Erica Barrett v. O’Reilly Automotive, Inc.

United States Court of Appeals
For the Eighth Circuit
___________________________
No. 23-2501
___________________________
Erica Barrett; Kathleen Vincent; Connie Enderle; Edward Q. Ingerson, II; Penny
Kenoyer; Gilbert Ontiveros
Plaintiffs - Appellants
v.
O’Reilly Automotive, Inc.; The Board of Directors of O’Reilly Automotive, Inc.;
O’Reilly Automotive 401(k) Plan Investment Committee; John Does 1-30
Defendants - Appellees
------------------------------
Chamber of Commerce of the United States of America
Amicus on Behalf of Appellee(s)
____________
Appeal from United States District Court
for the Western District of Missouri - Springfield
____________
Submitted: April 9, 2024
Filed: August 29, 2024
____________
Before BENTON, ARNOLD, and STRAS, Circuit Judges.
____________

-- 1 of 8 --

-2-
STRAS, Circuit Judge.
Cases like this one are nothing new. The basic allegation is that a retirement
plan’s fees are so high that its managers must be “asleep at the wheel.” Davis v.
Washington Univ. in St. Louis, 960 F.3d 478, 483 (8th Cir. 2020). But like most of
the cases that have come before, the complaint does not provide meaningful
benchmarks to evaluate the claim, so we affirm the district court’s1 decision to
dismiss.
I.
Like many companies, O’Reilly Automotive, Inc., offers a defined-
contribution plan that allows employees to sock away earnings with an eye toward
retirement. See Matousek v. MidAmerican Energy Co., 51 F.4th 274, 277–78 (8th
Cir. 2022) (explaining the mechanics of this kind of plan). The returns depend on
the choices made by two groups of people: the plan managers, who decide on the
mix of investments offered and the costs of participating; and the participants
themselves, who select how much to invest and where. See id.
Erica Barrett and other members of the second group accuse the first group of
breaching their fiduciary duty. See 29 U.S.C. § 1104(a)(1)(B). One part of their
claim is that the managers, either through incompetence or laziness, saddled them
with exorbitant recordkeeping expenses. See Davis, 960 F.3d at 482 (defining these
expenses as paying for “the day-to-day operations of the plan itself”). The other is
that the total cost was too high, in part because the investment funds in the plan had
inflated expense ratios. Taken together, these problems allegedly led to less money
in the participants’ pockets and more for the recordkeeper, T. Rowe Price, and the
individual fund managers.
1 The Honorable Brian C. Wimes, United States District Judge for the Western
District of Missouri.

-- 2 of 8 --

-3-
The district court granted the company’s motion to dismiss. It spotted the
same problem we have identified before: a lack of meaningful benchmarks
suggesting that the costs are too high for a plan of its size. See Matousek, 51 F.4th
at 279.
II.
We review the dismissal de novo, “accepting as true the allegations . . . in the
complaint and drawing all reasonable inferences in favor of the nonmoving party.”
Id. at 278 (citation omitted). “A complaint can only survive a motion to dismiss if
it contains sufficient factual matter to state a facially plausible claim for relief.” Id.
(citation omitted).
A.
The overarching theory is that, by allowing costs to soar, “the plan’s
fiduciaries [necessarily] have violated their duty of prudence, which is about how
they must act.” Id.; see Braden v. Wal-Mart Stores, Inc., 588 F.3d 585, 595 (8th Cir.
2009). Nothing in the complaint identifies any direct mismanagement of O’Reilly’s
retirement plan, like failing to hold meetings or rubber-stamping the work of the
recordkeeper. See, e.g., Tussey v. ABB, Inc., 746 F.3d 327, 336 (8th Cir. 2014)
(explaining that “significant allegations of wrongdoing . . . with respect to fees,”
including allegations that the fiduciaries used revenue-sharing to benefit themselves
at the plan’s expense, “state a claim for fiduciary breach”). Instead, the plaintiffs’
theory requires an inference of mismanagement from the high costs alone. See
Meiners v. Wells Fargo & Co., 898 F.3d 820, 822 (8th Cir. 2018) (noting that
“circumstantial allegations about [the fiduciary’s] methods” based on the
“investment choices a plan fiduciary made” can be enough to survive a motion to
dismiss); Davis, 960 F.3d at 482–83; see also Hughes v. Nw. Univ., 595 U.S. 170,
177 (2022) (explaining that reviewing courts must evaluate “the allegations as a
whole”).

-- 3 of 8 --

-4-
In a case like this one, “the key” to showing that imprudent management led
a plan to pay too much is “a meaningful benchmark” that provides “a sound basis
for comparison.” Matousek, 51 F.4th at 279–80 (citation omitted). “[T]he way to
plausibly plead a claim of this type,” in other words, “is to identify similar plans
offering the same services for less.” Id. at 279; see Albert v. Oshkosh Corp., 47 F.4th
570, 579–80 (7th Cir. 2022); Sweda v. Univ. of Pa., 923 F.3d 320, 330 (3d Cir. 2019).
Consider a simple example. Suppose that a complaint alleges that the annual
recordkeeping costs for a 100,000-participant plan are $7,000,000 and that those fees
are “excessive.” Without context, those numbers are meaningless. They may sound
like too much, but they would not be if other similarly sized plans charge $120 or
more per participant. Under those circumstances, the hypothetical plan would be
comparatively cheap at $70 per participant, and there would be no “plausible”
inference of mismanagement. Ashcroft v. Iqbal, 556 U.S. 662, 682 (2009).
The opposite would be true if similarly sized plans charge $40 per participant.
See Bell Atl. Corp. v. Twombly, 550 U.S. 544, 557 n.5 (2007) (explaining that
allegations must cross the line “between the factually neutral and the factually
suggestive”). In those circumstances, charging $70 per participant does not mean
that mismanagement exists, only that there is a “plausible” inference that it might,
because other comparable plans can provide the same services for less. Iqbal, 556
U.S. at 682. Without meaningful benchmarks, a complaint fails to meet basic
pleading requirements, at least in the absence of other non-conclusory allegations of
mismanagement.
The complaint in this case has benchmarks, but none are particularly
meaningful. See Matousek, 51 F.4th at 280; Davis, 960 F.3d at 486. The plaintiffs
rely on the cost figures from the plan’s “‘Annual Return/Report of Employee Benefit
Plan’—otherwise known as a Form 5500—which discloses the aggregate payments
made to the plan’s recordkeeper.” Matousek, 51 F.4th at 279; see Davis, 960 F.3d
at 484 n.3 (explaining that these documents are “embraced by the pleadings” and
can be considered on a motion to dismiss (citation omitted)). The complaint then

-- 4 of 8 --

-5-
divides the total costs by the number of participants, which provides the per-
participant fees. The math shows O’Reilly’s plan paid more than some others, in the
range of $47 to $88 annually, depending on the year. So far, so good.
The problem lies with what the complaint does next. It tries to draw an
inference of mismanagement from the higher fees but ignores the fact that the
“service codes” in the Form 5500s show that O’Reilly’s plan paid T. Rowe Price for
more than just recordkeeping. See Matousek, 51 F.4th at 279 (noting that service
codes explain what services the recordkeeper provided to the plan). That is, the per-
participant figure of $47 to $88 bundled in the cost of “other, non-recordkeeping
services” like investment-management and trustee fees. Id.
The comparator plans, on the other hand, either had no extra services listed or
included a different bundle. Without knowing what O’Reilly paid for recordkeeping
services alone, the basic plans offer no meaningful comparison. It would be like
trying to compare the costs of two otherwise identical grocery baskets, except one
contains filet mignon and the other does not. We would expect the one with the
steak to cost more, and the same goes for a plan that offers additional individualized
services.
Some of the other plans paid for extra services, just like O’Reilly did, but the
bundles were different. See Sweda, 923 F.3d at 330 (explaining that plaintiffs
relying on total compensation figures must point to plans that purchased the “same
services” (emphasis added)). The FedEx plan they rely on, for example, paid for
general consulting and investment advice for participants, but not for securities
brokerage or shareholder servicing, among other things. What the plaintiffs are
asking us to do is draw an inference of mismanagement from the differing costs of

-- 5 of 8 --

-6-
two grocery baskets with different items. We would not expect them to cost the
same, so their approach just doesn’t work.2
Had there been allegations that “the services purchased were sufficiently
similar to render the comparisons valid,” Mator v. Wesco Distrib., Inc., 102 F.4th
172, 188 (3d Cir. 2024), it might have “nudged their claim[] across the line from
conceivable to plausible,” Twombly, 550 U.S. at 570. But the complaint does not
even acknowledge that the other plans purchased different services, much less try to
explain why they were similar. Instead, they just assert, without further explanation,
that “the service codes from the . . . Form 5500 show[] that the plan received the
same type of services as the comparator plans.” And even then, this allegation is
conclusory and comes only in their briefing, long after they needed to raise it in their
complaint. See Dorothy J. v. Little Rock Sch. Dist., 7 F.3d 729, 734 (8th Cir. 1993)
(“[A]n attempt to amend one’s pleading in an appellate brief comes too late.”
(quoting Hanson v. Town of Flower Mound, 679 F.2d 497, 504 (5th Cir. 1982))).
As we explained in Davis, “[c]omparing apples and oranges”—or, as here,
different grocery baskets—“is not a way to show that one is better or worse than the
other.” 960 F.3d at 485. If the plaintiffs wanted to plead a plausible case of
mismanagement, the complaint needed meaningful benchmarks—ones containing a
similar bundle of services. There are none here.
B.
The plaintiffs’ other allegations rely on aggregate data from the Investment
Company Institute, which do not identify the cost structures of individual plans.
2 Nor does the plaintiffs’ reliance on a discovery stipulation from another case
work. Although the recordkeeper in that case stipulated that it charged between $14
and $21 per person, it does not tell us how much O’Reilly’s basic recordkeeping
expenses were. There is no way, in other words, “to make a like-for-like
comparison.” Matousek, 51 F.4th at 279.

-- 6 of 8 --

-7-
Rather, they show that the overall costs of the O’Reilly plan were higher than
average. And so were the expense ratios of the individual investment options within
it. From there, the plaintiffs say there is a plausible inference of mismanagement.
We considered and rejected this argument in Matousek. What is missing is
any explanation of what was aggregated or “the criteria used to sort” the plans and
the investments within them. Matousek, 51 F.4th at 281. “With so little information,
we have no way of knowing whether [the aggregated data] provide a sound basis for
comparison.” Id. (citation omitted); see Matney v. Barrick Gold of N. Am., 80 F.4th
1136, 1155 (10th Cir. 2023) (“A comparison to median expense ratios in broad
investment strategy categories, without more, does not provide the ‘meaningful
benchmark’ necessary to satisfy a plaintiff’s pleading burden . . . .”). “The bottom
line is that [it] fails ‘to connect the dots in a way that creates an inference of
imprudence.’” Matousek, 51 F.4th at 282 (quoting Davis, 960 F.3d at 486).
III.
Two loose ends remain. The first is the failure-to-monitor claim against
O’Reilly and its board of directors. Without “a plausible inference that the decision-
making process itself was flawed,” id. at 280, this derivative claim ends in the same
place: dismissal. See Allen v. Wells Fargo & Co., 967 F.3d 767, 777 (8th Cir. 2020);
Brown v. Medtronic, Inc., 628 F.3d 451, 461 (8th Cir. 2010).
The second is that the district court dismissed the complaint with prejudice
without giving the plaintiffs an opportunity to fix it. See Far E. Aluminium Works
Co. v. Viracon, Inc., 27 F.4th 1361, 1367 (8th Cir. 2022) (reviewing for an abuse of
discretion). Although litigants are “freely give[n] leave” to amend, see Fed. R. Civ.
P. 15(a)(2), they still must “follow [the] proper procedures,” Thomas v. United
Steelworkers Loc. 1938, 743 F.3d 1134, 1140 (8th Cir. 2014). And here, the
plaintiffs never formally requested leave to amend, much less “submitted an
amended complaint.” Matousek, 51 F.4th at 282 (citation omitted). The district

-- 7 of 8 --

-8-
court did not abuse its discretion in failing to give them a second chance they never
properly requested. See id.
IV.
We accordingly affirm the judgment of the district court.
_____________________________

-- 8 of 8 --