Frederick Rozo v. Principal Life Insurance Company

United States Court of Appeals
For the Eighth Circuit
___________________________
No. 21-2026
___________________________
Frederick Rozo
lllllllllllllllllllllPlaintiff - Appellant
v.
Principal Life Insurance Company
lllllllllllllllllllllDefendant - Appellee
------------------------------
Chamber of Commerce of the United States of America; American Benefits
Council; American Council of Life Insurers
lllllllllllllllllllllAmici on Behalf of Appellee
____________
Appeal from United States District Court
for the Southern District of Iowa - Central
____________
Submitted: May 12, 2022
Filed: September 2, 2022
____________
Before SMITH, Chief Judge, COLLOTON and SHEPHERD, Circuit Judges.
____________
SMITH, Chief Judge.

-- 1 of 14 --

Principal Life Insurance Company (Principal) offers a product called the
Principal Fixed Income Option (PFIO), a stable value contract, to employer-
sponsored 401(k) plans. Frederick Rozo, on behalf of himself and a class of plan
participants who deposited money into the PFIO, sued Principal under the Employee
Retirement Income Security Act of 1974 (ERISA), claiming that it (1) breached its
fiduciary duty of loyalty by setting a low interest rate for participants and (2) engaged
in a prohibited transaction by using the PFIO contract to make money for itself. The
district court1 granted summary judgment to Principal after concluding that it was not
a fiduciary. This court reversed, holding that Principal was a fiduciary. See Rozo v.
Principal Life Ins. Co., 949 F.3d 1071 (8th Cir. 2020). On remand, the district court
entered judgment in favor of Principal on both claims after a bench trial. Rozo
challenges the court’s judgment. We affirm.
I. Background
In a typical employer-sponsored 401(k) plan, the sponsor assembles a menu of
options for participants to choose from to place their retirement savings. The PFIO
is one of those options. It is a general-account backed group annuity contract that
consists of a series of “Guaranteed Interest Funds” (GIFs). During the class
period—from 2008 to November 2020—Principal created a new GIF every six
months and set the maturity for each at ten years. After a new GIF was created, a
portion of the money in each existing GIF was rolled forward into the new GIF. When
Principal created a new GIF, it determined for that GIF a “Guaranteed Interest Rate”
(GIR). Each GIR is fixed for the GIF’s ten-year life. This guarantee is the PFIO’s key
feature; it makes the PFIO attractive to participants who want to predictably grow
their retirement savings.
1The Honorable John A. Jarvey, then Chief Judge, United States District Court
for the Southern District of Iowa, now retired.
-2-

-- 2 of 14 --

Principal sets GIRs by subtracting “deducts” from the return it expects to earn
on assets that it holds. Deducts are Principal’s predictions about future risks and costs
that it will bear in connection with guaranteeing future payment over a GIF’s ten-year
life. Principal receives no fee for offering the PFIO; its only compensation is the
positive spread, if any, between the amount it promises to credit participants and the
amount its investments actually yield. Principal used 14 deducts to determine the
PFIO’s GIRs. The higher the amounts of the deducts, the less participants earn and
the more Principal makes. Over the class period, Principal reduced its deducts by
roughly 33 percent.
Participants earn interest at the “Composite Crediting Rate” (CCR), a weighted
average of all the GIRs. The CCR changes every six months when Principal
establishes a new GIF and GIR. Plan sponsors and participants are notified of each
new CCR before it takes effect. The CCR was between 1.10 percent and 3.50 percent
during the class period.
Rozo brought his claims under 29 U.S.C. §§ 1104(a)(1)(A) and 1106(b)(1).
Section 1104(a)(1)(A) requires “a fiduciary [to] discharge his duties with respect to
a plan solely in the interest of the participants and beneficiaries and—(A) for the
exclusive purpose of: (i) providing benefits to participants and their beneficiaries; and
(ii) defraying reasonable expenses of administering the plan.” Section 1106(b)(1)
prohibits a fiduciary from “deal[ing] with the assets of the plan in his own interest or
for his own account.”
Principal moved (1) to exclude the opinions and testimony of one of Rozo’s
experts, (2) to decertify the class, and (3) for summary judgment. Rozo moved to
exclude the opinions and testimony of one of Principal’s experts. The district court
concluded that Principal is not a fiduciary, granted summary judgment, and denied
the three other motions as moot. This court reversed and remanded, holding that
Principal acted as a fiduciary when it set the CCR. See Rozo, 949 F.3d at 1075–76.
-3-

-- 3 of 14 --

On remand, the district court held a bench trial in November 2020. The district
court made findings concerning nine deducts that Rozo challenged. Of those deducts,
five are at issue here: (1) the Surplus & Federal Income Taxes (FIT) deduct, (2) the
Additional Surplus deduct, (3) the Standard Expense Support (SES) deduct, (4) the
Full Service Accumulation (FSA) Pricing Support deduct, and (5) the Retirement and
Income Solutions (RIS) Risk Management deduct. The court found that those deducts
were reasonable and that they represented Principal’s reasonable expenses of
administering the PFIO.
On the disloyalty claim, the court first determined that “participant[s] . . .
ha[ve] an interest in payment of reasonable expenses of administering the plan” based
on the language of the statute, § 1104(a)(1), and that they have an interest in the
“soundness and stability” of the PFIO. Rozo v. Principal Life Ins. Co. (Dist. Ct. Op.),
No. 4:14-CV-00463-JAJ, 2021 WL 1837539, at *15 (S.D. Iowa Apr. 8, 2021). It
concluded “that Principal’s determination of the deducts . . . properly served the
interest of the participants.” Id. (internal quotation marks omitted). As to the CCR,
it determined:
It is in both the participants’ and Principal’s interest[s] to establish a
CCR that will appropriately account for Principal’s risks and costs in
offering the PFIO, not just so that the product can remain competitive
in the market, but so that Principal can make good on its guarantees
to participants.
Id. at *18. The court concluded that Principal set the best CCR that it could for
participants.
On the prohibited-transaction claim, the court first analyzed whether Principal
engaged in self-dealing. It then determined whether Principal instead received
“reasonable compensation for services rendered . . . in the performance of [its] duties
with the plan.” 29 U.S.C. § 1108(c)(2) (exemptions from prohibited transactions).
-4-

-- 4 of 14 --

“[T]he court [found] that Principal’s setting of the CCR was not dealing with the
assets of the plan in Principal’s own interest or for its own account.” Dist. Ct. Op.,
2021 WL 1837539, at *22. It alternatively held that “even if Rozo could establish
self-dealing . . . the court finds in Principal’s favor on its ‘reasonable compensation’
defense.” Id. at *23.
II. Discussion
After a bench trial, this court reviews legal conclusions de novo and
factual findings for clear error. Under the clearly erroneous standard,
we will overturn a factual finding only if it is not supported by
substantial evidence in the record, if it is based on an erroneous view
of the law, or if we are left with the definite and firm conviction that
an error was made.
Urb. Hotel Dev. Co. v. President Dev. Grp., L.C., 535 F.3d 874, 879 (8th Cir. 2008)
(internal quotation marks and citation omitted).
A. Disloyalty Claim
“To prevail on a claim of breach of fiduciary duty under ERISA, the plaintiff
must make a prima facie showing that a defendant acted as a fiduciary, breached his
fiduciary duties, and thereby caused a loss to the [p]lan.” Dormani v. Target Corp.,
970 F.3d 910, 914 (8th Cir. 2020) (internal quotation marks omitted). The issue in
this case is whether there was a breach. Rozo makes two arguments. First, he
contends that the district court erred by holding that Principal did not breach its duty.
Rozo avers “that Principal acted at least in part to advance its own interests, e.g., by
increasing profit,” and thus failed to act solely in participants’ interests. Appellant’s
Br. at 23 (emphasis omitted). Second, he argues that the court clearly erred by finding
that the deducts were reasonable and represented Principal’s reasonable expenses of
administering the PFIO.
Rozo contends that “if the fiduciary acts even ‘in part’ to further its own
interests, it breaches its duty.” Id. at 24–25 (citing Tussey v. ABB, Inc., 850 F.3d 951,
-5-

-- 5 of 14 --

957–58 (8th Cir. 2017); Leigh v. Engle, 727 F.2d 113, 129 (7th Cir. 1984); Donovan
v. Bierwirth, 680 F.2d 263, 271 (2d Cir. 1982)). The authorities relied on by Rozo are
distinguishable. In Leigh, the Seventh Circuit described a non-exhaustive list of
“several factors . . . relevant in deciding whether the plan administrators acted solely
in the interests of the plan beneficiaries.” 727 F.2d at 127. The only factor applicable
here is “the risk of conflicts between the interests of the fiduciaries and beneficiaries,”
which that court considered “the key warning signal for possible misuse of plan
assets.” Id.2 Leigh and Donovan, however, both “involved the commitment of plan
assets to corporate control contests in which the plan trustees’ jobs were at stake.”
Metzler v. Graham, 112 F.3d 207, 213 (5th Cir. 1997). Those cases present much
different scenarios from the one here.
In Donovan, the Second Circuit “accept[ed] the argument” that “despite the
words ‘sole’ and ‘exclusive’, . . . officers or directors [who were also trustees of a
company’s pension plan] d[id] not violate their duties by following a course of action
with respect to the plan which benefits the corporation as well as the beneficiaries.”
680 F.2d at 271. It also set forth the standard that “[fiduciaries’] decisions must be
made with an eye single to the interests of the participants and beneficiaries.” Id. The
Court also applies that standard but with the caveat that “incidental benefits . . . are
not impermissible.” Hughes Aircraft Co. v. Jacobson, 525 U.S. 432, 445 (1999)
(rejecting retired employees’ argument that their employer impermissibly benefitted
because it lowered its labor costs after it “effectively increas[ed] certain employees’
2The other two factors are not applicable. The first—“whether fiduciaries with
divided loyalties make an intensive and scrupulous investigation of the plan’s
investment options”—does not apply because Principal is not in a position to
investigate investment options for plan participants. Id. It merely offers one of those
options. The second—“the consistent management of plan assets in congruence with
the fiduciaries’ personal interests over a substantial period of time in control
contests”—does not apply because this case does not involve a contest for control of
a company. Id.
-6-

-- 6 of 14 --

wages through either providing increased retirement incentives or including those
employees in the [p]lan’s noncontributory [benefit] structure”).
In Tussey, we held that the company sponsor of multiple plans and its agents
(collectively, the company) breached its duties because it removed “fund A” from the
plans’ menu of options and redirected investments from fund A to “fund B” in order
to benefit the recordkeeper for the plans and the investment advisor for fund B. See
850 F.3d at 956–57. We rejected the company’s argument that “some [evidence]
. . . showed [the company] acting against [the interests of the recordkeeper and
investment advisor] in various ways” because “[the company’s] examples all relate[d]
to other investment decisions, not the [fund] swap.” Id. at 957. Rozo’s contention that
if a fiduciary acts even in part to further its own interests, it breaches its duty relies
in part on our rejection of the company’s argument in Tussey. Our rejection of that
argument, however, does not support Rozo’s contention. The facts are
distinguishable. That case centered upon the company’s fund swap. There are no
similar facts in this case.
Leigh and Donovan provide some general guiding principles on conducting the
breach-of-duty inquiry. But neither those cases nor Tussey involved a determination
of whether a fiduciary breached its duty of loyalty by setting a certain interest rate for
plan participants. Turning to more factually similar cases, some of our other sister
circuits have recognized that “ERISA does not create an exclusive duty to maximize
pecuniary benefits.” Collins v. Pension & Ins. Comm. of S. Cal. Rock Prods. & Ready
Mixed Concrete Ass’ns, 144 F.3d 1279, 1282 (9th Cir. 1998) (per curiam) (affirming
grant of summary judgment for plan administrator; holding that it did not breach its
duty by failing to increase benefits); see also Foltz v. U.S. News & World Rep., Inc.,
865 F.2d 364, 373–74 (D.C. Cir. 1989) (affirming judgment for employer; holding
that the employer did not breach its duty by deciding to treat stocks that it bought
back from retired employees as minority interests, even though that resulted in retired
employees’ stocks selling for less than the amount current employees’ stocks were
-7-

-- 7 of 14 --

later worth). The First Circuit has “balk[ed] at the notion that a fiduciary violates
ERISA’s duty of loyalty simply by picking ‘too conservative’ a benchmark for a
stable value fund.”3 Ellis v. Fid. Mgmt. Tr. Co., 883 F.3d 1, 9 (1st Cir. 2018)
(affirming summary judgment for plan administrator).
In addition to the Seventh and Second Circuits, see Leigh, 727 F.2d 113;
Donovan, 680 F.2d 263, other circuits have taken differing approaches. The Fourth
Circuit’s approach is to determine whether a substantial conflict of interest exists and,
if so, scrutinize the fiduciary’s actions more closely than when it acts solely in the
beneficiaries’ interests. See Doe v. Grp. Hospitalization & Med. Servs., 3 F.3d 80, 87
(4th Cir. 1993), abrogated on other grounds by Carden v. Aetna Life Ins. Co., 559
F.3d 256 (4th Cir. 2009). The Ninth Circuit considers the fiduciary’s “state of
mind”—chiefly, whether the fiduciary was “motivated by economic self-
interest”—which it has determined to be a fact issue for the district court’s resolution.
See Pilkington PLC v. Perelman, 72 F.3d 1396, 1402 (9th Cir. 1995). The First and
Fifth Circuits, respectively, also understand the duty of loyalty to “require . . . that the
fiduciary not place its own interests ahead of those of the [p]lan beneficiary,” Vander
Luitgaren v. Sun Life Assur. Co. of Canada, 765 F.3d 59, 65 (1st Cir. 2014)
(emphasis added), or “over the plan’s interest,” Metzler, 112 F.3d at 213 (emphasis
added).
3“A stable value fund is a portfolio of bonds that are insured to protect the
investor against a decline in yield or a loss of capital.”
Carol M. Kopp, Stable Value Fund Defined, Investmentopedia,
https://www.investopedia.com/terms/s/stable-value-fund.asp (last updated July 31,
2020). The PFIO is a stable value contract. Both types of products “use[] investment
contracts to help deliver the unique benefits for which stable value is known: capital
preservation, liquidity, and steady, positive returns.” Stable Value Inv. Ass’n, What
t y p e s o f c o n t r a c t s a r e u s e d i n s t a b l e v a l u e f u n d s ? ,
https://www.stablevalue.org/what-types-of-contracts-are-used-in-stable-value-funds/
(last updated July 23, 2021).
-8-

-- 8 of 14 --

Our circuit has not set forth factors for determining whether plan administrators
acted solely in participants’ interests. “Whether . . . fiduciaries’ actions constituted
a breach is a legal question we must answer de novo.” Tussey, 850 F.3d at 956. We
conduct an inquiry similar to those of our sister circuits. In assessing whether a
conflict of interest existed between Principal and the participants, we first determine
each of the parties’ interests. In determining the participants’ interests, we find
instructive the First Circuit’s analysis of the stable value fund issue in Ellis. The
participants here, like those in Ellis, have an interest in risk-averse “asset
preservation.” 883 F.3d at 9. The trade-off for risk avoidance is that they were “not
to expect robust returns.” Id. Principal, unpaid for offering the PFIO, is only
compensated for any positive spread between what it promises to credit participants
and what its investments actually yield. We agree with the district court “that there
is tension between [the parties’ interests], because the higher the deducts to the GIRs
for the PFIO, the lower the rate paid to participants will be, while—with the
exception of ‘pass through’ deducts—the higher the deducts, the higher Principal’s
revenue from the PFIO will be.” Dist. Ct. Op., 2021 WL 1837539, at *18. But that
tension does not inevitably result in the type of conflict of interest that establishes a
breach of the duty of loyalty. See Grp. Hospitalization, 3 F.3d at 87.
Because there is a conflict of interest, we scrutinize Principal’s actions more
closely, see id., and determine its state of mind when it set the PFIO’s CCR,
Pilkington, 72 F.3d at 1402. “[W]hat [the fiduciary] did, and why, are factual matters
on which we accept the district court’s findings unless they are clearly wrong.”
Tussey, 850 F.3d at 956. The district court determined that Principal set the CCR
according to a shared interest with participants—“to establish a CCR that will
appropriately account for Principal’s risks and costs in offering the PFIO.” Dist. Ct.
Op., 2021 WL 1837539, at *18. We agree that Principal and the participants share
that interest because “a guaranteed CCR that is too high threatens the long-term
sustainability of the guarantees of the PFIO, which is detrimental to ‘the interest of
the participants.’” Id.
-9-

-- 9 of 14 --

The question then becomes whether the court clearly erred by finding that
Principal set the CCR in participants’ interests. The court found the following:
The court [found] credible the testimony of Principal’s witnesses
that Principal’s actuaries who reviewed the deducts “tr[ied] to set the
best rate that [they could] for participants” while also appropriately
accounting for Principal’s anticipated costs and risks, to ensure Principal
could make good on its obligation to pay participants the PFIO’s
guaranteed rate regardless of future market conditions. . . .
[T]he primary support Rozo offers for his contention that the deducts
were excessive is the testimony of his expert, Dr. Kopcke. . . . The court
[found] his opinions wholly unpersuasive in light of the evidence of the
reasonable—indeed, meticulous—process Principal used to determine
the deducts. . . . [T]hat reasonable process provides an inference—here,
a strong one—that Principal’s motive was to act in “the interest of the
participants.”
Id. at *20 (second and third alterations in original). “When findings are based on
determinations regarding the credibility of witnesses, [Fed. R. Civ. P.] Rule 52
demands even greater deference to the trial court’s findings, and unless contradicted
by extrinsic evidence or internally inconsistent, such findings can virtually never be
clear error.” Adzick v. UNUM Life Ins. Co. of Am., 351 F.3d 883, 889 (8th Cir. 2003).
We hold that the court did not clearly err in finding that the deducts, and thus the
CCR, were set in participants’ interests.
We also hold that the court did not clearly err by finding that the deducts were
reasonable and set by Principal in the participants’ interest of paying a reasonable
amount for the PFIO’s administration. First, the court found that the RIS Risk
Management deduct was reasonable and a reasonable expense in the participants’
interest. The court concluded:
-10-

-- 10 of 14 --

Principal’s actuaries estimated this deduct through studies of its risk
management activities and stochastic analyses of the potential market
value losses associated with plan lapses (departures) for products with
a 12-month put. Principal refined its analysis of the 12-month put risk
to make it suitable for pricing and began to use that refined analysis
to compute the RIS Risk Management deduct for future GIRs from
2015 onward.
Dist. Ct. Op., 2021 WL 1837539, at *12.
Principal’s RIS division did operate with a “profit objective[]” as Rozo alleged.
Appellant’s Br. at 26 (quoting Appellant’s App’x at 285). But simply using that
division’s expertise did not show that Principal pursued a profit when it charged a
reasonable fee for the use of those services to participants via the deduct. We hold
that the court did not clearly err in its findings as to the RIS Risk Management deduct
because that conclusion was supported by substantial evidence. See Urb. Hotel, 535
F.3d at 879. The record shows the following: (1) a financial analyst involved in the
PFIO rate-setting process pre-2015 confirmed that Principal calculated what the
annual charges of those services are, and (2) Principal documented those calculations
and those documents were discussed at length at trial.
Second, Principal calculated the Surplus and FIT deduct to target a return of
its own funds set aside to back the PFIO. The court found that Principal’s target
return rate of 15 percent to 20 percent was reasonable and a reasonable expense in the
participants’ interest, rejecting Dr. Kopcke’s opinion that Principal should have aimed
for a rate of 10 percent to 12 percent. The court’s finding was not clearly erroneous.
Third, Principal implemented the Additional Surplus deduct after the 2008
financial crisis to increase the target rate in the Surplus and FIT deduct. It did so to
address increased risks and costs of offering the PFIO. The court found that the
deduct was reasonable and a reasonable expense in the participants’ interest, rejecting
-11-

-- 11 of 14 --

Dr. Kopcke’s opinion that the deduct should have been eliminated because the
Surplus and FIT deduct fully compensated Principal. As to both of those deducts,
Rozo argues that Principal’s target rate was higher than necessary because others in
Principal’s industry have gotten by with a lower rate. But as our sister circuits have
held, “ERISA does not create an exclusive duty to maximize pecuniary benefits.”
Collins, 144 F.3d at 1282; see also Foltz, 865 F.2d at 373–74.
The court’s conclusions as to the Surplus and FIT deduct and the Additional
Surplus deduct were supported by substantial evidence. The court heard multiple
witnesses testify that a 20-percent target rate became necessary after the 2008
financial crisis. The court noted, “In general, companies in the insurance and financial
industry target—and their investors expect—post-tax returns of 12% to 25% on their
capital.” Dist. Ct. Op., 2021 WL 1837539, at *4. Principal’s actual return on capital
averaged 11 percent and ranged from 5 percent to 16 percent, at the low end of or
lower than the industry range. We agree with the Ninth Circuit that “ERISA does not
create an exclusive duty to maximize pecuniary benefits.” Collins, 144 F.3d at 1282.
We hold that the court did not clearly err in its findings.
Fourth, Principal implemented the SES deduct to adjust for over-crediting
plans for its administrative services. This deduct equaled the amount credited to the
plan and participants in the form of a reduced administrative services fee. This deduct
was reasonable and constituted a reasonable expense in participants’ interests as pass-
throughs to the plan and participants. The district court found the same to be true for
the FSA Pricing Support deduct. We hold that the court did not clearly err in its
findings as to the SES and FSA Pricing Support deducts because its conclusions were
supported by substantial evidence. Specifically, the court relied on these facts: (1)
Principal did not make any profit from these deducts, (2) the amounts of the deducts
were well within the range of revenue sharing amounts for Principal’s other products,
(3) one witness testified that expense studies supported the SES deduct amount, and
-12-

-- 12 of 14 --

(4) another witness testified that the amount of the deduct was appropriate as a matter
of credible actuarial judgment.
The court did not clearly err in finding that Principal set the deducts in the
participants’ interest of paying a reasonable amount for the PFIO’s administration.
It also did not clearly err in finding that the CCR was set in the participants’ interest.
We accept both of these findings. Consequently, we hold that in setting the CCR,
Principal was not “motivated by economic self-interest,” Pilkington, 72 F.3d at 1402,
and that it did not either “place its own interests ahead of those of the [participants],”
Vander Luitgaren, 765 F.3d at 65, or “over the plan’s interest,” Metzler, 112 F.3d at
213. We affirm the court’s judgment in favor of Principal on the disloyalty claim.
B. Prohibited-Transaction Claim
Rozo argues that Principal engaged in prohibited self dealing because it
“generat[ed] revenue for itself from the plan contract.” Appellant’s Br. at 52. He also
argues that the reasonable-expense exemption from liability for self dealing does not
apply to Principal because it failed to establish that its compensation was reasonable
because the deducts and the CCR were not reasonable.
The reasonable-expense exemption from liability for self dealing “must be
proven by the defendant.” Braden v. Wal-Mart Stores, Inc., 588 F.3d 585, 601 (8th
Cir. 2009) (stating that defendant has burden to prove “statutory exemptions
established by § 1108”). To determine whether the exemption applies to Principal, we
first determine whether Principal proved that its compensation was reasonable. The
district court’s holding in Harley, which we affirmed, suggests that compensation is
reasonable if the amount was not “the result of any inflationary tactics” and the
fiduciary offers expert opinion that the amount was reasonable. Harley v. Minn.
Mining & Mfg. Co., 42 F. Supp. 2d 898, 911 (D. Minn. 1999), aff’d, 284 F.3d 901
(8th Cir. 2002). Here, the district court did not clearly err in finding that the
deducts—and thus the CCR—were reasonable. See supra Section II.A. Its findings
-13-

-- 13 of 14 --

were supported by witness testimony it deemed credible. See id. We hold that
Principal has met its burden of establishing that its compensation was reasonable.
We affirm the district court’s judgment in favor of Principal on the prohibited-
transaction claim because it is exempted from liability for receiving reasonable
compensation.
III. Conclusion
Accordingly, we affirm the district court’s judgment.
______________________________
-14-

-- 14 of 14 --