Fifth Third Bank of Western Ohio v. United States

PLAGER, Senior Circuit Judge. In this Wmstor-related case, the issue is whether the Government is liable for breach of contract resulting in alleged losses sustained by plaintiff Fifth Third Bank of Western Ohio (“Fifth Third”). The transactions at issue are familiar, arising out of the thrift industry problems in the 1980s. The United States Court of Federal Claims held a trial on the issue. After plaintiff presented its case-in-chief, the Government moved for judgment on the theory that the evidence did not establish the existence of the alleged contractual obligations. The trial court granted the Government’s motion; Fifth Third appeals the trial court’s judgment that the United States was not liable for breach of contract. Earlier in the proceedings the Government had moved for partial summary judgment with respect to plaintiffs damages claims. The court granted the motion in part, thereby precluding plaintiff from presenting certain damages theories at trial. Fifth Third appeals one aspect of this ruling — the trial court’s rejection of Fifth Third’s expectation damages model based on a hypothetical cost of replacing, or “covering,” goodwill lost as a result of breach. With regard to liability, the trial court erred in concluding that contracts did not exist between the United States and Fifth Third’s predecessor-in-interest, Citizens Federal Bank FSB (“Citizens”) regarding Citizens’ acquisition of four failing thrifts; 1 the contractual terms included permission for Citizens to use the purchase method of accounting to amortize supervisory goodwill over an extended period of time and to count supervisory goodwill toward capital reserve requirements. As has been explained in other Winstar-related cases, subsequent government activity caused these contracts to be breached. With regard to damages, the trial court correctly ruled that plaintiffs hypothetical cost of cover is not a proper measure of damages. Accordingly, the judgment of the Court of Federal Claims is affírmed-in-part and reversed-in-part. The case is remanded for further proceedings consistent with this opinion. BACKGROUND A. Regulatory Setting It has been more than twenty years since the critical events in the thrift indus *1224 try crisis of the 1980s occurred, and almost a decade since the law that governs these cases was first established. During that time the history and circumstances surrounding the thrift crisis and the ensuing-events, including enactment of the Financial Institutions Reform, Recovery, and Enforcement Act of 1989, Pub.L. No. 101— 73, 103 Stat. 183 (“FIRREA”), have been extensively discussed in opinions of the Supreme Court, see United States v. Winstar Corp., 518 U.S. 839, 116 S.Ct. 2432, 135 L.Ed.2d 964 (1996), and this court, see, e.g., Winstar Corp. v. United States, 64 F.3d 1531 (Fed.Cir.1995) (en banc); Glendale Fed. Bank, FSB v. United States, 239 F.3d 1374 (Fed.Cir.2001). We need provide only a brief overview here to serve as backdrop for our decision in this case. Rising interest rates and inflation during the late 1970s and early 1980s precipitated a crisis in the savings and loan industry. More than 400 thrifts failed between 1981 and 1983, and many additional thrifts were on the verge of insolvency. This situation threatened to exhaust the insurance fund of the Federal Savings and Loan Insurance Corporation (“FSLIC”). To deal with this crisis, the Federal Home Loan Bank Board (“FHLBB”), the agency responsible for regulating federally chartered savings and loan associations, sought out healthy financial institutions and outside investors for the purpose of having them purchase troubled thrifts through the use of “supervisory mergers.” The FHLBB offered financial incentives to induce such mergers and to prevent an acquiring thrift from becoming insolvent upon completion of the transaction. The incentive most pertinent to this case was the accounting treatment of what was called “supervisory goodwill,” basically the difference between the fair market value of the failing thrift’s liabilities assumed by an acquiring thrift and the fair market value of the failing thrift’s assets. In a merger utilizing supervisory goodwill, the FHLBB would permit the acquiring institution to count supervisory goodwill toward its reserve capital requirements and to use the purchase method of accounting to amortize this supervisory goodwill over an extended period of time, up to forty years. Additional incentives provided by the FHLBB in some supervisory mergers included cash contributions in the form of permanent credits to regulatory capital and forbearance agreements by the FHLBB not to enforce regulatory capital requirements for a specified period of time. However well-intended these various measures were, the problems in the savings and loan industry persisted. In 1989 Congress stepped in with the enactment of FIRREA. FIRREA abolished the FHLBB and FSLIC, transferred thrift insurance activities to the Federal Deposit Insurance Corporation (“FDIC”), established the Office of Thrift Supervision (“OTS”) as the new thrift regulatory agency, created the Resolution Trust Company (“RTC”) to liquidate failed thrifts, and made substantial changes in the regulation of the savings and loan industry. Directly to the point for purposes of this case, FIRREA also created a minimum capital requirement for thrifts, phased out the thrifts’ ability to count supervisory goodwill toward capital requirements, and limited the amortization period of supervisory goodwill. The impact of FIRREA, in particular the supervisory goodwill provisions, was swift and severe, and many thrifts quickly fell out of compliance with regulatory capital requirements, making them subject to seizure by thrift regulators. This course of events gave rise to hundreds of lawsuits filed by formerly healthy *1225 thrifts that had been acquirers of failing thrifts. The plaintiffs sought damages from the Government under several theories, including breach of contract. The thrifts alleged that Congress’s enactment of FIRREA breached the Government’s obligation to allow special accounting treatment of supervisory goodwill. In due course the matter reached the Supreme Court in United States v. Winstar Corp., 518 U.S. 839, 116 S.Ct. 2432, 135 L.Ed.2d 964 (1996). The appeal involved the cases of three acquiring thrifts taken as test cases; the cases had been consolidated for appeal. The Supreme Court affirmed this court’s en banc determination that the thrifts had indeed entered into enforceable contracts with the Government, and that the Government was liable for breach of contract as a result of the enactment and enforcement of FIRREA. In the years since the Supreme Court’s Winstar decision, appeals in numerous Winstar-related cases have come before this court. Because the background and the fact pattern of the Government’s involvement with the thrifts are well-known, in many cases the existence between the Government and the acquiring thrift of an agreement to allow supervisory goodwill is clear. The Government concedes liability, and the issue of contention between the parties has been the amount of damages. 2 In a relatively few cases, because of particular circumstances in the transactions at issue, the Government has contested whether the FHLBB and the thrift understood that supervisory goodwill would be part of the transaction. In assessing the issue of contractual obligation and breach, there is a difference between the issue of contract formation and the issue, once it is determined that the relationship is one of contract, of interpreting the dimension of particular contractual obligations. In this case, the parties do not dispute basic issues of offer, acceptance, and consideration, the essentials of contract formation. What is in dispute is whether a particular term is part of whatever contract existed between the parties, i.e., did the parties intend to bind themselves to long-term amortization of the regulatory goodwill created by the transactions in which they engaged. Often in cases such as this, with the facts of record and generally agreed, the matter is decided on summary judgment; this case is unusual in having the facts and the trial court’s conclusions determined only after a full trial. B. Factual History Citizens, a federally chartered mutual savings and loan association headquartered in Dayton, Ohio, acquired or merged with four failing thrifts between 1982 and 1985. As found by the trial court, each transaction at issue in this case followed the same general pattern. The Federal Home Loan Bank of Cincinnati (“FHLB-Cincinnati”) had identified Citizens as a healthy thrift and therefore a candidate for acquiring or merging with failed thrifts. In each of the four transactions, FHLB-Cincinnati contacted Citizens to propose a supervisory merger with the failing thrift. In each case Citizens and the failing thrift entered into a Purchase Agreement or a Merger Agreement expressly conditioned upon obtaining approval of the agreement by FHLBB. Citizens and the failing thrift then submitted an application for merger to FHLB-Cineinnati. FHLB-Cincinnati conditionally approved each transaction by issuing a Bank Board Resolution specifying the additional *1226 steps necessary for obtaining final approval. Each resolution required Citizens to furnish an opinion from its independent accountant justifying the use of the purchase method of accounting, describing any goodwill arising from the purchase, and substantiating the reasonableness and amounts of the goodwill and the related amortization period. Citizens complied with the requirements set forth in each resolution, including submission of an opinion letter from its accountant. FHLB-Cincinnati then submitted each proposed acquisition or merger to FHLBB headquarters for final approval. FHLBB approved each transaction, and Citizens accounted for each, including the use of the purchase method, the amortizing of goodwill over an extended period, and the counting of goodwill toward its regulatory capital requirements. At trial, plaintiff presented extensive testimony by Jerry Kirby, President and Chief Executive Officer of Citizens, and Lawrence Muldoon, FHLB-Cincinnati’s chief supervisory agent, describing the negotiations between Citizens and FHLB-Cincinnati with respect to the four transactions. The first transaction was Citizens’ acquisition of thirteen branches of Cardinal Federal Savings and Loan Association (“Cardinal”) in 1982. Mr. Kirby testified that the Cardinal transaction was initiated when Mr. McElheney, at the time a supervisory agent at FHLB-Cincinnati and Mr. Muldoon’s subordinate, contacted Citizens to propose that Citizens acquire the Cardinal branches. He explained that the FHLBB would permit Citizens to book Cardinal’s negative net worth as goodwill, which would count as an asset for regulatory capital purposes and could be amortized over an extended period of time. Mr. Kirby testified that he followed up with Mr. Muldoon to more fully understand how such an arrangement would work. Mr. Kirby asked Mr. Muldoon if Citizens could receive cash assistance. Mr. Muldoon replied that there was no cash assistance available, but Citizens could book supervisory goodwill as an asset and amortize it over an extended period of time. Mr. Muldoon’s testimony confirmed that he offered goodwill as a substitute for cash assistance. According to Mr. Kirby, he viewed the goodwill accounting treatment as the sine qua non of the Cardinal acquisition because without it Citizens would have immediately fallen out of capital compliance. Mr. Kirby testified that he discussed the proposed arrangement with his board of directors, although no written records of any meetings exist. Citizens and Cardinal entered into a formal Purchase Agreement expressly conditioned upon receiving FHLBB approval. Citizens submitted a merger application to FHLB-Cincinnati. The application included a pro forma financial statement based on the assumption of a thirty-year amortization period for goodwill. FHLB-Cincin-nati conditionally approved the transaction in Board Resolution V-O-M-82-7. As a condition for final approval, the Resolution required Citizens to submit an opinion from its independent accountant that: (a) indicates the justification under generally accepted accounting principles [GAAP] for use of the purchase method of accounting, (b) specifically describes any goodwill arising from the purchase to be recorded on Citizens’ books, and (c) substantiates the reasonableness and amounts of such goodwill and related 30 year amortization period and method. FHLB-Cincinnati sent a letter to the FHLBB recommending approval of the Cardinal acquisition. The letter stated that approximately $35 million of goodwill would result from the transaction and would be amortized over thirty years. Cit *1227 izens submitted the required accountant’s letter, which indicated that goodwill would be amortized over forty years, not thirty years. The FHLBB thereafter approved the transaction, and Citizens booked $37 million of goodwill using a 30-year amortization period. Citizens entered into three subsequent supervisory transactions — with Gateway Federal Savings and Loan Association (“Gateway”) in 1983, with Homestead Federal Savings and Loan Association (“Homestead”) in 1984, and with First Federal Savings and Loan Association (“First Federal”) in 1985. In each case, FHLB-Cincinnati identified the failing thrift, and Mr. Muldoon of FHLB-Cincin-nati contacted Mr. Kirby to propose that Citizens acquire or merge with the failing thrift. According to trial testimony, the discussions between Mr. Kirby and Mr. Muldoon leading up to each transaction followed a similar pattern. When considering whether to acquire each thrift, Mr. Kirby requested FSLIC cash assistance. Mr. Muldoon responded each time that no cash assistance was available, but that instead he could offer the same regulatory and accounting treatment of supervisory goodwill that had been used in the Cardinal transaction, i.e., use of the purchase method of accounting to book supervisory goodwill as an asset that would count toward regulatory capital requirements and be amortized over an extended period of time. Following the negotiations between Mr. Kirby and Mr. Muldoon regarding each potential transaction, Citizens and the FHLBB exchanged documents similar to those in the Cardinal transaction, as noted above. The documents, however, lacked some of the detail provided in the Cardinal documents. For example, the Board Resolutions for the Gateway, Homestead, and First Federal transactions required Citizens to provide an opinion from its independent accountant justifying the use of the purchase method of accounting, describing any goodwill arising from the transaction to be recorded on Citizens’ books, and substantiating the reasonableness and amounts of such goodwill and related amortization period and method. These Resolutions, however, did not specify the length of the amortization period. Nevertheless, it is undisputed that for each transaction Citizens booked supervisory goodwill as an asset and amortized it over a period of twenty years (for Homestead and First Federal) or ten years (for Gateway). C. The Trial After discovery in this case, both parties filed motions for summary judgment on liability. The trial court denied both motions. Fifth Third Bank of W. Ohio v. United States, 52 Fed. Cl. 264 (2002) (“Fifth Third I ”). The court granted the Government’s motion for reconsideration in order to address whether FHLB-Cin-cinnati had actual authority to bind the Government to a contract, an issue the trial court had erroneously deemed abandoned in its prior ruling. The court concluded that FHLB-Cincinnati possessed implied actual authority to bind the FHLBB to promises regarding the accounting treatment of supervisory goodwill. Fifth Third Bank of W. Ohio v. United States, 52 Fed. Cl. 637 (2002) (“Fifth Third II”). Meanwhile, the Government moved for summary judgment with respect to Plaintiffs damages claims. The trial court granted this motion in part, concluding that Plaintiff was not entitled to pursue various damages theories at trial, including its calculation of expectancy damages based on the doctrine of cover. Fifth Third Bank of W. Ohio v. United States, *1228 55 Fed. Cl. 223, 242-44 (2003) (“Fifth Third III”). The case proceeded to trial on the issues of liability and the remaining damages theories. At the close of Plaintiffs ease-in-chief, the Government moved for judgment on partial findings pursuant to Court of Federal Claims Rule 52(c). The trial judge orally granted the motion with respect to liability and denied it with respect to damages, and thereafter issued a written opinion. Fifth Third Bank of W. Ohio v. United States, 56 Fed Cl. 668 (2003) (“Fifth Third IV”). As discussed, Plaintiffs case included substantial oral and deposition testimony describing the communications between Citizens and FHLB-Cincinnati. The trial judge specifically found that “Mr. Kirby [Citizens’ President] thought that Citizens had contracts with the Government. Messrs. Muldoon, McElheney, and Thiemann [the Government officials] all thought that FHLBB had entered into contracts and that FHLBB had made a commitment to recognize goodwill for regulatory purposes that could not be withdrawn.” Id. at 694. Regarding the credibility of this testimony, the trial court noted that some of the testimony seemed to be rote, perhaps because of the long time since the events, but that the court “credits fully the integrity of these gentlemen.” Id. Despite these findings, the trial court concluded that the agreements between the parties were not contractual relationships, at least in part because the only written evidence corroborating the witness testimony was contained in what the court described as routine agency documents. Id. at 694-96. The Court of Federal Claims entered final judgment in favor of the United States. In this appeal, Fifth Third challenges the trial court’s Rule 52(c) judgment on liability and the trial court’s partial summary judgment precluding Fifth Third from presenting its cover damages theory at trial. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(3). DISCUSSION A. Liability The trial court cast the issue of liability in terms of contract formation — offer and acceptance, consideration, actual authority, and mutuality of intent. She posed the question as: “What evidence should be required to establish a commitment to provide something of value within the regulatory framework, along with the concomitant commitment not to regulate in contravention of that agreement?” Fifth Third IV, 56 Fed. Cl. at 692. On appeal plaintiff contends that the trial court erred by failing to accord proper weight to trial testimony regarding negotiations between Citizens and FHLB-Cincinnati officials. Plaintiff further argues that the trial court erred when it dismissed the documentary evidence relating to Citizens’ supervisory mergers as merely reflecting regulatory approval of the transactions rather than contractual commitments between the parties. The Government responds that the documents in evidence were indeed regulatory in nature and that the testimony was too general and vague to show that the written documents actually demonstrated the existence of contractual obligations. It is helpful to consider this case within the larger context of the earlier Winstar cases. In this court’s Winstar en banc decision, this court concluded that the Government formed contractual agreements regarding the treatment of supervisory goodwill with Winstar and two other institutions, Statesman Savings Holdings Corp. (“Statesman”) and Glendale Federal Bank (“Glendale”). Winstar, 64 F.3d at 1540- *1229 44. Analyzing not only the contemporaneous documents but also the circumstances surrounding the transactions, this court determined that the Government was contractually obligated to recognize supervisory goodwill resulting from the transactions as a capital asset and to permit such goodwill to be amortized over an extended period of time. Id. The Supreme Court considered the evidence and found no reason to question this court’s conclusion in that regard. Winstar, 518 U.S. at 861-68, 116 S.Ct. 2432. The Court also confirmed there were no constitutional or statutory obstacles to enforcing the agreements under ordinary contract principles. Id. at 871-911, 116 S.Ct. 2432. In each of the three transactions in Winstar, the relevant documents included either an Assistance Agreement or a Supervisory Action Agreement, each containing an integration clause incorporating contemporaneous documents such as the Bank Board Resolutions issued prior to each transaction. But such documents are not legal prerequisites to a contractual obligation. In California Federal Bank, FSB v. United States, 245 F.3d 1342 (Fed.Cir.2001), this court determined that contractual obligations existed even though there was no single document incorporating all the contract terms. Id. at 1346^47. We noted that our Winstar decision “did not rely exclusively on the assistance agreements to find a contract.” Id. at 1346. Instead, we agreed with the Court of Federal Claims that “[i]f the factual records of individual cases show intent to contract with the government for specified treatment of goodwill, and documents such as correspondence, memoranda and [FHLBB] resolutions confirm that intent, the absence of an [assistance agreement] or [supervisory action agreement] should be irrelevant to the finding that a contract existed.” Id. at 1347 (citation omitted) (alteration in original). This court has determined Government liability in other cases in which the Government has contested the issue. In Barron Bancshares, Inc. v. United States, 366 F.3d 1360 (Fed.Cir.2004), goodwill and capital credit provisions in the documents were “virtually identical” to those in the Winstar transactions. Id. at 1376. Nevertheless, the Court of Federal Claims had determined those provisions were not contractual obligations because there was no evidence they were the subject of any pre-contract negotiations between the parties, whereas other terms had been negotiated. Id. at 1372. We reversed, holding that the clear and unambiguous language of the integrated contract document was binding and precluded resort to parol evidence to alter its terms. Id. at 1375-76. In LaSalle Talman Bank, F.S.B. v. United States, 317 F.3d 1363 (Fed.Cir.2003), we affirmed the trial court’s conclusion that agreements between Taiman and the Government for specialized treatment of goodwill were contractual, although the terms were recorded in agency documents, such as Bank Board Resolutions, rather than executed contracts. The Bank Board Resolutions in that case indicated that Tai-man could use the purchase method of accounting for goodwill so long as it submitted a stipulation that goodwill would be determined and amortized in accordance with Bank Board Memorandum R-31b, which provided for amortization of supervisory goodwill over a period up to forty years. 3 Id. at 1370. *1230 More recently, this court decided LaVan v. United States, 382 F.3d 1340 (Fed.Cir.2004). In that case, FHLBB approval of a 35-year amortization period was reflected in both a Bank Board Resolution and an internal memorandum. Id. at 1346-47. We rejected the Government’s argument that the FHLBB was merely performing a regulatory function when it approved the transaction and agreed with the trial court that the parties had contracted for special goodwill treatment. Id. These cases reflect the relationships formed between the FHLBB and the thrifts as the FHLBB sought help from these institutions to solve the national savings and loan crisis. The cases are not identical, yet all arose in the same regulatory and economic environment, a backdrop against which these cases can be viewed. Despite different circumstances and variations in the documents containing the contract terms, in each of these cases we recognized there was a contractual agreement regarding the treatment of goodwill and its amortization period. Not surprisingly, considering the number of transactions nationally, there have been cases in which this court concluded based on the evidence presented that a contractual agreement with respect to goodwill did not exist. In D & N Bank v. United States, 331 F.3d 1374 (Fed.Cir.2003), this court found that D & N provided no evidence demonstrating that the parties intended to contract to permit special accounting treatment of goodwill. Notably, none of the documents proffered by D & N mentioned goodwill or the accounting treatment thereof. Id. at 1378. Furthermore, the FHLBB supervisory agent testified in his deposition that he could not recall any discussions with D & N about goodwill. Id. at 1379. The complete absence of evidence that the parties even contemplated an agreement regarding goodwill led us to find for the Government on the issue of liability. In another case, First Commerce Corp. v. United States, 335 F.3d 1373 (Fed.Cir.2003), the plaintiff institution began the process of acquiring a troubled thrift by submitting a bid letter to FHLB-India-napolis in which it requested permission to use the purchase method of accounting and amortize supervisory goodwill over a 25-year period. But in its formal merger application, First Commerce requested conventional GAAP treatment of goodwill rather than an extended amortization period. Id. at 1377. This evidence that First Commerce intended to go forward with the transaction without any agreement as to goodwill treatment caused this court to look closely at the documents in search of “mirror image” terms. Id. at 1381. The record showed that the FHLBB had approved the merger and issued a forbearance letter providing that First Commerce could amortize goodwill over a 25-year period. We determined that this qualified as a counteroffer and remanded for the trial court to consider whether First Commerce had accepted the counteroffer. Id. at 1381-82. On remand, the trial court readily found that First Commerce’s later conduct — -its actual use of a 25-year amortization period — constituted an unambiguous acceptance of the Government’s counteroffer, and the court granted summary judgment of liability in favor of First Commerce. First Commerce Corp. v. United States, 60 Fed. Cl. 570, 581-82 (2004). Anderson v. United States, 344 F.3d 1343 (Fed.Cir.2003), is another case in which evidence that parties had not contracted for special goodwill treatment *1231 prompted us to examine carefully the specific documents involved in the transaction. Westport, the acquiring institution, submitted an application in which it requested a 40-year amortization period for goodwill. Id. at 1347. In an internal FHLBB memorandum, the FHLBB staff recommended against granting the request for extended amortization of goodwill but recommended approval of other forbearance requests. Id. According to the memorandum, West-port’s CEO indicated that a goodwill forbearance was not necessary for completion of the transaction. Id. Consistent with the staff recommendation, the FHLBB issued a Forbearance Letter with no mention of goodwill amortization. Id. at 1348. The FHLBB also issued a Resolution, which mentioned goodwill only in the standard clause requiring an accountant’s letter; it made no reference to the use of the purchase method of accounting or to an extended period of amortization for goodwill. Id. at 1355. We found that the Forbearance Letter and the Resolution confirmed what the internal memorandum indicated, i.e., that the FHLBB had not agreed to extended amortization of goodwill as part of the transaction. Id. at 1358. Although in any given case it would be improper to presume the existence or non-existence of a contract, or the terms of a contract, it is true that, viewing the Winstar-related cases as a whole, the pattern of arrangements between the FHLBB and the acquiring financial institutions often had contractual dimensions; the Supreme Court and this court recognized that in the original Winstar cases. The FHLBB was dealing with the worsening crisis in the savings and loan industry by seeking healthy institutions to merge with or acquire failing thrifts and by offering incentives such as the use of supervisory goodwill. The healthy thrifts sought permission from and agreement with the FHLBB in order to safely undertake the salvage efforts the Government so eagerly desired. Since one of the main incentives offered by the FHLBB was the recognition of supervisory goodwill with an extended amortization period — the thread that runs through many of these cases — the question becomes what to believe regarding the parties’ understanding of special treatment of goodwill as a term of the contract. As shown by the cases just discussed, this court has ultimately found the Government liable for breach of contract in these Wins-tar-related cases when the evidence demonstrates that, in light of the discussions between the Government and the acquiring thrift with regard to protections affecting capital requirements, including supervisory goodwill, the parties agreed that the acquiring thrift was to be given the favorable accounting treatment of supervisory goodwill and its amortization. Turning to the case before us, Citizens completed four transactions over the course of four years. With respect to the initial transaction, the acquisition of the Cardinal branches in 1982, the record supports plaintiffs contention that there was a contractual agreement between Citizens and the FHLBB regarding special treatment of supervisory goodwill. According to testimony presented at trial, the parties negotiated the terms of the agreement— Mr. Kirby asked for cash assistance, and Mr. Muldoon offered instead that Citizens could book supervisory goodwill as an asset, which would count toward regulatory capital requirements, and Citizens could amortize the goodwill over an extended period of time. The documents of record— Citizens’ merger application, the Bank Board Resolution, and the conditional approval letter from FHLB-Cincinnati to the FHLBB — -confirm this agreement. Although the record is less than clear as to *1232 whether the parties agreed to a specific amortization period, it is clear enough that they agreed to an extended amortization period. The need to specify the exact length of the amortization period never became a relevant consideration; it got resolved by default in the later paperwork. Using the Cardinal transaction as a model, and at the suggestion of FHLB-Cincinnati, Citizens proceeded to complete three more transactions with failing thrifts. Although the written documents are not as detailed as those in the Cardinal transaction, it is evident from the pattern and circumstances of these transactions that the special treatment of supervisory goodwill was a central part of the agreements, just as it was for the Cardinal transaction. Trial testimony shows that before each transaction Mr. Kirby requested cash assistance, as he had done during the Cardinal negotiation, and Mr. Muldoon instead offered the same goodwill treatment used in the Cardinal transaction— use of the purchase method of accounting to book supervisory goodwill as an asset that would count toward capital reserve requirements and an extended amortization for goodwill. Both Mr. Kirby and Mr. Muldoon understood that the parties were agreeing to the same terms as those in the Cardinal acquisition, and there is no evidence that they intended to alter the arrangement. The clincher is the consequences, as the trial court explained, attendant on the execution of these transactions had supervisory goodwill not been a part. 4 At the time of the Cardinal acquisition, the regulatory capital requirement was four percent. Before the acquisition, Citizens’ regulatory capital was at 5.68 percent of its total capital reserve. After the transaction, with about $38 million in goodwill counting toward Citizens’ capital requirement, the ratio was 4.14 percent. Without the special goodwill treatment, however, Citizens’ post-transaction ratio would have been 0.06 percent, well below the required minimum. The other transactions were the same. Before the Gateway acquisition, Citizens was at 4.17 percent. After the acquisition the ratio was 3.74, still above the minimum three percent required at the time. Without the supervisory goodwill adjustment, the ratio would have been 0.18 percent. Before the Homestead transaction, Citizens’ regulatory ratio was 3.73; after it was 3.18, again above the minimum three percent required. Without the additional $35 million in supervisory goodwill on the books, however, the ratio would have been a negative 2.15 percent, which would not only have rendered Citizens out of capital compliance, but also insolvent. Finally, with the supervisory goodwill allowance provided by the First Federal transaction, Citizens was left with a comfortable 3.4 percent ratio; without it, the ratio was another negative 1.27 percent, out of compliance and insolvent. Messrs. Kirby and Muldoon testified that the transactions would have not occurred without an agreement for special treatment of goodwill because Citizens would have fallen out of capital compliance immediately after each transaction. Citizens would not have decided to complete transactions with such dire consequences; FHLB-Cincinnati would not have approved a transaction if the resulting thrift would not have been in capital compliance. As Justice Souter in Winstar observed, “[i]t would, indeed, have been madness for *1233 respondents to have engaged in these transactions with no more protection than the Government’s reading would have given them, for the very existence of their institutions would then have been in jeopardy from the moment their agreements were signed.” Winstar, 518 U.S. at 910, 116 S.Ct. 2432; see also Barron Bancshares, 366 F.3d at 1378 (finding additional support for existence of a contract in evidence that recapitalized thrift would have been out of compliance from its inception without special goodwill treatment); LaSalle Talman, 317 F.3d at 1370 (“[T]here would be little reason for any thrift to assume added liabilities if that assumption would place it in immediate danger of receivership and dissolution.”). Based on trial testimony regarding the parties’ negotiations, contemporaneous documents, and the circumstances surrounding the transactions at issue, it is apparent that Citizens and the Government intended to enter into a binding agreement governing the transactions at issue. It is further apparent that the agreement was understood to include Citizens’ use of the purchase method of accounting, amortizing of supervisory goodwill over an extended period of time, and counting of supervisory goodwill as an asset for purposes of meeting capital reserve requirements. Those issues were clearly on the table and available as far as the FHLBB was concerned, and were clearly key considerations in the decisions of the Citizens Bank management. The Government asserts that “Fifth Third’s allegation that it would have been ‘irrational,’ even ‘suicidal,’ for it to enter into these transactions without a long-term contract is questionable .... In any event, economic irrationality cannot create contracts.” Appellee’s Br. at 20-21. The latter is of course true, but not the point. More to the point is the fact that there is evidence in this case sufficient to show a contract for long-term amortization of regulatory capital binding on the Government, and, unlike the situation in Anderson, no evidence that Citizens failed to protect itself with enforceable contract rights. The Government further argues that the purpose of these transactions was to buy time until interest rates decreased, and the Government had no need to enter into long-term contracts in order to achieve this goal. That may be true, but it does not define the legal consequences of what the Government actually did. The argument that contract-based transactions were not necessary to the Government’s purpose was made in the original Winstar cases. It was not persuasive then, and the argument has not gained strength by repetition. The trial court reached the conclusion that there were no contractual obligations regarding supervisory goodwill in part because it failed to fully appreciate the context in which these agreements were reached and to give proper weight to the circumstances as well as the evidence, and in part because it misapprehended certain aspects of the relationship between the parties. In terms of context, these are not contracts for goods or services that the Government needs from time to time, and that stand on their own. They are part of a larger context in which the Government enlisted the aid of a major sector of the banking industry by initiating and conducting a nationwide program for protecting the Government’s position as a guarantor of the industry. Though as noted context alone does not create a presumption in favor of any particular outcome, context is relevant to the problem of contract interpretation. Beyond that, the trial court misapprehended the law with regard to certain specific issues. First, the trial court em *1234 phasized the need to memorialize any contract terms in an express written agreement. Fifth Third TV, 56 Fed. Cl. at 691-92. The trial court understood that a single integrated writing is not required in order to establish the existence of a contract in Winstar-related cases, see Cal. Fed., 245 F.3d at 1346-47, yet language in the trial court’s opinion suggests that the trial court believed that all contractual terms must be found in the written documents alone. See, e.g., Fifth Third IV, 56 Fed. Cl. at 692 (“[T]he Federal Circuit has not dispensed with the requirement that the parties produce a written memorialization of their commitment.”). To the extent the trial court applied that understanding of the law in this case, that was error. Evidence other than written documents, such as the testimony in this ease that contract terms were orally negotiated, may not be disregarded. Second, the trial court erred in characterizing the written documents in this case as “truly routine” and as “unadorned, non-customized agency documents.” Id. at 694-95. In some of the cases before this court, plaintiffs have argued that the FHLBB’s mere approval of a transaction demonstrated intent to contract regarding supervisory goodwill, even though there was specific evidence supporting the Government’s claim that it did not intend to agree contractually to special goodwill treatment. See Anderson, 344 F.3d at 1355; D & N Bank, 331 F.3d at 1378. In rejecting that argument, this court described the FHLBB documents approving the transactions as regulatory rather than contractual; we judged that mere approval of a merger by the FHLBB, acting solely in its regulatory capacity, did not create contractual obligations as such. Anderson, 344 F.3d at 1355-56; D & N Bank, 331 F.3d at 1378-79. Those cases, however, do not stand for the proposition that contractual terms cannot be found in agency regulatory documents. As the Government itself has noted, “[a] regulatory action rarely involves a simple affirmative or negative vote.” Appellee’s Br. at 19. In this case, for example, the Resolution approving the Cardinal acquisition was customized to include an extended amortization period for goodwill, a term the parties had negotiated previously. Since standardized agency documents are the way in which regulatory agencies typically memorialize their actions, the FHLBB can memorialize a contractual commitment in agency documents. Thirdly, the trial court was concerned that none of Plaintiffs witnesses “testified that commitments were made on behalf of the Government that the Government would not change the policy of allowing goodwill to count as regulatory capital.” Fifth Third IV, 56 Fed. Cl. at 694. The Supreme Court held, however, that such commitments are not required. The contracts in the Winstar, Glendale, and Statesman transactions did not purport! ] to bar the Government from changing the way in which it regulated the thrift industry. Rather, ... the Bank Board and the FSLIC were contractually bound to recognize amortization periods reflected in the agreements between the parties.... We read this promise as the law of contracts has always treated promises to provide something beyond the promisor’s absolute control, that is, as a promise to insure the promisee against loss arising from the promised condition’s nonoccurrence. Winstar, 518 U.S. at 868-69, 116 S.Ct. 2432. Thus, contractual liability does not require a promise by the Government not to change the regulations; it only requires a promise to recognize the extended amortization period agreed to by the parties. *1235 When the Government changed the regulations, it could no longer make good on its promise and was therefore in breach of the contract. Finally, the trial court discounted the testimony of the participants due to the “rote constancy” of Mr. Kirby’s explanation of the four deals. The “rote” character of the testimony is due to the fact that Mr. Muldoon set the pattern for each deal: on the key contract term, each was to be the same. How could the testimony, otherwise fully credited by the trial court, have been otherwise than by “rote”? By testifying honestly that each deal was exactly the same with regard to long-term amortization of regulatory capital, the evidence was repetitive, not “rote.” Ultimately, the trial court found the testimony of Messrs. Kirby and Muldoon and other FHLB-Cincinnati personnel to be truthful, but, giving their testimony little weight, concluded as a matter of legal interpretation of the transactions that no obligation existed on the part of the Government regarding supervisory goodwill, and therefore no breach of contract occurred. We conclude to the contrary. The totality of the evidence and the circumstances demonstrate that the parties intended to and did create contractual obligations which included the utilization of supervisory goodwill as an accounting treatment for capital compliance. The subsequent events surrounding the enactment and enforcement of FIRREA resulted in a breach of that promise. The trial court erred in concluding otherwise. B. Actual Authority As an alternative argument for affirmance, the Government posits that FHLB-Cincinnati did not have authority to bind the FHLBB to a contract involving supervisory goodwill, an issue the trial court decided against the Government on summary judgment. See Fifth Third II, 52 Fed. Cl. at 640-43. The trial court determined that FHLB-Cincinnati possessed implied actual authority to enter into contracts for treatment of goodwill because such authority was “integral to fulfilling [FHLB-Cincinnati’s] role in FHLBB’s policy to encourage the private acquisition of failing thrifts.” Fifth Third II, 52 Fed. Cl. at 643; see also H. Landau & Co. v. United States, 886 F.2d 322, 324 (Fed.Cir.1989) (“Authority to bind the Government is generally implied when such authority is considered to be an integral part of the duties assigned to a Government employee.” (citation and internal quotation marks omitted)). The Government disputes the trial court’s conclusion, arguing in effect that because some of these acquisition transactions did not include supervisory goodwill, citing D & N Bank and Anderson, permission to offer supervisory goodwill must have been specially required. We agree with the trial court that by April 1982, when FHLB-Cincinnati approved the first transaction in this case, FHLB-Cincinnati had at least implied actual authority to bind FHLBB to promises made to Citizens regarding the use of supervisory goodwill. In February 1982, the FHLBB delegated authority to its regional board principal supervisory agents 5 (“PSAs”) to “approve merger applications in which goodwill is included in assets” and to allow the PSAs “to agree to certain forbearances in approving supervisory mergers which are currently granted by the Board.” Delegation of Authority Regarding Merger Approvals, 47 Fed.Reg. 8152 (Feb. 25, 1982) (the “1982 delega *1236 tion”). The 1982 delegation noted that the FHLBB had provided the PSAs with Memorandum R-31b, which established guidelines for allowing the use of supervisory goodwill and an extended amortization period. Id. The 1982 delegation further indicated that merger applications raising significant policy issues for which the FHLBB had not established a formal position should be referred to the FHLBB. Id. at 8153. Like the trial court, see Fifth Third II, 52 Fed. Cl. at 642 & n. 3, we read the 1982 delegation as evidence of the FHLBB’s intention not to review individual mergers involving supervisory goodwill; the FHLBB gave the PSAs authority to follow the established policy regarding supervisory goodwill set forth in Memorandum R-31b. The trial court further noted that the Supreme Court’s Winstar decision recognized that the ability to offer supervisory goodwill as an asset for regulatory capital purposes and to allow extended amortization of goodwill was an essential tool for encouraging acquisition of failing thrifts. Id. at 642-43 (citing Winstar, 518 U.S. at 849-50, 863-64, 116 S.Ct. 2432). Considering all of the circumstances, we agree with the trial court that the authority to offer special treatment of goodwill as a contractual incentive was integral to the PSAs’ ability to encourage supervisory mergers, notwithstanding the few instances in which plaintiff institutions have been unable to prove the existence of such a contract. 6 C. Damages Fifth Third challenges one aspect of the trial court’s partial summary judgment on damages- — -the rejection of Fifth Third’s cover damages model. ,See Fifth Third III, 55 Fed. Cl. at 242-44. Fifth Third presented other damages theories at trial; on remand those claims will still be alive because the trial court denied the Government’s motion for judgment on partial findings with respect to damages. The issue of cover damages, however, is properly before us on appeal. The earlier order granting partial summary judgment as to some damages claims was not an ap-pealable judgment when entered, but when the trial court entered final judgment after granting the Government’s Rule 52(c) motion on liability, the earlier disposition merged into the final judgment and is reviewable. See Glaros v. H.H. Robertson Co., 797 F.2d 1564, 1573 (Fed.Cir.1986); 15B Charles Alan Wright & Arthur R. Miller and Edward H. Cooper, Federal Practice and Procedure § 3914.28 (2d ed.1992). Plaintiff proposed its cover damages theory as an alternative to lost profits. The lost profits theory, though not absolutely barred in Winstar-related cases as a matter of law, has not been susceptible of proof due to its speculative nature. See Cal. Fed. Bank v. United States, 395 F.3d 1263, 1270-73 (Fed.Cir.2005); Glendale Fed. Bank, FSB v. United States, 378 F.3d 1308, 1313 (Fed.Cir.2004). Plaintiffs cover damages theory suffers from the same problem. A cover damages claim measures the value of the goodwill destroyed when enactment of FIRREA caused the Government’s breach by determining the cost of *1237 substituting a different form of capital for the lost goodwill. Cover damages are thus a form of expectancy damages, restoring the plaintiff to the position it would have been in but for the breach by replacing, or covering, the value of an asset taken away by the breach. Plaintiffs proposed claim calculates the hypothetical cost of replacing goodwill capital with tangible capital in the form of preferred stock. According to the model, the replacement-of-capital cost includes transaction costs Citizens would have incurred in issuing preferred stock, and dividends Citizens would have been required to pay to investors who bought preferred stock. There are two problems with Plaintiffs particular cover damages theory. First, Citizens was a mutual organization at the time of the breach and therefore could not have issued preferred stock without first converting to stock form. Second, even if Citizens had been able to issue preferred stock as set forth in Plaintiffs model, Citizens did not actually do so. Plaintiffs cover damages theory is based entirely on hypothetical costs that were never actually incurred. In that regard, this case is distinguishable from Home Savings of America v. United States, 399 F.3d 1341, 1353-55 (Fed.Cir.2005), in which we affirmed the trial court’s cover damages award based on costs incurred when the plaintiff thrift actually raised new capital to replace lost supervisory goodwill. In contrast, Plaintiffs cover damages claim, like lost profits claims, is highly speculative, and we cannot fault the trial court’s decision to preclude Fifth Third from presenting its cover damages claim at trial. CONCLUSION The judgment of no liability is reversed. The trial court’s rejection of Fifth Third’s cover damages theory is affirmed. The matter is remanded to the trial court for determination of damages, if any, to be awarded. AFFIRMED-IN-PART, REVERSED-IN-PART, and REMANDED Footnotes [1] Fifth Third acquired Citizens in 1998 and became the successor-in-interest to Citizens' claims in this case. [2] The Court of Federal Claims has a special procedures order for Wiwsiar-related cases utilizing a short form motion for summary judgment on liability. [3] Bank Board Memorandum R-31b, issued in September 1981, set forth the FHLBB's guidelines on how the acquiring institution could count supervisory goodwill as an intangible asset using the purchase method of accounting. The Memorandum limited the amortization period of supervisory goodwill *1230 to forty years or less. See Winstar, 64 F.3d at 1541 & n. 4. [4] The following data are taken from the trial court’s factual findings, Fifth Third TV, 56 Fed. Cl. at 677 (Cardinal); id. at 679 (Gateway); id. at 681 (Homestead); id. at 682 (First Federal). [5] A 'Principal Supervisory Agent' was the president of the regional Federal Home Loan Bank in which the resulting institution in a proposed merger is a member. [6] The trial court interpreted the 1982 delegation as giving the PSAs express authority to enter into certain enumerated forbearance agreements, but not goodwill agreements. Fifth Third II, 52 Fed. Cl. at 642. Accordingly, the trial court went on to address whether the PSAs had implied actual authority. Because we accept the trial court’s implied authority analysis, we need not consider Plaintiffs alternative argument that the 1982 delegation gave the PSAs express actual authority or its theory that actual authority was established by ratification.