United States District Court, W.D. North Carolina, Charlotte Division
WILLIAM L. PENDER, ET AL., Plaintiff,
BANK OF AMERICA CORP., ET AL., Defendant.
C. MULLEN, UNITED STATES DISTRICT JUDGE
MATTER is before the Court following a bench trial held
November 7, 2016 - November 14, 2016. After hearing the
evidence presented at trial and reviewing both parties'
Proposed Findings of Fact and Conclusions of Law (Docs. Nos.
355, 356) the Court finds in favor of the Defendant, Bank of
America Corp. et al, on all issues for the reasons set forth
Overview of the Case
matter arises out of the decision by NationsBank, a company
that subsequently merged with Bank of America (“the
Bank”), to allow its employees to transfer their 401(k)
assets to a cash balance defined benefit plan (“the
Pension Plan”). Because the decade-long procedural
history in this case has been well documented elsewhere, the
Court will recite only the facts relevant to the present
proceeding. See Pender v. Bank of America, 2013 WL
4495153, No. 3:05-cv-00238-GCM (W.D. N.C. Aug. 19, 2013);
see also Pender v. Bank of Am. Corp., 756 F.Supp.2d
694, 696 (W.D. N.C. 2010), aff'd sub nom. McCorkle v.
Bank of Am. Corp., 688 F.3d 164 (4th Cir. 2012).
Fourth Circuit described the Pension Plan as follows:
[Under] [t]he 401(k) Plan[, ] participants' accounts
reflected the actual gains and losses of their
investment options. In other words, the money that 401(k)
Plan participants directed to be invested in particular
investment options was actually invested in those investment
options, and 401(k) Plan participants' accounts reflected
the investment options' net performance.
By contrast, Pension Plan participants' accounts
reflected the hypothetical gains and losses of their
investment options. Although Pension Plan participants
selected investment options, this investment was purely
notional. . . . Instead, the Bank invested Pension Plan
assets in investments of its choosing, periodically crediting
each Pension Plan participant's account with the greater
of (1) the hypothetical performance of the participant's
selected investment option, or (2) the Transfer Guarantee.
Pender v. Bank of America Corp., 788 F.3d 354,
358-359 (4th Cir. 2015) (footnote omitted) (emphasis in
the appearance of a Wall Street Journal article
covering the BAC transfers, the IRS opened its audit of the
Bank's retirement plans on or about July 20, 2000. During
the audit, the Bank and the IRS engaged in a series of
correspondences regarding the Internal Revenue Code's
(“IRC”) requirement of a separate account feature
for any employee 401(k) plan assets that are transferred into
a defined benefit plan such as the BAC plan. In these
correspondences, the Bank's position was that the
separate account feature was not violated if a defined
benefit plan such as the BAC plan provided a benefit not less
than the transferred 401(k) plan benefits, adjusted at a
‘going rate' for periods after the transfer. (Doc.
No. 295-29 at 1-2).
December 9, 2005, the IRS issued its Liability Technical
Advice Memorandum (“TAM”), which concluded that
the transfers of participants' 401(k) assets into the BAC
plan resulted in a loss of the separate account feature
required for defined contribution plans. (Doc. No. 295-5 at
26). The IRS reasoned that, “to preserve the separate
account feature, the separate defined contribution account
must be determined by the investment experience of the
contributions made on the participant's behalf.”
(Doc. No. 295-5 at 25). Thus, according to the IRS, the BAC
plan's hypothetical investment credits failed to preserve
the separate account feature.
ex parte settlement negotiations concerning this
alleged violation3 the IRS submitted that, in order to
restore the separate account feature, the Bank should pay
participants the greater of (a) the original TSA amount plus
BAC trust earnings (actual earnings) or (b) the hypothetical
TSA account balance. (Doc. No. 295-12 at 1). The Bank
disagreed with this restoration method and instead proposed
its “Rescission Plus” method. [See Doc. No.
295-35 at 4]. By this method: (1) the hypothetical balance of
participants' TSAs would be transferred out of the BAC
trust and into individual 401(k) plan accounts; (2) the
balance of the TSA would be maintained as a sub-account
within participants' 401(k) plan accounts; (3) the
restored funds would actually be invested at the direction of
the individual participants; (4) the balance guarantee in the
BAC plan would be maintained to ensure that no participant
received less than his initial TSA balance in the course of
shifting the 401(k) assets out of the BAC trust, into
individual accounts; and (5) a minimum rate of return would
be guaranteed to BAC plan participants. (Doc. No. 295-7 at
a few days following a July 20, 2007 settlement meeting
between the Bank and the IRS, the parties reached an
agreement to settle the ongoing audit. The determination
letters the IRS issued in connection with the Closing
Agreement stated that they related only to the status of the
Bank plans under the IRC and did not amount to a
determination regarding the application of other federal
statutes. (Doc. No. 295-8). Under the settlement, the Bank
paid 10 million dollars to the U.S. Treasury and spent
approximately 10 million dollars applying its Rescission Plus
method to shift participants' 401(k) assets out of the
commingled trust, back into separate accounts. (Doc. No.
295-7 at 3, 9).
Closing Agreement with the IRS, the Bank established a new
special purpose defined contribution plan. Effective April
15, 2009, for participants who still had TSA accounts under
the BAC plan, the Bank implemented steps to transfer BAC
participants' TSA account balances out of the BAC plan,
into individual accounts in the name of each participant.
(Doc. No. 295-7 at 7-10). The transferred TSA balance
reflected a participant's originally transferred 401(k)
balance plus hypothetical investment credits to date. After
this transfer from the BAC plan occurred, a participant's
TSA assets would actually be invested in the options a
participant chose and would receive investment credits based
on the actual performance of those options. (Id.).
addition to delivering on its guarantee against investment
loss under the BAC plan, the Bank amended its BAC plan to
guarantee a minimum rate of return on transferred 401(k)
assets that were invested in the BAC plan. For participants
who had not received their benefit payment before January 1,
2007, this minimum rate of return was 11.6%. According to the
Bank, 11.6% represented the difference between (a) the rate
of return the Bank earned by investing participants'
401(k) assets in the BAC trust between July 1, 1998 and
December 31, 2006 and (b) the average hypothetical return
earned by participants during the same period. (Id.
benefit calculation method for participants who received
their benefit payment before January 1, 2007 was different.
The guarantee against investment loss remained. But the
guaranteed minimum rate of return was not 11.6%. Rather, the
minimum rate of return for such a participant was calculated
by (a) taking the actual return the transferred BAC assets
made between the participant's original 401(k) transfer
date and the date on which the participant received payment
from the BAC plan and (b) comparing that actual return to the
average hypothetical return earned by all participants over
the same period. The participant's guaranteed minimum
rate of return was equal to the positive difference, if any,
between (a) the actual rate of return on participants'
transferred 401(k) assets and (b) the average hypothetical
rate of return for the period when the participant's
401(k) assets were invested in the BAC trust. (Id.
Fourth Circuit Remand
most recent opinion directed at an issue in this case, the
Fourth Circuit held “that Plaintiffs have both
statutory and Article III standing” and remanded this
case for further proceedings. Pender v. Bank of America
Corp., 788 F.3d 354, 358-359 (4th Cir. 2015).
court first held that the Plaintiff has statutory standing to
bring their claim under ERISA § 502(a)(3). Id.
at 363. For Section 502(a)(3) to apply to these facts, the
transfers must have violated a covered ERISA provision and
the Plaintiff must seek “‘appropriate equitable
relief' within the meaning of the statute.”
Id. at 363.
transfers violated a covered ERISA provision, Section
204(g)(1), because the transfers eliminated the defined
contribution plan's separate account feature. This
feature “constitutes an ‘accrued benefit'
that ‘may not be decreased by amendment of the
plan'” under ERISA § 204(g)(1). Id.
the only remaining question was whether Plaintiffs sought
relief that was equitable in nature. The court found that the
Plaintiffs seek the following relief, which constitutes
appropriate equitable relief as used in Section 502(a)(3):
Here, Plaintiffs seek the difference between (1) the actual
investment gains the Bank realized using the assets
transferred to the Pension Plan, and (2) the transferred
assets' hypothetical investment performance, which the
Bank has already paid Pension Plan participants. In other
words, Plaintiffs seek the profit the Bank made using their
assets. This is the hornbook definition of an accounting for
Pender, 788 F.3d at 364.
court explained that an accounting for profits is “a
restitutionary remedy based upon avoiding unjust enrichment,
” which “holds the defendant liable for his
profits, not for damages.” Id. at 364-5.
Because this type of relief is quintessentially equitable,
Plaintiffs could proceed with their claims under §
502(a)(3). Id. at 367.
the court went through Article III standing analysis and
found that the Plaintiffs satisfied all requirements.
Id. at 366.
addition, the court addressed Defendants' argument that
the case was moot because they had restored the separate
account features of Plaintiffs' accounts and because
Plaintiffs had suffered no monetary harm as a result of the
temporary elimination. Id.; see also Id. at
366 (“Requiring a financial loss for disgorgement
claims would effectively ensure that wrongdoers could profit
from their unlawful acts as long as the wronged party suffers
no financial loss. We reject that notion.”). The panel
The Bank rightly notes that its closing agreement with the
IRS restored Plaintiffs' separate account feature. That
restoration, however, did not moot the case. Plaintiffs
contend that the Bank retained a profit, even after it
restored the separate account feature to Plaintiffs and paid
a $10 million fine to the IRS. Defendants do not rebut this
argument, noting only that there has been no discovery to
this effect. If an accounting ultimately shows that the Bank
retained no profit, the case may well then become moot.
Id. at 368.
Fourth Circuit vacated this Court's grant of summary
judgment “based on its erroneous standing
determination” and remanded for further proceedings
without additional instructions on how the required
accounting for profits should be calculated. See Id.
March 10, 2016 this Court issued an order on how best to
implement the instructions set out by the Fourth Circuit: the
analysis of whether or not the Bank retained a profit must be
conducted in the aggregate. See Doc. 347. The Court
concluded that the Fourth Circuit decision did not call for
“60, 000 separate and distinct”
account-by-account examinations of “the profits or
losses derived from each separate transaction.” Doc.
347 at 8-9. The Court also rejected Plaintiffs' proposal
to count “gains” and not “losses, ”
holding that there was “no basis for finding that a
subset of the Plaintiff class is equitably entitled” to
a temporary surplus generated using their assets, when
“all members of the class suffered the same injury -
the temporary loss of their separate account feature-and
received all of their promised benefits.” Id.
at 7. The order set a bench trial to be held on the issue of
whether, after it restored the separate account feature and
paid a $10 million fine to the IRS, the Bank nevertheless
profited from its transfer strategy. Id.
party was allowed to present two experts to testify on their
behalf. The Court carefully considered the four experts'
testimony, the documents admitted into evidence, and the
parties' respective Findings of Fact and Conclusions of
Law. For the reasons indicated herein the Court finds that
the Defendants established that it did not retain a profit
from the Pension Plan, even after it restored the separate
account feature to Plaintiffs and paid a $10 million fine to
holding the bench trial and reviewing the parties'
arguments and relevant case law, the Court finds in favor of
Bank of America Corp et al on all counts.
Defendants' Experts' Testimony was More Credible than
Plaintiff's Experts' Testimony
the Plaintiffs and the Defendants experts have presented a
coherent and facially plausible story for their parties. The
Plaintiffs' expert, Lawrence Deutsch, argues that since
the Plaintiffs' transferred assets were comingled with
other assets in the Pension Plan, the transferred assets
should be considered undifferentiated Plan assets. So, all
investment returns on all assets in the Plan should be used
to calculate the profit and the transferred funds would be
assigned a pro rata share of those returns. Mr. Deutsch's
calculation finds the investment gains retained by the Bank
from the transferred assets are $379 million. The
Plaintiffs' other expert Clark Maxam argues that the Bank
must disgorge that greater of the aggregate gains the Bank
still retains from the transferred accounts or the market
interest the Bank hypothetically would have paid to receive a
loan of the transferred assets. Dr. Maxam calculated the
retained interest savings of the transferred assets to be
Defendants' expert Russell Wermers opines that through
the use of accepted investment return benchmarks, he can
assess whether the Plan's Investment Strategy could have
produced a profit. Dr. Wermers found that during the transfer
period equities significantly declined, while fixed income
investments substantially increased. Since the Plan's
Investment Strategy caused the Plan to invest more heavily in
equities than the hypothetical investments made by
participants, Dr. Wermers found that the transfer strategy
did not result in a profit for the Bank. The Defendant's
other expert, David Andreasen, argues that the returns of the
Investment Strategy can be calculated by tracking the returns
for each month for the Equity Hedge strategy and the
overweighting equity strategy. The participants'
hypothetical equity and fixed income investments were also
tracked each month and Mr. Andreasen argues that whether the
Pension Plan retained any profit from the transferred assets
should be calculated by comparing these two values. Mr.
Anderson's calculations show a loss of $272 million as a
result of the transfer assets.
four experts' testimony are not contradicted by objective
evidence and so this Court is in the position where it must
make a determination to credit the testimony of either the
Plaintiffs' or Defendants' experts based on the
Court's understanding of and belief in what was said at
trial. On that basis, the Court finds that the
Defendants' experts provided evidence at trial that is
more credible than the testimony provided by the
Findings of Fact
reviewed and carefully considered the evidence and arguments
presented at trial, the Court makes the following findings of
core of the Plan's Investment Strategy was to invest the
assets used to fund the TSAs more heavily in equities than
participants invested their hypothetical accounts, on the
theory that equities would be expected to outperform fixed
income options over the long term. The Plan did this by
matching or “hedging” participant equity
investments with Plan equity investments and investing
approximately 60% of participant fixed income investments in
fact, the Investment Strategy failed. During the transfer
period, equity markets experienced historic downturns, and
the Plan's greater allocation to equity investments
caused its investment returns to be significantly less than
the aggregate returns credited to participant accounts. The
Court finds, having observed the testimony of the witnesses,
assessed their credibility, and considered the entirety of
the evidence, that Defendants did not retain a profit as a
result of the transfer. To the contrary, the evidence
persuasively shows that the Plan experienced a net investment
loss as a result of the Investment Strategy applicable to the
TSAs, because it was more weighted in equities during a
period when equity markets significantly underperformed fixed
Defendants have provided calculations of the amount of the
Plan's losses resulting from the transfer including, as
set forth in more detail below, calculations based on
contemporaneous records of investment returns maintained in
the ordinary course of business. Those calculations show that
the Plan incurred an investment loss of approximately $143
million attributable to the transfer.
addition, Defendants paid participants approximately $108
million in Transfer Guarantee payments and made more than $21
million in payments to the IRS and to restore separate
accounts in the TSA Plan as required by the Closing
Agreement. In total, Defendants' calculation of the
losses attributable to the transfer exceeds $270 million.
Plaintiffs have criticized various aspects of the investment
loss calculations presented by Defendants' expert, Dr.
Russell Wermers, and David Andreasen, a senior Vice President
of the Bank responsible for Pension Plan investments. But
Plaintiffs have failed to quantify or credibly explain how
their criticisms would turn a failed Investment Strategy from
a loss into a profit. What matters here is whether Defendants
retained a profit as a result of the transfer; the precise
amount would be relevant only if the Court found that the
evidence demonstrates a profit. The Court finds the evidence
does not support a conclusion that Defendants realized a
Plaintiffs' expert witnesses presented two alternative
analyses allegedly showing a profit. Plaintiffs' expert
Lawrence Deutsch opined that the Plan actually profited by
more than $454 million. The difference between Mr.
Deutsch's analysis and Defendants' analysis turns on
two major points. First, Mr. Deutsch calculates profit based
on all investment returns earned on all assets in the
Plan-including investment returns on legacy fixed-benefit
obligations that pre-date the transfer. For the reasons
discussed below, the Court rejects this approach, finding it
to be a less accurate and reliable means of measuring whether
there was any profit retained from the transfer because it
captures investment returns that the Plan would have earned
even if the transfer did not occur. In addition, as a matter
of equity, the Court finds that this proposed methodology is
inappropriate and inferior to calculating profit based on the
actual Investment Strategy utilized with respect to the TSAs.
This methodology also appears to be at odds with the type of
assessment contemplated by the Fourth Circuit and the method
for determining investment return “spread” that
the IRS approved.
Second, Mr. Deutsch's profit measure rests on an
individual-by-individual calculation of participants'
hypothetical returns and excludes from the calculation
individual participants for whom there was a negative
“spread”-i.e., whose hypothetical
investments outperformed the Plan, causing the Plan to incur
losses. The Court has already rejected this
approach as inconsistent with the Fourth Circuit's
ruling. But even if it were not, the Court finds that this
approach to calculating “profit” would not serve
the purposes of equity here and would instead be punitive in
nature, particularly given that participants have already
received all benefits which they were promised.
Plaintiffs' second expert, Dr. Clark Maxam, offers an
entirely separate theory. Dr. Maxam opines that the Plan was
unjustly enriched by the imputed “use value” of
the transferred assets, which Dr. Maxam states was “at
least $346 million.” As set forth below, Dr.
Maxam's imputed use value theory is not a reliable or
appropriate way of measuring “profit, ”
particularly under the facts and circumstances of this case.
In addition, the Court finds that such a methodology would
not serve the purposes of equity as compared to the method
proposed by Defendants, which focuses on actual profit. For
that and other reasons, the Court does not accept the
measurement of unjust enrichment based on use value.
sum, the evidence demonstrates that the Plan suffered a loss
and that Plaintiffs' various analyses are flawed.
Accordingly, the Court finds that Defendants did not retain a
profit as a result of the transfer. Having found that there
was no profit, the Court need not make findings regarding the
other equitable defenses that Defendants have raised in this
litigation (which were not the subject of this
Evidence Demonstrates That There Was No Retained Profit
Plan's Heavy Concentration In Equities During A Time When
Equities Underperformed Fixed Income Investments Compels A
Finding That The Plan Experienced A Loss
Defendants' expert, Dr. Wermers, opined that the Plan did
not retain a profit, and he presented various explanations
and analyses in support of that conclusion. The Court finds
Dr. Wermers' testimony credible and his analyses to be
persuasive and helpful.
Wermers is a Professor of Finance at the Smith School of
Business at the University of Maryland and Director of the
Center for Financial Policy at the University of Maryland.
Dr. Wermers' expertise includes quantitative equity
strategies, investment manager performance, and measuring
performance of actively managed pension plan sub-portfolios.
He has taught courses on Quantitative Equity Portfolio
Management, Corporate Finance Theory, Security Analysis, and
Investment Theory, among other topics. He has written
academic papers on pension plans that have focused on
benchmarking and measuring the performance of their actively
managed sub-portfolios in different asset classes, and
co-authored a scientific textbook on how to measure the
performance of portfolio managers. He has provided advisory
services to the Quantitative Strategies group at Goldman
Sachs Asset Management, as well as the Office of Financial
Research of the United States Treasury Department.
Wermers opined that the outcome of the transfer strategy can
be determined even without directly analyzing the specific
hypothetical investment elections participants made and the
specific investments the Plan made. Specifically, Dr. Wermers
assessed whether the Plan's Investment Strategy applied
to the TSAs could reasonably have produced a profit in light
of the performance of accepted investment return benchmarks
during the relevant period. Using the investment return
benchmarks specified in the Plan's Investment Policy
Statements, Dr. Wermers evaluated: the performance of the
Plan's domestic equity investments by looking to the
Russell 3000 Index; the performance of the Plan's
international equity investments by looking to the MSCI EAFE
Index; and the performance of the Plan's fixed income
investments by looking to the Lehman Brothers Aggregate Bond
Wermers observed that during the transfer period, equities
experienced significant declines, while fixed income
investments experienced substantial increases. Indeed,
between July 1, 1998 through March 31, 2009, the difference
between equities and fixed income investments was very
significant. The Russell 3000 (Equity) Index
declined 11.2%, and the MSCI EAFE (Equity) Index
declined 3.5%, while the Lehman Brothers Aggregate
Bond Index (a fixed income index) increased 81.1%
during the same period.
Because the Plan's Investment Strategy intentionally
caused the Plan to invest more heavily in equities than the
hypothetical investments made by participants, given the
relative performance of equity and fixed income investments
during the transfer period, it is clear that the transfer
strategy resulted in a loss to the Plan, not a profit, as Dr.
Wermers testified and the Court finds.
Wermers further confirmed this conclusion by constructing a
model based on the Plan's Investment Strategy and
participant-directed accounts. The model hedged participant
equity investments and invested participant fixed income
investments in a 60%/40% mix of equity and fixed income
investments, in accordance with the Plan's Investment
Policy Statements. Dr. Wermers then used the model to test
the Plan's net investment performance for every possible
participant-directed allocation, from 100% equity to 100%
fixed income, based on the performance of the Plan's
investment benchmarks during the transfer period.
Using this approach, Dr. Wermers concluded that there is no
participant-directed allocation that would have resulted in
an aggregate investment gain for the Plan. For every possible
hypothetical investment combination that participants
directed, the Plan's Investment Strategy called for the
Plan to invest in, at a minimum, a share of equities
equal to participant hypothetical equity
investments, and most often a higher share of
equities (if participants allocated any portion of their
hypothetical investments to fixed income options).
Because equities substantially underperformed fixed income
investments during the relevant period, and because the
Plan's investments were more heavily concentrated in
equities than were participant-directed hypothetical
investments, it is unsurprising that the Plan would have
experienced a net loss.
Plan's Records Show A Substantial Loss From The Transfer
Defendants' analysis, based on contemporaneous records
that the Plan maintained in the ordinary course of business,
produced calculations and estimates consistent with the
conclusion that the Plan realized a loss on the transfer as a
result of the underperformance of equities during the
order to track the performance of the Investment Strategy,
the Plan, through its investment consultant, Callan
Associates, measured returns each month separately for (a)
its Equity Hedge strategy and (b) its 60/40 strategy. The
Plan's investments, including investments made pursuant
to the Equity Hedge strategy, were not tracked and investment
returns were not measured on an individual
Plan also tracked each month, on an aggregate basis,
participants' hypothetical equity and fixed income
investments. These were recorded in monthly “trial
balance” documents, which are documents that reflect
Plan liabilities, including changes in such liabilities, and
which were maintained by the Plan's recordkeeper in the
ordinary course of business.
These contemporaneous records were compiled into the Return
Data Spreadsheet, which was created and admitted as a summary
for this litigation. The Return Data Spreadsheet used these
primary inputs to calculate the monthly and cumulative
percentage returns of the Pension Plan's Investment
Strategy, on the one hand, and participants' hypothetical
investment performance, on the other. The Plan's net
investment performance was determined by subtracting
aggregate participant percentage returns (i.e.,
liabilities) from the Plan's percentage returns
attributable to the Investment Strategy (i.e.,
assets). The net investment returns on a percentage basis
were then multiplied by the aggregate balance of the TSAs
Using the data in the Return Data Spreadsheet, the
Bank-through Mr. Andreasen, with assistance from Callan
Associates-calculated the cumulative totals in the Return
Data Spreadsheet. This calculation shows that the Plan
experienced a cumulative net investment loss on the
Investment Strategy related to TSAs.
Beyond this investment loss, the Bank also accounted for
other payments associated with the transfer. It subtracted
the total amount of Transfer Guarantee payments made by the
Pension Plan to participants. It subtracted the total amount
of the additional Transfer Guarantee payments made by the
Pension Plan to participants pursuant to the Closing
Agreement. It subtracted the $10 million IRS payment pursuant
to the Closing Agreement. And finally, it subtracted expenses
for the restoration of separate accounts as required by the
Closing Agreement. Combining these figures, the Bank's
calculations showed a cumulative loss of approximately $272
million resulting from the transfer.
discussed below, Plaintiffs dispute the Plan's cumulative
investment loss, as reflected in Row A. There is no dispute,
however, as to the accuracy of the figures contained in Rows
B through E, or that these payments were actually made.
Instead, Plaintiffs argue that Rows B through E ...