Anchor Savings Bank, FSB v. United States

BRYSON, Circuit Judge. This is one of the last of the “Winstar ” cases arising out of the savings and loan crisis of the late 1970s and early 1980s. See United States v. Winstar Corp., 518 U.S. 839, 116 S.Ct. 2432, 135 L.Ed.2d 964 (1996). During those years, high interest rates and inflation placed hundreds of savings and loan institutions, or “thrifts,” in severe financial distress. In order to prevent the thrifts’ collapse and the resulting burden on the federal government, which insured many of the thrifts’ depositors, the government developed a plan to induce healthy financial institutions to take over the failing thrifts through so-called “supervisory mergers.” Because the troubled thrifts were unattractive investments on their own, the government offered significant incentives to the acquiring institutions. Those incentives included cash and cash substitutes in the form of what was called “supervisory goodwill.” Supervisory goodwill was an accounting credit equal to the negative net worth of the thrift. Pursuant to the supervisory merger agreements, the acquiring institution was permitted to treat supervisory goodwill as an asset and to amortize the goodwill over a period of many years. That arrangement enabled the acquiring institution to satisfy its regulatory capital requirements while working to integrate and rehabilitate the failing thrift. Anchor Savings Bank was among the institutions that contracted with the government in the 1980s to acquire several failing thrifts. Anchor was a relatively strong institution that had already engaged in significant expansion of its business and was positioning itself to become a *1359 major player in the mortgage banking industry. Between 1982 and 1985, Anchor acquired the assets and assumed the liabilities of four failing thrifts in a series of supervisory mergers arranged by the government. As part of the transactions, the government promised Anchor that it could use more than $550 million in supervisory goodwill in calculating its regulatory capital and that it could amortize that supervisory goodwill over a period of 25 to 40 years. As the government understood, the major attraction of the acquisition agreements to Anchor was the favorable regulatory treatment of supervisory goodwill. Without those forbearances, Anchor would have failed to satisfy its regulatory requirements as a result of acquiring so much liability. In June 1988, Anchor purchased Residential Funding Corporation (“RFC”), a mortgage banking company. That purchase was consistent with Anchor’s long-term business plan to become more involved in mortgage banking as a way to insulate itself from operating deficits created by the “interest rate spread” — the difference between the high interest rates it had to pay on deposits at the time and the low interest rates it was receiving on the fixed-rate mortgages in its loan portfolio. Anchor had already begun to implement its plan through its 1983 supervisory merger with mortgage-banking enterprise Suburban, which became Anchor Mortgage Services (“AMS”). AMS acquired and sold whole mortgage loans, but it retained the servicing rights on those loans so as to generate regular fees for the bank independent of interest rates. Like AMS, RFC specialized in acquiring whole mortgage loans and reselling them in the secondary market. However, RFC served a niche market as a “conduit” specializing in wholesale originations of jumbo mortgages for resale as “private-label” mortgage-backed securities (“MBS”). 1 RFC performed “master servicing” for the MBS, generating steady servicing fees. RFC was an industry leader at the time Anchor purchased it. In the first quarter of 1988, RFC was the largest issuer of private MBS in the nation. RFC generated over $10.5 million in net profit in its first year under Anchor and $7.8 million in net profit during the first seven months of the following year. The business was highly successful and fit well with Anchor’s long-term business plans^ — so well, in fact, that Anchor largely discontinued its operation of AMS in favor of RFC. In mid-1989, Anchor’s CEO wrote that RFC “continues to fly” and was “authorized to double its volume in 1990.” At about the same time, Anchor and RFC developed a business plan designed to expand RFC’s business into other areas. On August 9, 1989, Congress enacted the Financial Institutions Reform, Recovery, and Enforcement Act, Pub.L. No. 101-73, 103 Stat. 183 (1989) (“FIR-REA”). The new statute — and particularly its implementing regulations, which were announced in October 1989 — effectively terminated the favorable treatment of supervisory goodwill that had been promised to Anchor at the time of the *1360 supervisory mergers. The sudden eradication of more than half a billion dollars of regulatory capital caused Anchor to fall out of capital compliance by more than $800 million. Facing the threat of seizure and liquidation by the government, Anchor scrambled to raise the necessary capital through a swift series of asset sales. Those sales resulted in the divestiture of RFC and a majority of Anchor’s branch offices. Anchor sold RFC in March 1990 to General Motors Acceptance Corporation (“GMAC”) for $64.4 million. Under GMAC’s ownership, RFC continued to operate with largely the same management, and it continued to implement the Anchor-developed plan to expand its business. Unlike some other thrifts at the time, Anchor survived FIRREA, and by July 1993 it received a “well capitalized” rating. At that point, it was able to resume its long-term business plans. In January 1995, Anchor merged with Dime Savings Bank of New York. Like post-FIRREA Anchor, Dime lacked a sophisticated mortgage banking operation. Accordingly, in October 1997 the Anchor/Dime entity acquired the North American Mortgage Company (“NAMCO”) for $351 million. Like RFC, NAMCO engaged mostly in wholesale mortgage origination and was a major player in the secondary mortgage market. NAMCO also provided and serviced individual mortgages, generating regular fees. Unlike RFC, however, NAMCO operated primarily in the market for mortgages that met the underwriting criteria of government-sponsored entities (the Federal National Mortgage Association, the Federal Home Loan Mortgage Corporation, and the Government National Mortgage Association). Meanwhile, on January 13,1995, Anchor filed suit in the Court of Federal Claims, alleging that the adoption of FIRREA and its implementing regulations breached the government’s obligations under the supervisory merger contracts. In accordance with the Supreme Court’s decision in Winstar, which had held that those actions could constitute a breach of contract by the government, the trial court concluded that the United States had breached its supervisory merger contracts with Anchor. The trial court then conducted a five-week trial on damages. Following the trial, the court entered an award of $356,454,910.91 in damages to Anchor and issued a detailed opinion explaining its decision. The damages award consisted of lost profits from RFC’s operations after Anchor sold RFC to GMAC; mitigation costs for Anchor’s purchase of NAMCO to replace RFC; expectancy damages for stock proceeds Anchor would have received from a post-FIRREA stock offering if it had retained RFC; damages from the sale of portions of Anchor’s branch network; increased FDIC insurance premiums; and “wounded bank” damages. Anchor Sav. Bank, FSB v. United States, 81 Fed.Cl. 1, 153 (2008). More than 90 percent of the award was attributable to Anchor’s sale of RFC. On appeal, the government challenges the trial court’s award of damages related to the sale of RFC. The government argues that the trial court erred in four principal respects: (1) it improperly applied the law of foreseeability and erred in finding that the type and magnitude of damages from Anchor’s sale of RFC were reasonably foreseeable; (2) it erred in finding that the government’s breach caused Anchor to sell RFC; (3) it erred by measuring damages based on RFC’s profits after it was sold to GMAC, rather than RFC’s market value at the time of the breach or the sale; and (4) it erred in finding that the NAMCO purchase constituted mitigation for the loss of RFC. In its *1361 cross-appeal, Anchor argues that the trial court made a calculation error that erroneously reduced Anchor’s damages by more than $63 million. Anchor has also filed a conditional cross-appeal in which it contends that if this court does not affirm the award of expectancy damages, it should reverse the trial court’s decision denying Anchor’s alternative claim for reliance damages. I Damages for breach of contract are designed to make the non-breaching party whole. One way to accomplish that objective is to award “expectancy damages,” i.e., the benefits the non-breaching party would have expected to receive had the breach not occurred. Glendale Fed. Bank, FSB v. United States, 239 F.3d 1374, 1380 (Fed.Cir.2001). Expectancy damages “are often equated with lost profits, although they can include other damage elements as well.” Id. To recover lost profits for breach of contract, the plaintiff must establish by a preponderance of the evidence that (1) the lost profits were reasonably foreseeable or actually foreseen by the breaching party at the time of contracting; (2) the loss of profits was caused by the breach; and (3) the amount of the lost profits has been established with reasonable certainty. Cal. Fed. Bank v. United States, 395 F.3d 1263, 1267 (Fed.Cir.2005); Energy Capital Corp. v. United States, 302 F.3d 1314, 1324-25 (Fed.Cir.2002). Each of those inquiries presents a question of fact as to which we exercise “clear-error” review. Landmark Land Co. v. FDIC, 256 F.3d 1365, 1379 (Fed.Cir.2001); Bluebonnet Sav. Bank, F.S.B. v. United States, 266 F.3d 1348, 1356-58 (Fed.Cir.2001). Although we have noted that the lost profits theory of damages in Winstar cases often fails because it is too speculative, “[w]e have not ... barred as a matter of law the use of expectancy/lost profits theory.” Glendale Fed. Bank, FSB v. United States, 378 F.3d 1308, 1313 (Fed.Cir.2004). In its appeal of the lost profits award relating to Anchor’s sale of RFC, the government challenges the trial court’s conclusions as to each of the three elements of a lost profits claim set forth above. A The government first argues that the trial court erred in finding that the damages relating to the RFC sale were reasonably foreseeable. The government contends that the court should have found those damages to be available only if the government could have foreseen that Anchor would purchase and then sell RFC or an asset like RFC. In the trial court’s view, the proper question was whether a reasonable person in the government’s position could have foreseen the general type of use Anchor made of its supervisory goodwill and that a profitable enterprise would be sacrificed if Anchor lost that supervisory goodwill. It was enough, the court found, that the government could reasonably have foreseen that its breach would force Anchor to divest itself of profitable assets purchased in reliance on the benefits conferred by the contracts. 2 Anchor, 81 Fed. Cl. at 80-81. *1362 According to the government, the breaching party must be able to foresee the particular asset or type of asset to be purchased and sold in order to appreciate the risk of loss flowing from a breach. The government argues that when the contract was formed neither Anchor nor any other thrift had entered into RFC’s line of business, which was novel and considered risky at that time. Therefore, the government contends that it could not have foreseen: (1) that Anchor would purchase that type of asset, let alone that it would later sell that asset to attain capital compliance in the wake of FIRREA; and (2) that RFC would be so profitable, thus substantially increasing the magnitude of the damages flowing from a contract breach. The test that the government proposes is too narrow. In previous Winstar cases we have recognized that the particular details of a loss need not be foreseeable, as long as the plaintiff bank’s “need to raise capital in the event of a breach was foreseeable.” See Fifth Third Bank v. United States, 518 F.3d 1368, 1376 (Fed.Cir.2008) (plaintiff was entitled to compensation for damages incurred in generating regulatory capital, through branch sales, to replace lost goodwill and was not required to establish the foreseeability of the poor economic conditions that created difficulties in raising capital). “If it was foreseeable that the breach would cause the other party to obtain additional capital, there is no requirement that the particular method used to raise that capital or its consequences also be foreseeable.” Citizens Fed. Bank v. United States, 474 F.3d 1314, 1321 (Fed.Cir.2007) (plaintiff was entitled to compensation for negative tax consequences incurred in raising capital to replace lost goodwill; it did not need to prove that the specific tax consequences were foreseeable). The government contends that the loss at issue in this case was not foreseeable because Anchor did not own RFC at the time it entered into the relevant contracts. It is logical, however, to apply the reasoning of Fifth Third Bank and Citizens Federal Bank to situations in which the government’s breach causes a bank to sell an asset that was purchased prior to FIR-REA in reliance on the supervisory goodwill obtained in the mergers. First, the importance of supervisory goodwill to the merger agreements is undeniable. Goodwill was critical to the government’s ability to induce banks to enter into the supervisory mergers; without that inducement, the banks would have had little incentive to purchase failing thrifts with substantial net liabilities. See, e.g., Winstar, 518 U.S. at 921, 116 S.Ct. 2432 (Scalia, J., concurring) (characterizing favorable regulatory treatment as “an essential part of the quid pro quo ” of the merger contracts with the government); Cal. Fed. Bank, FSB v. United States, 245 F.3d 1342, 1349 (Fed.Cir.2001) (“The continued use of supervisory goodwill as regulatory capital for the entire 35-40 year amortization period initially promised was ... a central focus of the contract and the subject of the government’s breach.”). Second, the government expected the infusion of regulatory capital in the form of supervisory goodwill to allow the acquiring institutions to make profitable investments that could rehabilitate the failing thrifts. As the trial court found, “the government needed and expected an acquiring thrift to leverage its goodwill into profitable investments because retained earnings were essential to replace the regulatory capital that a thrift lost each year when its goodwill was amortized.” Anchor, 81 Fed.Cl. at 78. By providing that supervisory goodwill would be amortized over time, the government’s contracts encoui’aged institutions such as Anchor to invest early and aggressively so *1363 as to take advantage of the supervisory goodwill before it was gone. Under these circumstances, it was reasonable for the trial court to apply the foreseeability rule in a manner that encompassed both the purchase and the ultimate sale of assets. The trial court thus properly required only a general showing that (1) the government could reasonably have foreseen that the influx of supervisory goodwill under the contracts would cause the acquiring institution to make investments in order to generate profit and rehabilitate the failing acquired thrifts; and (2) the government could reasonably have foreseen that a breach of contract would cause the acquiring institution to sell off those very investments in order to raise capital to meet regulatory requirements. The government argues that our decision in Old Stone Corp. v. United States, 450 F.3d 1360 (Fed.Cir.2006), stands for the proposition that damages cannot be awarded unless the loss that actually occurred — here, the loss of the profits that RFC would have generated if it had not been sold — was foreseeable. Old Stone, however, involved quite different facts, and the analysis employed by the court in that case does not preclude a damages award here. Like this case, Old Stone involved supervisory mergers that generated supervisory goodwill for the acquiring bank. This court in Old Stone allowed a recovery of the payments made by the plaintiff to replace the regulatory capital eliminated by FIRREA. However, the court denied an award of damages flowing from the bank’s seizure, which resulted from “other problems” that were unrelated to the loss of supervisory goodwill. According to the bank’s theory, it was entitled to a damages award equal to the full amount that it had contributed to the acquired thrifts in the supervisory mergers because (1) the breach had deprived the bank of regulatory capital; (2) when the bank encountered the “other problems” unrelated to FIR-REA, it was forced to sell valuable assets that it would not have had to sell if it had retained the regulatory capital from the supervisory mergers; and (3) the valuable assets, if they had been retained, would have been sufficient to enable the bank to avoid seizure. 450 F.3d at 1376. The Old Stone court pointed out that there was no testimony in that case “that would suggest that the seizure of the thrift by itself was a foreseeable result of the shrinkage,” and that the trial court did not find that the forced shrinkage of the bank had “a foreseeable relationship to the seizure.” 450 F.3d at 1376. Because the plaintiff had “failed to establish that [the] extended chain of causation” leading to the seizure of the bank was foreseeable at the time of contract formation, this court rejected the plaintiffs reliance damages claim. Id. The court acknowledged that recovery can be predicated on the need to replace regulatory capital or on the failure of a thrift due to a deficiency in regulatory capital. Id. at 1376-77. In the case before it, however, the court held that the bank’s theory of foreseeability — “that the seizure resulted from the fact that the replacement capital was unavailable to resolve other problems not caused by FIR-REA” — was both unsupported by the evidence and too attenuated to support a damages award. Id. at 1377. In the course of its opinion, the court in Old Stone noted that “even if the need for replacement capital was foreseeable, that hardly establishes that the adverse consequences alleged to flow from the need to make infusions were foreseeable.” The court then quoted from the Restatement of Contracts, which provides that the “mere circumstance that some loss was foreseeable, or even that some loss of the same *1364 general kind was foreseeable, will not suffice if the loss that actually occurred was not foreseeable.” Restatement (Second) of Contracts § 351 cmt. a (1981), quoted in Old Stone, 450 F.3d at 1376. While the quoted comment from the Restatement limits damages awards when unforeseeable events result in enhanced losses to the non-breaching party, it does not suggest that the specific loss in question must have been within the contemplation of the parties at the time of contracting. The Restatement makes that point clear, stating that “the party in breach need not have made a ‘tacit agreement’ to be liable for the loss. Nor must he have had the loss in mind when making the contract, for the test is an objective one based on what he had reason to foresee.” Restatement § 351 cmt. a. As a leading commentator has explained, summarizing the foreseeability limitation on expectancy damages, [t]he magnitude of the loss need not have been foreseeable, and a party is not disadvantaged by its failure to disclose the profits that it expected to make from the contract. However, the mere circumstance that some loss was foreseeable may not suffice to impose liability for a particular type of loss that was so unusual as not to be foreseeable. E. Allan Farnsworth, Farnsworth on Contracts § 12.14, at 262 (3d ed.2004). The principle recited in Old Stone is consistent with the generally recognized rule that foreseeability for purposes of determining contract damages requires “merely that the injury actually suffered must be one of a kind that the defendant had reason to foresee and of an amount that is not beyond the bounds of reasonable prediction.” 11 Joseph M. Perillo, Corbin on Contracts § 56.7, at 108 (rev. ed.2005). “Just as reason to foresee does not mean actual foresight, so also it is not required that the facts actually known to the defendant are enough to enable the defendant to foresee that a breach will cause a specific injury or a particular amount in money.” Id.; see also Farns-worth, supra, § 12.14, at 260-61 (“There is no requirement that the breach itself or the particular way that the loss came about be foreseeable.”). Contrary to the government’s contention, Old Stone does not depart from those general principles and impose a restrictive test of foreseeability in which the specific mechanism of loss must be foreseeable. As this court stated in Citizens Federal Bank, “If it was foreseeable that the breach would cause the other party to obtain additional capital, there is no requirement that the particular method used to raise that capital or its consequences also be foreseeable.” 474 F.3d at 1321. In light of the applicable test, the trial court did not commit clear error in finding that Anchor’s loss was foreseeable. As the trial court found (and the government does not dispute), the government expected that Anchor would “take advantage of the special treatment of supervisory goodwill to acquire additional assets, grow the bank, and generate new profits that would fill the capital void created by the annual amortization of goodwill.” Anchor, 81 Fed.Cl. at 79. The evidence also supports the court’s conclusion that it was foreseeable that the withdrawal of favorable regulatory treatment for goodwill would deprive Anchor of capital, that Anchor would have difficulty or be delayed in raising the required capital, and that Anchor would therefore have “to sacrifice assets and forgo the very profit opportunity that made the contracts appealing in the first place.” Id. at 78-79; see also Restatement § 351 cmt. e (where “the lender has reason to foresee that the borrower will be unable to borrow elsewhere or will be delayed in borrowing elsewhere, the *1365 lender may be liable for much heavier damages based on the borrower’s inability to take advantage of a specific opportunity ..., his having to postpone or abandon a profitable project ..or his forfeiture of security for failure to make prompt payment”). The evidence also shows that the specific type of investment made by Anchor was among the options available to Anchor at the time of contracting. As the market for private MBS issuers was emerging and thriving during that period, it was reasonable to expect that thrifts would gravitate toward that industry, as they were already familiar with many aspects of that trade. The government even sold Anchor a mortgage-banking enterprise (Suburban, later AMS) through one of the early supervisory mergers. Anchor described that acquisition in its 1983 Annual Report as a means of “positioning itself for a major entrance into nationwide mortgage origination and secondary market activity.” A memorandum from the Federal Home Loan Bank Board to Anchor describing the large amount of new capital needed to sustain and grow Suburban is a further indication that the government was aware that Anchor was becoming more involved in that type of capital-intensive business. In light of that evidence, we reject the government’s contention that the trial court committed clear error in finding the type of Anchor’s damages to be foreseeable. We also sustain the trial court’s finding that the magnitude of Anchor’s damages was foreseeable. See Anchor, 81 Fed.Cl. at 79 (finding that the “sheer volume of goodwill” rendered the magnitude of the lost profits from RFC’s sale foreseeable). This case involved an exceptionally large amount of supervisory goodwill. The court found that the contracts, which provided Anchor with more than $550 million in regulatory capital, permitted it to leverage more than ten billion dollars in assets while still meeting its regulatory capital requirements. Id. Moreover, as the court noted, the supervisory mergers at issue did not involve “small ‘mom and pop’ thrifts or single-branch institutions”; the agreements constituted “big business.” Id. Anchor was expected to contribute nearly $80 million to Suburban alone over a six-year period, and the acquired thrift was expected to experience nearly $100 million in losses during that same period. In light of that substantial anticipated commitment, it was reasonable for the trial court to conclude that the government expected that Anchor would use each dollar of regulatory capital allotted to create a dollar of capital through future profits. Id. at 81. The trial court thus did not commit clear error in finding that it was foreseeable that the breach would result in lost profits to Anchor in an amount commensurate with the ultimate award for lost profits. 3 B The government next asserts that the trial court erred when it found that the breaching provisions of FIRREA, which deprived Anchor of more than $500 million *1366 in regulatory capital, caused Anchor to sell RFC. The government asserts that, even absent the breach, Anchor would have sold RFC because the risk-based capital requirements that were enacted as part of FIRREA (and did not breach the supervisory merger agreements) would have made it impossible for Anchor to operate RFC profitably. Under the risk-based capital provisions, a thrift was required to “risk-weigh” its assets, so that a smaller percentage of “riskier” assets could be counted toward its regulatory capital obligations. That requirement was a particular problem for RFC, which relied on a senior/subordinated MBS structure in which the securities in the subordinated class absorbed a large percentage of the risk, were of non-investment grade, and were frequently held for extended periods for “credit enhancement” purposes. The trial court disagreed with the government’s contention that it was the risk-based capital requirements that caused Anchor to sell RFC, rather than the changes in the treatment of supervisory goodwill. The court made detailed findings as to that issue, devoting 16 pages of its opinion to its findings on causation and another 20 pages to its discussion of the expert testimony regarding Anchor’s ability to retain and operate RFC in a hypothetical non-breach world. Anchor, 81 Fed.Cl. at 59-75, 98-117. Causation is an “intensely factual determination.” Cal. Fed. Bank, 395 F.3d at 1270. While a “causal connection between the breach and the loss of profits must be definitely established,” id. at 1268, the breach need not be the sole factor or the sole cause of the plaintiffs loss. Although a plaintiff may recover only for those losses that would not have occurred but for the breach, “[t]he existence of other factors operating in confluence with the breach will not necessarily preclude recovery based on the breach.” Id. We find no clear error in the trial court’s determination that Anchor established a causal connection between the government’s breach and the sale of RFC. The court’s findings are well supported by the record and reflect a careful weighing of documentary and expert evidence. The court relied on contemporaneous documents and “highly credible” testimony from Anchor and RFC officials, which revealed that Anchor never planned to sell RFC unless absolutely necessary, because RFC was performing well beyond Anchor’s expectations and provided Anchor with its pick of “spectacular” quality loans. Anchor, 81 Fed.Cl. at 60-61. Even with the threat of FIRREA looming in mid-1989, Anchor’s contemporaneous documents describe the sale of RFC as merely a “[b]ack up plan” and note that RFC “continues to fly” and is “authorized to double its volume in 1990.” The documents support the trial court’s conclusion that only after the full extent of FIR-REA’s impact became known — including its effect in creating an enormous capital shortfall for Anchor — did Anchor’s management conclude that the sale of RFC was necessary as part of company-wide “triage” efforts. Id. at 60, 68. The evidence also supports the trial court’s finding that the “risk-based capital provisions required the sale of RFC only because of Anchor’s capital deficiency resulting from the breach.” Anchor, 81 Fed.Cl. at 70. In other words, but for the breach, Anchor’s supervisory goodwill would have given it an ample capital cushion with which to continue operating RFC even under the risk-based capital regime. The government’s own expert acknowledged that, absent the breach, Anchor would have been well capitalized (and able to absorb additional assets) when the risk- *1367 based capital provisions went into effect. Anchor’s MBS expert, Mr. Lederman, whose testimony the trial court deemed “extremely credible,” explained that Anchor and RFC could have employed other credit enhancement options (besides retaining the subordinated securities) by selling the subordinated pieces into the secondary market or using pool insurance to credit enhance the securities. The trial court was entitled to weigh the competing expert testimony on that issue and find Anchor’s evidence more persuasive. The government asserts that in a non-breach world Anchor and RFC would not or could not have implemented Mr. Leder-man’s suggestions of eliminating the risky “B” pieces and purchasing pool insurance as a means of dealing with the risk-based capital requirements. Contrary to the government’s contention, not only were Anchor and RFC aware of Mr. Leder-man’s options for alternative credit enhancement, but they had actually used both measures prior to the sale of RFC. RFC sold $66 million in “B” pieces in early 1990. Anchor, 81 Fed.Cl. at 71 (citing Mr. Lederman’s testimony that his own investment company participated in the purchase of all of those “B” pieces). In addition, Mr. Lederman testified that RFC had used pool insurance in 1989, after the enactment of FIRREA. Id. Documents from March and December 1989 confirm that RFC considered, and ultimately used, pool insurance in at least 37 different transactions. Finally, it was not error for the trial court to discount the government’s evidence purporting to show that Anchor sold RFC because of the risk-based capital requirements. The government introduced evidence that Anchor’s executives had made public statements, including a 1991 filing with the Securities and Exchange Commission, suggesting that it was the risk-based capital requirements that caused RFC’s sale. The trial court, however, noted that it had difficulty evaluating those documents, particularly in light of credible contrary testimony by Anchor’s witnesses, because the government never presented the documents to Anchor’s witnesses at trial. Moreover, the court observed that nearly all of Anchor’s public statements (including its SEC filing) were made after the breach and must be read in the context of Anchor’s severely impaired capital position. The post-breach statements thus shed no light on whether the risk-based provisions would have forced the sale in the absence of any loss of supervisory goodwill. In fact, the statements are consistent with the court’s finding that, while the risk-based capital requirements operated in confluence with the breach, Anchor’s post-breach, impaired capital condition was the but-for cause of RFC’s sale. 4 We thus uphold the trial court’s finding of a causal connection between the government’s breach of contract and Anchor’s sale of RFC. C The government argues that the trial court erred as a matter of law by measur *1368 ing damages based on the profits RFC earned after Anchor sold it. According to the government, damages should have been measured based on RFC’s market value at the time.of the contract breach or RFC’s sale. The trial court first addressed whether to measure damages based on RFC’s fair market value at the time of its sale to GMAC shortly after the breach. The court noted that, in early 1990, Goldman Sachs valued RFC at approximately $60-70 million, which is consistent with the $64.4 million price paid for RFC in March 1990. Nevertheless, the court concluded that the Goldman Sachs appraisal did not reflect RFC’s fair market value because in the tumultuous post-FIRREA climate all prospective buyers knew that Anchor was desperate to sell RFC quickly. Anchor, 81 Fed.Cl. at 86-87. The court then addressed whether to measure damages based on RFC’s post-breach profits as a GMAC subsidiary. The court concluded that the latter method was more reliable, particularly because, after its acquisition by GMAC, RFC continued to operate under largely the same management and in accordance with the business plan developed under Anchor. The court concluded that “where the goal of the contracts was to enable the plaintiffs to generate ongoing profits that would, over time, fill the shortfall between the various acquired institutions’ assets and liabilities, it seems especially and unreasonably static and wooden to limit the expectation interest to the then-present value of individual assets.” Anchor, 81 Fed.Cl. at 89. The government argues that the trial court erred as a matter of law by considering post-breach evidence of damages, when the only relevant evidence concerns RFC’s market value at the time of the March 1990 sale. The government cites this court’s decision in First Federal Lincoln Bank v. United States, 518 F.3d 1308 (Fed.Cir.2008), for the proposition that damages for the loss of “income-generating property” must be measured by the asset’s market value as of the time the property is lost, not by the loss of the profits the asset could have produced in the future. Id. at 1317. According to the government’s theory, the market valuation should already reflect RFC’s expected future risks and future profit stream. While the trial court did not consider Lincoln, which issued just nine days before its own opinion, we conclude that the trial court’s approach is consistent with our precedent, including Lincoln. In Lincoln, this court reviewed a damages award to a thrift that had prevailed on its Winstar-related breach-of-contract claim. 518 F.3d at 1311. The damages were allegedly incurred when the thrift reduced its deposit base to raise regulatory capital following the enactment of FIR-REA. Although the trial court permitted the plaintiff to recover the lost value of the deposits as of the time of trial, it rejected the plaintiffs theory that it should recover for future lost profits that would have been generated by the forgone deposits. On appeal, the parties disputed only the ruling as to the claim for the lost value of the deposits. Characterizing that claim as one “for loss of income-generating property,” this court stated that “damages for lost income-producing property is properly determined as of the time the property is lost (usually the time of the breach) because the market value of the lost property reflects the then-prevailing market expectation as to the future income potential of the property.” Id. at 1317. For support, the court cited the Second Circuit’s decision in Schonfeld v. Hilliard, which stated that *1369 [w]hen the defendant’s conduct results in the loss of an income-producing asset with an ascertainable market value, the most accurate and immediate measure of damages is the market value of the asset at the time of breach — not the lost profits that the asset could have produced in the future. 218 F.3d 164, 176 (2d Cir.2000). Based on that language, the government argues that Lincoln adopted the rule that damages for the loss of income-generating assets must always be measured by the market value of the lost asset at the time it is lost, and never by evidence of post-breach profits generated by the asset after it has been purchased by a third party. The government’s interpretation of Lincoln is incorrect for two reasons. First, Lincoln recognized two permissible methods of measuring damages: (1) the market value of a lost income-producing asset (“lost asset” or “lost asset value” damages); and (2) future lost profits that could have been derived from the lost income-producing asset (“lost profits” damages). Both Lincoln and Schonfeld discuss the award of lost asset damages as an alternative to lost profits damages. See Lincoln, 518 F.3d at 1317 (“First Federal has [correctly] recognized that ‘The Trial Court Awarded Damages Based on the Value of the Foregone Deposits, Not the Earnings that the Foregone Deposits Would have Generated.’ ”); Schonfeld, 218 F.3d at 176 (“Although lost profits and ... lost asset damages are both consequential, rather than general, in nature, courts have universally recognized that they are separate and distinct categories of damages.”). Neither decision mandates that one measurement method must invariably be used, as opposed to the other. Second, the procedural posture and factual circumstances of Lincoln differ from those in the present case. In Lincoln, only lost asset damages were at issue on appeal. The plaintiff did not appeal the denial of lost profits damages, which the trial court had considered and rejected as speculative. Lincoln, 518 F.3d at 1315, 1317. Therefore, Lincoln’s focus on a market value measurement stemmed from the posture of that case, not from any conclusion that market value was the only permissible method of measuring damages under the law. 5 Moreover, rather than favoring one method of measuring damages over the other, Lincoln focused more specifically on the proper timing of the damages measurement, ruling that the trial court had erred in valuing the lost deposits as of the date of trial rather than the date the breach. Id. at 1317-18. Neither Lincoln nor any of our other Winstar decisions bars the court from considering post-breach evidence in determining the quantum of a lost profits award. As we have repeatedly recognized, the rule favoring the measurement of damages as of the time of the breach “does not apply ... to anticipated profits or to other expectancy damages that, absent the breach, would have accrued on an ongoing basis over the course of the contract. In those circumstances, damages are measured throughout the course of the contract.” Energy Capital Corp., 302 F.3d at 1330; see also Lincoln, 518 F.3d at 1316. Likewise, our decision in Fifth Third Bank, issued shortly after Lincoln, recognized that “strict application of the [time-of-breach] rule may not result in the most accurate assessment of expectancy damages.” 518 F.3d at 1377. Thus, where it is necessary to fashion an appropriate *1370 award, a court “may consider post-breach evidence when determining damages in order to place the non-breaching party in as good a position as he would have been had the contract been performed.” Id.; see also Cal. Fed. Bank, 245 F.3d at 1349-50 (reversing award of summary judgment for defendant and remanding because plaintiffs evidence, including the actual post-sale performance of assets the plaintiff was forced to sell, created a genuine issue of material fact as to the existence and quantum of lost profits). That is particularly apt to be true where, as here, the court is required to choose between (1) equivocal evidence as to the market value of an income-generating asset many years earlier, unenhanced by interest, and (2) reliable evidence as to the actual earnings the asset would have produced over the pertinent period. Notably, this is not a case in which the government’s breach forced Anchor to sell RFC but left it free to invest the proceeds of the sale in another equally profitable income-generating asset, so that the forced sale would in theory produce little by way of damages. In this case, by requiring Anchor to sell RFC in order to buttress its capital accounts, the breach deprived Anchor of the profits it would have obtained from retaining RFC while at the same time preventing Anchor from investing the proceeds of the RFC sale in a similarly profitable enterprise. In effect, that meant that the proceeds of the RFC sale lost much of their value as a potential source of profit, and thus that the difference between the fair market value of RFC and the proceeds from the sale was not necessarily a reliable measure of Anchor’s loss from the breach. The trial court, which presided over the five-week-long trial on damages and was intimately familiar with the circumstances of the breach and the parties’ competing arguments, considered the two permissible methods of calculating damages — lost asset value and lost profits. It also reviewed extensive documentary and testimonial evidence and weighed the relevant expert analysis. Ultimately, the court concluded that the most accurate approach was to base the award of damages on RFC’s actual post-breach profits under GMAC. We hold that the court did not abuse its discretion in measuring Anchor’s damages by the post-breach profits generated by RFC. 6 Quite apart from the unsuitability of RFC’s fair market value to provide full compensation to Anchor for the breach, the evidence supports the trial court’s finding that the $64.4 million estimate of RFC’s value was likely a substantial undervaluation of RFC’s fair market value on the date of sale. As the Supreme Court has held, “fair market value presumes conditions that, by definition, simply do not obtain in the context of a forced sale.” BFP v. Resolution Trust Corp., 511 U.S. 531, 538, 114 S.Ct. 1757, 128 L.Ed.2d 556 (1994). Those conditions include a price “fixed by negotiation and mutual agreement, after ample time to find a purchaser, as between a vendor who is willing (but not compelled) to sell and a purchaser who desires to buy but is not compelled to take the particular ... piece of property.” Id. at 538, 114 S.Ct. 1757. Here, the evidence suggests that “RFC was sold in a veritable fire sale” and that there was “blood in the *1371 water,” because potential buyers knew of Anchor’s need to sell RFC. Anchor, 81 Fed.Cl. at 68, 87. Contrary to the government’s argument, the trial court reasonably concluded that assessing RFC’s fail-market value at the time of the breach would have been a more speculative means of measuring Anchor’s damages than looking to RFC’s actual post-breach profits. The record also reflects that the trial court carefully weighed the pertinent expert testimony. While acknowledging that Anchor’s damages model was not a perfect proxy for the profits that an Anchor-owned RFC (as opposed to a GMAC-owned RFC) would have earned, the court concluded that it nevertheless provided “a concrete, reasonable and appropriate model” by which to develop a “sound and appropriate” damages award. Anchor, 81 Fed.Cl. at 107-08. The evidence showed that, even after its sale to GMAC, RFC’s management team remained in place and that RFC followed the same business plan it had developed under Anchor, including entering into all the new lines of business it had planned to enter under Anchor. Id. at 118. Where “reasonable certainty” as to lost profits was established, such as for the 1990 to 1995 period, the trial court used Anchor’s damages model. However, where the model broke down, such as for 1996 and 1997, the trial court denied lost profits, instead crediting the government’s argument that RFC’s large post-1995 increase in volume, if risk-weighted under FIRREA, would have caused Anchor to incur a capital deficiency. Id. at 121. We conclude that the trial court’s findings and analysis are reasonable and well supported by the record. Accordingly, we uphold the trial court’s award of the lost profits damages attributable to Anchor’s forced sale of RFC. II The government challenges the trial court’s decision to award damages based on the amount paid by Anchor/Dime to purchase NAMCO in 1997. The government asserts that the trial court erred when it found that Anchor’s purchase of NAMCO constituted a compensable form of mitigation for the loss of RFC. While the government does not dispute either the trial court’s application of the mitigation doctrine or its calculation of mitigation costs, the government contests the court’s finding that NAMCO constituted a commercially reasonable substitute for RFC. As the trial court noted, our decision in Hughes Communications Galaxy, Inc. v. United States, 271 F.3d 1060 (Fed.Cir.2001), likened damages for substitute goods or performance to the Uniform Commercial Code concept of “cover.” Regarding the types of goods or services that constitute “cover,” or mitigation, we held: The substitute goods or services involved in cover need not be identical to those involved in the contract, but they must be “commercially usable as reasonable substitutes under the circumstances.” Whether cover provides a reasonable substitute under the circumstances is a question of fact. Id. at 1066 (citing U.C.C. § 2-712 cmt. 2 (1997)). Under that standard, a court must determine what a reasonable person in the non-breaching party’s position would consider to be commercially useable as a substitute, in light of the particular circumstances of the transaction. Following that approach, the trial court focused on the reasonableness of NAMCO as a substitute for RFC, given Anchor’s business purposes and needs. We conclude that the court’s findings are not clearly erroneous. The trial court found that NAMCO was an appropriate substitute for RFC. Ac *1372 cording to the court, the two companies “are functionally similar, operate in the same market, require similar skills to effectively operate, and provide many similar operational benefits.” Anchor, 81 Fed.Cl. at 126. Both originated mortgages nationally and in large volume; both sold a majority of mortgages into the secondary market, while providing Anchor the opportunity to cherry-pick a preferred handful of mortgages to be held in portfolio; both generated servicing rights that led to steady servicing fee income; and both were a source of interest-free financing for Anchor. Id. at 130. Moreover, NAMCO and RFC were similar in size with respect to the volume of their originations and earnings. Id. Even the government’s expert, Dr. Carrón, conceded that NAMCO and RFC brought the same functional attributes and benefits to Anchor and served the same needs. Id. at 132. The trial court gave weight to the similarities between the companies, while regarding the “slightly different product focuses” as less significant. Id. at 132. The government maintains that NAM-CO and RFC were fundamentally different businesses that operated in different markets. RFC issued private-label MBS (for which RFC provided credit enhancement), whereas NAMCO issued only agency-sponsored MBS (for which government-sponsored entities provided credit enhancement). RFC did not directly make loans, whereas NAMCO did so for a certain percentage of its mortgages. RFC maintained no direct contact with borrowers, whereas NAMCO maintained contact with some borrowers to service the loans it originated. According to the government, the trial court’s conclusion that those differences were only a matter of “product focus” fails to acknowledge that NAMCO and RFC were in different businesses altogether. The trial court’s findings as to the significance of the differences between the two companies are not clearly erroneous. The court considered and analyzed each point raised by the government. It noted, for example, that most of NAMCO’s mortgages were wholesale originations, like RFC’s mortgages, and that both NAMCO and RFC engaged in servicing and generating servicing income, albeit at different ends of the “servicing spectrum.” See Anchor, 81 Fed.Cl. at 130-32. The court also pointed to considerable evidentiary support for its conclusion that the similarities between NAMCO and RFC outweighed the differences between the two companies. See id. at 126, 129-30 (discussing documents and testimony describing NAMCO’s large geographic presence, significant proportion of wholesale mortgage originations, and reliable non-interest revenue streams). Under the mitigation standard articulated by this court, the substitute services “need not be identical,” but merely a “reasonable” replacement in light of the particular circumstances. Hughes, 271 F.3d at 1066. While NAMCO and RFC may not have served identical markets or used identical strategies, they were both major players in the secondary mortgage market that Anchor sought to re-enter, generating interest-rate-independent servicing fees and allowing Anchor to diversify its portfolio and its risk. The court therefore permissibly concluded that NAMCO was a reasonable commercial substitute for RFC, and its purchase thus qualified as mitigation for the loss of RFC. III In light of our conclusion that the trial court did not err in measuring and awarding expectancy damages, we need not address Anchor’s conditional cross-appeal seeking reliance damages. Anchor’s remaining argument on cross-appeal concerns a purported “calculation error” by *1373 the trial court in the portion of the award relating to mitigation damages. After concluding in its mitigation analysis that NAMCO was a reasonable substitute for RFC, the trial court quantified Anchor’s costs associated with acquiring NAMCO. The court agreed with the government that the purchase price of $851 million was partially duplicative of Anchor’s 1990 to 1997 lost profits claim. That is because NAMCO traditionally distributed very little of its profits to shareholders as dividends, and the amount of shareholders’ equity consisting of retained earnings increased each year. The trial court credited the government’s expert’s conclusion that [i]f the court were to award plaintiff damages based on RFC’s lost profits from 1990 to 1997, it would be duplica-tive to also reimburse the full purchase price of NAMCO since much of that purchase price simply liquidated the shareholder equity (including retained earnings) that built up over the same 1990-97 period. In other words, such a damage award would provide plaintiff with RFC’s profits and NAMCO’s profits over the same period. Anchor, 81 Fed.Cl. at 133. As a result, the court reduced the $351 million purchase price to $185.9 million, which was shown (in Dime’s 1997 Annual Report) to be the “premium” that Anchor paid, over and above accumulated assets. The trial court concluded that the $165.1 million reduction constituted a “conservative best estimate” of the offset for NAMCO’s retained earnings (i.e., the duplicate profit for 1990-97). Id. at 134. The court did not address or make any adjustment for the fact that it had only awarded RFC-related lost profits for 1990 to 1995. Anchor argues that because the court declined to award any RFC-related lost profits damages for the years 1996 and 1997, it should not have reduced NAMCO’s purchase price by the amount of NAM-CO’s retained earnings through 1997. Anchor asserts that a proper offset would reduce NAMCO’s purchase price only by its retained earnings through 1995. According to Anchor, there is no need for a remand on this issue, because the proper “correction” amount is clear from the record. Anchor calculates that amount by subtracting NAMCO’s retained earnings through 1995 ($101,909,000, according to NAMCO’s 1996 Annual Report) from the trial court’s “best estimate” of retained earnings through 1997 ($165,100,000). Thus, Anchor requests that we increase its damages by $63,191,000, for a total award of mitigation costs of $249,091,000. It appears that the trial court may have reduced Anchor’s mitigation costs to avoid a “double counting” that did not actually occur. As we understand it, the trial court determined that Anchor would have been entitled to recover the full purchase price of NAMCO, were it not for a concern about double counting profits from both RFC and NAMCO during “the same 1990-97 period.” See Anchor, 81 Fed.Cl. at 133. However, because no lost profits were awarded for 1996 and 1997, there could have been no double recovery that would justify extending the offset of mitigation damages to those years. Nevertheless, we are not sufficiently confident that our assessment comports with the trial court’s methodology and intent, or that Anchor’s proposed correction would appropriately and reliably “correct” the error, if any. In fact, Anchor’s calculation mixes a precise figure (derived from the NAMCO annual report) with an admittedly imprecise “estimate” by the trial court. Thus, while it appears possible that a correction is warranted, it also appears possible that no correction is required— either because the trial court’s mitigation estimate was “close enough” or because the trial court’s full 1990-97 offset was *1374 made deliberately and appropriately in the first instance. On the information before us, we cannot make that determination. We therefore remand to the Court of Federal Claims to allow that court to determine whether an error was made in offsetting Anchor’s mitigation costs by NAMCO’s retained earnings through 1997 (rather than through 1995) and, if so, how to correct that error. AFFIRMED IN PART and REMANDED IN PART. Footnotes [1] “Private-label” mortgage-backed securities are so designated because they are not backed by government-sponsored entities and therefore must be credit-enhanced by the issuer in order to receive an "investment grade” (AA or AAA) rating by one of the nationally recognized credit rating agencies. Credit enhancement typically occurs through a "senior-subordinated structure,” in which the security is divided into two or more classes, with the subordinated class, or "B piece,” absorbing a disproportionately larger share of any pool losses, so that the senior class, or “A piece,” is insulated from loss and therefore qualifies for a higher credit rating. [2] The trial court found that even under the heightened burden of foreseeability urged by the government, "[f]he record indicates that it was in fact reasonable at the time of contracting for defendant to foresee that Anchor might become heavily engaged in the type of business in which RFC specialized, and that a breach might force Anchor from that market.” Anchor, 81 Fed.Cl. at 81. Because we conclude that the plaintiff need not satisfy that heightened burden, we do not address the question whether the court's findings on that issue are supportable. [3] We also reject the government’s argument that RFC’s post-sale profits were not foreseeable because they exceeded contemporaneous estimates of RFC's value, based on a preliminary, non-binding 1988 merger offer of $170 million for Anchor as a whole. The trial court found "little evidence to suggest that the $170 million ... offered for the entire Anchor franchise was probative of the bank's value at the time.” Anchor, 81 Fed.Cl. at 54. According to the court, the government did not present credible evidence that Anchor could not have exacted a higher price, or that no other potential buyers might have made an offer. In addition, a shareholder lawsuit challenged the offer as "substantially below [Anchor’s] fair or inherent value.” [4] For a similar reason, we reject the government’s argument that the trial court's position is internally inconsistent because "the court found that a capital shortage forced Anchor to cease an activity that the trial court found Anchor could have continued without additional capital.” The court’s suggestion that Anchor could have eliminated most of its recourse, and thus could have continued to operate RFC absent a breach, does not compel the conclusion that the breach had no effect on Anchor's ability to retain RFC. As the court found, Anchor's breach-induced capital deficit necessitated a swift shedding of assets, which included assets that Anchor could have retained if it had had to contend only with FIRREA’s risk-based capital requirements. [5] In Schonfeld, the Second Circuit likewise discussed the measurement of lost asset damages only after reviewing and affirming the trial court's determination that the plaintiffs separate lost profits claim was too speculative. 218 F.3d 164, 172-76. [6] The government argues that Lincoln also bars the trial court’s award of $42 million for lost stock proceeds, which was based on RFC's post-sale profits. Because we disagree with the government's interpretation of Lin-coin, and because the government has not offered any other specific reason to question that component of the court’s damages award, we reject the government's argument.