NOT RECOMMENDED FOR PUBLICATION File Name: 21a0299n.06
Case No. 20-5710
UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT
FILED Jun 28, 2021 CYNTHIA PITTMAN, Individually and on ) DEBORAH S. HUNT, Clerk Behalf of All Others Similarly Situated, et al., ) ) Plaintiffs-Appellants, ) ON APPEAL FROM THE UNITED ) STATES DISTRICT COURT FOR v. ) THE EASTERN DISTRICT OF ) TENNESSEE UNUM GROUP, et al., ) ) Defendants-Appellees. ) )
BEFORE: COLE, STRANCH, and THAPAR, Circuit Judges.
THAPAR, Circuit Judge. Unum Group is an insurance company that thought it could profit
from the sale of “long-term-care” policies. These policies provide support for individuals as they
age, often by paying for nursing-home and other care. But Unum’s calculations turned out to be
off, and the policies generated large losses. Unum announced that it had to divert some of its cash
into the reserve account used to fund these policies. The stock price fell, and investors sued. They
alleged that the company misled them in violation of the Securities Exchange Act of 1934. In a
thorough opinion, the district court analyzed the plaintiffs’ allegations and held that they were
insufficient to state a claim for securities fraud. We agree and affirm. Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
I.
A.
Unum Group sells disability, life, accident, critical-illness, and other types of insurance. It
previously sold long-term-care (LTC) policies. These policies are designed to cover costs
associated with aging, such as assisted-living and nursing-home care.
LTC policies became increasingly popular in the 1970s. At the time, insurance companies
(including Unum) thought the policies would turn a large profit. But as it turns out, the “major
pricing assumptions” underlying LTC policies were “woefully inaccurate.” R. 37, Pg. ID 302.
Policy holders began living longer and requiring more care, fewer individuals let their policies
lapse, and interest rates declined (reducing insurance companies’ investment returns). As a result,
most LTC policies were not profitable. And many companies stopped issuing them. But these
companies still faced significant liabilities from their existing policies.
Unum’s results were in line with industry trends. It stopped issuing new LTC policies by
2012. Yet it continues to manage a block of already-issued LTC policies. These policies cover
nearly one million lives and generate over $600 million in income from premiums each year. To
cover future liabilities associated with its LTC policies, Unum established a reserve fund. The
challenge was in determining how much money to put into this fund. Unum relied on various
factors to set its original reserve amount. These factors included annual premiums paid, interest
earned on reserves, policy lapse rates, and benefits paid out.
In 2014, Unum concluded that its LTC reserves were inadequate. So it increased them by
$698 million. It also updated its reserve assumptions, hoping to avoid future increases. When
Unum did so, it informed investors that it was going to use a formula called an “interest-adjusted
loss ratio” to monitor the reserves. Unum publicly released its loss ratio each quarter. It told
-2- Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
investors that if the ratio stayed in the 85–90% range, it would not have to increase reserves. If,
however, the ratio exceeded 90% “for a prolonged period of time,” Unum said it would have to
reconsider its “reserve assumptions.” R. 37, Pg. ID 336.
At first, Unum’s loss ratio remained in range. But signs of trouble emerged in 2016: The
loss ratio exceeded 90% for two quarters. It dropped back down into an acceptable range for the
next three, but then things took a turn for the worse. In the final two quarters of 2017, Unum’s
loss ratio twice exceeded 90%. And when the same thing happened in the first quarter of 2018,
the situation came to a head. Unum told investors that it would have to reevaluate its LTC reserves.
Investors recognized that a large reserve charge was likely, and the company’s shares fell nearly
17%. (Later that year, Unum increased its LTC reserves by $750 million.)
B.
In the aftermath of these stock losses, some investors filed proposed class-action lawsuits.
The district court consolidated the cases and appointed lead counsel to represent a putative class
of investors. The proposed class covered those who acquired Unum stock from October 27, 2016,
to May 1, 2018.
Plaintiffs allege that Unum Group and four of its corporate officers engaged in securities
fraud. See 15 U.S.C. § 78j(b) (direct liability); 15 U.S.C. § 78t(a) (controlling-person liability).
How? By misleading investors about the health of Unum’s LTC insurance business and reserves.
More specifically, plaintiffs allege that Unum misled investors about: (1) its success in obtaining
state approval for premium increases; (2) the types of policies in its LTC block; (3) its loss ratio,
including the relevance of that ratio for evaluating its reserves; and (4) its conformity with certain
accounting practices.
-3- Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
Plaintiffs argue that Unum’s alleged misstatements and omissions violated section 10(b)
of the Exchange Act, as implemented by SEC Rule 10b-5.1 See 15 U.S.C. § 78j(b). SEC Rule
10b-5 says that it is unlawful to “make any untrue statement of a material fact or to omit to state a
material fact necessary in order to make the statements made, in the light of the circumstances
under which they were made, not misleading.” 17 C.F.R. § 240.10b-5(b).
To establish a violation of this provision, a plaintiff must show: “(1) a material
misrepresentation or omission by the defendant; (2) scienter; (3) a connection between the
misrepresentation or omission and the purchase or sale of a security; (4) reliance upon the
misrepresentation or omission; (5) economic loss; and (6) loss causation.” Matrixx Initiatives, Inc.
v. Siracusano, 563 U.S. 27, 37–38 (2011). If a plaintiff fails to plausibly allege even one of these
elements, a securities-fraud claim cannot proceed.
In response to the complaint, Unum filed a motion to dismiss. The district court granted
the motion. It held that the plaintiffs failed to adequately allege at least two elements: (1) a
material misstatement or omission, and (2) scienter, which is the required mental state for a finding
of liability.
This appeal followed.
II.
We review the district court’s decision de novo, accept the plaintiffs’ factual allegations as
true, and construe the complaint in their favor. In re Omnicare, Inc. Sec. Litig., 769 F.3d 455, 469
(6th Cir. 2014). Because the district court correctly held that the plaintiffs failed to adequately
1 Plaintiffs also argue that the defendant corporate officers are qualifying “controlling person[s]” and thus liable for the company’s fraudulent acts under section 20(a) of the Exchange Act. See 15 U.S.C. § 78t(a). To establish controlling-person liability under section 20(a), a plaintiff must first show a primary violation of securities law. Doshi v. Gen. Cable Corp., 823 F.3d 1032, 1045 (6th Cir. 2016). Since plaintiffs’ section 20(a) claims are contingent on their section 10(b) claim succeeding, and since their section 10(b) claim does not, we provide no independent analysis of plaintiffs’ section 20(a) claims.
-4- Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
allege scienter, we affirm. We need not, and thus do not, address the district court’s conclusion
that plaintiffs also failed to adequately allege a material misstatement or omission.
Before proceeding to our analysis, a bit of background on scienter is helpful. Scienter is
“a mental state embracing intent to deceive, manipulate, or defraud.” Tellabs, Inc. v. Makor Issues
& Rights, Ltd., 551 U.S. 308, 319 (2007). To plead scienter in a securities-fraud case, a plaintiff
must “state with particularity facts giving rise to a strong inference,” 15 U.S.C. § 78u–4(b)(2)(A),
that the defendant acted with either (1) a “knowing and deliberate intent to manipulate, deceive,
or defraud,” or (2) recklessness. Doshi, 823 F.3d at 1039. An inference of scienter is strong only
if the inference is “cogent and at least as compelling as any opposing inference of nonfraudulent
intent.” Tellabs, 551 U.S. at 314. It is not enough that the inference is “merely plausible or
reasonable.” Id.; see also City of Monroe Emps. Ret. Sys. v. Bridgestone Corp., 399 F.3d 651, 683
(6th Cir. 2005).
We evaluate scienter by looking at “all the allegations holistically.” Tellabs, 551 U.S. at
326. In doing so, we consider a list of nine factors:
(1) insider trading at a suspicious time or in an unusual amount; (2) divergence between internal reports and external statements on the same subject; (3) closeness in time of an allegedly fraudulent statement or omission and the later disclosure of inconsistent information; (4) evidence of bribery by a top company official; (5) existence of an ancillary lawsuit charging fraud by a company and the company’s quick settlement of that suit; (6) disregard of the most current factual information before making statements; (7) disclosure of accounting information in such a way that its negative implications could only be understood by someone with a high degree of sophistication;
-5- Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
(8) the personal interest of certain directors in not informing disinterested directors of an impending sale of stock; and (9) the self-interested motivation of defendants in the form of saving their salaries or jobs.
Doshi, 823 F.3d at 1039–40 (quoting Helwig v. Vencor, Inc., 251 F.3d 540, 552 (6th Cir. 2001) (en
banc)). This list of factors, “while not exhaustive,” is “at least helpful in guiding securities fraud
pleading.” Helwig, 251 F.3d at 552.
In analyzing scienter, the district court noted that the plaintiffs had not presented any
evidence as to the first, second, fourth, seventh, and eighth factors. So it focused its analysis on
the remaining factors. And it found that those factors provided little or no support for scienter.
Thus, it concluded that any inference of scienter was not “cogent and at least as compelling as any
opposing inference of nonfraudulent intent.” Tellabs, 551 U.S. at 314.
The plaintiffs urge us to reach a different result. They point to five pieces of evidence that
they say support scienter. But after reviewing the evidence holistically, we conclude that the
district court was correct: Any inference of scienter is not at least as compelling as an opposing
inference of nonfraudulent intent.
1. The plaintiffs first argue that Unum’s close attention to its LTC business and the overall
importance of Unum’s LTC reserves support an inference of fraudulent intent. The theory seems
to be that because Unum’s executives had extensive knowledge of the company’s LTC business,
it is more likely that the company acted fraudulently. But the fact that executives are intimately
familiar with a core component of their business does little to suggest fraudulent intent. So this is
not a scienter-bolstering fact.
-6- Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
2. Next, the plaintiffs contend that Unum made public statements that were at odds with
factual information known by the company. And when a company shows “disregard of the most
current factual information before making statements,” this supports an inference of scienter.
Helwig, 251 F.3d at 552 (the sixth factor). The plaintiffs allege that Unum’s statements were in
conflict with the facts in two ways.
First, Unum’s officers “continually told investors that they had a granular view of Unum’s
all-important LTC block and the underlying factors affecting its reserves.” Appellant Br. 65. And
these executives undoubtedly had access to “financial reporting . . . at all times through its
software.” Id. at 66. Thus, plaintiffs argue, Unum must have known its LTC block was “troubled”
even though it “reassured” investors that it was not. Id. But “fraudulent intent cannot be inferred
merely from the [executives’] positions in the [c]ompany and alleged access to information.” PR
Diamonds, Inc. v. Chandler, 364 F.3d 671, 688 (6th Cir. 2004). And since that is the gist of what
plaintiffs allege here, this argument falls short.
Second, the plaintiffs contend that Unum mischaracterized its requests for rate increases as
being on track. (Rate increases generally require approval from state regulatory bodies.) But such
statements were not misleading because Unum revealed the approval percentages for investors to
make their own judgments. For example, Unum told investors in October 2016 that it had achieved
over two-thirds of its assumed rate increases. R. 37, Pg. ID 314 (“Less than a third of those rate
increase assumptions remain outstanding.”). And it told investors in May and December 2017 that
it had achieved “85% to 90%” of its assumed increases. Id. at 318, 320. What’s more, these
statistics made clear that at least some of Unum’s rate requests remained outstanding (the 10% to
15% gap). And plaintiffs do not allege that the disclosures were false. So any investor buying
stock during this period would have been in a position to draw his or her own conclusions from
-7- Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
the evidence. Unum’s characterization of “the statistical facts already disclosed” is not securities
fraud. In re Sanofi Sec. Litig., 155 F. Supp. 3d 386, 404 (S.D.N.Y. 2016). At worst, the statements
amount to corporate puffery or undue optimism, neither of which gives rise to liability. In re Ford
Motor Co. Sec. Litig., 381 F.3d 563, 570–71 (6th Cir. 2004).
3. The plaintiffs also argue that a California lawsuit filed against Unum in May of 2013
bolsters the case for scienter. See Don v. Unum Life Ins. Co. of Am., No. BC509235 (Cal. Super.
Ct. May 17, 2013). That lawsuit alleged that Unum improperly calculated benefits for its LTC
policyholders. Unum settled the suit in November 2015.
The “existence of an ancillary lawsuit charging fraud by a company and the company’s
quick settlement of that suit” could support an inference of scienter. Helwig, 251 F.3d at 552 (the
fifth factor). But those criteria are not met here. As the district court recognized, the allegations
of wrongdoing in the California suit relate to an entirely different course of conduct
(miscalculating rate-of-inflation increases and improperly determining policy anniversaries). And
that course of conduct ended before the class period in this suit even began. Since the evidence in
the two suits is not “closely related,” the settlement of the first suit is not persuasive evidence of
scienter in the second. See City of Monroe, 399 F.3d at 685. What’s more, despite plaintiffs’
assertion that settlement talks began a few months after the first lawsuit was filed, the case did not
actually settle until November 2015—roughly two and a half years after the suit was filed. That
is not the kind of “quick settlement” that generates a meaningful inference of scienter. Helwig,
251 F.3d at 552.
4. Additionally, the plaintiffs argue that the closeness in time between allegedly false
statements and the later disclosure of inconsistent information supports a finding of scienter. Id.
(the third Helwig factor). They allege that the timing of Unum’s announcement in early May 2018
-8- Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
that it was reassessing its LTC reserves supports scienter because of its temporal proximity to two
events.
The more proximate of the two events was Unum’s filing of a proxy statement on April 12,
2018. Plaintiffs allege that statements in the proxy improperly highlighted Unum’s “purported
success in managing Unum’s Closed Block operations, and repeated Unum’s inflated fiscal 2017
results.” Appellant Br. 72 (emphasis omitted). We have previously held that the passage of just
six weeks between an allegedly false statement and the disclosure of contrary information weighs
“minimally” in favor of scienter. Dougherty v. Esperion Therapeutics, Inc., 905 F.3d 971, 981
(6th Cir. 2018). So the slightly shorter gap in this case (about three weeks) would also provide
some support for scienter.
Plaintiffs fault the district court for largely overlooking this point. Unum, for its part,
suggests the district court may have overlooked the point because the proxy statement “is not the
subject of any one of [the] myriad misstatement claims” in the complaint. Appellee Br. 50
(emphasis omitted). Unum has a point: Plaintiffs only mention the April 2018 proxy statement
twice in their complaint, and briefly at that. But even assuming the plaintiffs adequately raised the
issue below, and even assuming the plaintiffs could show a true inconsistency between the May
statements and the April proxy, the timing would provide only a moderate boost for scienter.
The second event, even less proximate, was Unum’s filing of its Form 10-K on February
21, 2018. According to plaintiffs, the form contained “false and misleading statements about
Unum’s purported success in achieving rate increases, its reserves’ adequacy, and the
sustainability of its LTC ratio.” Appellant Br. 71–72. But plaintiffs never identify the specific
statements they contend were inaccurate in the filing, nor do they develop more specific arguments
based on the filing. Thus, they have forfeited the issue. See In re Darvocet, Darvon, &
-9- Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
Propoxyphene Prods. Liab. Litig., 756 F.3d 917, 932 (6th Cir. 2014). And in any event, having
been made ten weeks prior to the early May disclosure, these allegedly inaccurate statements offer
at most “minimal[]” support for scienter, see Dougherty, 905 F.3d at 981, if any at all, see City of
Monroe, 399 F.3d at 688 (statements made over sixteen weeks prior to a disclosure are too remote
to support a scienter inference).
5. Finally, the plaintiffs argue that Unum’s executives had a financial incentive to engage
in fraud, and that this bolsters the case for scienter. More specifically, they allege that Unum’s
Chief Executive Officer and Chief Financial Officer were “motivated to engage in [fraud] because
portions of their compensation during the Class Period were specifically tied to the performance,
or perceived performance, of Unum’s LTC business.” R. 37, Pg. ID 398. Plaintiffs base this
allegation on statements from Unum’s compensation committee explaining that it awarded both
executives bonus compensation (in part) for their role in “developing and executing the company’s
[LTC] strategy,” maintaining the block’s “stability,” and ensuring that financial performance was
“within expectations.” Id. at 398–99. The district court held that the bonus compensation did not
favor scienter because the bonus-based claims were too general and too speculative to support a
specific inference of scienter.
Our starting point is Helwig. There we explained that “the self-interested motivation of
defendants in the form of saving their salaries or jobs” tends to support a finding of scienter.
Helwig, 251 F.3d at 552 (the ninth factor). In applying this principle to executive compensation,
we have explained that merely alleging that an executive’s “compensation is directly tied to the
company’s performance” is not enough to bolster an inference of scienter. Dougherty, 905 F.3d
at 981. However, where a bonus is “directly tied” to fraudulent metrics promoted by an executive,
that fact bolsters scienter. Frank v. Dana Corp., 646 F.3d 954, 960, 962 (6th Cir. 2011). At best,
- 10 - Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
this case lies somewhere between Dougherty and Frank, and any link between the allegations and
the bonuses would be quite modest. The bonuses were not tied to a single key metric, but rather
(in part) to the executives’ overall performance in managing the LTC block, which was their job.
And given the paucity of other evidence supporting scienter, the bonus compensation would not
be enough to push plaintiffs’ case for scienter over the finish line.
C.
After weighing this evidence, we conclude that any whiff of scienter is not “at least as
compelling as any opposing inference of nonfraudulent intent.” Tellabs, 551 U.S. at 314. As
Unum argues, the facts in this case are more consistent with a company playing “a constant game
of catch-up than with fraud.” Appellee Br. 55–56 (cleaned up) (quoting Sjunde AP-Fonden v. Gen.
Elec. Co., 417 F. Supp. 3d 379, 399 (S.D.N.Y. 2019)). Unum raised its reserves multiple times
over a short period. Other major insurance companies had to increase their reserves too. When
Unum increased its reserves in 2014, it admitted that it was not immune to future reserve charges.
And it told investors that “if the current environment persists longer than the three- to five-year
time horizon, we will probably have to revisit the assumption again.” R. 54-4, Pg. ID 548.
What’s more, Unum also told investors that if its loss ratio exceeded 90% “for a prolonged
period of time,” it would have to reconsider its “reserve assumptions.” R. 37, Pg. ID 336. It then
disclosed its loss ratio each quarter, readily revealing when that ratio exceeded the 90%
benchmark. And when Unum’s loss ratio exceeded 90% for three straight quarters, the company
did what it had promised to do: It revisited its reserve assumptions. This is “hardly the sort of
behavior one would expect from the perpetrator of securities fraud.” SEC v. Steadman, 967 F.2d
636, 642 (D.C. Cir. 1992).
- 11 - Case No. 20-5710, Pittman, et al. v. Unum Group, et al.
* * *
For all of these reasons, we hold that the plaintiffs failed to adequately allege an inference
of scienter that is “cogent and at least as compelling as any opposing inference of nonfraudulent
intent.” Tellabs, 551 U.S. at 314. We thus affirm the judgment of the district court dismissing
plaintiffs’ complaint for failure to state a claim.
- 12 -