United States Court of Appeals
for the Federal Circuit
______________________
AGILITY DEFENSE & GOVERNMENT SERVICES,
INC.,
Plaintiff-Appellant
v.
UNITED STATES,
Defendant-Appellee
______________________
2016-1068
______________________
Appeal from the United States Court of Federal
Claims in Nos. 1:13-cv-00055-TCW, 1:13-cv-00097-TCW,
Judge Thomas C. Wheeler.
______________________
Decided: February 6, 2017
______________________
WALTER BRAD ENGLISH, Maynard, Cooper & Gale,
P.C., Huntsville, AL, argued for plaintiff-appellant. Also
represented by EMILY J. CHANCEY, JON DAVIDSON LEVIN,
JOHN ANDREW WATSON, III.
MICHAEL DUANE AUSTIN, Commercial Litigation
Branch, Civil Division, United States Department of
Justice, Washington, DC, argued for defendant-appellee.
Also represented by BENJAMIN C. MIZER, ROBERT E.
KIRSCHMAN, JR., KENNETH M. DINTZER.
______________________
2 AGILITY DEFENSE v. US
Before MOORE, WALLACH, and TARANTO, Circuit Judges.
MOORE, Circuit Judge.
Agility Defense (“Agility”) appeals from the Court of
Federal Claims (“Claims Court”)’s denial of its claim for
an equitable adjustment arising out of its fixed price
indefinite delivery contract with the Defense Logistics
Agency (“DLA”)’s Defense Reutilization and Marketing
Service (“DRMS”). For the reasons discussed below, we
reverse and remand.
BACKGROUND
DLA is an agency of the United States Department of
Defense that provides supplies for the military. DRMS is
a primary level field activity of DLA that disposes of
surplus military property at Defense Reutilization and
Marketing Offices (“DRMOs”) after the military departs
an area of operations. Property that cannot be reutilized
is demilitarized and/or reduced to scrap. Property re-
duced to scrap can be sold on the market.
Historically, the government operated all DRMOs, but
in 2006, DLA’s Director determined that DRMS could not
sustain its workload unless it brought in outside contrac-
tors. DRMS issued a Request for Proposal (“RFP”) in
January 2007. The RFP sought performance of DRMO
activities for up to five years. Three offerors responded to
DRMS’s RFP.
During solicitation, DRMS issued several amend-
ments relevant to anticipated workload and costs. In
Amendment 002 on February 26, 2007, in response to a
request for “workload history and projection by category
and location,” DRMS stated: “Workload history and
current inventory levels can be found at
http://www.drms.dla.mil/newproc/index.html and link to
‘DRMS Information for Southwest/Central Asia.’ Addi-
tional workload data will be provided via amendment.
AGILITY DEFENSE v. US 3
The Government does not have workload projections.”
J.A. 810. The referenced website showed DRMS’s histori-
cal workload by line item and scrap weight. Line items
are the number of military property items received at
each DRMO for processing. Scrap weight is the amount of
scrap processed at each DRMO. DRMS updated its
website approximately biweekly to reflect the line items
received, scrap weight, and scrap sales during the prior
weeks.
In Amendment 004 on June 20, 2007, DRMS respond-
ed to a request for an estimate on workload, stating this
time, “[w]e anticipate an increase in property turn-ins.”
J.A. 945. Amendment 004 added clause H.19, titled
“DRMO Workload Changes,” which contemplated that
“the contractor may experience significant workload
increases or decreases” and outlined a process for the
contractor to “renegotiate the price” if workload increased.
J.A. 836–37. As originally drafted, to warrant a pricing
adjustment under clause H.19, the contractor had to
experience an increased workload 150% above the work-
load it experienced the previous three months. Amend-
ment 004 also added that the contractor to whom the
contract is awarded may sell any scrap, and the contrac-
tor “is entitled to all sales proceeds” from the scrap sales
to “offset some of the costs incurred in performing this
contract.” J.A. 893–94.
On July 24, 2007, DRMS issued Amendment 007. In
response to an offeror’s request for an estimate of scrap
sales, DRMS directed offerors to an attachment projecting
scrap quantities for the duration of the contract (hereinaf-
ter, the “Amendment 007 Chart”). The Amendment 007
Chart projected a stable workload for the first two years
and then “workload declines” for option years three
through five, down 75%, 50%, and 30%, respectively.
J.A 990–91. With Amendment 007, DRMS specified that
contractors would keep their scrap proceeds “without any
type of reduction in payments,” asking contractors to
4 AGILITY DEFENSE v. US
describe their anticipated proceeds “to demonstrate the
Government received consideration for providing the
scrap.” J.A. 1010–11.
Agility submitted its initial proposal on August 2,
2007, reflecting a $20,342,608 offset for expected scrap
revenues during the life of the contract. After receiving
final proposed revisions from Agility on September 24,
2007, on November 29, 2007, DRMS awarded its first-ever
contract to Agility to operate six DRMOs for one base year
with four option years at a fixed price of $45,233,914.92
per year. The other two offerors proposed prices well
above Agility’s, at $68,394,500.47 and $71,507,029.78,
respectively.
In early 2008, DRMS issued its first Task Orders,
which incorporated a workload baseline dated August 4,
2007 for each DRMO (hereinafter, “the Baseline Data”).
DRMS retrieved the Baseline Data from the same website
it referred offerors to in Amendment 002 to view DRMS’s
historical workload data. The Baseline Data detailed the
received line items and scrap weight during the periods
July 13 to July 19 and July 20 to July 26, 2007. The first
Task Order requested work at the DRMO in Arifjan,
Kuwait, with a period of performance from March 3, 2008
to March 2, 2009.
Upon commencing work in Arifjan, the largest of the
six DRMOs, Agility immediately fell behind. It inherited
a backlog of approximately 70,000 line items, which when
compared to the Baseline Data would have equated to the
line items received over approximately 30 weeks. From
the start of Agility’s performance at Arifjan, the volume of
line items received at Arifjan was also greater than
Agility anticipated. Over the next several months, Agility
began performance at the other five DRMOs, where it also
encountered backlogs at each location other than the
DRMO in Speicher, Iraq. In short, the workload from the
outset was substantially higher than predicted. After
AGILITY DEFENSE v. US 5
receiving a June 2008 letter from DRMS expressing its
concerns, Agility stated it would increase staffing at the
DRMOs “by more than 50% at no additional cost to the
government.” J.A. 1693.
It was around this time that Agility requested clarifi-
cation from DRMS regarding when it could invoke
clause H.19 to request compensation for its increased
workload. The parties disputed whether clause H.19’s
requirement that workload must increase “by more than
150% above the average workload at the DRMO location
for the preceding three (3) consecutive months,” J.A. 836,
permitted Agility to compare its workload to the Baseline
Data or required Agility to compare its workload to what
it experienced upon beginning performance. Agility
explained that if it was required to carry an increased
workload for several months before initiating its request
for increased compensation for additional staffing, it
would be overwhelmed and unable to meet DRMS’s needs.
DRMS expressed that Agility could only invoke
clause H.19 if its workload exceeded the average work it
experienced the three preceding months, and opined that
Agility had not met the requirements of clause H.19.
DRMS argued that clause H.19 only allowed a contractor
to ask for an increase if the workload was originally low
and then increased by 150%. It did not, according to
DRMS, allow the contractor to ask for an increase if the
workload was from the outset 150% or more higher than
predicted.
After months of discussion, DRMS and Agility agreed
to modify clause H.19 in March 2009. Instead of requir-
ing the parties to react to a surge, the modification per-
mitted a pricing adjustment if DRMS or Agility
anticipated “an average monthly workload increase of
scrap or line items at any DRMO location by more than
25% above the monthly average of [fiscal year 2008] scrap
or line items received . . . .” J.A. 1621. As amended,
clause H.19 also required Agility to maintain its current
6 AGILITY DEFENSE v. US
level of staffing. Agility never submitted a formal request
for costs under either the original or amended
clause H.19.
The parties terminated their contract for convenience
in June 2010. Agility thereafter requested funding for its
additional costs associated with performance with the
contract. It submitted two claims for increased costs to
the contracting officer, claiming DRMS provided inaccu-
rate workload estimates during solicitation. Agility’s
claims requested $4,359,071.79 covering the period before
the parties modified clause H.19 and $1,602,148.67 cover-
ing the period after the parties modified clause H.19. The
contracting officer awarded Agility only $236,363.93 for
its first claim and nothing for the second, determining
Agility could not recover the remainder because it had not
satisfied the requirements of clause H.19 and noting that
Agility received an offset from its scrap sales. Agility
pursued its claims in the Claims Court. 1
The Claims Court found that, with the exception of
the DRMO in Speicher, Agility experienced workloads
“much greater” than the Baseline Data during the base
year of performance:
DRMO Annualized Actual Percent of
Baseline Workload Annualized
(received (received Baseline
line items) line items)
Speicher 12,768 9,561 74.9%
Victory 13,992 21,899 156.5%
1 Agility pursued three theories of recovery in the
Claims Court and on appeal to this court: (1) constructive
change of contract; (2) negligent estimate; and (3) breach
of warranty of reasonable accuracy. Because Agility’s
theories of recovery are alternative to one another, we
reach only Agility’s claim of negligent estimate.
AGILITY DEFENSE v. US 7
Al Asad 14,616 24,392 166.9%
Anaconda 19,994 71,653 358.4%
Arifjan 109,560 242,401 221.3%
Bagram 6,480 15,364 237.1%
However, it denied Agility’s claims, holding that DRMS’s
conduct was acceptable because it provided Agility with
reasonably available historical data. The Claims Court
did not reach what impact, if any, clause H.19 had on
Agility’s claims. It found that Agility provided “no cause-
and-effect links to isolate its damages” and referenced the
revenue Agility received from its scrap sales to hold the
equities did not weigh in Agility’s favor. J.A. 16–18.
Agility timely appealed. We have jurisdiction under 28
U.S.C. § 1295(a)(3).
DISCUSSION
We review the Claims Court’s conclusions of law de
novo and its findings of fact for clear error. Mass. Bay
Transp. Auth. v. United States, 254 F.3d 1367, 1372 (Fed.
Cir. 2001). Contract interpretation is a matter of law we
review de novo. Id.
A contractor can recover damages from the govern-
ment for increased costs it incurred in performing a
contract under a negligent estimate theory, which re-
quires the contractor to show by preponderant evidence
that the government’s estimates were “inadequately or
negligently prepared, not in good faith, or grossly or
unreasonably inadequate at the time the estimate was
made.” Medart, Inc. v. Austin, 967 F.2d 579, 581 (Fed.
Cir. 1992). The parties agree that DRMS’s contract with
Agility is a requirements contract. In a requirements
contract, Federal Acquisition Regulation (“FAR”) 16.503
requires DRMS to provide offerors with a realistic esti-
mate of workload. 48 C.F.R. § 16.503. FAR 16.503(a)(1)
reads:
8 AGILITY DEFENSE v. US
For the information of offerors and contractors,
the contracting officer shall state a realistic esti-
mated total quantity in the solicitation and result-
ing contract. This estimate is not a representation
to an offeror or contractor that the estimated
quantity will be required or ordered, or that condi-
tions affecting requirements will be stable or
normal. The contracting officer may obtain the es-
timate from records of previous requirements and
consumption, or by other means, and should base
the estimate on the most current information
available.
The Claims Court rejected Agility’s claim of negligent
estimate, finding, “[r]ather than carelessly form[] esti-
mates by asking DRMOs to guess their upcoming needs,
DRMS provided objective, historical workload data from
which the offerors could extrapolate future needs.”
J.A. 15–16. It noted that DRMS informed offerors that
property turn-ins would increase, and found “the offerors
in this case were well aware of volume variations in the
processing of surplus property.” J.A. 16. It cited Medart,
where we applied FAR 16.503 to a scenario in which a
contractor sought reimbursement where the government’s
estimated needs varied significantly from those actually
required. 967 F.2d at 580. We held that the government
did not negligently estimate its needs when it provided
the contractor with historical data from the prior year.
Id. at 581–82. Relying on Medart, the Claims Court found
that DRMS “used reasonably available historical data and
did not negligently estimate its needs.” J.A. 17.
We hold that these findings are clearly erroneous for
two primary reasons. First, the Claims Court ignored
that DRMS did not only provide historical data; it also
estimated its requirements via the Amendment 007
Chart. Second, the Claims Court failed to address evi-
dence indicating that DRMS’s historical data was not “the
AGILITY DEFENSE v. US 9
most current information available.” See
FAR 16.503(a)(1).
First, the Amendment 007 Chart provided by DRMS
is itself an estimate of projected requirements. The RFP
expressly states that “[t]he scope of this contractual effort
includes all tasks DRMS performs in support of the
Department of Defense mission.” J.A. 752. One such task
requires the contractor to “receive, segregate, store and
dispose of scrap material and items downgraded to scrap.”
J.A. 779. By “providing quantity estimates of scrap
commodities” for “the purpose of assisting offerors in
preparing their proposals,” DRMS provided an estimate of
its requirements for processing scrap. See J.A. 1003. The
Amendment 007 Chart is also relevant to the projected
amount of property turn-ins or line items. As noted by
the government, because scrap is created from processing
the military property, the weight of scrap would be ex-
pected to correlate with the number of line items. See
Appellee’s Br. 32 (“As an intuitive matter, if Agility
processed more property than anticipated, corresponding-
ly its [scrap] sales proceeds increased.”). DRMS’s con-
tracting officer likewise testified that he knew of no way
to project scrap weight without projecting the amount of
property turn-ins. Thus, by projecting stable and then
declining scrap weight in the Amendment 007 Chart,
DRMS estimated that property turn-ins would, to some
extent, remain constant and then decline. It was clear
error for the Claims Court not to treat the Amend-
ment 007 Chart as an estimate.
Second, the fact that DRMS “provided objective, his-
torical workload data” does not end the inquiry as to
whether it provided a realistic estimate. See J.A. 15–16.
Medart does not hold, and we do not hold now, that
providing an offeror with historical data is reasonable per
se. See J.A. 14 (citing Medart, 967 F.2d at 582). In
Medart, we recognized that FAR 16.503 explicitly states
that the government “may obtain the estimate from
10 AGILITY DEFENSE v. US
records of previous requirements.” 967 F.2d at 582. But
an important distinction between Medart and the case
before us is that in Medart, the contractor did not present
evidence that “the most current information available” to
the government was something other than its historical
requirements. See FAR 16.503(a)(1). Instead, the con-
tractor’s arguments in that case rested on information it
contended the government could have uncovered to devel-
op a more accurate estimate. See Medart, 967 F.2d
at 581–82. We rejected that argument, holding the gov-
ernment “need not search for or create additional infor-
mation” and need only use “information that was
reasonably available.” Id. at 582.
Here, unlike Medart, Agility presented evidence that
DRMS possessed information regarding its anticipated
requirements above and beyond its historical require-
ments. Agility presented a memorandum dated Novem-
ber 16, 2007, before DRMS awarded Agility its contract,
indicating DRMS was aware of planned troop movement
and a “surge of equipment and material that will be
turned over to DRMS as units depart.” J.A. 1303–04.
The memorandum explicitly states the anticipated surge
is “[o]ne of the key reasons the contract was contemplat-
ed” and concedes “DRMS is not staffed or equipped to
handle this requirement.” J.A. 1304. In fact, DRMS
considered this very information in connection with its
decision to award the contract to Agility. Id.; see also
J.A. 1179 (evaluating the sufficiency of Agility’s proposal
as compared to its “projected workload”).
Because DRMS anticipated increased workload, simp-
ly providing offerors with historical workload was not “the
most current information available” sufficient to provide a
realistic estimate under FAR 16.503. DRMS should have
based its estimate on its anticipated “surge” in workload.
Although DRMS informed offerors that it “anticipate[d]
an increase in property turn-ins,” it did so on June 20,
2007, J.A. 945, before it seemingly changed course on
AGILITY DEFENSE v. US 11
July 24, 2007 with its updated estimate in the Amend-
ment 007 Chart, projecting stable workload for two years
followed by “workload declines.” J.A. 980, 990–91. DRMS
was not obligated to guarantee the accuracy of its esti-
mates or perfectly forecast its requirements, and Agility
ultimately bore the risk associated with any variance in
workload from a realistic estimate. See Medart, 967 F.2d
at 581. But it was clearly erroneous for the Claims Court
to find that DRMS complied with the requirements of
FAR 16.503 by providing historical data.
In addition to the Claims Court’s clear error in finding
DRMS did not provide a negligent or inadequate estimate,
the Claims Court clearly erred in finding that “Agility
points to no specific cause-and-effect links to isolate its
damages.” J.A. 16–17. The Claims Court cited no evi-
dence and provided no reasoning in support of this find-
ing. The government argues that Agility failed to show
that it actually relied on the Amendment 007 Chart or
DRMS’s historical data based on the dates DRMS provid-
ed these estimates to Agility. Appellee’s Br. 36. It argues
the Amendment 007 Chart was provided just eight days
before Agility submitted its initial proposal, and the
Baseline Data was provided two days after Agility sub-
mitted its proposal. Id. Neither of the government’s
arguments are supported by the record.
As for the Amendment 007 Chart, Agility’s initial
proposal expressly states that Agility “used the workload
data provided by the government in Amendment 007 to
determine manning requirements and the amount of
potential revenue from the sale of scrap.” J.A. 1063. A
witness for Agility testified that Agility used the Amend-
ment 007 Chart in formulating its proposal. This is the
very purpose for which DRMS provided the Amend-
ment 007 Chart to offerors: “For the purpose of assisting
offerors in preparing their proposals.” J.A. 1003.
12 AGILITY DEFENSE v. US
As for the Baseline Data, we note that only Agility’s
initial proposal was submitted before the Baseline Data;
Agility submitted its final revisions on September 24,
2007, well after DRMS provided the Baseline Data on
August 4, 2007. Regardless, the Baseline Data was
retrieved from the same website to which DRMS directed
offerors in order to view historical data during Amend-
ment 002. DRMS provided this website on February 26,
2007, over five months before Agility submitted its initial
proposal. Agility’s initial proposal expressly states that
“[m]anpower estimates . . . are based on the historical size
of the workload at each facility.” J.A. 1175. A witness for
Agility testified that Agility visited DRMS’s website “on a
regular basis” and that it factored DRMS’s historical
workload data into its proposal. J.A. 127:16–23, 133:16–
18. And again, this was the very purpose for which the
government provided its historical data, to provide “the
offerors who were going to bid on the contract a view of
what was going on at the site and the most current infor-
mation we had available to us . . . .” J.A. 508:2–6.
FAR 16.503 requires the government to provide a re-
alistic estimate to offerors in requirements contracts
because “presumably contractors rely on the proffered
estimates in formulating their bids.” Medart, 967 F.2d
at 581. The evidence overwhelmingly shows that Agility
relied on DRMS’s estimates when formulating its pro-
posal. The Claims Court found, and the parties do not
dispute, that Agility experienced workloads “much great-
er” than the workload data DRMS provided to Agility.
J.A. 8. And Agility’s claim for damages seeks to recover
the costs it incurred from performing in excess of DRMS’s
negligent estimates. It was clear error for the Claims
Court to find there was no causal link between DRMS’s
estimates and Agility’s damages.
The Claims Court declined to address whether clause
H.19, which provides a mechanism for the parties to
adjust the contract price if specified changes in workload
AGILITY DEFENSE v. US 13
are observed, forecloses Agility’s claims. J.A. 12. As an
issue of contract interpretation, we hold that Agility’s
claim of negligent estimate is not foreclosed by
clause H.19.
The government does not contend that Agility’s claim
for compensation arises under clause H.19. Instead, it
argues clause H.19 is Agility’s only means of recovery.
We disagree. Under FAR 16.503, DRMS was required to
provide Agility with a realistic estimate of its require-
ments. DRMS failed to provide a realistic estimate. We
see no reason why DRMS’s negligence in providing an
adequate estimate during solicitation should be excused
by its inclusion of a provision directed to workload chang-
es upon performance. See J.A. 836. Agility’s claim for
relief is rooted in DRMS’s violation of FAR 16.503, leading
to a large disparity between pre-contract estimates and
actual workloads during the performance period. Agility’s
claim does not involve the limited subject of H.19, namely,
sufficiently large changes in workload levels that might
occur between earlier and later times entirely within the
performance period. On the merits of Agility’s claim of
negligent estimate, clause H.19 bears no relevance.
Finally, we address the effect of Agility’s receipt of
scrap sales on its claim for an equitable adjustment. The
Claims Court explained that the contract provision per-
mitting Agility to retain scrap sale revenue somewhat
mitigated Agility’s risk, stating, “if contract quantities
were higher than expected, theoretically the contractor’s
revenue from the sale of scrap would be higher.” J.A. 2.
It found that, despite the increased workload, Agility
realized less scrap proceeds than it had projected. None-
theless, the Claims Court held, “[a]lthough Agility faced
workloads significantly in excess of what it anticipated,
Agility still received over $44 million in scrap proceeds
over the 27 months of the contract. The fact that the
scrap proceeds were ‘22.9% lower’ than Agility’s projec-
tions does not move the equities in Agility’s favor.”
14 AGILITY DEFENSE v. US
J.A. 18. Under the parties’ contract, however, Agility’s
receipt of scrap proceeds does not limit Agility’s recovery.
The government argues Agility improperly seeks to
recover both its costs associated with increased workload
while retaining all of its proceeds from scrap sales. Appel-
lee’s Br. 32. It contends DRMS anticipated that the very
reason the contractor would retain scrap proceeds was to
“offset some of the costs incurred in performing the con-
tract.” Id. (quoting J.A. 894). It suggests that by pro-
cessing more property than anticipated, Agility must have
received more sales proceeds. Id. The government’s
arguments are without merit.
The parties’ contract expressly contemplated that
Agility would retain its scrap proceeds. DRMS’s RFP
stated that the contractor “is entitled to all sales pro-
ceeds” from scrap sales. J.A. 894. During solicitation,
DRMS explained that the contractor would retain such
scrap proceeds “‘free and clear’ without any type of reduc-
tion in payments.” J.A. 1010–11. While DRMS stated
such scrap sales were intended to “offset some of the costs
incurred in performing this contract,” J.A. 894, the only
impact Agility’s scrap proceeds were to have on the con-
tract price was the offset Agility offered to DRMS in its
proposal. Agility proposed an offset of $20,342,608 from
scrap proceeds over the life of the contract, offering an
offset of $5,730,312 for both the base year and first option
year. The government does not dispute that Agility
honored this commitment and DRMS received an offset of
$11,460,624 during its two years of performance. That
Agility received scrap proceeds above what it offered to
DRMS does not mean Agility would recover more than it
is entitled to by prevailing on its negligent estimate claim.
Under the contract, Agility is entitled to its scrap pro-
ceeds above the offset it offered DRMS regardless of its
workload. Moreover, as found by the Claims Court, the
scrap proceeds Agility actually retained were below the
amount it projected. Agility’s scrap proceeds do not offset
AGILITY DEFENSE v. US 15
the additional cost it incurred in performing the contract.
There is nothing in the record to suggest that Agility
benefited from DRMS’s failure to provide a realistic
estimate as required by FAR 16.503.
We thus hold that the Claims Court’s findings that
DRMS did not inadequately or negligently prepare its
estimates and that Agility did not rely on those estimates
are clearly erroneous. We hold Agility’s receipt of scrap
sales and the parties’ agreement to clause H.19 do not
preclude Agility from recovering under this claim. We
reverse the Claims Court’s denial of Agility’s negligent
estimate claim and remand for calculation of Agility’s
equitable adjustment.
CONCLUSION
For the foregoing reasons, we reverse the Claims
Court’s denial of Agility’s claim for increased costs and
remand for further proceedings consistent with this
opinion.
REVERSED AND REMANDED
COSTS
Costs to Agility.