same time, and both made partner in the same year. And Bard, Anspach concluded, had always been in the 10% group, proof sufficient that Clark belonged there too. In recommending to Semo and Bard that Clark be placed in that group, Anspach forwarded to them a memo written three months after Clark made partner that showed that she and Bard were in the 10% group. Bard had always thought that he and Clark had been in the 10% group during all the years they had worked together at the firm. In Bard's mind, the memo confirmed this view. The memo also reinforced the shared belief of Semo and Bard that the 20% group was reserved for Semo, who was more senior than Clark and Bard.

As it turns out, Anspach was mostly right but partly wrong. He was right that Clark and Bard had both been in the 10% group for most of their time at the firm. But he was wrong in reporting that Clark and Bard had been in the 10% group for all of their years at the firm. For some reason not offered by any of the parties, Bard was placed in the 20% group for a single year in 2001, though neither Bard nor Semo had requested, approved, or even known of the assignment.

Clark argues that the district court erred in concluding that Semo and Bard were entitled to rely on Anspach's recommendation. Although she never makes clear why, she seems to assume that Semo and Bard had an absolute duty to look behind Anspach's advice and conduct their own investigation to see if it was grounded in fact. But, as we have already established, Clark is wrong to the extent she suggests fiduciaries have such an unyielding obligation. Clark's argument turns on the fact that Anspach's advice was based, in part, on a mistake about who was grouped where in 2001. Even so, Semo and Bard were justified in relying on Anspach's advice. At the time it was given, they had no reason to know or even suspect Anspach's mistake. He had been the plan's counsel since the early 1990s. There was no reason to think he was unfamiliar with its details. His recommendation appeared to be based on a reasonable investigation, was accompanied by supporting documentation, and was consistent with the understanding that Semo and Bard had about the way the plan's groups were structured. Nothing about Anspach's advice would have suggested to Semo and Bard the need to investigate further.

III

For the reasons stated above, and for the reasons stated in the district court's opinions, we affirm.