2023-1337, 2023-1338

Appeals from the Armed Services Board of Contract Appeals in No. 59222, Administrative Judge David D’Alessandris, Administrative Judge Cheryl L. Scott, Administrative Judge Owen C. Wilson, Administrative Judge Richard Shackleford.

Decided: December 5, 2025

BORISLAV KUSHNIR, Commercial Litigation Branch, Civil Division, United States Department of Justice, Washington, DC, argued for appellant. Also represented by BRIAN M. BOYNTON, PATRICIA M. MCCARTHY; ALEXANDER MARTIN HEALY, Contract Disputes Resolution Center, Defense Contract Management Agency, Hanscom Air Force Base, MA.

JEFFREY A. HALL, Bartlit Beck LLP, Chicago, IL,Page 2 argued for cross-appellant. Also represented by CINDY L. SOBEL; DHANANJAY S. MANTHRIPRAGADA, Gibson, Dunn & Crutcher LLP, Los Angeles, CA; LINDSAY MIRIAM PAULIN, Washington, DC.

Before MOORE, Chief Judge , DYK and CUNNINGHAM, Circuit Judges .

DYK, Circuit Judge .

Pratt & Whitney (“Pratt”) builds aircraft engines, both for commercial airline customers and for the United States. This case involves cost calculations for cost-plus contracts between Pratt and the United States and specifically cost allocation between Pratt's government and private contracts. The government contends that it overpaid Pratt on the cost-plus contracts because Pratt incorrectly calculated the government's share of indirect costs (e.g., overhead). The Armed Services Board of Contract Appeals (the “Board”) ruled for the government in part and for Pratt in part. The government appealed, and Pratt cross-appealed. We hold that we lack jurisdiction over the Board’s decision concerning the base for allocation of overhead costs because the Board’s decision in this regard was not a final decision. However, we hold that the Board’s decision that there was an enforceable agreement between the parties as to the inclusion of so-called “Drag” in the overhead pool was final, and that the agreement is invalid. We therefore dismiss in part, reverse in part, and remand.

The facts underlying this case are complex but are virtually undisputed by the parties. During the relevant period, Pratt had a number of cost-plus contracts with the government as well as commercial engine contracts with private companies. Under the government contracts, the government was required to reimburse Pratt for overhead costs properly allocated to the government contracts. The Page 3Cost Accounting Standards (“CAS”) govern allocability of overhead costs. Pursuant to 48 C.F.R. § 9904.418 (“CAS 418”), a portion of the overhead pool is allocated to the government based on the proportion of material costs Pratt incurs for the government engine program as opposed to material costs associated with the commercial engine program. See Rumsfeld v. United Techs. Corp. , 315 F.3d 1361, 1369 (Fed. Cir. 2003).¹

To calculate Pratt’s overhead cost allocation to its government contracts, two different calculations must be undertaken. The first concerns which costs are properly included in Pratt’s overhead pool. The second relates to the allocation base of the overhead pool and how much of the pool is attributed to the government engine program versus the commercial engine program under CAS 418. This calculation defines the proportion of the overhead pool for which the government is responsible. The government bears a larger burden of the indirect costs when the total size of Pratt’s overhead pool increases and when the government’s portion of the base increases.

The long-running controversy in this case arises from Pratt's use of unconventional agreements with its parts suppliers. Instead of simply purchasing commercial engine parts from parts suppliers, Pratt has entered into “collaboration agreements” with its suppliers under which the suppliers (the “Collaborators”) are paid a percentage share ofPage 4 the engine program revenues in exchange for the provision of parts.

The first computation involves the question whether so-called Drag costs are properly included in the overhead pool. Drag is calculated through a fixed percentage of a Collaborator’s revenue share of commercial program profits rather than a function of the actual cost of overhead incurred by Pratt.² In the government’s view, Drag represents money that Collaborators paid to Pratt to cover the Collaborators’ share of commercial program expenses, and Drag costs should be removed from the overhead pool as an already-recovered expense. In this appeal, the resolution of this question depends on the validity of an agreement between the government and Pratt.

The second, more complex computation relates to the base calculation. If the costs related to the government program are larger, then the government pays more of the overhead pool. If the costs related to the commercial program are larger, then the government pays less of the overhead pool. As we surmised in Rumsfeld ,

It is therefore in the government’s interest to maximize the amount of material costs associated with commercial contracts in order to increase the value of the commercial program's portion of the overhead pool and inversely decrease the government’s share of overhead; and it is in Pratt’s interest to minimize the commercial program material costs, decrease the value of the commercial program's portion of the overhead pool and increase the government’s share of overhead.

Page 5Id. at 1363-64.

For over 30 years, the parties have disputed how to calculate the cost of the commercial engine parts acquired under the collaboration agreements. Initially, Pratt treated these parts as having no direct material cost, so the cost of the parts was not factored into the calculation of the commercial contracts’ share of the overhead pool. At the same time, Pratt reduced the total overhead pool by the Drag figure. In 1992, the contracting officer found that Pratt’s accounting methods violated CAS 418 because Pratt did not assign a cost to the engine parts that Pratt acquired from the Collaborators for the commercial engine program. According to the contracting officer, the cost of the engine parts was equivalent to the revenue share owed to the Collaborators. In 1996, another contracting officer issued a decision incorporating the first decision and establishing quantum for the violation.

Pratt appealed these decisions to the Board, which determined that Pratt’s accounting practices did not violate CAS 418. The government then appealed to this court, and we determined that Pratt’s failure to account for a cost of the engine parts violated CAS 418 and caused “a substantial distortion in overhead allocation” between the commercial and government engine programs. Id. at 1372. We vacated the Board’s decision and remanded for the Board to determine the proper method to determine the cost of the engine parts. Id. at 1377.

After our decision in Rumsfeld , on June 5, 2006, the parties reached a settlement agreement and also entered an agreement (the “Drag agreement”) whereby in the future Pratt would include the parts' material cost in the overhead pool base but would no longer reduce the overhead pool by the Drag figure. The Drag agreement provided in relevant part:

The [contracting officer determined] that Pratt's discontinuance of its prior practice of crediting

Page 6DRAG to its overhead pools was compliant with both CAS and the Federal Acquisition Regulation [(“FAR”)] from January 1, 2005 to the date of this agreement. The Government will not require such credits prospectively, provided that Pratt continues to include a cost for collaboration parts in its allocation base or bases and that the “DRAG” provisions of the collaboration agreements do not materially change.

J.A. 5496.³

The Drag agreement did not resolve the dispute. In 2013, a new contracting officer issued a decision revisiting the issue of how to calculate the government's share of Pratt’s overhead expenses. The contracting officer determined once again that Pratt’s accounting practices violated CAS. One dispute concerned Pratt’s use of “Manufacturing Target Cost” (“MTC”) to estimate the cost of the parts. This number represented how much Pratt expected it would cost to manufacture the parts itself. The contracting officer determined that Pratt’s use of an estimate of material costs of the commercial engine parts rather than the actual costs was impermissible. To measure material costs, the contracting officer instead used gross revenue share, which is the percentage of the list price Pratt owed to a Collaborator when a commercial engine was sold. In reaching this conclusion, the contracting officer referenced our decision in Rumsfeld , stating that we “defined the cost of collaboration material as revenue share payments,” which the contracting officer apparently determined to mean gross revenue share. J.A. 5710.

The contracting officer also determined that the Drag agreement was not valid and therefore was not binding on Page 7the government. The contracting officer ruled that Pratt was required to remove any Drag costs from the overhead pool because those costs were reimbursed by the Collaborators. The contracting officer calculated a total debt of $210,968,414 owed to the United States.

Pratt once more appealed to the Board. The Board sustained Pratt’s appeal in part, finding that the Drag agreement was a valid agreement and that Pratt was not required to eliminate Drag from the overhead pool. But the Board concluded that Pratt improperly calculated its material costs under CAS 418, rejecting Pratt’s MTC method of calculating the material cost of the parts. At the same time, the Board also rejected the contracting officer’s finding that gross revenue share was the proper measure of material cost. Instead, the Board determined that the proper measure was net revenue share. This was because Pratt rarely collects the full list price in its sales of commercial engines, and it deducts certain items (including Drag and certain discounts called Fleet Introductory Assistance) from the payments it ultimately sends to Collaborators. Therefore, as the Board viewed it, the total payment that Pratt actually sent to Collaborators, net revenue share, was the proper basis for determining the material cost of the collaborator parts. The Board found that net revenue share represents what Pratt “ actually pays to its collaborators for the parts it receives.' J.A. 99 (internal quotation marks omitted); see also Rumsfeld , 315 F.3d. at 1371 (discussing “material cost” as “the outlay for materials ‘purchased’”).

The Board remanded the issue of quantum to the parties; in other words, the Board remanded to have the parties agree on quantum. If the parties failed to do so, the Board would be required to determine quantum itself. See Teller Env't Sys. v. United States , 802 F.2d 1385, 1390 (Fed. Cir. 1986) (noting that a remand to the parties to negotiate quantum is not “a transfer, or a cessation, of the [B]oard’s jurisdiction over the case” but a “direction to thePage 8 parties to negotiate . . . and to come back to the [B]oard . . . for resolution of any dispute if agreement cannot be reached'). The government appeals, and Pratt cross-appeals.

In every appeal, the first and fundamental question is that of jurisdiction. Steel Co. v. Citizens for a Better Env’t , 523 U.S. 83, 94 (1998) (quoting Great S. Fire Proof Hotel Co. v. Jones , 177 U.S. 449, 453 (1900)). This court has exclusive jurisdiction over “an appeal from a final decision of an agency board of contract appeals” pursuant to 41 U.S.C. § 7107(a)(1). 28 U.S.C. § 1295(a)(10) (emphasis added). The government characterizes this appeal as addressing two separate claims. The first claim addresses the calculation of Collaborator commercial part costs for the purpose of determining the government’s allocation base of Pratt’s overhead pool (the “CAS 418 Claim”). The second claim addresses whether Pratt is required to reduce the total overhead pool by the amount it receives in Drag payments from Collaborators (the “Drag Claim”). The government asserts that the Board’s decision as to the CAS 418 Claim was a final decision, even though quantum remained to be determined. It also argues the Board’s Drag Claim decision was itself a separate and final decision. The government argues that if we find we do not have jurisdiction over the CAS 418 Claim, which required a quantum calculation, we can still exercise jurisdiction over the Drag Claim, which was fully resolved by the Board.

Although “finality in this context is a flexible concept,” England v. Contel Advanced Sys., Inc. , 384 F.3d 1372, 1378 (Fed. Cir. 2004), there are long-settled rules that guide our analysis. The “relevant inquiry” centers on “the contracting officer’s decision, for this determines the extent of the Page 9contractor’s right of appeal and the [B]oard’s jurisdiction.” Brownlee v. DynCorp , 349 F.3d 1343, 1347 (Fed. Cir. 2003) (quoting Dewey Elecs. Corp. v. United States , 803 F.2d 650, 655 (Fed. Cir. 1986)). Where, as here, the contracting officer made a determination both as to liability and quantum, we have long held that a Board decision that determines only liability and does not reach quantum is not final. See Teledyne Cont’l Motors v. United States , 906 F.2d 1579, 1583 (Fed. Cir. 1990); AAA Eng’g & Drafting, Inc. v. Widnall , 129 F.3d 602, 604 (Fed. Cir. 1997); Kinetic Builder’s Inc. v. Peters , 226 F.3d 1307, 1312–13 (Fed. Cir. 2000); United Pac. Ins. Co. v. Roche , 294 F.3d 1367, 1370 (Fed. Cir. 2002).

With respect to the CAS 418 Claim, the government argues that, despite the Board’s remand for determination of quantum, we can exert jurisdiction over this appeal in its entirety based on our decision in Southern California Edison Co. v. United States , 226 F.3d 1349 (Fed. Cir. 2000) (“ SCE '). In SCE we held that, when a lower court remands a case to an agency, that order is an appealable final order where the court rules on an issue of law and remands for the agency to apply that legal conclusion such that the agency’s remaining task was merely ministerial. 226 F.3d at 1354-55 (citing Sullivan v. Finkelstein , 496 U.S. 617, 619 (1990)). There, we said that “[s]uch orders are final and appealable because they represent ‘the final disposition of the proceeding respecting the Secretary’s practice’ and ‘compel [ ] action . . ., on remand, contrary to the Secretary’s prior ruling.’” Id. at 1355 (quoting Travelstead v. Derwinski , 978 F.2d 1244, 1248 (Fed. Cir. 1992) (second and third alterations in original)).

We do not find the logic of SCE to be applicable to the present case. SCE was a case that dealt with the finality of a Court of Federal Claims (“Claims Court”) decision in a breach of contract action brought against the government. Id. at 1354. There, once the Claims Court rendered its decision, there were no further issues for the Claims Court to Page 10resolve. This was so because the decision that we were reviewing in SCE originated in the Claims Court, not the agency, and we were reviewing a decision of the Claims Court that was final even though the decision was characterized as a remand to the agency and further calculation by the agency was required. That is not this case. Here, we are being asked to directly review an agency determination, the decision of the Board. The Board’s decision as to the CAS 418 Claim is not final. Instead, the Board retained jurisdiction over the case when it remanded the matter to the parties to negotiate quantum. See Teller , 802 F.2d at 1390. If the parties are unable to settle the matter on quantum, then the Board must conduct further quantum proceedings.

The parties have not cited to any cases where we have applied SCE to a proceeding involving a remand from the Board, nor are we aware of any. We have consistently held a Board decision such as this one that does not resolve quantum is not a final decision for the purposes of our appellate jurisdiction. Kinetic Builder's , 226 F.3d at 131213; United Pac. Ins. Co. , 294 F.3d at 1370; see also Teller , 802 F.2d at 1389-90; Teledyne , 906 F.2d at 1583; AAA Eng’g , 129 F.3d at 604. Indeed, in SCE , we described and approved the holding of Teledyne , 226 F.3d at 1355, and the government concedes that 'read in isolation,' these cases suggest that the Board’s decision is not final. Appellant’s Br. 27. We continue to follow the long-settled rule that, where the contracting officer made a determination as to both liability and quantum, and the Board has only made a determination as to liability, the Board’s decision is not final for the purposes of our appellate jurisdiction. Therefore, we do not have jurisdiction over the CAS 418 Claim.

The government separately argues we have jurisdiction over the Board's decision with respect to the DragPage 11 Claim. It is established that where the contracting officer has been presented with multiple entitlement claims, “not based on a common or related set of operative facts,” and some of those claims are remanded to the parties for negotiation of quantum while others are finally decided, 'the Board decision on the non-remanded claims is deemed final.” Kinetic Builder’s , 226 F.3d at 1313 (citing Dewey Elecs. , 803 F.2d at 654); accord Elkins v. Gober , 229 F.3d 1369, 1373–74 (Fed. Cir. 2000). Therefore, if the CAS 418 Claim and the Drag Claim are separate claims, as the government contends, then we can assert jurisdiction over the Drag Claim even though the CAS 418 Claim was not fully resolved.

The Drag Claim concerns whether Drag costs are properly included in the overhead pool. The Board decision turned on the validity of the provisions of the Drag agreement and whether the government contracting officer had the authority to enter into the agreement. Therefore, the Drag Claim turns on factual questions unrelated to the CAS 418 Claim. See Kinetic Builder’s , 226 F.3d at 1314– 15 (finding two claims related to the same contract separate where they “do not involve proof of a common set of operative facts”). We conclude that the operative facts for these two claims are distinct and constitute separate claims.

The Board’s decision on the Drag Claim is final. While a remand of the Board decision would require a calculation, the Board decision does not require a new calculation to determine the size and makeup of the overhead pool. Instead, the Board affirmed Pratt’s ongoing practice of calculating the overhead pool without regard to the Drag payments. Because the Board's decision as to the Drag Claim was separate and final, not requiring a further calculation, it was appealable, and we have jurisdiction over the Board’s decision on the Drag Claim.

Page 12As the parties note, this case is also similar to Orlando Helicopter Airways v. Widnall , 51 F.3d 258 (Fed. Cir. 1995). In Orlando Helicopter , we considered whether we had jurisdiction over a partial summary judgment ruling by the Board. 51 F.3d at 260. There, a contractor asserted a single claim that an investigation into the contractor initiated by the Department of Justice ('DOJ') and subsequent actions by the contracting officer had caused the contractor to incur additional costs alleged to be compensable by the government. Id. at 259. At summary judgment, the Board determined that the costs incurred as a result of the DOJ investigation were not compensable under the sovereign acts doctrine, but that the contractor could pursue its claims before the Board for costs incurred as a result of the contracting officer’s actions. Id. at 260.

We determined that the Board's grant of summary judgment with regard to the DOJ investigation costs was a final decision, not requiring any further calculation, and we had jurisdiction over its appeal because it “determine[d] the rights and obligations of the parties” and was “wholly separate and distinct from any issues which may remain before the Board.” Id. at 261. Here, the Board's decision on the Drag Claim determined rights and obligations of the parties separate and distinct from the issues presented by the CAS 418 Claim, which required a quantum computation before the decision was final, and we have jurisdiction over the Drag Claim.

Having established that we have jurisdiction over the Drag Claim, we now consider the merits of the government’s appeal. The Board determined that the Drag agreement permitting Pratt to cease reducing the overhead pool by the Drag figure was valid and enforceable. The parties do not appear to dispute that the plain language of the Drag agreement permits Pratt to discontinue this practice. However, the government argues that the Drag agreement Page 13was invalid and unenforceable against the government because it violated the FAR.

Whether an enforceable contract exists is a mixed question of law and fact. See Ransom v. United States , 900 F.2d 242, 244 (Fed. Cir. 1990). We review the Board’s conclusions of law de novo. Triple Canopy, Inc. v. Sec’y of the Air Force , 14 F.4th 1332, 1337-38 (Fed. Cir. 2021). We may only set aside the Board's findings of fact if the Board's findings are “(A) fraudulent, arbitrary, or capricious; (B) so grossly erroneous as to necessarily imply bad faith; or (C) not supported by substantial evidence.' 41 U.S.C. § 7107(b)(2).

As the Board acknowledged, '[a] Government agent must have actual authority to bind the Government to a contract.' Liberty Ammunition, Inc. v. United States , 835 F.3d 1388, 1401 (Fed. Cir. 2016) (citing Trauma Serv. Grp. v. United States , 104 F.3d 1321, 1325 (Fed. Cir. 1997)); accord Boyd v. United States , 134 F.4th 1348, 1352 (Fed. Cir. 2025). Here, the Drag agreement was explicitly entered into “under authority of” the FAR. J.A. 5495. Generally, a government official lacks actual authority to enter into a contract that violates the FAR. Johnson Mgmt. Grp. CFC v. Martinez , 308 F.3d 1245, 1256 n.2 (Fed. Cir. 2002) (citing McDonnell Douglas Corp. v. United States , 670 F.2d 156, 159 (Cl. Ct. 1982)). Without specific authorization from the agency head or his designee (and there has been no showing of such authorization here), 'any deviation from regulatory requirements is a violation of the regulations and beyond the authority of the contracting officer.” Id. The question then is whether the Drag agreement violates the FAR.

The government argues that the Drag agreement violates the FAR because it conflicts with the FAR’s requirements for advance agreements codified at 48 C.F.R. § 31.109 and it conflicts with the FAR’s credits provision.

Page 14Because we agree with the government’s first argument, we need not address the second.

The FAR's advance agreements provision notes that “contracting officers and contractors should seek advance agreement on the treatment of special or unusual costs.” 48 CFR § 31.109(a). This provision establishes an optional pathway for contractors and contracting officers to negotiate the future treatment of costs where “the reasonableness, the allocability and allowability . . . of certain costs may be difficult to determine” under standard FAR provisions. Id. However, any advance agreement reflecting a negotiated treatment of unusual costs “must be in writing, executed by both contracting parties, and incorporated into applicable current and future contracts. An advance agreement shall contain a statement of its applicability and duration.' Id. § 31.109(b). A contracting officer is not authorized “to agree to a treatment of costs inconsistent with this part” of the FAR governing advance agreements. Id. § 31.109(c).

The Drag agreement here, on its face, states that it was executed under the authority of § 31.109, the advance agreements provision. The Board also correctly noted that there was no evidence that the Drag agreement was incorporated into any of Pratt’s government contracts and that the Drag agreement lacks a statement of its duration. The undisputed facts establish how the Drag agreement does not comply with the FAR. The Board agreed. J.A. 90. Nonetheless, the Board concluded that the Drag agreement did not need to satisfy the FAR’s requirements so long as it satisfied the general requirements for forming a contract. Pratt argues that this is permissible because contracting officers are granted broad license 'to settle CAS-related disputes with contractors.” Cross-Appellant’s Br. 65.

We disagree. It is undeniable that the Drag agreement is an advance agreement. The Drag agreement, on its face, states that it is “entered into . . . under authority of [FAR]Page 15 31.109,” the advance agreements provision. J.A. 5495. The Drag agreement is a negotiated agreement to 'prospectively” agree to the treatment of costs, namely Pratt’s inclusion of Drag costs in the overhead pool. J.A. 5496. We are aware of no holding that gives the contracting officer authority to disregard the FAR restrictions because a settlement agreement is involved, and the Board cites to none. To be sure, there is a strong public interest in enforcing settlements, especially in complex litigation. See, e.g. , FTC v. Actavis, Inc. , 570 U.S. 136, 154 (2013); Baseload Energy, Inc. v. Roberts , 619 F.3d 1357, 1361 (Fed. Cir. 2010). However, a settlement agreement is a contract, so its enforceability first turns on basic principles of contracting, such as whether there was authority to bind the parties. See, e.g. , Sweeney v. U.S. Postal Serv. , 159 F.3d 1342, 1344 (Fed. Cir. 1998) (“A settlement agreement is a contract between the parties.' (citing Mahboob v. Dep't of Navy , 928 F.2d 1126, 1128 (Fed. Cir. 1991))); Slattery v. Dep't of Just. , 590 F.3d 1345, 1347 (Fed. Cir. 2010) (“A settlement agreement is a contract.”); Bell BCI Co. v. United States , 570 F.3d 1337, 1341 (Fed. Cir. 2009) (“Because a release is contractual in nature, it is interpreted in the same manner as any other contract term or provision.”).

Because the FAR was the source of the contracting officer’s authority and did not authorize the agreement, the Board was required to find, as a matter of law, that the contracting officer lacked the authority to execute the Drag agreement on behalf of the government. See Johnson Mgmt. Grp. , 308 F.3d at 1256 n.2. Because the contracting officer lacked authority to bind the government, the Drag agreement cannot be enforced against it.⁴

Page 16We recognize that Pratt argues that even without the Drag agreement, it is not required to deduct Drag from the overhead pool. The government argues that Drag costs must be excluded from the overhead pool because they were reimbursed. In this respect, the government relies on the FAR’s credits provision, codified at 48 C.F.R. § 31.2015. We are skeptical a reimbursed cost could be properly included in the overhead pool, but we decline to reach this issue. Nor do we decide Pratt’s argument that there has not been a showing that the Drag payments correspond to any costs included in the overhead pool. On remand, the Board will need to address and resolve these issues in the first instance.

For the foregoing reasons, we dismiss this appeal with respect to the CAS 418 Claim. With respect to the Drag Claim, we conclude that we have jurisdiction, and we reverse the Board’s determination that the Drag agreement was an enforceable contract. We remand for further proceedings consistent with this opinion.

Costs to neither party.