Dairyland Power Cooperative v. United States

United States Court of Appeals
for the Federal Circuit
__________________________

DAIRYLAND POWER COOPERATIVE,
Plaintiff-Cross Appellant,

v.
UNITED STATES,
Defendant-Appellant.
__________________________

2010-5110, -5111
__________________________

Appeal from the United States Court of Federal
Claims in case no. 04-CV-106, Judge Edward J. Damich.
__________________________

Decided: June 24, 2011
__________________________

ROBERT L. SHAPIRO, Hughes Hubbard & Reed, LLP, of
Washington, DC, argued for plaintiff-cross appellant.

HAROLD D. LESTER, JR., Assistant Director, Commer-
cial Litigation Branch, Civil Division, United States
Department of Justice, of Washington, DC, argued for
defendant-appellant. With him on the brief were TONY
WEST, Assistant Attorney General, JEANNE E. DAVIDSON,
Director, ALAN J. LO RE, Assistant Director, and SCOTT
SLATER, Trial Attorney. Of counsel on the brief were
PATRICK B. BRYAN, Trial Attorney, and JANE K. TAYLOR,
DAIRYLAND POWER COOPERATIVE v. US 2

Office of General Counsel, United States Department of
Energy, of Washington, DC.
__________________________

Before RADER, Chief Judge, GAJARSA and PROST, Circuit
Judges.
PROST, Circuit Judge.

This case concerns the Department of Energy’s
(“DOE’s” or “the government’s”) breach of its obligation to
accept spent nuclear fuel from the nation’s nuclear power
utilities. Liability is not at issue. The parties dispute
various aspects of the U.S. Court of Federal Claims’
damages award.

First, the government contends that the trial court
erred in awarding damages based on testimony that
absent breach, the plaintiff would have successfully
bargained its way to the front of DOE’s fuel acceptance
queue and would have transferred away all spent nuclear
fuel in the first year of performance. Relatedly, Dairyland
cross-appeals the amount of damages award, contending
that the trial court erred in reducing the damages
awarded by the cost of purchasing the exchange. Second,
the government argues that the trial court erred in
awarding the plaintiff damages to compensate for various
indirect overhead costs it claims were caused by the
breach. Third, the government contests the trial court’s
award of plaintiff’s investment in an industry consortium
to build a private spent fuel storage facility, particularly
because, the government points out, plaintiff received
significant equity in the venture for its investment. See
generally Dairyland Power Coop. v. United States, 90 Fed.
Cl. 615 (2009) (“Trial Op.”).
3 DAIRYLAND POWER COOPERATIVE v. US

We hold that the Court of Federal Claims did not
commit reversible error in three of these four issues. We
therefore affirm the award of damages based on plaintiff’s
“exchange” model and the award of indirect costs, as well
as the cross-appealed discounting of plaintiff’s damages.
Regarding the plaintiff’s investment in a private venture
to build a spent fuel storage facility, we hold that the
court was required to only award the cost of that
investment to the extent it was made for mitigation, and
not as a speculative venture for profit. We vacate the
award of damages for the investment in the private fuel
storage venture, and we remand for determination of the
extent to which the investment was mitigation and the
extent (if any) to which it was speculation.

I. BACKGROUND

This appeal, like a number of others recently before or
pending with this court, concerns the government’s
liability for damages in connection with its failure to
develop a permanent solution for the storage of spent
nuclear fuel (“SNF”). 1 From 1967 to 1987, Plaintiff
1 See, e.g., Dominion Res., Inc. v. United States,
Nos. 2009-5031, -5032, 2011 WL 1532145 (Fed. Cir. Apr.
25, 2011); Energy Nw. v. United States, No. 2010-5112,
2011 WL 1312306 (Fed. Cir. Apr. 7, 2011); S. Nuclear
Operating Co. v. United States, No. 2008-5020, 2011 WL
832912 (Fed. Cir. Mar. 11, 2011); Carolina Power & Light
Co. v. United States, 573 F.3d 1271 (Fed. Cir. 2009); Pac.
Gas & Elec. Co. v. United States, 536 F.3d 1282 (Fed. Cir.
2008); Yankee Atomic Elec. Co. v. United States, 536 F.3d
1268 (Fed. Cir. 2008); Ind. Mich. Power Co. v. United
States, 422 F.3d 1369 (Fed. Cir. 2005); see also N. States
Power Co. v. United States, No. 2008-5037 (Fed. Cir.
argued Apr. 4, 2011); Pac. Gas & Elec. Co. v. United
States, No. 2010-5123 (Fed. Cir. argued Mar. 10, 2011); S.
Cal. Edison Co. v. United States, No. 2010-5147 (Fed. Cir.
argued Mar. 9, 2011); Sys. Fuels v. United States, No.
DAIRYLAND POWER COOPERATIVE v. US 4

Dairyland Power Cooperative (“Dairyland”) operated a
nuclear power plant in Genoa, Wisconsin called the La
Crosse Boiling Water Reactor. The reactor is no longer
active, but Dairyland maintains 38 metric tons of spent
uranium there in a wet storage pool. The fact that there
is SNF stored on-site prevents Dairyland from
permanently decommissioning the La Crosse plant.

In 1983 Dairyland, along with the nation’s other
operating nuclear utilities, entered into a contract with
the Department of Energy (“DOE”) to address the
question of what to do with SNF. See Standard Contract
for Disposal of Spent Nuclear Fuel and/or High-Level
Radioactive Waste, 10 C.F.R. § 961.11 (1983) (“Standard
Contract”). The Standard Contract obliged DOE to accept
possession of and title to the signatory utilities’ SNF no
later than January 31, 1998. Id. art. II. DOE had a
mandate from Congress to take responsibility for long-
term storage of contract holders’ SNF. See generally
Nuclear Waste Policy Act of 1982, 42 U.S.C. § 10101 et
seq. (2006).

The Standard Contract provided that DOE would
accept a certain amount of SNF from various utilities
each year until all the SNF from all the signatory utilities
had been removed. Standard Contract, art. II. While the
contract did not set forth a detailed schedule for this
removal, it stated generally that acceptance priority
rankings would be assigned based on “the date of
discharge of such material [e.g., SNF] from the civilian

2010-5116 (Fed. Cir. argued Feb. 8, 2011); Vt. Yankee
Nuclear Power v. United States, No. 2011-5033 et al. (Fed.
Cir. docketed Dec. 16, 2010); Yankee Atomic Elec. Co. v.
United States, No. 2011-5020 et al. (Fed. Cir. docketed
Nov. 9, 2010).
5 DAIRYLAND POWER COOPERATIVE v. US

nuclear power reactor.” Id., sec. VI.B.1(a). This became
known as the “oldest fuel first” priority ranking. The
Standard Contract required DOE to issue an annual
capacity report (“ACR”) that would project in a more
detailed fashion DOE’s acceptance of SNF from the
utilities. Id., sec. IV.B.5(b).

The schedules on which DOE would accept spent fuel
from the utilities were known as “delivery commitment
schedules.” See Standard Contract, sec. V.B. The
Standard Contract permitted the utilities to negotiate
with each other to adjust the delivery commitment
schedules proposed by DOE:

E. Exchanges

Purchaser [i.e., the utility] shall have the right to
determine which SNF and/or HLW [high-level
radioactive waste] is delivered to DOE; provided,
however, that Purchaser shall comply with the
requirements of this contract. Purchaser shall
have the right to exchange approved delivery
commitment schedules with parties to other
contracts with DOE for disposal of SNF and/or
HLW; provided, however, that DOE shall, in
advance, have the right to approve or disapprove,
in its sole discretion, any such exchanges. . . .

Id. sec. V.E.

DOE was unable to meet its contractual obligation to
take possession of the utilities’ SNF by January 31, 1998,
and as a result, partially breached the Standard Contract.
Pac. Gas & Elec. Co. v. United States, 536 F.3d 1282,
1284, 1287 (Fed. Cir. 2008). Due to DOE’s breach, Dairy-
land has had to maintain the 38 metric tons of SNF in its
DAIRYLAND POWER COOPERATIVE v. US 6

wet storage pool. Had DOE performed, the parties agree
that, based on the ACR, the last of the SNF at the La
Crosse plant would have been removed in late early 2006.
Trial Op., 90 Fed. Cl. at 615. The cost of maintaining the
SNF, according to Dairyland’s estimate, is about $29.8
million along with approximately $6.1 million in general
overhead costs.

Dairyland also sought a solution for storing SNF off-
site. It became associated with a venture to privately
develop an SNF repository known as Private Fuel
Storage, LLC (“PFS”). PFS was formed by a consortium
of eleven nuclear utilities (Dairyland included) in order to
locate, license, build, and operate such a facility.
Dairyland decided to become a shareholder in PFS as a
means of sharing with other nuclear operators the cost
associated with such a project.

Dairyland structured its participation in PFS as
follows. In 1995, it set up Genoa Fuel Tech, Inc. (“GFT”)
as a wholly-owned subsidiary incorporated in Wisconsin.
This permitted Dairyland to avoid potential legal liability
and unfavorable tax treatment associated with its
investment in PFS. See Trial Op., 90 Fed. Cl. at 647.
Dairyland contributed about $8.7 million into GFT, which
GFT then invested in PFS in exchange for shares.
Dairyland also incurred about $2.3 million in various
other costs to support GFT, most of which were
administrative expenses. Dairyland also allocated about
$1 million in general and administrative overhead costs to
its involvement with GFT and PFS.

Dairyland sued for breach of contract in the Court of
Federal Claims in January 2004. The court entered
summary judgment of liability. Dairyland Power Coop. v.
United States, No. 04-106C, Dkt. #35 (Fed. Cl. Apr. 27,
7 DAIRYLAND POWER COOPERATIVE v. US

2006) (unreported) (citing Energy Nw. v. United States, 69
Fed. Cl. 500 (2006)). The case proceeded to damages.

Shortly before trial, the court denied the government’s
plea for summary judgment that Dairyland could not
recover damages for its investment in PFS. Dairyland
Power Coop. v. United States, 82 Fed. Cl. 379 (July 2,
2008) (“SJ Denial”). It held a bench trial, and awarded
Dairyland about $37.6 million. Trial Op., 90 Fed. Cl. at
618. The award included roughly $16.6 million for
Dairyland’s cost of maintaining SNF on-site from late
1998 to 2006 and roughly $12 million for its investment in
PFS. About $6.1 million of the award was for various
overhead costs associated with Dairyland’s mitigation. 2

Post-trial, the court denied a motion from Dairyland
to reconsider the amount awarded for SNF maintenance
costs (Dairyland had sought over $33 million). Dairyland
Power Coop. v. United States, No. 04-106C, slip op., 2010
WL 637793 (Fed. Cl. Feb. 22, 2010) (unreported) (“Denial
of Reconsid.”). The government timely appealed the
award of SNF maintenance costs, costs of investment in
PFS, and the award of overhead costs generally.

2 The Court of Federal Claims’ discussion of dam-
ages included a discussion of SNF maintenance costs, a
discussion of PFS investment costs, and a discussion of
overhead and general and administrative (G & A) costs.
The sums awarded in the SNF maintenance and PFS
investment categories each included recovery of overhead
and G & A expenses, i.e., some of the $16.6 million
awarded for SNF maintenance was for overhead and G &
A costs, and likewise for the $12 million awarded for PFS
investment. The cited $6.1 million figure for overhead
and G & A costs is thus not a standalone award, but the
sum of the overhead and G & A costs awarded in other
categories. We perceive no double counting in the trial
court’s opinion.
DAIRYLAND POWER COOPERATIVE v. US 8

Dairyland cross-appealed concerning the amount of its
SNF maintenance award. We have jurisdiction over these
appeals from a final judgment of the Court of Federal
Claims under 28 U.S.C. § 1295(a)(3).

II. DISCUSSION

A. Standard of Review

We review the judgments of the Court of Federal
Claims to determine if they are incorrect as a matter of
law or premised on clearly erroneous factual
determinations. Whitney Benefits, Inc. v. United States,
926 F.2d 1169, 1171 (Fed. Cir. 1991). We review that
court’s legal determinations de novo. Dehne v. United
States, 970 F.2d 890, 892 (Fed. Cir. 1992).

B. The Court of Federal Claims Committed No Error in
Adopting Dairyland’s “Exchanges” Model

Several of the issues on appeal concern the trial
court’s award of damages for Dairyland’s cost of SNF
maintenance from the end of 1998 to the close of damages,
December 31, 2006. See Trial Op., 90 Fed. Cl. at 635.

As discussed, there is no dispute that, under the
“oldest fuel first” rubric of the Standard Contract,
Dairyland was not entitled to have the last of its fuel
removed until 2006. Nevertheless, the Court of Federal
Claims awarded damages for Dairyland’s storage costs
from 1999 to 2006 because it credited Dairyland’s
evidence that it would have participated in exchanges to
eliminate all of its SNF in 1998. Trial Op., 90 Fed. Cl. at
627–36. This had the effect of making all Dairyland’s
costs from that time to the end of the damages period
recoverable.
9 DAIRYLAND POWER COOPERATIVE v. US

The court’s holding for Dairyland in this respect was
largely based on Dairyland’s argument that an
“exchanges market” would have existed. Dairyland
presented legal argument and fact and expert testimony
that, had the government not breached, utilities who
believed their delivery commitment schedules were
unacceptably late would have negotiated with those who
held year-one delivery commitments. This “exchanges
market” did not actually emerge because the government
breached, but Dairyland contended that it was a
foreseeable aspect of the non-breach world, pointing
particularly to the Standard Contract’s “Exchanges”
provision. Standard Contract, sec. V.E. Dairyland’s
expert modeled this market, and opined that Dairyland
would have successfully bargained with the other utilities
for year-one acceptance of all Dairyland’s SNF, i.e., DOE
would have removed 100% of Dairyland’s 38 tons of SNF
by the end of 1998. Dairyland’s expert also testified that
Dairyland, as a shutdown reactor with a small amount of
SNF, “would have been willing to pay among the most for
each ton of 1998 acceptance allocations.” Trial Op., 90
Fed. Cl. at 633. The trial court agreed, and so computed
Dairyland’s damages as covering the period from late
1998 to 2006.

On appeal, the government argues that the trial
court’s award was clearly erroneous. The government
cites a number of alleged deficiencies. For example, the
government complains that Dairyland did not identify the
specific utilities it would have obtained year-one delivery
commitment schedules from. The government also argues
that local communities might have pressured utilities
having year-one delivery commitment schedules to use
those schedule slots themselves (and so remove waste
from the area), and that Dairyland’s proof fails to take
this pressure into account. The government also contends
DAIRYLAND POWER COOPERATIVE v. US 10

that Dairyland’s proof ignores the possibility that utilities
having year-one delivery commitment schedules might try
to extract high prices for them. The government argues
that the court’s award fails to reflect such motivations by
the “seller” utilities. And the government argues that
Dairyland did not put forward sufficient evidence of any
pre-litigation intent to press for early removal of SNF
from the La Crosse plant. The government views these
purported deficiencies as undermining Dairyland’s case to
the point that it was clear error for the Court of Federal
Claims to credit Dairyland’s evidence and argument, and
to hold that SNF storage costs from late 1998 through the
end of the damages period were recoverable.

This court reviews the factual findings of the Court of
Federal Claims only for clear error. The trial court’s
opinion discusses in detail the evidence presented by
Dairyland, as well as the government’s arguments in
opposition. The court viewed the government as
essentially pointing out possible problems with
Dairyland’s proof, but without showing how those
problems could be resolved or that they affected the
outcome of the analysis. Rather than undermining
Dairyland’s case, the government’s proffer was deemed
“not particularly helpful to the Court.” Trial Op., 90 Fed.
Cl. at 633. We find no error in this conclusion, as it
appears to have been grounded in proper weighing of the
evidence.

Concerning, for example, the government’s complaint
that Dairyland failed to identify specific utilities with
whom it would contract for accelerated delivery
commitment schedules, the court concluded that, even
assuming some utilities refused to contract, “the
Government did not make any showing that this would
have substantially affected Mr. Graves’s [Dairyland’s
11 DAIRYLAND POWER COOPERATIVE v. US

expert witness] conclusions.” Id. at 634. The government
has pointed to nothing in the record to reveal clear error
in this conclusion.

Concerning the government’s argument that
Dairyland had not shown pre-litigation intent to exchange
for earlier acceptance, the court noted that Dairyland and
the DOE had taken some steps pre-breach in the direction
of setting up a market for exchanges. Id. at 632. The
court noted the government’s complaint and stated that
evidence of actual negotiation “would certainly have
strengthened [Dairyland’s] claim,” but cited other
evidence put forward by Dairyland as “convincing
evidence of its preexisting interest in utilizing the
exchanges process.” Id. Again, the government’s
arguments on appeal do not demonstrate clear error in
the trial court’s reasoning.

As to the government’s contention that local politics
would have prevented utilities possessing year-one
delivery commitment schedules from bargaining, after
review of the record we find no basis for concluding that
such considerations should have barred the trial court’s
findings. The government’s main evidence for this
contention, testimonial suggestions that local politics
would have been a factor in non-breach-world
negotiations over delivery commitment schedules, was
before the court and, in our view, properly weighed. We
therefore defer to the trial court in its determination that
the evidence of Dairyland’s ability to negotiate for year-
one delivery commitment schedules was sufficient to
prove the issue.

In a related vein, the government also contends that
the model proposed by Dairyland was too speculative to
be the basis of a damages award. The government points
DAIRYLAND POWER COOPERATIVE v. US 12

particularly to Dairyland’s claim that, once Dairyland had
completed negotiations for year-one delivery commitment
schedules for all its SNF, DOE would have approved the
transaction. See Standard Contract, sec. V.E. Citing this
court’s precedent, the government contends that this is a
step too far in the modeling of hypothetical behavior, and
argues that there is no certainty at all as to whether DOE
would or would not have approved the transaction.
Appellant Br. 25 (citing San Carlos Irrigation & Drainage
Dist. v. United States, 111 F.3d 1557, 1561–62 (Fed. Cir.
1997)). Certainly the Standard Contract reserves to DOE
the right to disapprove any proposed exchange of delivery
commitment schedules, “in its sole discretion.” Id. The
government further points out that some of the SNF at
the La Crosse plant is “failed fuel,” and argues that DOE
might have been reticent to accelerate its removal. 3

The question of whether Dairyland’s model is or is not
too speculative to be reliable is, again, a fact issue on
which we owe deference to the Court of Federal Claims.
That court’s opinion demonstrates that it considered the
government’s arguments and found them unpersuasive.
Trial Op., 90 Fed. Cl. at 633. The evidence highlighted by
the government on appeal does not demonstrate clear
error. While considerations such as DOE’s discretion to
approve such transactions and worries about the presence
of failed fuel are certainly relevant, they are not
overriding concerns sufficient to make the court’s finding
clearly erroneous.

3 Failed fuel’ means spent nuclear fuel that, while
otherwise meeting the Standard Contract’s specifications,
is contained in defective fuel assemblies (e.g. structural
deformity, damage requiring special handling).” Dairy-
land Power Coop. v. U.S., 79 Fed. Cl. 722, 727 n.5 (2007).
13 DAIRYLAND POWER COOPERATIVE v. US

San Carlos Irrigation is consistent with our
reasoning. There, as here, this court affirmed the trial
judge’s finding on whether certain claimed damages were
too speculative to be awarded. In that case, the trial
judge had denied a plaintiff’s attempt to recover the value
of water it would have received if the government had not
breached and if there had been an excess of water in that
year and if the government had elected to designate part
of that excess “surplus” and convey it to the plaintiff. San
Carlos Irrigation, 111 F.3d at 1561–62. This court
affirmed (albeit on slightly different grounds than the
trial court), and stated, “Too many contingencies—
including, most importantly, the discretion of the agency
to dispose of excess water—exist in the causal chain[.]”
Id. at 1563.

As this court found no error in the denial of damages
in San Carlos Irrigation, we find no error in the damages
award here. If anything, San Carlos Irrigation reminds
us that the question of whether the “causal chain” is
sufficiently well-formed is one of fact, to be made in the
first instance by the trial judge and reversed by this court
only for clear error. As that circumstance is not presented
here, we affirm the trial court’s determination that
Dairyland was entitled to damages for its storage of SNF
for the entire period from 1999 to 2006.

C. The Court of Federal Claims Committed No Error in
Offsetting Dairyland’s SNF Maintenance Award to
Account for the Cost of Exchanges

We next turn to Dairyland’s cross-appeal and examine
the trial court’s limitation of awarded damages.
Dairyland asked the trial court to award it $33,282,048 to
recover its SNF storage costs. Though adopting most of
Dairyland’s methodology for computing SNF storage
DAIRYLAND POWER COOPERATIVE v. US 14

damages, the trial court disagreed that the full amount
should be awarded. Instead, the court awarded Dairyland
precisely half—$16,641,024. Trial Op., 90 Fed. Cl. at 636.

The reason was that although Dairyland’s damages
proof relied on the premise that Dairyland would
negotiate for year-one delivery commitments, the $33.3
million figure did not account for the cost of those
commitments. Id. at 635–36. The court noted that
acquiring year-one delivery commitments would confer
great benefits to Dairyland. It reasoned that the sellers
(utilities holding year-one delivery commitments) would
know their value:

The Court is convinced that all utilities would
have assigned significant value to their
allocations, and would have behaved as
sophisticated and well-advised negotiators.
Buyers would only have induced sellers to part
with their allocations by offering to share the
benefits of such a bargain. Thus, as Dr.
Neuberger [the government’s expert] testified,
“you could have situations [resulting in] any price,
up to and including the willingness to pay
amounts.” Both buyers and sellers would have
come to the negotiating table demanding no less
than a fair share of the benefits of the bargain.

Id. at 635 (second alteration in original) (citation
omitted).

Dairyland argued that if the cost of these allocations
were to be taken into account, it should be about $2
million. Citing trial testimony, the trial court rejected
that figure as “requir[ing] substantial upward
adjustment.” Id. It reasoned that the buyer and the
15 DAIRYLAND POWER COOPERATIVE v. US

seller of the year-one delivery commitment would split the
benefits evenly, and the cost of the commitments would be
exactly half their value to Dairyland. It wrote, “The
Court will offset Dairyland’s SAFSTOR [i.e., SNF
maintenance] damages by that amount [i.e., half] and
award Dairyland the other half, or $16,641,024.” Id. at
636.

On appeal, Dairyland presents two arguments
attacking this conclusion, one contending that there was
legal error and one that there was factual error.

First, Dairyland argues for legal error because it
views the cost of the first-year delivery commitments as a
cost only deferred by the government’s breach, and not
avoided. Dairyland argues that, while a non-breaching
party’s recovery can in some cases be offset to account for
costs it avoided because of the breach, it is inappropriate
to offset for costs that are not avoided but only deferred.
See Carolina Power, 573 F.3d at 1277. It argues that the
government’s breach merely deferred Dairyland’s cost of
accelerating acceptance, reasoning that when the
government ultimately performs, Dairyland will have to
negotiate again for year-one delivery commitments from
the other utilities.

The trial court disagreed, and so do we. See Denial of
Reconsid., slip op. at 3. The cost of accelerating
acceptance in 1998 is not one that will necessarily recur
when and if the government eventually performs. The
former is a past cost that, as has now been established,
would have occurred in the absence of breach. The latter
is an elective future cost that Dairyland might or might
not take up, depending on its situation. The trial court
did not find that Dairyland will in the future be saddled
with this cost, and we see no error there. “Carolina
DAIRYLAND POWER COOPERATIVE v. US 16

Power properly urges caution when speculating about the
future in a case of partial breach—usually, the proper
approach is to wait for those events to actually occur, and
to resist premature conclusions.” Energy Nw., slip op. at
12 (citing Carolina Power, 573 F.3d at 1277). We
therefore hold that there was no error of law in the trial
court’s treatment of Dairyland’s damages request.

Dairyland next argues that there was an error of fact,
and that the $16.6 million reduction in its award was
excessive and based on a misinterpretation of the
evidence. On factual questions, we defer to the trial court
absent clear error. Dairyland points to its expert’s
testimonial opinion that the “exchanges market” was one
with multiple sellers, which would essentially compete
with one another to sell their allocations to Dairyland.
Cross-Appellant Br. 43. It argues that this negotiation
would bring the final price of year-one delivery
commitment schedules to “something near the marginal
bid price” which, in Dairyland’s view, was dramatically
lower than $16.6 million.

In a similar line, Dairyland argues that the court
misinterpreted a statement by its expert that “the cost of
exchanges could have ranged as high as $21.2 million.”
Cross-Appellant Br. 39. Dairyland argues that this
testimony was limited to a damages analysis not
applicable in this case. 4 Dairyland also complains that to
the extent there was any uncertainty about the amount
Dairyland would have had to pay to accelerate SNF

4 Dairyland’s expert had apparently offered opin-
ions concerning economic outcomes if the delivery com-
mitment schedules of the 1991 Annual Capacity Report
were used as a starting point, rather than the 1987 An-
nual Capacity Report that the trial court ultimately used.
Cross-Appellant Br. 39.
17 DAIRYLAND POWER COOPERATIVE v. US

acceptance, that uncertainty should be resolved against
the government.

The trial court’s order denying reconsideration
addresses each of these arguments. Denial of Reconsid.,
slip op. at 3–4. On review, we do not find clear error in
the court’s treatment of any of them. On the question of
whether the presence of multiple sellers would drive down
the price, the court noted that there were also multiple
buyers, whose presence in the market Dairyland’s expert
had not sufficiently addressed. Concerning the testimony
of Dairyland’s expert, the court confirmed that its holding
applied the appropriate ACR and gave proper weight to
the expert’s opinion. And the court rejected Dairyland’s
implication that the government had failed to carry any
burden of proof properly assigned to it. In the court’s
view, the evidence “allowed it to fashion a fair damages
award with ‘reasonable certainty.’” Id. at 3. That award
was the $16.6 million award now on appeal.

“This court affords the Court of Federal Claims wide
discretion in assessing an appropriate quantum of
damages.” Carolina Power, 573 F.3d at 1276. As the
primary finder of fact, part of the trial court’s role is to
fashion a fair assignment of responsibility—even where,
as here, the solution departs from the specific relief
requested by either party. Because we find no error to
justify departing from the outcome the court has reached,
we affirm its reduction of Dairyland’s award.

D. The Court of Federal Claims Committed No Error in
Awarding Overhead and G & A Costs

The government also argues that the trial court erred
in including within its damages award “indirect overhead
DAIRYLAND POWER COOPERATIVE v. US 18

and G & A costs” incurred by Dairyland in support of its
post-breach activities. Appellant Br. 40.

The trial court based its award of overhead damages
on testimony that analyzed Dairyland’s accounts and
determined the portion of its overhead expenses
attributable to the breach. Trial Op., 90 Fed. Cl. at 638.
The court accepted this as “a foundation sufficient to
award overhead and G & A costs to Plaintiff.” Id. The
government disputes this holding. It argues that
Dairyland failed to show that these costs were actually
caused by the government’s breach, and suggests that the
costs—at least some portion of them—would have been
incurred absent breach. It points out that Dairyland
offered the trial court only a post hoc computation of its
overhead and G & A costs (Dairyland does not allocate
such costs during the normal course of business). The
government also recites various cost categories which, in
its view, should not have been included in the
computation, such as “costs associated with Dairyland’s
performance administration, plant operations, integrated
planning, marketing and development, procurement, and
headquarters cafeteria cost centers[,]” “management and
other administrative tasks supporting Dairyland as a
whole” “[and] interest and depreciation on Dairyland’s
headquarters facility.” Appellant Br. 43.

It is well-settled that a plaintiff is entitled to recover
costs actually caused by the defendant’s breach. Energy
Nw., slip op. at 17 (citing Ind. Mich. Power, 422 F.3d at
1373). The government argues that Dairyland failed to
prove causation; the trial court disagreed. In matters
concerning the sufficiency of proof on such a question, we
defer to the judgment of the trial court absent clear error.
Here, the court agreed with Dairyland that these
expenses represented “resources that are consumed
19 DAIRYLAND POWER COOPERATIVE v. US

because of the consumption of breach related activities[.]”
Trial Op., 90 Fed. Cl. at 638 (quoting Dairyland’s expert
witness). In other words, to execute on its mitigation
activities, Dairyland provided its employees a variety of
overhead services. To award a portion of those expenses
attributable to the breach was consistent with this court’s
precedent, and we see no error in the court’s acceptance of
Dairyland’s evidence. See Energy Nw., slip op. at 17–18
(affirming award of overhead expenses where plaintiff
had proved the proper apportionment of those expenses
via expert testimony); Carolina Power, 573 F.3d at 1276–
77.

We therefore agree with the Court of Federal Claims’
overall reasoning in its award of overhead and G & A
expenses, and we affirm its award of those expenses as
part of the costs of SNF storage. 5

E. The Court of Federal Claims Was Required to Award
the Costs of Dairyland’s PFS Investment Only to the
Extent Those Costs Were Taken for Mitigation

Finally, we turn to the government’s contention that
the trial court erred in awarding Dairyland the entire cost
of its investment in PFS, including both money
contributed via GFT into PFS, expenses incurred on
GFT’s behalf for PFS, and (as already discussed) various
overhead and G & A expenses.

5 As discussed infra, we are remanding for further
development the question of how much of the investment
in PFS is recoverable. We expect remand proceedings will
include determination of how much of the overhead and
G & A expenses associated with PFS are recoverable. We
therefore vacate the award of overhead and G & A ex-
penses as to the PFS investment.
DAIRYLAND POWER COOPERATIVE v. US 20

The government does not contend that it was
unreasonable or unforeseeable that Dairyland would seek
interim off-site storage for its SNF. Nor does it argue
that Dairyland was not entitled to collaborate with other
nuclear utilities to aggregate the costs of such storage.
The problem with the trial court awarding Dairyland the
full cost of its participation in PFS, according to the
government, concerns the size and the specific manner of
that participation.

The government presents two arguments.

1. No Error in Taking GFT’s Investment in PFS as
the Baseline for Computing Damages

First, the government points out that the actual
investor in PFS was not Dairyland but its wholly-owned
subsidiary GFT. Dairyland owns no PFS shares in its
own right and has contributed no capital directly to PFS.
The government argues that the trial court improperly
conflated the two entities, Dairyland and GFT. To the
extent a claim exists for investment in PFS, the
government would limit that claim to GFT alone. It
argues that if Dairyland has any claim at all, it is for its
investment in GFT—not GFT’s investment in PFS.
Appellee Br. 27. The government argues that Dairyland
failed to press this claim—and argues that it would not be
recoverable anyway, as it views use of a wholly-owned
intermediary such as GFT as unforeseeable at the time of
contracting.

The trial court rejected this line of argument and so
do we. The record demonstrates—and the government
does not dispute—that GFT had no practical purpose but
to be a conduit for Dairyland’s investment in PFS. See SJ
Denial, 82 Fed. Cl. at 383 (“GFT had no assets other than
21 DAIRYLAND POWER COOPERATIVE v. US

those provided by Dairyland for the sole purpose of
investing in PFS.”). The parties agree that the amount
Dairyland contributed into GFT equals the amount GFT
invested in PFS. See id.; see also Oral Argument at 2:18,
Dairyland Power Coop. v. United States, (No. 2010-5110, -
5111), available at
http://oralarguments.cafc.uscourts.gov/Audiomp3/2010-
5110.MP3 (government conceding that the amounts are
equal). And there is no dispute that GFT was at all times
wholly owned and controlled by Dairyland.

The authority cited by the government is
unpersuasive. Several are cases where individual
shareholders attempted to assert claims properly
belonging to the corporation, and were rebuffed. See, e.g.,
S. Cal. Fed. Sav. & Loan Ass’n v. United States, 422 F.3d
1319, 1328 (Fed. Cir. 2005); S.R. Weinstock & Assocs. v.
United States, 223 Ct. Cl. 633, 680 (Ct. Cl. 1980). Those
cases are not applicable here, both because Dairyland’s
relationship with GFT is far different from that of an
individual shareholder and because Dairyland, not GFT,
has a privity relationship with the government. Nor does
the government’s citation to American Capital Corp. v.
Federal Deposit Insurance Corp., 427 F.3d 859, 864–66
(Fed. Cir. 2006), aid its cause. There, this court rejected a
parent corporation’s attempt to assert a claim properly
belonging to its subsidiary in a Winstar-type case. See
generally United States v. Winstar Corp., 518 U.S. 839
(1996) (holding the government liable for contractual
breach associated with certain financial reform
legislation). In American Capital, again unlike here, it
was the subsidiary that had the contractual relationship
with the government, and this court held that the parent
could not assert the subsidiary’s cause of action. 472 F.3d
at 866–67. Here, it is the parent with the privity
relationship, seeking to recover its own investment,
DAIRYLAND POWER COOPERATIVE v. US 22

computed based on the flow of capital through a specially-
created subsidiary. American Capital is therefore
distinguishable. We see no problem with the trial court’s
method of computing Dairyland’s damages by looking at
the investment made by GFT on Dairyland’s behalf.

2. Investment in PFS is Recoverable Only to the
Extent It Was for Mitigation

The government’s second argument fares better. It
argues that Dairyland’s investment in PFS was more
profit speculation than mitigation, and so should not be
recoverable as a matter of law. The government points
out that PFS was conceived as a for-profit venture
(though it has yet to actually turn a profit). It notes that
Dairyland stands to profit if and when PFS becomes
commercially successful. And it argues that the size of
Dairyland’s investment in PFS far outstrips Dairyland’s
actual need for off-site interim storage.

As already discussed, Dairyland has 38 metric tons of
SNF to store. The government cites testimony that, using
standard storage casks of the type it expects PFS will use,
this would require six casks. By contrast, the government
argues that PFS’s interim storage facility, when built, will
have a storage capacity of 4,000 casks. Appellee Br. 33–
34. The government points to trial testimony that
Dairyland was a 13.5% owner in PFS. From this, the
government reasons that Dairyland had proportional
ownership of 13.5% of PFS’s expected 4,000 storage casks,
about 540 total casks. The government points out that
this amount of storage dramatically exceeds Dairyland’s
storage requirements. From this, the government argues
that the investment in PFS is not recoverable or, if it
were, the trial court should have either performed an
accounting of the value of Dairyland’s PFS stake or
23 DAIRYLAND POWER COOPERATIVE v. US

ordered disgorgement of the shares to avoid unjust
enrichment. The government further notes that this
court has previously held a nuclear utility’s investment in
PFS to be nonrecoverable as speculative and
unforeseeable. Ind. Mich. Power Co. v. United States, 422
F.3d 1369, 1376 (Fed. Cir. 2005).

Dairyland opposes, arguing first that Indiana
Michigan should be limited to its facts. It argues that the
evidence here demonstrates, and the trial court held, that
it was both reasonable and foreseeable for Dairyland to
collaborate with other nuclear utilities to reduce the
overall cost of interim off-site storage. In that light,
Dairyland argues that the exact manner of the
collaboration—in this case, equity participation in PFS—
is not required to be foreseeable. Dairyland further
argues that the ownership structure and division of
revenue in PFS is more complex than the government’s
arithmetic suggests. Finally, Dairyland notes that PFS’s
facility has not yet been built, nor has it ever turned a
profit.

The government is correct that expectation damages
are available to compensate a plaintiff for the cost of
actions taken in mitigation, and not for speculative
ventures. This is an extension of the requirement that
damages are recoverable only to the extent the non-
breaching party can show that the damages were actually
caused by the breach. Ind. Mich. Power, 422 F.3d at
1376; see also Energy Nw., slip op. at 12–13; S. Nuclear,
slip op. at 13.

While we decline to enter into fact-finding, we agree
with the government that the facts of this case urge
caution. The government having raised the specter of a
bounty accruing to Dairyland from its PFS investment,
DAIRYLAND POWER COOPERATIVE v. US 24

Dairyland had the burden to prove how much, if any, of
its PFS investment was speculative as opposed to
mitigation-oriented. The government, of course, was
entitled to contest that proof, and the trial court to
determine which party the evidence best favored.

Faced with these arguments, the Court of Federal
Claims concluded that it was not required to apply this
level of detailed inquiry to the causation analysis. Trial
Op., 90 Fed. Cl. at 651. We disagree as a matter of law
and so vacate the award of damages for the PFS
investment and remand for further development. In so
doing, we emphasize that we draw no conclusions as to
the ultimate outcome of the trial court’s inquiry. It is that
court’s role, not ours, to determine in the first instance
the amount to offset Dairyland’s award of its PFS
investment to account for speculation (if indeed there was
speculation). 6

III. CONCLUSION

For the foregoing reasons, we affirm the Court of
Federal Claims’ award of damages based on Dairyland’s
“exchange” model and its reasoning in awarding overhead
and G & A costs. We also affirm its discounting of
damages for the cost of the year-one delivery commitment
schedules, which Dairyland raised on cross-appeal. We
vacate those portions of the award concerning Dairyland’s
investment in PFS and remand for further proceedings.

6 As noted supra note 5, the Court of Federal
Claims’ determination on remand should include an
assessment of the extent, if any, to which the award of
overhead and G & A expenses associated with the PFS
investment should be offset.
25 DAIRYLAND POWER COOPERATIVE v. US

COSTS

Each party shall bear its own costs.

AFFIRMED-IN-PART, VACATED-IN-PART, AND
REMANDED