The issue in this case is how to measure the damages sustained by savings and loan institutions as a result of the breach by the United States of contracts it made with these organizations during the savings and loan crisis of the late 1970s and early 1980s. In Winstar Corp. v. United States, 64 F.3d 1531 (Fed. Cir. 1995), this court upheld the judgment of the Court of Federal Claims holding the United States liable for breach of contract as a consequence of the enactment of the Financial Institutions Recovery, Reform and Enforcement Act. The judgment of this court was affirmed by the Supreme Court, Winstar Corp. v. United States, 518 U.S. 839 (1996), and the matter was remanded to the Court of Federal Claims for a trial on damages.

The trial in this case, the first of more than 120 Winstar-related cases, ultimately took more than 150 days conducted over a period of fourteen months, and produced more than 20,000 pages of trial transcript. The trial court made a conscious decision that the trial in this case would serve to air and test many of the models and theories at issue in most of the other cases. We have the benefit of a carefully reasoned and extensive opinion from the trial judge.

Though we decide only the case before us, we too are sensitive to the precedential value of the decision in this case. For that reason we have considered carefully the arguments of both parties, as well as those of the several amici. Though factual differences may require some fine-tuning of results, we expect that the guidance given to the trial court in this opinion will assist the court, and the parties, in disposing of the remaining cases as well.

On April 12, 1999, the trial court entered a judgment awarding plaintiff Glendale Federal Bank, FSB ("Glendale") $908,948,000 in restitution and non-overlapping reliance damages for the breach of contract, and denying the Government's motion for summary judgment based on a special plea of fraud. The Government appeals the trial court's judgment. Glendale cross-appeals, stating that, should we reverse the award of restitution damages, it is entitled to additional reliance damages which were denied it because they overlapped the trial court's grant of restitution.

We conclude that the trial court properly resolved the issue regarding the fraud plea. However, because we see the question of the proper damages theory somewhat differently, we vacate the trial court's damages award and remand the case with instructions for further proceedings consistent with this opinion.

We provide a brief background of the case for completeness and to provide the setting for the decision to follow. Much may already be familiar to the reader; the facts regarding the crisis that generally affected the savings and loan industry, and that led up to the formation and breach of the contract between Glendale and the Government, are covered extensively in prior opinions of this court, the Court of Federal Claims, and the United States Supreme Court.1

The Federal Savings and Loan Insurance Corporation ("FSLIC")2 was established by Congress to insure the deposits of savings and loan institutions (also known as "S&Ls" or "thrifts"), and the Federal Home Loan Bank Board ("FHLBB")3 was established to safeguard the soundness of those institutions. Pursuant to this regulatory scheme, thrifts that desired to provide depositors with the insurance guaranteed under the program were required to maintain a certain level of capital.

Interest rates soared during the late 1970s and early 1980s, causing the savings and loan industry serious economic problems. At that time, thrifts' liabilities were principally short-term deposits, while their assets were primarily long-term fixed-rate mortgages. When interest rates soared, the value of the long-term fixed-rate assets plummeted, and the thrifts had to pay higher rates on their liabilities. As a result, many thrifts experienced difficulty remaining solvent and many became insolvent.

The Government's insurance fund lacked sufficient funds to liquidate even a small percentage of the thrifts that became insolvent. Consequently, the FSLIC and the FHLBB began to consider proposals for outside investors and thrifts to acquire other thrifts through mergers to prevent an exhaustion of the insurance fund.

Mergers were attractive to solvent thrifts because they enabled the thrifts to acquire previously prohibited interstate branches, to acquire high-quality assets that suffered only from the current interest rate squeeze, and to transform an insolvent thrift's net liabilities into an intangible asset called "supervisory goodwill." "Supervisory goodwill" was the accounting term for the difference between the market value of the acquired entity's liabilities and the market value of the acquired entity's assets. This difference was conceived of as part of the cost to the acquiring thrift. The accounting device was important to the acquiring thrift because the FHLBB allowed thrifts to include supervisory goodwill in the calculation of its regulatory capital requirements. Further, regulators in some instances would permit the goodwill to be amortized over a period of forty years.

In 1981, the Government and Glendale, an S&L institution primarily doing business in California, entered into negotiations concerning Glendale's possible acquisition of First Federal Savings and Loan Association of Broward County, Florida ("Broward"). The FHLBB found that Broward's financial condition was deteriorating and projected that Broward's regulatory capital would fall to zero by June 1982. Although Glendale was also losing money, the federal regulators determined that Glendale was nevertheless a financially strong institution, had no material management problems, and had the strength to absorb Broward and remain viable for some years.

In November 1981, the FHLBB approved Glendale's acquisition of Broward in a voluntary merger, and the Government and Glendale entered into a contract memorializing the arrangement. At the time of the merger, the market value of the liabilities of Broward exceeded the market value of its assets by $734 million. Pursuant to its contract with the Government, Glendale was permitted to book Broward's resulting market value deficit or net excess liabilities (negative net worth) as supervisory goodwill, an asset for purposes of meeting regulatory capital requirements. This supervisory goodwill was to be amortized over forty years, or until 2021.

Not long after the merger, interest rates declined, and the asset squeeze experienced by the industry eased. For a variety of reasons, some in the Government began to doubt the wisdom of the contracts that had been made with the salvaging S&Ls. In August of 1989, Congress enacted the Financial Institutions Recovery, Reform and Enforcement Act ("FIRREA"). FIRREA, among other things, greatly restricted the use of goodwill and other intangible assets in the calculation of regulatory capital.

Beyond that, FIRREA and its implementing regulations repudiated the Government's obligation to recognize Glendale's goodwill as an asset for purposes of regulatory capital over the contract's forty-year amortization period by requiring Glendale to deduct goodwill in determining its regulatory capital on a greatly accelerated schedule. In addition to mandating the phase-out of goodwill as an asset includable in calculating a thrift's regulatory capital ratios, FIRREA and its implementing regulations also changed the minimum requirements for capital.

A thrift, such as Glendale, that depended on supervisory goodwill to meet the new capital requirement could either: 1) increase its capital-to-assets ratio through retained earnings; or 2) increase its capital-to-assets ratio by reducing the size of the institution or by raising capital. When FIRREA was enacted, Glendale carried $536 million on its books in supervisory goodwill from its acquisition of Broward. In response to the phase-out of goodwill, Glendale reduced its size from $25.6 billion in total assets to $14.4 billion and, in 1990, ceased most high credit-risk lending. Although shrinking reduced its losses, Glendale nonetheless failed the risk-based capital requirement in March of 1992 and the core capital requirement in December 1992.

To return to compliance, in 1993 Glendale raised $451 million from new investors. Glendale incurred $24.2 million in transaction costs in this recapitalization. In 1994, Glendale sold its Florida Division, including Broward. In 1995, Glendale sold University Savings, its Washington subsidiary. In 1998, Glendale merged with California Federal Bank.

Glendale brought this action claiming the enactment of FIRREA breached an agreement by the Government to give Glendale favorable regulatory treatment in connection with its merger with Broward. As noted, the courts that reviewed the matter held that FIRREA's enactment rendered the Government liable to Glendale for breach of contract; the matter was remanded to the trial court for a trial on damages.

The trial court: 1) denied the Government's motion for summary judgment which raised a special plea in fraud, based on 28 U.S.C. § 2514 (1994); 2) denied Glendale's claim for expectancy damages because it was speculative and not proven to a reasonable certainty; 3) awarded Glendale $509,921,000 in restitution damages, which was the amount by which Broward's liabilities exceeded its assets on the date of the merger, less the value of the benefits Glendale received from the contract; 4) awarded Glendale $18,400,000 in restitution damages, which was the amount that Glendale paid to the Government under the interest shifting provision of the contract; 5) denied Glendale's restitution claim for interest earned on the assets of the FSLIC fund in the amount of $1.11 billion; 6) found that Glendale was entitled to $380,787,000 in non-overlapping (post-breach) reliance damages; and 7) denied Glendale's claim for reliance damages which overlapped with its restitution award.

The Government has appealed the trial court's denial of its summary judgment motion based on a special plea in fraud, the trial court's award of restitution damages, and its award of non-overlapping reliance damages. Glendale has conditionally cross-appealed the trial court's denial of reliance damages which overlap with the restitution award; should this Court reverse the trial court's award of restitution damages, Glendale claims the benefit of the denied damages. Glendale has not appealed the trial court's denial of Glendale's claim for expectancy damages and restitution damages for the interest earned on the assets of the FSLIC fund in the amount of $1.11 billion.

The trial court's legal conclusions are reviewed independently. Hendler v. United States, 175 F.3d 1374, 1378-79 (Fed. Cir. 1999). Findings of fact are subject to the clearly erroneous standard. Id. at 1378.

First, we will briefly dispose of an issue raised by the Government. The Government has appealed the trial court's denial of its motion for summary judgment based on a special plea in fraud under 28 U.S.C. § 2514 (1994). That statute provides, "[a] claim against the United States shall be forfeited to the United States by any person who corruptly practices or attempts to practice any fraud against the United States in the proof, statement, establishment, or allowance thereof." 28 U.S.C. § 2514 (1994). This court has explained that "[t]o prevail under [28 U.S.C. § 2514], the government is required to establish by clear and convincing evidence that the contractor knew that its submitted claims were false, and that it intended to defraud the government by submitting those claims." Commercial Contractors, Inc. v. United States, 154 F.3d 1357, 1362 (Fed. Cir. 1998) (emphasis added).

The Government argues that Glendale's suit is an effort to defraud the Government. The Government's claim is based upon conflicting statements made by two Glendale officers concerning Glendale's business intentions. Their statements made earlier in this protracted litigation conflict with statements they made more recently in support of Glendale's damages theories.

Though the trial court styled its resolution of this question as a denial of a motion for summary judgment, it is clear from the treatment of the issue in the trial court's opinion that the matter was fully considered at trial. Accordingly, because the question of whether the conflict in the testimonies evidences an intent by Glendale to defraud the Government is a question of fact, we review the trial court's findings under the clearly erroneous standard. The trial court considered the Government's arguments concerning the conflicting statements in its discussion and rejection of Glendale's claim for expectancy damages. Glendale Fed. Bank v. United States, 43 Fed. Cl. 390, 400 (1999). After examining and hearing the evidence, and observing Glendale's employees and their demeanor, the trial court found that the statements "do not rise to the level of a showing of fraud...." Id. The trial court reasoned that while the statements were inconsistent, "[t]his does not mean fraud occurred, but rather more likely, that Glendale [officials] cannot help but think about what [they] would have done with the benefit of hindsight [and i]t also means that post hoc reconstructions of this sort are likely to be colored by knowledge of what actually happened...." Id. The trial court's findings are not clearly erroneous and the trial court's resolution of the issue is affirmed.

The primary issue in this case is damages, how much and why. A review of basic principles of contract law and damages will illuminate the issues and their resolution.

Much of law and legal liability turns on fault, on the notion of wrong-doing, and of providing retribution for and discouraging such conduct. Contract law, and the consequences of breach of contract, do not. When parties enter into a contract, and promise each other that they will perform in accordance with the terms of their agreement, they retain a choice: when the time comes they can perform, and accept whatever benefits and losses the contract gives them, or they can refuse to perform and pay the consequences. The consequences the law imposes are for the purpose of making the non-breaching party whole, not for the purpose of punishment, or retribution, or deterrence. (Throughout this discussion we are assuming that the breach involved is a material breach, as it is in the case before us.)

One way the law makes the non-breaching party whole is to give him the benefits he expected to receive had the breach not occurred. See Restatement (Second) of Contracts § 344(a) (1981). The benefits that were expected from the contract, "expectancy damages," are often equated with lost profits, although they can include other damage elements as well. See Restatement (Second) of Contracts § 347. The problems of proof attendant on the burden placed on the non-breaching party of establishing lost profits--on establishing what might have been--are well recognized. Even with a generous standard of proof applied in such cases, see, e.g, San Carlos Irrigation & Drainage Dist. v. United States, 111 F.3d 1557, 1563 (Fed. Cir. 1997); Neely v. United States, 285 F.2d 438, 443 (Ct. Cl. 1961), the proof problems can in some situations prove to be insurmountable.

In the case at hand, Glendale sought expectancy damages; the trial court concluded that on the facts of the case Glendale was not entitled to expectancy damages. This conclusion appears to rest upon problems of proof, coupled with certain evidentiary issues related to inconsistent positions taken at different stages of the litigation by Glendale's key witnesses. In this appeal, Glendale does not challenge that ruling by the trial court.

When proof of expectancy damages fails, the law provides a fall-back position for the injured party--he can sue for restitution.4 The idea behind restitution is to restore--that is, to restore the non-breaching party to the position he would have been in had there never been a contract to breach. See Acme Process Equip. Co. v. United States, 347 F.2d 509, 528 (Ct. Cl. 1965); Far West Fed. Bank, S.B. v. Trinity Ventures, Ltd., 119 F.3d 1358, 1367 (9th Cir. 1997); In re First Penn Corp., 793 F.2d 270, 272 (10th Cir. 1986) ("The object of restitution is to return the parties to the position that existed before the transaction occurred."); 5 Arthur Linton Corbin, Corbin on Contracts § 1102, at 548 (1964); John D. Calamari & Joseph M. Perillo, The Law of Contracts, § 15.4 (4th ed. 1998) ("The basic aim of restitution is to place the plaintiff in the same economic position as the plaintiff enjoyed prior to contracting."); cf. Restatement (Second) of Contracts § 384, cmt. a ("A party who seeks restitution of a benefit that he has conferred on the other party is expected to return what he has received from the other party. The objective is to return the parties, as nearly as practicable, to the situation in which they found themselves before they made the contract."). In other words, the objective is to restore the parties to the status quo ante. See Acme Process, 347 F.2d at 528; Far West Fed. Bank, 119 F.3d at 1367; First Penn, 793 F.2d at 272. This is typically not as good as lost profits, from the viewpoint of the non-breacher, but a lot better than nothing.

Restitution is sometimes described in terms of taking from the breaching party any benefits he received from the contract and returning them to the non-breaching party. See Restatement (Second) of Contracts § 344(c) ("[The judicial remedies serve to protect the promisee's restitution] interest in having restored to him any benefit that he has conferred on the other party."). That requires determining what benefit from the contract the breaching party has received, and restoring that to the nonbreaching party. This approach makes good sense when viewed, for example, from the perspective of a typical contract for the sale of goods. B contracts to buy 1000 widgets from S, for $10 a widget. B gives S $1000 down, the balance of $9000 to be paid on delivery. S defaults on the contract; B sues for restitution. B gets his $1000 back. The amount that S is wrongfully benefited by the contract is taken from S, and restored to B.

In the case before us, the trial court, faced with a situation in which expectancy damages were ruled out but with a legitimate claim by the plaintiff for a remedy for the Government's breach, fashioned a restitutionary remedy based on the assumed risk that Glendale undertook when it acquired Broward. This was done by taking the value of Broward's obligations or debts at the time the contract was made, and subtracting from it Broward's then-assets. The resulting figure, a negative $798,291,000, was deemed to be the amount that the Government benefited from the contract, and to which Glendale was entitled as restitutionary damages.

Glendale supports the trial court's conclusion by arguing that Glendale's assumption of the market value of Broward's net liabilities was a benefit to the Government because Glendale thus relieved the Government of its imminent responsibility for those liabilities. Glendale also argues that Broward's net liabilities were Glendale's costs of the merger because when Glendale performed in November 1981, all of the liabilities assumed from Broward immediately became legally binding obligations of Glendale. Glendale further asserts that Glendale's cost of performance or the Government's benefit is to be measured as of the time of its performance (November 1981) and that subsequent events, such as the fall in the interest rates, are irrelevant, except to calculate the offset amount of Glendale's benefit. The trial court found, argues Glendale, that Glendale's expert's offset calculations included all benefits that Glendale's Florida franchise actually received from interest rate declines after the merger. Glendale additionally argues that it is not necessary that the Government's benefit in November 1981 be a cash gain; the Government's benefit is measured by the value of Glendale's performance in November 1981. Thus, whether the Government, absent the merger, would have actually liquidated Broward is irrelevant to the measure of restitution.

The Government's response is that the real figure should be zero. This is because, according to the Government, the amount of Broward's excess liabilities was not a cost to Glendale of performing under the contract because Glendale never actually paid Broward's excess liabilities, and in no scenario would Glendale ever pay Broward's excess liabilities because if interest rates did not decline then Broward would fail, Glendale would fail, and the FSLIC would have had to pay both Broward's and Glendale's depositors. The Government also argues that the Government never received a $798,291,000 benefit from the contract because the Government would not have liquidated Broward in the absence of a merger with Glendale, and the Government would have been responsible for Glendale and Broward if interest rates had not fallen and both institutions failed.

We conclude that the arguments by both Glendale and the Government are flawed. In the first place, it is clear that the Government's promise that was breached had substantial value. As was developed in the oral argument with Government counsel, given the choice between purchasing a failing thrift without the Government's promise regarding supervisory goodwill and purchasing one with it, a reasonable banker would surely take the latter. Whether viewed as a grant of a special bookkeeping procedure for these cases, or as a promise to waive the regulatory requirement on capital reserves for a specified number of years, there can be no doubt that, for a banker interested in purchasing a failing thrift, the promise was of substantial value toward accomplishing that goal.

At the same time, the action taken by the purchasing S&L in acquiring the failing thrift did not result in the Government, specifically the FSLIC, saving the dollar value of the net obligations of the thrift. For one, it is not at all clear that but for Glendale's purchase of Broward the Government would have been called upon to make up that deficit then and there. Glendale was only one of a number of potential acquirers of Broward. Alternatively, rather than approve a merger, the Government had open to it the option of hiring new and better management to run Broward and make a go of it, just as Glendale itself did. In a very real sense, what the Government received in exchange for its promise was time--time to deal with other failing S&Ls, time to see what the market would do before having to commit substantial resources to the problem. Though the value of time was more than zero, there is no proof of what in fact it was worth.

It is important to remember that, even after Glendale's merger with Broward, the Government was not free of potential liability for the failing thrift. Had interest rates not come down, and Broward, and perhaps Glendale as well, failed, the Government's contingent liability would have matured, and the FSLIC would have had to step in at that time and assume the very losses that Glendale now claims were benefits the Government received.

Fortunately for both the industry and the Government, the facts of the case are that interest rates did come down, market forces allowed the thrift industry to escape impending doom, and neither Glendale nor the Government was called upon to pay the potential losses the fear of which was the motivation for the scenario in the first place.

This case, then, presents an illustration of the problem in granting restitution based on an assumption that the non-breaching party is entitled to the supposed gains received by the breaching party, when those gains are both speculative and indeterminate. We do not see how the restitution award granted by the trial court, measured in terms of a liability that never came to pass, and based on a speculative assessment of what might have been, can be upheld; accordingly we vacate the trial court's damage award on this theory.

This does not mean that Glendale is without a remedy. Glendale recognized the problems in the restitution award, and cross-appealed, arguing that, should the court reject that award, Glendale nevertheless would be entitled to damages on a reliance theory. Indeed, the trial court recognized this third category of damages, known as reliance damages, and added specified reliance damages to the total award it granted plaintiff.

The underlying principle in reliance damages is that a party who relies on another party's promise made binding through contract is entitled to damages for any losses actually sustained as a result of the breach of that promise. See Restatement (Second) of Contracts § 344(b) ("[Judicial remedies serve to protect the promisee's] reliance interest, which is his interest in being reimbursed for loss caused by reliance on the contract by being put in as good a position as he would have been in had the contract not been made."); 3 Dan B. Dobbs, Law of Remedies § 12.3(1) (2d ed. 1993) ("The reliance recovery is a reimbursement for losses the plaintiff suffers in reliance on the defendant's contractual promise."). As a general proposition, these damages are available for injuries resulting from activities that occurred either before or after the breach. See Calamari & Perillo, The Law of Contracts, § 14.9 ("[A] party may recover expenses of preparation of part performance, as well as other foreseeable expenses incurred in reliance upon the contract."); Restatement (Second) Contracts § 350 cmt. g, illus. 18; cf. Restatement (Second) Contracts § 349(b) ("As an alternative to the measure of damages stated in § 347, the injured party has a right to damages based on his reliance interest, including expenditures made in preparation for performance or in performance...." (emphasis added)).

In the case at bar, the trial court awarded reliance damages for post-breach events, which the court denominated non-overlapping reliance damages. More specifically, it awarded Glendale damages for the following: lost historic cost of funds advantage over its competitors in the amount of $335,400,000; increased deposit insurance premiums in the amount of $11,480,000; increased Office of Thrift Supervision assessments in the amount of $4,675,000; transaction costs from the sale of University Savings in the amount of $4,472,000; transaction costs from the 1993 recapitalization in the amount of $24,235,000; and custodial fees paid to the Federal Home Loan Bank in the amount of $524,932.

Note that, unlike restitution in which the critical event that fixes the damages is when the contract was entered into, in reliance damages the critical event is the breach itself. Damages consequent to that breach may be the result of actions before or after the breach. Though the trial court limited Glendale's reliance damages to post-breach damages, since otherwise there would be a problem with duplication of the restitution elements, there is nothing inherent in the concept to prevent a court from awarding reliance damages for pre-breach activities from which ascertainable losses as a result of the breach can be shown.

In a case like the one before us, for all the reasons we have explained, we conclude that, for purposes of measuring the losses sustained by Glendale as a result of the Government's breach, reliance damages provide a firmer and more rational basis than the alternative theories argued by the parties. We recognize the appeal in the restitution approach, but we find that keying an award to a liability that was at most a paper calculation, and which ignores the reality of subsequent events as they impacted on the parties, and particularly the plaintiff, is not justifiable. Reliance damages will permit a more finely tuned calculation of the actual losses sustained by plaintiff as a result of the Government's breach.

We appreciate the enormous effort and dedication to these cases by then-Chief Judge Loren Smith in the course of the fourteen months during which he organized the many related cases and finally brought this one to completion. We trust that much of the extensive record before the court will lend itself to re-evaluation in the light of this opinion. As did then-Chief Judge Smith, we urge the parties to these cases to pursue settlements consistent with the facts and the law as they are now understood. Equitable resolution of these cases is in the interests of the institutions and their shareholders without further unnecessary litigation and its attendant time and expense; it is also in the interests of the United States to settle these cases equitably and fairly, so the cost to the taxpayers can be concluded without further delay.

Accordingly, we vacate the trial court's award of damages, and remand the matter to the court for a determination of total reliance damages to which plaintiff may be entitled. Because the matter is being returned to the trial court for further proceedings, we do not now undertake a piecemeal review of the specific items of reliance damages already awarded; that review will await the final award review when and if it is appealed.

The judgment of the trial court is affirmed-in-part, vacated-in-part, and remanded for proceedings consistent with this opinion.

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

No costs.