NOTICE: This opinion is subject to
formal revision before publication in the preliminary print of the
United States Reports. Readers are requested to notify the Reporter
of Decisions, Supreme Court of the United States, Washington,
D. C. 20543, of any typographical or other formal errors, in
order that corrections may be made before the preliminary print
goes to press.
SUPREME COURT OF THE UNITED STATES
_________________
No. 18–1116
_________________
INTEL CORPORATION INVESTMENT POLICY COMMITTEE,
et al., PETITIONERS
v. CHRISTOPHER M. SULYMA
on writ of certiorari to the united states
court of appeals for the ninth circuit
[February 26, 2020]
Justice Alito delivered the opinion of the
Court.
The Employee Retirement Income Security Act of
1974 (ERISA) requires plaintiffs with “actual knowledge” of an
alleged fiduciary breach to file suit within three years of gaining
that knowledge rather than within the 6-year period that would
otherwise apply. §413(a)(2)(A), 88Stat. 889, as amended, 29
U. S. C. §1113. The question here is whether a plaintiff
necessarily has “actual knowledge” of the information contained in
disclosures that he receives but does not read or cannot recall
reading. We hold that he does not and therefore affirm.
I
A
Retirement plans governed by ERISA must have
at least one named fiduciary, §1102(a)(1), who must manage the plan
prudently and solely in the interests of participants and their
beneficiaries, §1104(a). Fiduciaries who breach these duties are
personally liable to the plan for any resulting losses. §1109(a).
ERISA authorizes participants and their beneficiaries, as well as
co-fiduciaries and the Secretary of Labor, to sue for that relief.
§1132(a)(2).
Such suits must be filed within one of three
time periods, each with different triggering events. The first
begins when the breach occurs. Specifically, under §1113(1), suit
must be filed within six years of “the date of the last action
which constituted a part of the breach or violation” or, in cases
of breach by omission, “the latest date on which the fiduciary
could have cured the breach or violation.” We have referred to
§1113(1) as a statute of repose, which “effect[s] a legislative
judgment that a defendant should be free from liability after the
legislatively determined period of time.”
California Public
Employees’ Retirement System v.
ANZ Securities, Inc.,
582 U. S. ___, ___ (2017) (slip op., at 5) (internal quotation
marks omitted).
The second period, which accelerates the filing
deadline, begins when the plaintiff gains “actual knowledge” of the
breach. Under §1113(2), suit must be filed within three years of
“the earliest date on which the plaintiff had actual knowledge of
the breach or violation.” Section 1113(2) is a statute of
limitations, which “encourage[s] plaintiffs to pursue diligent
prosecution of known claims.”
Id., at ___ (slip op., at 5)
(internal quotation marks omitted).
The third period, which applies “in the case of
fraud or concealment,” begins when the plaintiff discovers the
alleged breach. §1113. In such cases, suit must be filed within six
years of “the date of discovery.”
Ibid.
B
Respondent Sulyma worked at Intel Corporation
from 2010 to 2012. He participated in two Intel retirement plans,
the Intel Retirement Contribution Plan and the Intel 401(k) Savings
Plan. Payments into these plans were in turn invested in two funds
managed by the Intel Investment Policy Committee.[
1] These funds mostly comprised stocks and
bonds. After the stock market decline in 2008, however, the
committee increased the funds’ shares of alternative assets, such
as hedge funds, private equity, and commodities. These assets
carried relatively high fees. And as the stock market rebounded,
Sulyma’s funds lagged behind others such as index funds.
Sulyma filed this suit on behalf of a putative
class in October 2015, alleging primarily that the committee and
other plan administrators (petitioners here) had breached their
fiduciary duties by overinvesting in alternative assets.
Petitioners countered that the suit was untimely under §1113(2).
Although Sulyma filed it within six years of the alleged breaches,
he filed it more than three years after petitioners had disclosed
their investment decisions to him.
ERISA and its implementing regulations mandate
various disclosures to plan participants. See generally 29
U. S. C. §§1021–1031; see also
Gobeille v.
Liberty Mut. Ins. Co., 577 U. S. ___, ___–___ (2016).
Sulyma received numerous disclosures while working at Intel, some
explaining the extent to which his retirement plans were invested
in alternative assets. In November 2011, for example, he received
an e-mail informing him that a Qualified Default Investment
Alternative (QDIA) notice was available on a website called
NetBenefits, where many of his disclosures were hosted. See App.
149–151; see also 29 CFR §§2550.404c–5(b)–(d) (2019) (QDIA
notices); §2520.104b–1(c) (regulating electronic disclosure). This
notice broke down the percentages at which his 401(k) fund was
invested in stocks, bonds, hedge funds, and commodities. See App.
236. In 2012, he received a summary plan description explaining
that the funds were invested in stocks and alternative assets,
id., at 227, and referring him to other documents—called
fund fact sheets—with the percentages in graphical form. See 29
U. S. C. §§1022, 1024(b) (summary plan descriptions); see
also App. 307 (June 2012 fact sheet for his 401(k) plan fund);
id., at 338 (June 2012 fact sheet for his retirement
contribution plan fund);
id., at 277–340 (other fact sheets
provided during his tenure at Intel). Also in 2012, he received
e-mails directing him to annual disclosures that petitioners
provided for both his plans, which showed the underlying funds’
return rates and again directed him to the NetBenefits site for
further information. See 29 CFR §2550.404a–5; see also App. 242–243
(retirement contribution plan annual disclosure);
id., at
250–251 (401(k) plan annual disclosure).
Petitioners submitted records showing that
Sulyma visited the NetBenefits site repeatedly during his
employment.
Id., at 258–276. But he testified in his
deposition that he did not “remember reviewing” the above
disclosures during his tenure.
Id., at 175; see also
id., at 183, 193, 196–197. He also stated in a declaration
that he was “unaware” while working at Intel “that the monies that
[he] had invested through the Intel retirement plans had been
invested in hedge funds or private equity.”
Id., at 212. He
recalled reviewing only account statements sent to him by mail,
which directed him to the NetBenefits site and noted that his plans
were invested in “short-term/other” assets but did not specify
which. See,
e.g.,
id., at 375.
The District Court granted summary judgment to
petitioners under §1113(2), reasoning that “[i]t would be improper
to allow Sulyma’s claims to survive merely because he did not look
further into the disclosures made to him.” 2017 WL 1217185, *9 (ND
Cal., Mar. 31, 2017). The Ninth Circuit reversed. As relevant
here,[
2] the court construed
“actual knowledge” to mean “what it says: knowledge that is actual,
not merely a possible inference from ambiguous circumstances.” 909
F.3d 1069, 1076 (2018) (internal quotation marks omitted). Although
Sulyma “had sufficient information available to him to know about
the allegedly imprudent investments” more than three years before
filing suit, the court held that his testimony created a dispute as
to when he actually gained that knowledge.
Id., at 1077.
Several Circuits have likewise construed
§1113(2) to require “knowledge that is actual,”
id., at
1076, but one has construed it to require only proof of sufficient
disclosure.[
3] We granted
certiorari, 587 U. S. ___ (2019), to resolve whether the
phrase “actual knowledge” does in fact mean “what it says,” 909
F. 3d, at 1076, and hold that it does.
II
A
“We must enforce plain and unambiguous
statutory language” in ERISA, as in any statute, “according to its
terms.”
Hardt v.
Reliance Standard Life Ins. Co.,
560 U.S.
242, 251 (2010). Although ERISA does not define the phrase
“actual knowledge,” its meaning is plain. Dictionaries are hardly
necessary to confirm the point, but they do. When Congress passed
ERISA, the word “actual” meant what it means today: “existing in
fact or reality.” Webster’s Seventh New Collegiate Dictionary 10
(1967); accord, Merriam-Webster’s Collegiate Dictionary 13 (11th
ed. 2005) (same); see also American Heritage Dictionary 14 (1973)
(“In existence; real; factual”);
id., at 18 (5th ed. 2011)
(“Existing in reality and not potential, possible, simulated, or
false”). So did the word “knowledge,” which meant and still means
“the fact or condition of being aware of something.” Webster’s
Seventh New Collegiate Dictionary 469 (1967); accord,
Merriam-Webster’s Collegiate Dictionary 691 (2005) (same); see also
American Heritage Dictionary 725 (1973) (“Familiarity, awareness,
or understanding gained through experience or study”);
id.,
at 973 (2011) (same). Thus, to have “actual knowledge” of a piece
of information, one must in fact be aware of it.
Legal dictionaries give “actual knowledge” the
same meaning: “[r]eal knowledge as distinguished from presumed
knowledge or knowledge imputed to one.” Ballentine’s Law Dictionary
24 (3d ed. 1969); accord, Black’s Law Dictionary 1043 (11th ed.
2019) (defining “actual knowledge” as “[d]irect and clear
knowledge, as distinguished from constructive knowledge”).[
4] The qualifier “actual” creates that
distinction. In everyday speech, “actual knowledge” might seem
redundant; one who claims “knowledge” of a topic likely means to
suggest that he actually knows a thing or two about it. But the law
will sometimes impute knowledge—often called “constructive”
knowledge—to a person who fails to learn something that a
reasonably diligent person would have learned. See
id., at
1043. Similarly, we held in
Merck & Co. v.
Reynolds,
559 U.S.
633 (2010), that the word “discovery,” when used in a statute
of limitations without qualification, “encompasses not only those
facts the plaintiff actually knew, but also those facts a
reasonably diligent plaintiff would have known.”
Id., at
648. The addition of “actual” in §1113(2) signals that the
plaintiff ’s knowledge must be more than “potential, possible,
virtual, conceivable, theoretical, hypothetical, or nominal.”
Black’s Law Dictionary 53 (4th ed. 1951). Indeed, in
Merck,
we cited §1113(2) as evidence of the “linguistic distinction”
between “ ‘
actual knowledge’ ” and the
“hypothetical” knowledge that a reasonably diligent plaintiff would
have. 559 U. S., at 646–647 (quoting §1113(2); emphasis in
original).
Congress has drawn the same distinction
elsewhere in ERISA. Multiple provisions contain alternate 6-year
and 3-year limitations periods, with the 6-year period beginning at
“the date on which the cause of action arose” and the 3-year period
starting at “the earliest date on which the plaintiff acquired
or should have acquired actual knowledge of the existence of
such cause of action.” §§1303(e)(6), (f )(5) (emphasis added);
accord, §§1370(f )(1)–(2), 1451(f )(1)–(2). ERISA also
requires plaintiffs challenging the suspension of benefits under
§1085 to do so within “one year after the earliest date on which
the plaintiff acquired or should have acquired actual knowledge of
the existence of such cause of action.” §1085(e)(9)(I)(iv). Thus,
Congress has repeatedly drawn a “linguistic distinction” between
what an ERISA plaintiff actually knows and what he should actually
know.
Merck, 559 U. S., at 647. And when Congress has
included both forms of knowledge in a provision limiting ERISA
actions, it has done so explicitly. We cannot assume that it meant
to do so by implication in §1113(2). Instead we “generally
presum[e] that Congress acts intentionally and purposely when it
includes particular language in one section of a statute but omits
it in another.”
BFP v.
Resolution Trust Corporation,
511 U.S.
531, 537 (1994) (internal quotation marks omitted).
Petitioners dispute the characterization of
anything less than actual knowledge as constructive knowledge,
arguing that the latter term usually refers to information that a
plaintiff must seek out rather than information that is sent to
him. But if a plaintiff is not aware of a fact, he does not have
“actual knowledge” of that fact however close at hand the fact
might be. §1113(2). And Congress has never added to §1113(2) the
language it has used in other ERISA limitations provisions to
encompass both what a plaintiff actually knows and what he
reasonably could know.
As presently written, therefore, §1113(2)
requires more than evidence of disclosure alone. That all relevant
information was disclosed to the plaintiff is no doubt
relevant in judging whether he gained knowledge of that
information. See Part III,
infra. To meet §1113(2)’s “actual
knowledge” requirement, however, the plaintiff must in fact have
become aware of that information.
B
Petitioners offer arguments for a broader
reading of §1113(2) based on text, context, purpose, and statutory
history. All founder on Congress’s choice of the word “actual.”
As for text, petitioners do not dispute the
normal definitions of “actual,” “knowledge,” or “actual knowledge.”
They focus instead on the least conspicuous part of the phrase “had
actual knowledge”: the word “had.” §1113(2). Once a plaintiff
receives a disclosure, they argue, he “ha[s]” the knowledge that
§1113(2) requires because he effectively holds it in his hand.
Ibid. In other words, he has the requisite knowledge because
he could acquire it with reason- able effort. That turns §1113(2)
into what it is plainly not: a constructive-knowledge
requirement.
Petitioners’ contextual argument fails for the
same reason. As they point out, ERISA’s disclosure regime is meant
to “ensur[e] that ‘the individual participant knows exactly where
he stands with respect to the plan.’ ”
Firestone Tire &
Rubber Co. v.
Bruch,
489 U.S.
101, 118 (1989) (quoting H. R. Rep. No. 93–533, p. 11
(1973)). This is the reason for ERISA’s requirements that
disclosures be written for a lay audience. See,
e.g., 29
U. S. C. §1022(a). Once plan administrators satisfy their
obligations to impart knowledge, petitioners say, §1113(2)’s
knowledge requirement is satisfied too. But that is simply not what
§1113(2) says. Unlike other ERISA limitations periods—which also
form §1113(2)’s context—§1113(2) begins only when a plaintiff
actually is aware of the relevant facts, not when he should be. And
a given plaintiff will not necessarily be aware of all facts
disclosed to him; even a reasonably diligent plaintiff would not
know those facts immediately upon receiving the disclosure.
Although “the words of a statute must be read in their context,”
Davis v.
Michigan Dept. of Treasury,
489 U.S.
803, 809 (1989), petitioners’ argument again gives the word
“actual” little meaning at all.
Petitioners also argue that §1113(2)’s plain
meaning undermines its purpose of protecting plan administrators
from suits over bygone investment decisions. If a plan participant
can simply deny knowledge, they say, administrators will rarely get
the benefit of §1113(2). But even if this is true, as it may well
be, we cannot say that heeding the clear meaning of the word
“actual” renders the statute so “ ‘[in]coherent’ ” that
it must be disregarded.
Kingdomware Technologies, Inc. v.
United States, 579 U. S. ___, ___ (2016) (slip op., at
8).
For one thing, plan participants are not the
only potential plaintiffs subject to §1113. The Secretary of Labor,
for example, may also sue imprudent fiduciaries for the benefit of
plan participants. See §1132(a)(2). And the United States
represents that the Secretary will have a hard time doing so within
§1113(2)’s timeframe if deemed to have actual knowledge of the
facts contained in the many reports that the Department receives
from ERISA plans each year. See Brief for United States as
Amicus Curiae 27–28. Moreover, the statute’s repose period
will still protect defendants from suits filed more than six years
after the alleged breach. See §1113(1).
Petitioners may well be correct that heeding the
plain meaning of §1113(2) substantially diminishes the protection
that it provides for ERISA fiduciaries, but by the same token,
petitioners’ interpretation would greatly reduce §1113(1)’s value
for beneficiaries, given the disclosure regime that petitioners
themselves emphasize. Choosing between these alternatives is a task
for Congress, and we must assume that the language of §1113(2)
reflects Congress’s choice. If policy considerations suggest that
the current scheme should be altered, Congress must be the one to
do it. See,
e.g.,
Azar v.
Allina Health
Services, 587 U. S. ___, ___ (2019).
Finally, petitioners argue that the plain
meaning of “actual knowledge” renders an
earlier version of
§1113(2) incoherent. As originally enacted, the §1113(2)
limitations period began either when the plaintiff gained actual
knowledge of the alleged breach or when “a report from which [the
plaintiff] could reasonably be expected to have obtained knowledge
. . . was filed with” the Secretary of Labor. 29
U. S. C. §1113(2) (1976 ed.). That latter,
constructive-knowledge clause was later repealed. See Omnibus
Budget Reconciliation Act of 1987, §9342(b), 101Stat. 1330–371.
According to petitioners, if “actual knowledge” means what it says,
then the original version of §1113(2) charged plan participants
with learning what was sent to the Secretary but not what was sent
to them.
The version at issue here, however, is the
current one—from which Congress removed any mention of constructive
knowledge. “When Congress acts to amend a statute, we presume it
intends its amendment to have real and substantial effect.”
Intel Corp. v.
Advanced Micro Devices, Inc.,
542 U.S.
241, 258–259 (2004) (internal quotation marks omitted). Section
1113(2)’s history thus more readily suggests that the current
version does in fact require actual knowledge.
III
Nothing in this opinion forecloses any of the
“usual ways” to prove actual knowledge at any stage in the
litigation.
Farmer v.
Brennan,
511 U.S.
825, 842 (1994). Plaintiffs who recall reading particular
disclosures will of course be bound by oath to say so in their
depositions. On top of that, actual knowledge can be proved through
“inference from circumstantial evidence.”
Ibid.; see also
Staples v.
United States,
511
U.S. 600, 615–616, n. 11 (1994) (“[K]nowledge can be inferred
from circumstantial evidence”). Evidence of disclosure would no
doubt be relevant, as would electronic records showing that a
plaintiff viewed the relevant disclosures and evidence suggesting
that the plaintiff took action in response to the information
contained in them. And though, “[a]t the summary judgment stage,
facts must be viewed in the light most favorable to the nonmoving
party,” that is true “only if there is a ‘genuine’ dispute as to
those facts.”
Scott v.
Harris,
550 U.S.
372, 380 (2007) (quoting Fed. Rule Civ. Proc. 56(c)). If a
plaintiff ’s denial of knowledge is “blatantly contradicted by
the record,” “a court should not adopt that version of the facts
for purposes of ruling on a motion for summary judgment.” 550
U. S., at 380.
Today’s opinion also does not preclude
defendants from contending that evidence of “willful blindness”
supports a finding of “actual knowledge.” Cf.
Global-Tech
Appliances, Inc. v.
SEB S. A.,
563 U.S.
754, 769 (2011).
In the case before us, however, petitioners do
not argue that “actual knowledge” is established in any of these
ways, only that they need not offer any such proof. And that is
incorrect.
* * *
For these reasons, we affirm.
It is so ordered.
