MICHEL, Circuit Judge.

In this Fifth Amendment regulatory takings case, Plaintiffs Cienega Gardens, et al. ("the Owners") appeal the United States Court of Federal Claims' grant of summary judgment to the government. Cienega Gardens v. United States, No. 94-1C (Fed.Cl. Jan. 8, 2002) (order granting summary judgment). Specifically, the Owners contend that when Congress enacted the Emergency Low Income Housing Preservation Act of 1987, Pub.L. No. 100-242, tit. II, 101 Stat. 1877 (1988) (codified at 12 U.S.C. § 1715l note (1988)) (hereinafter "ELIHPA"), and the Low-Income Housing Preservation and Resident Homeownership Act of 1990, Pub.L. No. 101-625, tit. VI, 104 Stat. 4249 (1990) (codified at 12 U.S.C. §§ 4101 et seq. (1994)) (hereinafter "LIHPRHA"), it abrogated the Owners' contractual rights to prepay their forty-year mortgage loans after twenty years. In doing so, the Owners argue, the government prevented them from regaining possession and control of their real estate because only extinguishment of their mortgages released the Owners from the Department of Housing and Urban Development ("HUD") low-rent housing programs. This harmed the Owners because, under the programs, rental rates were held below market rates. On exiting the programs, however, the Owners could charge market rentals or sell their properties.

Because the parties agreed with the trial court that its decision in Alexander Investment v. United States, 51 Fed. Cl. 102 (2001), controls this case, the trial court granted summary judgment solely on the basis of that decision. This appeal is, therefore, in essence, a review of the legal conclusions by the same judge in the Alexander Investment decision. We conclude that on this record the trial court erred in holding: (1) that developers who voluntarily participated in the HUD housing programs had no vested property rights despite their two agreements — one with the lenders and one with HUD — and despite their ownership in fee simple of the land; and (2) that if any taking occurred, it could not, as a matter of law, be a compensable taking. Under constitutional, real estate, and contract law, we conclude a property right vested in the Owners that was temporarily taken. We also conclude that there is no reason this taking is not, as a matter of law, compensable under the Takings Clause of the Fifth Amendment to the United States Constitution. We further hold with respect to at least the subset of Owners for whom there is a well-developed record before us, that they are entitled to "just compensation." We therefore reverse the trial court's grant of summary judgment of no taking as to that subset of plaintiffs and for the other plaintiffs remand for further proceedings, as more fully set forth, post.

BACKGROUND

The Owners' original complaint included claims for breach of contract, for just compensation under the Takings Clause of the Fifth Amendment, and for allegedly unlawful administrative actions. This appeal is one in a series of proceedings, including two prior appeals decided by this court: Cienega Gardens v. United States, 194 F.3d 1231 (Fed. Cir. 1998) (" Cienega IV"), which reversed a breach of contract judgment in the plaintiffs' favor for lack of privity with the government, thereby overturning a damage award of some three million dollars stemming from a damage trial for a subset of plaintiffs; and Cienega Gardens v. United States, 265 F.3d 1237 (Fed. Cir. 2001) (" Cienega VI"), which affirmed the dismissal of per se takings claims but held that the regulatory takings claims were ripe for adjudication because it would have been futile to require plaintiffs to await a HUD ruling on applications for prepayment permission because, under the undisputed facts, HUD lacked statutory authority to grant such permission. Our remand in Cienega VI led to the summary judgment now appealed.

There are forty-two plaintiffs in this case. This includes four Model Plaintiffs who had a damages trial on their breach of contract claims (after the entire group of plaintiffs had previously won a summary judgment motion on liability for the breach of contract claims in front of the trial court). These are the only plaintiffs for whom there is a well-developed record. The other thirty-eight have never been given a trial on any claim and it is unclear whether there has even been discovery with respect to any of their claims. There are also a number of related cases involving similarly-situated plaintiffs. See, e.g., Chancellor Manor v. United States, 51 Fed. Cl. 137 (2001); Anaheim Gardens v. United States, No. 93-655C, 1997 WL 580613 (Fed.Cl. 1997) (granting a stay for part of the case pending the outcome of the Cienega litigation). The former case is presently under submission after argument on March 3, 2003.

The Owners are real estate developers who received loans from private lenders to construct housing projects that for a period of years would be under the housing programs established by sections 221(d)(3) and 236 of the National Housing Act (generally codified at 12 U.S.C. §§ 1701 et seq.). HUD provided participants' mortgage insurance, which facilitated low-interest, forty-year mortgages. In return, each participant entered into a "Regulatory Agreement" with HUD.

Each Regulatory Agreement placed a variety of restrictions on the Owners, including restrictions on the income levels of tenants, allowable rental rates, and the maximum rate of return on initial equity that the Owners could receive from their housing projects. The Regulatory Agreement also prohibited sale or further mortgage of the property without HUD approval and required each participant to submit to extensive HUD audits, inspections, and management reviews. The Regulatory Agreements were to remain in effect only "so long as the contract of mortgage insurance continues in effect." The Regulatory Agreements also referenced the relevant provisions of the National Housing Act.

HUD's regulations then in effect included recognition of the Owners' rights to prepay their forty-year mortgages after twenty years and the mortgage documents themselves were reviewed by HUD and drafted to conform to existing HUD regulations. The mortgage trust notes provided that the Owners could not prepay their mortgages (without HUD approval) during the first twenty years of the mortgage, but could do so "without such approval" after twenty years. Though not a party to the mortgage contracts, HUD reviewed, endorsed, and approved them and their terms mirrored HUD regulations. Thus, under the original regulatory scheme, the Regulatory Agreements and their attendant restrictions were to remain in effect for at least twenty years, at which point the Owners would have the option of prepaying the mortgages and thereby dissolve the mortgage insurance and hence the restrictive Regulatory Agreements.

The relevant HUD regulations also provided that they could be amended but not to the prejudice of the lenders. The mortgage documents themselves, however, contained no explicit reference to the amendment provision. The Regulatory Agreements each mention either section 221(d)(3) or section 236 and the corresponding regulations in their preambles but do not otherwise cite any further provision of the statutes or regulations.

As the Owners' participation in the housing programs approached the twenty-year mark, it became clear to Congress that large numbers of owners would prepay their mortgages and remove their properties from the federally-assisted low-income housing pool. H.R. Conf. Rep. No. 100-426, at 192 (1987). Loss of this federally-assisted, low-income housing "would inflict unacceptable harm on current tenants and would precipitate a grave national crisis in the supply of low income housing that was neither anticipated nor intended when contracts for these units were entered into." ELIHPA § 202(a)(4). To avert the problem, Congress enacted ELIHPA in 1987 as a temporary measure. Under ELIHPA, even after twenty years, all housing program participants had to obtain HUD approval in order to prepay their mortgages despite their mortgage contracts containing a provision guaranteeing prepayment "without [HUD] approval." Moreover, HUD could grant approval only if prepayment would not "materially increase economic hardship for current tenants" by increasing monthly rental payments by more than 10 percent or "involuntarily displace current tenants (except for good cause) where comparable and affordable housing is not readily available." Id. § 225. Indeed, HUD was expressly prohibited from approving any application that did not meet the criteria listed in the Act. Id.

LIHPRHA was enacted three years later to replace ELIHPA. It extended indefinitely the prohibition on prepayment without HUD approval. Thus, these statutes annulled the provision of the mortgage trust notes that prepayment was allowed after twenty years "without [HUD] approval." The restrictions imposed by ELIHPA and LIHPRHA were essentially lifted by the Housing Opportunity Program Extension (HOPE) Act of 1996. Thus, this case involves the economic effects of ELIHPA and LIHPRHA during a period of up to eight years.

The immediate effect of ELIHPA and LIHPRHA was to nullify the Owners' option to prepay their mortgages. Cienega IV, 194 F.3d at 1235. This lawsuit arose from the Owners' resulting inability to exit the housing programs after twenty years and thereby regain normal rights of ownership. The Court of Federal Claims in Cienega III found that the Model Plaintiffs "intended from the beginning to prepay their existing mortgage balances on the original prepayment date and to convert their subsidized properties to conventional ones" and that "[t]his enticement and corresponding ability to convert to conventional became a material term of the contract." 38 Fed. Cl. at 75. The court also concluded that ELIHPA and LIHPRHA caused the Model Plaintiffs an aggregate loss of $3,061,107 on the four properties. Id. at 89. The damage award, however, rested on a breach of contract theory, not on a takings theory.

After the second remand from this court (holding that the regulatory takings claims of the plaintiffs' cases were ripe for adjudication), the Court of Federal Claims judge ordered the parties to advise him whether judgment should be entered for the government on the basis of a decision in another case. In Alexander Investment, a case with similarly-situated plaintiffs, the same trial judge, Judge Hodges, had ruled that the plaintiffs had no property interests that could have been taken by the government because (1) the contractual prepayment rights in the mortgage trust notes were not vested as of the time of the alleged taking, and (2) HUD had reserved the right to amend its regulations. Id. He also ruled in the alternative, that even if the plaintiffs had rights that amounted to property interests, a compensable regulatory taking had not occurred, as a matter of law under Penn Central Transportation Co. v. New York City, 438 U.S. 104, 98 S.Ct. 2646, 57 L.Ed.2d 631 (1978). Alexander Investment, 51 Fed. Cl. at 108-10.

More specifically, the trial court's determination that the plaintiffs had no property rights relied heavily on the assumption that the plaintiffs' early-1970's contract prepayment rights had not vested when ELIHPA and LIHPRHA were enacted in 1987 and 1990 (apparently because the initial twenty-year period had not yet expired) and the fact that HUD's regulations were explicitly amendable. Id. at 110. In addition, the trial court held that the plaintiffs did not have a property interest in their contractual rights to prepay their mortgages because their rights and obligations arose only as a consequence of a regulatory scheme established by Congress that HUD (and Congress) always retained the right to amend. Id.

In response to the judge's order, the Owners conceded that, except as to the question of economic impact, the legal conclusions expressed by the court in Alexander Investment would mandate judgment for the government in this case. Based on these submissions, the judge entered summary judgment for the government without further analysis. (No summary judgment motion was filed by either party, but rather, the court acted sua sponte.) This appeal reviews that grant of summary judgment.

For additional treatment of the background facts of this case and the legislation involved, see Cienega IV, 194 F.3d 1231, and Cienega VI, 265 F.3d 1237.

DISCUSSION

Ordinarily this court examines the Court of Federal Claims' findings of fact for clear error and reviews legal conclusions de novo. Bass Enters. Prod. Co. v. United States, 133 F.3d 893, 895 (Fed. Cir. 1998). Whether or not a taking has occurred is a question of law based on factual underpinnings. Id. Summary judgment, however, is, of course, in all respects reviewed de novo. Ethicon Endo-Surgery, Inc. v. U.S. Surgical Corp., 149 F.3d 1309, 1315 (Fed. Cir. 1998). We have jurisdiction over this appeal pursuant to 28 U.S.C. § 1295(a)(3).

There are two principal questions in this appeal. First: did a property interest vest in the Owners, which was then taken by the enactment of ELIHPA and LIHPRHA? For any Fifth Amendment takings claim, the complaining party must show it owned a distinct property interest at the time it was allegedly taken, even for regulatory takings. See, e.g., Wyatt v. United States, 271 F.3d 1090, 1097 (Fed. Cir. 2001) (holding that "the existence of a valid property interest is necessary in all takings claims"). Second: if plaintiffs owned such a vested property interest, did the regulatory restrictions in the new statutes go "too far" in constraining them, thereby entitling the plaintiffs to compensation? For this inquiry, the complaining party must offer evidence to show that the government has improperly shifted a public burden to a small class of private parties. See Armstrong v. United States, 364 U.S. 40, 49, 80 S.Ct. 1563, 4 L.Ed.2d 1554 (1960) ("The Fifth Amendment's guarantee that private property shall not be taken for a public use without just compensation was designed to bar Government from forcing some people alone to bear public burdens which, in all fairness and justice, should be borne by the public as a whole.").

The Court of Federal Claims in Alexander Investment answered both of these principal questions in the negative. We hold that the Court of Federal Claims erred in granting summary judgment in this case because the conclusion that housing program participants did not suffer a compensable taking was incorrect.

I.

Addressing the first principal question, we hold that the Owners did have vested property interests at the time of the passage of ELIHPA and LIHPRHA. Specifically, in abrogating the Owners' contractual prepayment rights, the statutes intentionally defeated the Owners' real property rights to sole and exclusive possession after twenty years and to convey or encumber their properties after twenty years. The Owners thus had property interests and these were based on the interaction of both real property rights and contractual rights. Also, the Owners' property interests vested long before the statutes were enacted, for they arose immediately upon execution of the mortgage loan agreement and the purchase of the land, both occurring in the 1970's.

A. 1.

Considering the real property interests first, as titleholders to land on which the apartment buildings were erected, the plaintiffs had fee simple ownership. An owner of land in fee simple generally has inherent rights to rent his or her land at any price he or she can command. See Bd. of County Supervisors v. United States, 48 F.3d 520, 527 (Fed. Cir. 1995) (holding that because there were no explicit conditions on the transfer of land to a Virginia county, the county took the land in fee simple and "was free as a matter of property law to do with the property what it wished"). Precedent shows that the ability to exercise every one of the "sticks" (rights) in the "bundle" of fee simple rights at the time of a taking is not a prerequisite to establishing a valid property interest under the Fifth Amendment. For example, under Wyatt and Cavin v. United States, 956 F.2d 1131 (Fed. Cir. 1992), a plaintiff is disqualified from claiming a Fifth Amendment taking only if he or she has no "valid property interest," Wyatt, 271 F.3d at 1096, or if he or she is "[w]ithout undisputed ownership" of the property at the time of the takings, Cavin, 956 F.2d at 1134. Present possessory rights are, thus, not necessary. It is also clear that fee owners who transfer a leasehold interest are still entitled to compensation. Alamo Land Cattle Co. v. Arizona, 424 U.S. 295, 303, 96 S.Ct. 910, 47 L.Ed.2d 1 (1976) (explaining that a lessor is entitled to compensation for the taking of leased land regardless of whether the lessor has possession of the land at the time it is taken — the measure of damages simply does not include the value of the leasehold interest if the lessor does not have possession). In fact, "[e]very sort of [real property] interest the citizen may possess" counts as a property interest under the Fifth Amendment. United States v. Gen. Motors Corp., 323 U.S. 373, 378, 65 S.Ct. 357, 89 L.Ed. 311 (1945).

In this case, the Owners contractually deferred enjoyment of a limited number of rights in their land. They relinquished only the right to exclude certain tenants, and only for a fixed period of time — twenty years. They never conveyed to HUD any interest in their land; they merely contracted with the government not to assert rights that they otherwise could. The owners are not somehow deprived of their Fifth Amendment rights merely because they temporarily relinquished some of their rights of fee simple ownership. Their retained rights put them well within the categories shown by precedent to invoke the Takings Clause of the Fifth Amendment.

2.

Turning next to the plaintiffs' contractual rights, there is also ample precedent for acknowledging a property interest in contract rights under the Fifth Amendment. See, e.g., Lynch v. United States, 292 U.S. 571, 579, 54 S.Ct. 840, 78 L.Ed. 1434 (1934) ("The Fifth Amendment commands that property be not taken without making just compensation. Valid contracts are property, whether the obligor be a private individual, a municipality, a State or the United States."); United States v. Petty Motor Co., 327 U.S. 372, 381, 66 S.Ct. 596, 90 L.Ed. 729 (1946) (holding that plaintiff was entitled to just compensation for government's taking of option to renew a lease); United States Trust Co. of N.Y. v. New Jersey, 431 U.S. 1, 19 n. 16, 97 S.Ct. 1505, 52 L.Ed.2d 92 (1977) ("Contract rights are a form of property and as such may be taken for a public purpose provided that just compensation is paid."). Here the Owners had unequivocal contractual rights after twenty years to prepay their mortgages; thus they had a property interest in those rights — both in the subject matter of the contract (the real property rights) and in the contract itself.

That the Owners' contract rights arose from the interaction of several different documents does not change this conclusion. The Regulatory Agreements — the documents that restricted the Owners' use and alienation of their properties — terminated when the Owners' mortgages terminated. Under the mortgage loan notes, after twenty years the mortgages could be paid "without [HUD] approval." Thus, the mortgage contracts were the basis of the time constraints in the Regulatory Agreements that were essentially land use contracts. This interrelation was obvious from the start of the Owners' participation in the HUD programs. When Congress enacted ELIHPA and LIHPRHA, it intentionally deferred the Owners' ability to exit the housing programs and make more profitable use of their land from twenty years to forty years (or whenever in between those dates HUD consented). The significance of the contract right here lies not in the right to prepay as such, but the exercise of the contract right as a prerequisite of the Owners' right to exit the program.

3.

Having concluded that the Owners did have a property interest in the contractual right to prepay and exit the housing programs, we next focus on the question whether this property interest vested before enactment of the statutes. The plaintiffs' real property rights automatically vested upon their taking title to the property. Their contract rights vested when the contracts were signed, because there was no explicit contract provision to the contrary. As already explained, the plaintiffs in this case contracted to waive certain ownership rights and to be entitled to end the waivers after twenty years. There was no condition, precedent or subsequent, in their contracts. Their contracts, and thereby their consequent property interests were, thus, unquestionably vested by the time Congress enacted ELIHPA and LIHPRHA more than a decade later. Contrary to the view of the government, that the contract rights were not assertable until twenty years passed did not mean they were not vested.

B.

On the question whether the Owners had vested property interests, the government's position is that "enforceable rights sufficient to support a taking claim against the United States cannot arise in an area voluntarily entered into and one which, from the start, is subject to pervasive Government control, such as the Section 221(d)(3) and 236 insured housing programs." Because the prepayment right at issue in this case was created by the Owners' private contracts with private lenders, the position urged by the government means that all such contract rights are purely illusory if they concern activity "subject to pervasive Government control." To understand what is wrong with this argument it is necessary to understand the true scope of the effect implied by this viewpoint. The government, essentially, asks us to hold that nothing in the Owners' private mortgage agreements has any force and effect. The government, thus, advocates a legal regime that eviscerates century-old understandings of the stable and enduring nature of contract and real property rights. We take particular issue with the government's position that the Owners had no property interest because "[t]he prepayment provision contained in the Owners' deed-of-trust notes did not exist independently of, but rather was subject to, limited by, and totally dependent upon, [the] regulatory regime, including the power of the Government to change the prepayment rules expressly set forth in regulations reserving to HUD that authority" (emphasis in the original). This interpretation of the implications of a regulatory program on the existence of property interests is unwarranted. The government can point to nothing in either the mortgage contracts or the Regulatory Agreements that supports the position that contract and real property rights were premature, suspended or diluted almost to the point of meaninglessness. Instead, the government boldly asserts that this court should read into private contracts limitations that are not even hinted at in their text. The government's position is unconvincing. The government's contention is, in effect, that Congress could retroactively alter the Owners' mortgage contracts in any way it chose without any recourse for the Owners. That cannot be, and is not, the law.

C.

Yet the trial court accepted this very same argument. The court further explained that "[p]laintiffs did not have compensable property rights for purposes of the Fifth Amendment because regulations authorizing the program reserved to the Department of Housing and Urban Development the right to amend those regulations at any time." Alexander Investment, 51 Fed. Cl. at 103. In support of this holding, the trial court relied primarily on three cases. Id. at 107-08. The present case is, however, easily distinguishable from each of these cases.

First, in Bowen v. Public Agencies Opposed to Social Security Entrapment, 477 U.S. 41, 106 S.Ct. 2390, 91 L.Ed.2d 35 (1986), the State of California and several of its public agencies challenged an amendment to the Social Security Act preventing states and agencies from withdrawing from the social security program despite an inter-sovereign agreement that allowed such withdrawals upon giving at least two years' notice. The plaintiffs alleged that the nullification of their option to terminate the agreement effected a taking. The Supreme Court held in Bowen that there was no taking specifically because the agreement between the governments did not rise to the level of property.

Bowen is distinguishable from this case because the agreement in Bowen was not a contract involving a private party and not one for which general public contract law would apply. The terms of the states' agreements with the federal government in Bowen were specifically read by the Court to "reserve the Government's right to modify its terms by subsequent legislation" because the agreements stated explicitly that they had to conform to the requirements of the corresponding statute. United States v. Winstar Corp., 518 U.S. 839, 877, 116 S.Ct. 2432, 135 L.Ed.2d 964 (1996) (explaining the basis of the Bowen decision). The Court in Bowen also explained that the termination provision in the states' agreements "constituted neither a debt of the United States, see Perry v. United States, [ 294 U.S. 330, 55 S.Ct. 432, 79 L.Ed. 912 (1935)], nor an obligation of the United States to provide benefits under a contract for which the obligee paid a monetary premium, see Lynch v. United States, [ 292 U.S. 571, 54 S.Ct. 840, 78 L.Ed. 1434]." Bowen, 477 U.S. at 55, 106 S.Ct. 2390. And also that "[t]he termination clause was not unique to this Agreement; nor was it a term over which the State had any bargaining power or for which the State provided independent consideration. Rather, the provision simply was part of a regulatory program over which Congress retained authority to amend in the exercise of its power to provide for the general welfare." Id. In other words, the Bowen plaintiffs' participation in the social security system was really part of a legislative scheme that the plaintiffs chose to participate in "as is" — not a situation in which there were true contracts creating enforceable rights.

In contrast, here the Owners' private mortgage loan notes did not reserve to Congress the prerogative of modifying the notes, mention any likelihood of Congress' doing so, allocate the risk of such changes to the Owners, or even mention HUD's reservation of the right to amend in its separate regulations. See 24 C.F.R. § 236.249. Nor did the Regulatory Agreements contain any direct reference to the power to amend; they contained only a general reference to the regulations. Also in contrast to the Bowen plaintiffs, the Owners did have bargaining power over the prepayment term in their mortgage contracts, in the sense that prepayment was a fundamental consideration of the deal to which they agreed. The prepayment term set the termination date for a contract of finite duration and finite return on investment. To pretend that there was a meeting of the minds sufficient to form a contract that did not include an agreement upon the term to limit the duration is disingenuous. The Owners did not need to participate in the program — it was a profit-making venture for them, not a means of receiving needed social services offered only by the government as in Bowen. The Owners could simply have agreed to larger mortgages with private lenders if the economics of the offer by the government was not financially advantageous. The prepayment term was part of what made the offer a better choice than the alternatives. The Owners' contracts were also more like the contracts in Lynch because the Owners were making payments that gave rise to obligations — in return for mortgage insurance, the Owners made payments to the government in the form of below-market rate rental charges to the government's chosen tenants. Therefore, unlike in Bowen, the contracts in this case and the economic circumstances of the transactions contained no hint that the rights the contracts described were conditional, much less illusory.

The Owners also do not attempt to remove their contracts from the reach of constitutional power in arguing that, unlike the Bowen plaintiffs, they had a property interest, as the government erroneously implies in its brief. The Owners do not challenge Congress' authority to enact ELIHPA or LIHPRHA, only its right to refuse to pay just compensation.

Not only is Bowen distinguishable, but more importantly, the rule the trial court derived from Bowen is an unjustified extrapolation. The rule that the trial court attributes to Bowen is that a background of regulation precluded owners who were participating in the housing program from having any property right in prepayment provisions because no such right could have vested in the face of the government's alleged authority to declare it null and void at any time. Alexander Investment, 51 Fed. Cl. at 108. The trial court erroneously links the vesting of property rights to the plaintiffs' expectations. Id. Not only is this inappropriate, but it misconstrues what would be a reasonable expectation under the circumstances. The government has not shown that the plaintiffs had reason to think that the prepayment right in their private contracts would be subject to seizure, much less seizure without compensation. The prepayment provisions were found by the trial court to be the very enticement that the government used to draw the Owners into the program. Cienega III, 38 Fed. Cl. at 75. At the time the Owners entered into their mortgage contracts, HUD regulations authorized prepayment after twenty years; thus, the government's approval of the mortgage contract term was explicit. See, e.g., 24 C.F.R § 236.30 (1970). The government has not shown that the Model Plaintiffs had an objective reason to think that it would abrogate that provision. ( See further discussion of the Owners' expectations in section II(C), post.)

Nor is the regulated nature of federal housing programs a talisman that automatically prevents vesting of the right to prepay. See United Nuclear Corp. v. United States, 912 F.2d 1432 (Fed. Cir. 1990) (reversing a holding that a leasehold interest to mine on some land did not constitute property for purposes of the Fifth Amendment because mining is a regulated industry). Agreements between private parties concerning real property give rise to protected property interests, irrespective of whether the subject matter of the contracts is under the government's regulatory jurisdiction. Moreover, the industry of private mortgage loans is not at all one that is highly regulated. The trial court's assertion that owners could not acquire vested rights is unsound.

This case is instead comparable to Winstar, 518 U.S. 839, 116 S.Ct. 2432, 135 L.Ed.2d 964. Though Winstar was a breach of contract case and not a takings case, it showed that the abrogation by legislation of clear, unqualified contract rights requires a remedy, even in a highly regulated industry, there banking, because the contracts embodied the commitments of the contracting parties. Id. at 896-97, 116 S.Ct. 2432. The Owners' contractual prepayment rights are such clear, unqualified rights. The purpose of contracts is precisely to fix obligations and entitlements so that they will not be affected by subsequent background changes. Nor were the Owners' rights subject to divestiture simply because of their voluntary participation in a housing program under government control. To hold otherwise would mean that Congress could have changed the mortgage contracts in any way to affect any of the rights established by the contracts — including changing the contracts to extend their term from forty to, for example, eighty years — and the Owners would be without a remedy. Again, this is not and cannot be the law.

The trial court also relied heavily on Omnia Commercial Co. v. United States, 261 U.S. 502, 43 S.Ct. 437, 67 L.Ed. 773 (1923). Alexander Investment, 51 Fed. Cl. at 107. In Omnia the plaintiff had a contract to buy steel from a steel company at a below-market price. 261 U.S. at 502, 43 S.Ct. 437. Six months later, the government requisitioned all of the company's steel production, and ordered the company not to supply steel to the plaintiff. Id. The Supreme Court held that there was no taking and no right to compensation because the plaintiff's loss was no more than a consequential one resulting from the lawful exercise of the police power of the government. Id. at 511, 43 S.Ct. 437. The government took steel from the steel company, but did not succeed to the plaintiff's rights and duties under the contract. Id.

The trial court cited Omnia for the principle that "consequential loss or injury resulting from lawful governmental action" does not amount to a taking. Alexander Investment, 51 Fed. Cl. at 107 (quoting Omnia, 261 U.S. at 510, 43 S.Ct. 437). Citation of this principle in the context of ELIHPA and LIHPRHA is inapt. The trial court implied that, under Omnia, if the government is not a party to a contract, the effect of legislation on a private contract can only be consequential and there can be no taking of a contract right. Alexander Investment, 51 Fed. Cl. at 107. Omnia does not support any such rule. The proposition in Omnia about consequential loss or injury refers to legislation targeted at some public benefit, which incidentally affects contract rights, not, as in this case, legislation aimed at the contract rights themselves in order to nullify them. See Cienega III, 38 Fed. Cl. at 74 (explaining that the purpose of ELIHPA and LIHPRHA was to extend the duration of the Owners' duty to provide low-income housing under the Regulatory Agreements and citing ELIHPA). The Supreme Court explained in Winstar that "governmental action will not be held against the Government for purposes of the impossibility defense so long as the action's impact upon public contracts is . . . merely incidental to the accomplishment of a broader governmental objective." 518 U.S. at 898, 116 S.Ct. 2432 (citation omitted, emphasis added). "The greater the Government's self-interest, however, the more suspect becomes the claim that its private contracting partners ought to bear the financial burden of the Government's own improvidence, and where a substantial part of the impact of the Government's action rendering performance impossible falls on its own contractual obligations, the defense will be unavailable." Id.; see Sun Oil Co. v. United States, 215 Ct.Cl. 716, 768, 572 F.2d 786 (1978) (rejecting sovereign acts defense where the Secretary of the Interior's actions were "directed principally and primarily at plaintiffs' contractual right"). The enactment of ELIHPA and LIHPRHA directly and intentionally abrogated the contracts. The effect on the contracts is, therefore, not merely consequential. Where Congress' actions have the effect of "keep[ing] [the contract] alive for the use of the government" rather than "bring[ing] the contract to an end," a court should conclude that there has been a taking. Cf. Omnia, 261 U.S. at 513, 43 S.Ct. 437. Omnia is relevant to this case only because it confirms that contract rights can be property within the meaning of the Fifth Amendment and require compensation "if taken for public use." Id. at 508, 43 S.Ct. 437.

Finally, the trial court's reliance on Mitchell Arms, Inc. v. United States, 7 F.3d 212 (Fed. Cir. 1993), is misplaced. Alexander Investment, 51 Fed. Cl. at 107. The plaintiff in Mitchell Arms was a federally-licensed arms importer who purchased Yugoslavian assault rifles notwithstanding the earlier receipt of notification that its import permit had been suspended. When the Customs Service later refused entry of the rifles for lack of a valid import permit, the plaintiff filed suit for just compensation. The plaintiff's only asserted taking was of its "investment-backed reliance on the issued import permits," maintaining that this reliance "constituted 'property' within the meaning of the Fifth Amendment." Mitchell Arms, 7 F.3d at 215. This court rejected that assertion, reasoning that in revoking the import permit, the government did not take the rifles, and the plaintiff could have done anything it wished with them except import them into the United States. Id. at 217. The trial court cited Mitchell Arms for the proposition that an exercise of governmental power is not a compensable taking "when the contract is voluntarily entered into and in an area subject to pervasive government control." Alexander Investment, 51 Fed. Cl. at 107 (citing Mitchell Arms, 7 F.3d at 216).

What makes reliance on Mitchell Arms inapposite is this court's underlying rationale. According to this court in Mitchell Arms, "the Fifth Amendment concerns itself solely with the 'property,' i.e., with the owner's relation as such to the physical thing and not with other collateral interests which may be incident to his ownership." 7 F.3d at 217 (quoting United States v. Gen. Motors Corp., 323 U.S. at 378, 65 S.Ct. 357). The plaintiff's expectation in Mitchell Arms that it would be allowed to import, then sell those weapons in the United States was a collateral interest that "comes into being only upon the issuance of an import permit." Id. (citation omitted). It was, therefore, not truly property. Id. In addition, Mitchell Arms stated that "[t]he reason 'enforceable rights sufficient to support a taking claim' cannot arise in [an area subject to pervasive government control] is that when a citizen voluntarily enters such an area, the citizen cannot be said to possess 'the right to exclude.'" Id. at 216 (quoting Hendler v. United States, 952 F.2d 1364, 1374 (Fed. Cir. 1991)).

The holding that there was no property interest in Mitchell Arms is, thus, limited to those cases in which the interest at issue does not inhere to some property that the plaintiff owns independently. Indeed, in Mitchell Arms this court specifically distinguished the case at hand from United Nuclear Corp. v. United States, 912 F.2d 1432 (Fed. Cir. 1990), because "the interest affected [in United Nuclear] — the right to mine — was inherent in the ownership rights" that United Nuclear held "independent of their denied permits" whereas Mitchell Arms' expectation of selling the assault rifles in domestic commerce "was not inherent in its ownership of the rifles." Mitchell Arms, 7 F.3d at 217 (citation and punctuation omitted). Although a twenty-year prepayment right does not inhere in a forty-year mortgage, an express provision in the mortgage loan notes here explicitly creates such a prepayment right. Additionally, although the Owners' right to exclude others from their land may have been waived for a fixed period of twenty years, their right to exclude inhered in their titles to real property, not in any grant by the government. Also, the property rights asserted in this case were not generated by the regulatory regime itself, as with the system of arms import permits. Finally, the regulatory regime for arms and other areas that require permits involves a blanket prohibition and individual exceptions. The housing programs at issue here were part of a totally different regime in which there was not only blanket permissiveness, but also inducements.

II.

As we hold that each of the Owners did have a vested property interest that was affected by the enactment of ELIHPA and LIHPRHA, we must next decide whether the enactments constituted a compensable taking. A regulatory action only becomes a compensable taking under the Fifth Amendment if the government interference has gone "too far." Pa. Coal Co. v. Mahon, 260 U.S. 393, 415, 43 S.Ct. 158, 67 L.Ed. 322 (1922) ("The general rule at least is, that while property may be regulated to a certain extent, if regulation goes too far it will be recognized as a taking."). Under the Fifth Amendment's guarantee that private property shall not be taken for a public use without just compensation, the designation that a government action has gone "too far" refers to circumstances where the government has forced "some people alone to bear public burdens which, in all fairness and justice, should be borne by the public as a whole." Armstrong, 364 U.S. at 49, 80 S.Ct. 1563 (holding that building material suppliers were entitled to compensation where a shipbuilder's contract with the government specifically showed that the shipbuilder retained title to the uncompleted ships during their construction, because the suppliers' lien interests in the building materials could attach while the contractor held title and the government's action in taking title then also took the suppliers' lien property interests).

Whether a given regulation goes "too far" for purposes of the Fifth Amendment is determined by an "ad hoc, factual inquiry." Penn Central, 438 U.S. at 124, 98 S.Ct. 2646. Under Penn Central, courts use a three-factor analysis to assess claimed regulatory takings: (1) character of the governmental action, (2) economic impact of the regulation on the claimant, and (3) extent to which the regulation interfered with distinct investment-backed expectations. Loveladies Harbor v. United States, 28 F.3d 1171, 1176-77 (Fed. Cir. 1994) (citing and explaining Penn Central, 438 U.S. at 124, 98 S.Ct. 2646).

The trial court held that the taking of owners' property interests failed, by definition, to be compensable. Alexander Investment, 51 Fed. Cl. at 110. This was error. For the reasons discussed below, we hold that Congress' enactment of ELIHPA and LIHPRHA did go "too far" with respect to the Model Plaintiffs and therefore the Model Plaintiffs are entitled to compensation.

We are able to make this decision for the Model Plaintiffs outright, instead of simply remanding the case for further analysis for several reasons: (1) the extensive fact-finding already completed for these plaintiffs in Cienega I and Cienega III; (2) our ability to construe the relevant contracts, regulations, and legislation as a matter of law because there are no disputed facts pertaining to these sources, only disputed implications; (3) the lack of arguments by the government controverting the plaintiffs' specific arguments with respect to each of the three Penn Central factors and the lack of arguments identifying errors in the earlier fact finding.

A. Character of Governmental Action

"The first criterion require[s] that a reviewing court consider the purpose and importance of the public interest reflected in the regulatory imposition. In effect, a court [must] balance the liberty interest of the private property owner against the Government's need to protect the public interest through imposition of the restraint." Loveladies, 28 F.3d at 1176. The plaintiffs identify two rights that ELIHPA and LIHPRHA curtailed, namely, their right to exclude low-rent tenants and their right to sell or lease their properties to parties not included in the programs. With respect to the first right, they argue that by requiring the Owners to continue to rent their properties at below-market rents to government-approved low-income tenants beyond twenty years, ELIHPA and LIHPRHA eviscerated the Owners' right to exclude, and this is "one of the most essential sticks in the bundle of rights that are commonly characterized as property." Kaiser Aetna v. United States, 444 U.S. 164, 177, 100 S.Ct. 383, 62 L.Ed.2d 332 (1979).

The plaintiffs also argue that ELIHPA and LIHPRHA had the character of a taking because the statutes authorize the continuing physical occupation of particular developers' properties to address a societal shortage of low-income housing and that this is intrusive beyond the level of traditional governmental limits on land titles. Under Penn Central, "[a] 'taking' may more readily be found when the interference with property can be characterized as a physical invasion by the government . . . than when interference arises from some public program adjusting the benefits and burdens of economic life to promote the common good." 438 U.S. at 124, 98 S.Ct. 2646 (citation omitted). The plaintiffs analogize the government in this case to a holdover tenant — HUD effectively rented apartment buildings from the Owners beyond the term of the agreed leasehold and then sublet apartments to low-income tenants. We agree that the enactment of ELIHPA and LIHPRHA could fairly be characterized as akin to this type of physical invasion. See, e.g., United States v. Gen. Motors Corp., 323 U.S. at 380, 65 S.Ct. 357 (holding that a tenant was entitled to compensation for being required to give up tenancy of property even beyond the fair market rental value because "[t]he right to occupy, for a day, a month, a year, or a series of years, in and of itself and without reference to the actual use, needs, or collateral arrangements of the occupier, has a value.").

As for the abrogation of the second right, the right to transfer, the plaintiffs argue that ELIHPA and LIHPRHA had the character of a taking under Hodel v. Irving, 481 U.S. 704, 715-16, 107 S.Ct. 2076, 95 L.Ed.2d 668 (1987), which held that a statute abrogating the common-law right to devise to one's heirs even a small portion of a property was a taking — even absent a showing of economic harm or interference with investment-backed expectations — because the character of the government regulation was "extraordinary."

We agree with the plaintiffs' characterization of the effects of the statutes. The character of the government's action is that of a taking of a property interest, albeit temporarily, and not an example of government regulation under common law nuisance or other similar doctrines, which we would treat differently. Unquestionably, Congress acted for a public purpose (to benefit a certain group of people in need of low-cost housing), but just as clearly, the expense was placed disproportionately on a few private property owners. Congress' objective in passing ELIHPA and LIHPRHA — preserving low-income housing — and method — forcing some owners to keep accepting below-market rents — is the kind of expense-shifting to a few persons that amounts to a taking. This is especially clear where, as here, the alternative was for all taxpayers to shoulder the burden. Congress could simply have appropriated more money for mortgage insurance and thereby induced more developers to build low-rent apartments in the public housing program to replace housing, such as the plaintiffs', that was no longer part of the program.

The government offers little to controvert these arguments as to the character of the government action. The government tries to distinguish the case from one in which the government is a holdover tenant by saying that there was no twenty-year term in this case, only an indeterminate term up to forty years. Since the mortgage loan notes expressly, and the Regulatory Agreements by generic reference to both the regulations and the mortgage loan notes, gave the "landlords" the right to terminate low-rent tenancies after twenty years, this contention is simply inaccurate.

The government also argues that there was no unfair burden allocation because the Owners in this case were treated just like all of the other owners in the program, but in drawing conclusions from this circumstance, the government ignores the fact that the relevant class for comparing treatment or allocation of any burden in a takings analysis is that of the class of persons disturbed by the lack of affordable housing — presumably all of society — and the group of people available to remedy that problem — a group much larger than just those involved in these two particular housing programs. The government maintains that the statutes' imposition is industry-wide but this assertion also is inaccurate since it did not affect all landlords.

The government also has a number of separate, additional arguments about the character of ELIHPA and LIHPRHA. They are all unconvincing and do not in any way diminish the Model Plaintiffs' arguments. For example, the government argues that prohibiting prepayment does not have the character of a compensable taking because the legislation did not appropriate the owners' titles or dispossess them in any way; it merely denied them a future enhancement in the value of their property. As already described, precedent does not support this argument. Dispossession is not required for a regulatory taking. Nor is divesting title.

The government also emphasizes that the prohibition was not permanent. Again, we must respond that a taking need not be permanent to be compensable. Also, no mention is made of the fact that the prohibition of prepayment could last as long as another twenty years — until expiration of the forty-year mortgage. Another twenty years in the housing program, double the time agreed, may not have been permanent but it was substantial.

Finally, the government argues that the Owners themselves created the potential housing shortage by their intention to exercise their prepayment right and that somehow prevents the legislation from having the character of a taking. This argument is unsound as a matter of logic. Though the Owners' withdrawal from participation in a remedy for a problem could exacerbate effects of the problem, it does not make the Owners the cause of that problem.

The analysis of this factor by the trial court is equally unhelpful and unconvincing. The trial court cited Penn Central for the principle that "[i]nterference from a public program that adjusts the benefits and burdens of economic life to promote the common good normally is not a taking." Alexander Investment, 51 Fed. Cl. at 108 (citing Penn Central, 438 U.S. at 124, 98 S.Ct. 2646). The court then qualified this rule by differentiating between interference to "prevent injury to the public welfare" and "merely bestowing upon the public a nonessential benefit." Alexander Investment, 51 Fed. Cl. at 108 (quoting Radioptics, Inc. v. United States, 223 Ct.Cl. 594, 621 F.2d 1113, 1127 (Ct.Cl. 1980)). Only the latter requires compensation. Id. The conclusion that the court then derived from these principles without further analysis or support is that ELIHPA and LIHPRHA were the routine exercise of power to regulate for the common good and because of that they could not have had the character of a compensable taking. Alexander Investment, 51 Fed.Cl. at 108.

The court's conclusion ignores the fact that this is not a case in which the burden for remedying a societal problem has been imposed on all of society. Congress' purpose in enacting the statutes may have been entirely legitimate but the government has not shown that the actions Congress took — the enactment of ELIHPA and LIHPRHA — were within its powers to exercise without also granting compensation. The disproportionate imposition on the Owners of the public's burden of providing low-income housing is not rendered any more acceptable by worthiness of purpose. The principles, as used by the trial court in this context, misconstrue what type of government action counts as a compensable taking by focusing on the purpose rather than the nature of the action.

We conclude, as matter of law, that the government's actions in enacting ELIHPA and LIHPRHA, insofar as they abrogated the Model Plaintiffs' contractual rights to prepay their mortgages and thereby exit the housing programs, had a character that supports a holding of a compensable taking.

B. Economic Impact

"The second criterion, economic impact of the regulation on the claimant, [is] intended to ensure that not every restraint imposed by government to adjust the competing demands of private owners [will] result in a takings claim." Loveladies, 28 F.3d at 1176; see also Pa. Coal, 260 U.S. at 413, 43 S.Ct. 158 ("Government hardly could go on if to some extent values incident to property could not be diminished without paying for every such change in the general law."). What has evolved in the case law is a threshold requirement that plaintiffs show "serious financial loss" from the regulatory imposition in order to merit compensation. Loveladies, 28 F.3d at 1177. There is not, however, "an automatic numerical barrier preventing compensation, as a matter of law, in cases involving a smaller percentage diminution in value." Yancey v. United States, 915 F.2d 1534, 1539, 1541 (Fed. Cir. 1990) (affirming the conclusion that there had been a compensable taking despite arguments that the diminution in value of 77% was too small).

The Model Plaintiffs freely conceded at oral argument that not every owner who participated in the housing programs suffered a compensable taking. Whether or not there was a compensable taking, the Model Plaintiffs admitted, depends on whether, after twenty years, a program participant was not allowed to prepay its mortgage and, as a result, suffered a financial loss during the years when ELIHPA and LIHPRHA were in effect. The implication of this admission, which is also a correct statement of the law, is that specific findings of fact about the effects of the legislation on the plaintiffs are necessary to complete the analysis of the economic impact factor. The trial court erred by not using specific facts pertaining to these plaintiffs to analyze this factor.

With respect to the Model Plaintiffs we do not need to remand because the trial court already made findings of fact that are dispositive of the question of economic impact. The parties offered extensive evidence and opposing experts in the damages trial. See Cienega III, 38 Fed. Cl. at 74-82. The fact-finding in that trial was sufficient in scope and depth to permit an economic impact analysis here because the trial court awarded damages for breach of contract based on a lost profits theory. It required the Model Plaintiffs to prove, and made findings of fact for each of three prongs in a lost profits test: (A) causation, (B) contemplation, and (C) certainty. Cienega III, 38 Fed. Cl. at 73. The lost profits proof, thus, also led to many findings with direct relevance to a Penn Central economic impact analysis. Indeed, "lost profits" are, by definition, a measurement of what a party would have received absent the breaching party's action; in other words, in this case they involved only those losses that the court determined the government was directly responsible for and the number is offset by expenses the party would have incurred to receive its profits, to a reasonable degree of certainty. Its analysis overlaps in many ways with the analysis of the "economic impact" of the same actions. See generally Chain Belt Co. v. United States, 127 Ct.Cl. 38, 115 F.Supp. 701 (1953). "Economic impact" requires similar evidence to quantify the harm to the plaintiffs resulting from the government's actions to determine whether it amounts to "serious financial loss" (although the evidence required for sufficient "economic impact" may actually be less stringent than that required for loss profits).

The government has not challenged the court's findings or their use in this appeal even though the plaintiffs clearly argued the appropriateness of our using them. In the absence of any such arguments by the government and our own determinations that the findings were not clearly erroneous and are an appropriate foundation for the analysis of "economic impact," we conclude that there is no impediment to our analyzing this factor for the Model Plaintiffs without resorting to a remand.

In addition to the explicit findings on lost profits, we can also appropriately derive other facts from the record because, in this case, the trial court specifically made findings about the credibility of the Model Plaintiffs' expert's method of calculation. Cienega III, 38 Fed. Cl. at 75. After trial, the court found that the damage calculations of the plaintiffs' expert, Dr. Peiser, were credible because they were based on estimates that were "actually conservative." Id. Specifically "[t]he court [found] Dr. Peiser's economic model more comprehensive and economically rational than that of either of defendant's experts, Messrs. Cunningham and Nevin." Id. Thus, although the Cienega III opinion does not include many numerical values, we can derive from the record specific numbers from the analysis that the trial court has certified as credible. As the government in its brief also uses some of these values, we conclude that they are undisputed.

Thus, using the trial court's findings we can be confident of the plaintiffs' expert's calculation that the aggregate amount of the Model Plaintiffs' annual earnings at the time of their respective prepayment eligibility dates totaled $45,741. This number reflects the restrictions on profits from rentals imposed on the properties under paragraph 6(e) of the Regulatory Agreements. The Model Plaintiffs' actual equity in their properties — the agreed-upon market values less their mortgage balances — shortly after their respective prepayment dates was calculated by the plaintiffs' expert as $17,452,045. The Model Plaintiffs were, thus, on average, limited to an annual return of approximately 0.3% on their real equity in their properties.

By comparing this rate of return to low-risk Fannie Mae bonds, which, according to Dr. Peiser, would have generated an 8.5% rate of return, we can make a rough estimate of the Model Plaintiffs' percentage loss of return. Indeed, doing so, we calculate that the Model Plaintiffs would have received, by exiting the programs and reinvesting their money, on average, at least, 28 times greater return than they did have by being forced to stay in the programs. (An 8.5% rate of return is about 28 times more than a 0.3% rate of return.) Another way to think of this is that a 0.3% rate of return is only about 4% of an 8.5% rate of return so the Model Plaintiffs earned a return that was about 96% less than what they could have earned by investing their money elsewhere. This calculation of an approximately 96% percent loss of return on equity offers us a way to understand that a 0.3% rate of return actually demonstrates that the abrogation of the Model Plaintiffs' prepayment rights had sufficient economic impact to merit compensation even under stringent conceptions in the case law of the percent diminution necessary to merit compensation. The loss of 96% of the possible rate of return on the investment is, even under the most conservative view, a "serious financial loss."

In opposing this interpretation of the economic impact, the government first suggests that the plaintiffs needed to claim that the prepayment restrictions denied them all economically beneficial use of their property in order for the takings to be compensable. It is clearly not the law that only such 100% value regulatory takings are compensable. In Florida Rock Industries, Inc. v. United States, 18 F.3d 1560, 1570 (Fed. Cir. 1994), this court specifically distinguished between "partial takings" and "mere diminutions"; the former is compensable even though it does not take 100% of the value of the property interest (i.e., it is not "categorical"). "Nothing in the Fifth Amendment limits its protection to only 'categorical' regulatory takings, nor has the Supreme Court or this court so held." Id.; see, e.g., Yancey, 915 F.2d at 1540 (holding that a federal government quarantine that had caused the owner of a flock of healthy breeder turkeys to market them for slaughter with a resulting loss to the owner of 77% of the breeder flock's value was a compensable taking even though the owner did receive some value in return for the slaughtered turkeys).

The government also argues in the alternative that the present market value of the assets of an enterprise is not the only acceptable basis of calculating a reasonable return. The government does not point to any specific errors in the predicate numbers or the methods the plaintiffs used to calculate their losses; it merely suggests that other methods of calculation or ways of measuring economic impact would be valid. In particular, the government maintains that ELIHPA and LIHPRHA did not limit the plaintiffs' rate of return inappropriately because of the low financial risk they undertook by participating in the section 221(d)(3) and 236 insured-housing programs. In making this argument the government cites the plaintiffs' discussion of the method of setting a fair rate of return for regulated public utilities. For regulated utilities, the returns to the equity owner should be commensurate with returns on investments in other enterprises having corresponding risks. Fed. Power Comm'n v. Hope Natural Gas Co., 320 U.S. 591, 603, 64 S.Ct. 281, 88 L.Ed. 333 (1944) (holding that a return rate of 6.5% was "just and reasonable"). Regulated utilities are not, however, sufficiently analogous to make the holding in Federal Power relevant. And even if they were, the risks involved in the participation in the housing programs were sufficiently high to merit greater return than 0.3%. Section 221(d)(3) and 236 projects that failed to receive sufficient rents to cover their costs were subject to foreclosure, just like any other mortgaged property. Program participants were, thus, subject to the risk that they would lose their initial investments and any further equity that had accrued. Some did. Therefore to claim that the ventures were low risk is unjustified.

The government also emphasizes that one of the Model Plaintiffs, Pico Plaza, did not prepay its HUD-insured loan when given the opportunity by the HOPE Act, and that Pico Plaza's inaction would not have been logical if it were true that LIHPRHA so diminished the value of the property. This inference, however, does not necessarily follow. There could be reasons that Pico Plaza did not prepay that have nothing to do with whether or not it suffered "a serious financial loss" because of the regulatory imposition.

Finally, the government cites Tahoe-Sierra Preservation Council, Inc. v. Tahoe Regional Planning Agency, 535 U.S. 302, 122 S.Ct. 1465, 1484, 152 L.Ed.2d 517 (Apr. 23, 2002) for the principle that "[l]ogically, a fee simple estate cannot be rendered valueless by a temporary prohibition on economic use, because the property will recover value as soon as the prohibition is lifted." This citation is inapt because of a fundamental error in the government's characterization of this lawsuit. The Owners' theory of recovery is not that their fee simple estates were taken or their land rendered "valueless." The Owners' entitlement to compensation is based on the taking of the real property interests reflected in the mortgage loan notes and the Regulatory Agreements. The difference is that the Owners' loss of the contractual prepayment rights was both total and immediate. They were barred from the unregulated rental market and other more lucrative property uses. Moreover, the Supreme Court's holding in Tahoe-Sierra was based on the interpretation of the land use regulation case, Lucas v. South Carolina Coastal Council, 505 U.S. 1003, 112 S.Ct. 2886, 120 L.Ed.2d 798 (1992), as holding that a regulatory taking that destroys 100% of a property interest's value is categorical, like a taking by physical invasion, and automatically counts as a compensable taking. Based on this interpretation, the specific holding in Tahoe-Sierra was that the Court would not deem a regulation resulting in the temporary valuelessness of a fee simple estate any form of categorical taking under Lucas. Tahoe-Sierra, 122 S.Ct. at 1483 (citing Lucas, 505 U.S. at 1017, 112 S.Ct. 2886). Rather, the Court explained that its affirmance of no taking was based on the rule that "[a]nything less than a 'complete elimination of value,' or a 'total loss,' the Court acknowledged [in Lucas], would require the kind of analysis applied in Penn Central," Tahoe-Sierra, 122 S.Ct. at 1483 (citing Lucas, 505 U.S. at 1019-20 n. 8, 112 S.Ct. 2886), and the plaintiffs had explicitly disavowed a theory of recovery based on Penn Central. Tahoe-Sierra, 122 S.Ct. at 1485. The holding of Tahoe-Sierra, thus, does not preclude recovery by plaintiffs who suffered less than a total loss but who do argue for recovery under a Penn Central analysis.

With comparable arguments, the trial court concluded that this factor counted against owners' claims because the "change in regulation did not remove the property from their possession" and, as a result, "any economic impact could not be more than mere diminution in value." Alexander Investment, 51 Fed.Cl. at 109, 110. The first statement fails to consider the correct property interest: the contractual right to prepay and then repossess the real property. The second statement is based on cases holding that diminutions in value approaching 75% or greater do not "necessarily" result in a regulatory taking. Id. at 109 (citing Euclid v. Ambler Realty Co., 272 U.S. 365, 47 S.Ct. 114, 71 L.Ed. 303 (1926) (no taking despite a 75% diminution), and Hadacheck v. Sebastian, 239 U.S. 394, 36 S.Ct. 143, 60 L.Ed. 348 (1915) (no taking despite 87.5% diminution)). It is perfectly true that no percentage diminution in value necessarily results in a compensable regulatory taking, but that is not the same as saying that below a certain percentage diminution, a taking can never be compensable, or even that an assessment of the economic impact below that percentage can never favor a conclusion that compensation is merited. Euclid and Hadacheck did not, as a matter of law, create an automatic numerical barrier preventing compensation in cases involving a smaller percentage diminution in value. See Yancey, 915 F.2d at 1541 (specifically interpreting these cases and holding that they are no barrier to compensation where it is found that there was "severe economic impact"). In any event, 96% diminution suffices in comparison with the percentages in the cited cases.

This court's assessment of the economic impact is that the Model Plaintiffs' expert's calculations (and the finding by the trial court about the credibility of their expert's methods) proved sufficient financial loss on the part of the Model Plaintiffs for this factor to favor compensation for each of them, especially in view of the lack of any specific challenge by the government of the trial court's findings or of the Model Plaintiffs' methods and data. Without additional information about the rest of the plaintiffs, this court cannot make any determination regarding them.

C. Reasonable Investment-Backed Expectations

The purpose of consideration of plaintiffs' investment-backed expectations is to limit recoveries to property owners who can demonstrate that "they bought their property in reliance on a state of affairs that did not include the challenged regulatory regime." Loveladies Harbor, 28 F.3d at 1177. This factor also incorporates an objective test — to support a claim for a regulatory taking, an investment-backed expectation must be "reasonable." Ruckelshaus v. Monsanto Co., 467 U.S. 986, 1005, 104 S.Ct. 2862, 81 L.Ed.2d 815 (1984). Here, the issue is not an imposition of a new "regulatory regime," but legislative abrogation of the key rule of a pre-existing regime. The critical question is whether a reasonable developer confronted with the particular circumstances facing the Owners would have expected the government to nullify the twentieth-year prepayment right in the mortgage contract and in the regulations. This is an objective, but fact-specific inquiry into what, under all the circumstances, the Owners should have anticipated.

We are able to complete an analysis of this Penn Central factor for the Model Plaintiffs based on earlier fact-finding by the trial court, a number of undisputed facts, our own authority to construe contracts, statutes, and regulations as a matter of law, and the lack of specific arguments by the government alleging error in the plaintiffs' arguments. This is again possible only for the Model Plaintiffs because the record with respect to the rest of the plaintiffs is not sufficient to support any conclusions.

Before examining the critical question about reasonableness for the Model Plaintiffs, we start with an analysis of the Model Plaintiffs' actual expectation because we require actual expectation of, or reliance on the government not nullifying the Model Plaintiffs' contractual and regulatory rights as a threshold matter. The Model Plaintiffs' expectation about prepayment would not really be "investment-backed" unless they actually believed in a certain outcome and entered the program in reliance on it.

1.

The trial court made findings of fact in earlier phases of the lawsuit that have direct relevance to the question of the actual expectations of the Model Plaintiffs. First, the trial court found that "[t]he evidence presented during the trial strongly indicates that plaintiffs intended from the beginning to prepay their existing mortgage balances on the original prepayment date and to convert their subsidized properties to conventional ones," and that the right to do so was "an essential aspect of Sections 221(d)(3) and 236." Cienega III, 38 Fed. Cl. at 75 (emphasis added). The trial court also found as a fact that the right to prepay and terminate the HUD restrictions without HUD approval after twenty years was one of the primary incentives HUD offered precisely to encourage their voluntary participation in the public housing programs. Id. We think the clear implication of these findings is that the Model Plaintiffs did not expect cancellation of their primary incentive or the nullification of their long-term plan.

Second, the trial court's findings in Cienega III also show that the Model Plaintiffs' business plan anticipated a higher rate of return after the twenty years upon entry into the commercial market, sale of their properties, or refinancing of their properties. The trial court found that each of the Model Plaintiffs' properties had "great potential for refinancing well above the amount owed and reinvesting the proceeds in other investment opportunities or for conversion and resale as condominium complexes." Id. at 76 (citation omitted). In view of these findings, we conclude that it would be only normal business judgment under the circumstances for the Model Plaintiffs to plan from the start to terminate their Regulatory Agreements as soon as possible to take advantage of more profitable opportunities. These findings, therefore, further establish that the Model Plaintiffs expected all along to be able to exit the housing programs after twenty years.

Similarly revealing is the court's finding that the locations the Model Plaintiffs selected for the properties they purchased also showed their expectation that they would be able to prepay and exit the HUD programs after twenty years. Id. at 75. "After careful study, the court observed that the four representative model properties are all located in middle class neighborhoods and, in general, are equal to or surpass the quality of the neighboring properties in each area." Id. Finally, in a separate opinion, the trial court remarked that the Model Plaintiffs' expectations "were based on express language contained in their deed of trust notes and authorized by HUD," Cienega I, 33 Fed. Cl. at 222, and were "consistent . . . with contemporaneous HUD regulations," id. Thus, they fully expected to retain the right to prepay after twenty years.

The government does not argue that the court's findings were clearly erroneous, but rather refers to the Owners' now-dismissed breach of contract claims contending that any expectations the Owners had depended on their mistaken belief that they had secured a vested contractual right to prepay that protected them from any regulatory policy changes. Because that was the basis for their expectation, the government argues that the Owners cannot maintain that they entertained an expectation of compensation for any loss resulting from regulatory change. We disagree. In any event, their underlying premise was not false.

The Model Plaintiffs, we conclude, are the type of landowners who bought their property and entered into contracts in reliance on a different regulatory regime. The trial court found as a fact that the Model Plaintiffs "would not have entered into the agreements with HUD but for" the benefits in the form of high rents and high market value that they would receive upon prepayment at twenty years and said that it "simply [did] not believe the plaintiffs entered into the HUD program without intending to reap a commensurate economic advantage as of the 20 year prepayment date." Cienega III, 38 Fed. Cl. at 75. The Model Plaintiffs' very participation in the program thus signifies their understanding that their option to exit the housing programs at twenty years would remain in effect. The trial court makes it clear that they would not have participated in the programs otherwise so no extrapolation on our part is necessary to know what the Model Plaintiffs' actual expectations were. We also see no clear error in the trial court's findings. The effect of ELIHPA and LIHPRHA — the Owners being indefinitely constrained by the Regulatory Agreements after twenty years — was contrary to what they were explicitly guaranteed by the mortgage contract and the regulations that were referenced in the Regulatory Agreement. ELIHPA and LIHPRHA cancelled the consideration for their entering the program. The Model Plaintiffs agreed to enter the programs in reliance on prepayment as the consideration for signing the Regulatory Agreement. The later legislation thus conflicted with the Model Plaintiffs' investment-backed expectations.

2.

Next, we must determine whether a reasonable developer in the Model Plaintiffs' circumstances would believe that the twentieth-year prepayment right was guaranteed by the regulations and that HUD "authorized" and endorsed mortgage contracts expressly including it. We start from the assumption that a contract term that was both material, in that it gave the duration of a business arrangement with limited profit, and apparently justified, in that it was consistent with (and identical to) the relevant regulation, if not also explicitly prescribed by the agency, is one that gives rise to reasonable expectations of fulfillment of the contract term. We then consider whether there are grounds for changing this supposition because additional circumstances or facts show that a reasonable developer should not have expected fulfillment of the contract term despite its presence in the contract.

In contrast to the government's argument, a Regulatory Agreement referring to regulations that contain a provision that they are amendable does not mean that program participants are reasonably on notice for every possible change and therefore could not have had a reasonable investment-backed expectation of no change to the prepayment right. Some changes would have been outside of the realm of the reasonably expected. See United Nuclear, 912 F.2d at 1436 ("The fact that United agreed that the leases would be subject to future regulations does not indicate that United fairly can be said to have anticipated that the Secretary would apply a new policy requiring tribal approval of mining plans to leases entered into almost six years earlier, in reliance on which United had expended some $5 million."). Abrogating the prepayment right was such an unexpected change. It was reasonable for the Model Plaintiffs to have expected the twentieth-year prepayment right to remain unchanged because it was a material inducement to the Model Plaintiffs signing the mortgage contracts and Regulatory Agreements, and the government presented no evidence of notice to the contrary. Once a plaintiff has shown that its expectation is reflected by a material contract term of which the government is aware, the government cannot establish a lack of reasonable expectations simply by showing that the regulations were amendable by HUD or nullifiable by Congress. We conclude that a prepayment term would be presumptively material for any housing program participant with similar documents. This is because by fixing an exit date from participation in the programs, the prepayment right would be a significant factor in the calculation of the total profit that could be expected over the lifetime of the investment in the property. The prepayment right specifies at what year revenue from the property could potentially be increased through higher rental rates, sale, or refinancing. This total profit calculation would, in turn, determine whether a developer decided to participate in the program in the first place. Because without the prepayment right the developers might have earned more profit investing elsewhere and therefore have declined to enter the programs, abrogation of this right would not reasonably be expected simply because the regulations were amendable or subject to legislative alteration. Moreover, the mortgage contract and the regulations incorporated by reference into the Regulatory Agreements both explicitly established a prepayment right which allowed them to end the mortgage contracts and, thereby the Regulatory Agreements. HUD reviewed, approved, and endorsed the deed of trust notes, including the riders containing the express prepayment right. The Model Plaintiffs, thus, also had every indication from HUD that this term in their private mortgage contracts was acceptable and would continue to control the duration of the Regulatory Agreements.

Similarly, no legislation was cited that contradicts the reasonableness of the Model Plaintiffs' expectation that the prepayment right would be undisturbed. At the time of the Model Plaintiffs' entry into the housing programs the only legislative reference to their ability to exit from the housing programs were broad delegations by Congress to HUD that, for example (for the section 221(d)(3) program), the Secretary "may at any time, under such terms and conditions as he may prescribe, consent to the release of the mortgagor from his liability under the mortgage or the credit instrument secured thereby, or consent to the release of parts of the mortgaged property from the lien of the mortgage." Housing Act of 1954, amended by Housing Act of 1964, Pub.L. No. 88-560, tit. I, § 114, 78 Stat. 778-79 (codified as amended at 12 U.S.C. § 1715l(e)(2)(2000)). The explicit legislative delegation to HUD also, thus, makes reliance on the mandatory prepayment provision in HUD's then-regulations reasonable; Congress broadly authorized such a regulation. The regulations were self-executing and gave HUD no discretion to specify who could prepay. Rather, they mandated that certain categories of housing program participants could prepay after twenty years at their own discretion. See, e.g., 24 C.F.R. § 221.524 ("A mortgage indebtedness may be prepaid in full and the Commissioner's controls terminated without the prior consent of the Commissioner. . . .").

We conclude, on this well-developed record, our own analysis of the explicit language of the relevant documents and relevant legislation, and given the limited arguments advanced by the government, described post, that the Model Plaintiffs' expectations were not only distinct and backed by investment, but were also reasonable.

3.

The government focuses its arguments mainly on one theory: that the Owners could not have formed a reasonable expectation because the Owners could not reasonably have assumed that if faced with a shortage in the supply of low-income rental housing, Congress would respond with a solution other than abrogating the right of prepayment. To this end, the government points out the trial court's statements that the "[p]laintiffs knew that they were entering a sensitive and highly regulated field that was subject to continuing congressional interest and attention" and they "could have anticipated that Congress might concern itself with the possibility of a low-income housing shortage and act to prevent or delay such a shortage." Alexander Investment, 51 Fed.Cl. at 110. Indeed, the court also suggested that owners could have "insisted on contract language that would have shifted the risk of later congressional action to the Government." Id. The housing programs themselves, the government argues, were entirely creatures of statute and regulation and therefore the Owners would have had no reasonable expectation that Congress would not change provisions of the programs' requirements during the first twenty years of each Owner's participation.

We, however, do not think the reasoning used by the trial court or anything else in the record compels the result sought by the government. Certainly, Congress could have met the need for replacement housing by funding new mortgage insurance to induce new participation under these very programs. We know that this is what Congress had done before because these housing programs were not new, but were instead long-standing programs (the first relevant legislation is from 1961; see ante note 5). We conclude then that it was reasonable to assume that Congress would continue down the path it had long pursued to alleviate a housing shortage. At the very least, one would not reasonably expect Congress to make legislative changes that would actually discourage parties from participating in the programs in the future.

Nor is the fact that the industry is regulated dispositive. A business that operates in a heavily-regulated industry should reasonably expect certain types of regulatory changes that may affect the value of its investments. But that does not mean that all regulatory changes are reasonably foreseeable or that regulated businesses can have no reasonable investment-backed expectations whatsoever. See United Nuclear, 912 F.2d at 1436.

The government cites Branch ex. rel. Maine National Bank v. United States, 69 F.3d 1571, 1581 (Fed. Cir. 1995), for the proposition that investment-backed expectations are "greatly reduced in a highly regulated field." This is an accurate statement of law but its simplicity of language conceals a complexity of analysis. The statement is not a rule dictating how we must proceed, but rather, a generalized description of prior decisions. For expectations to be "greatly reduced" does not mean that the reasonableness of every expectation of the status quo is by definition eliminated. It is our task then to determine whether the expectations the Model Plaintiffs had were reasonable ones. The range of expectations that is reasonable may be reduced in proportion to the amount of regulation, but this is not a blanket rule that disqualifies parties' expectations without inquiry. Also, what "highly regulated" and "field" are is not self-defining. Branch concerned the banking industry, and the power of the government to allocate the burdens of bank failures in a way that protects the public, regardless of the principle of limited corporate liability. Id. All this shows is that at the extremes, where history shows consistent, intrusive and changing government regulation of all facets of all transactions even arguably within a field, for example, banking, the effect of being in so highly a regulated field is clear. We have no evidence that the housing programs involved here were part of such an extreme field and therefore cannot, as the government urges, rely solely on the fact of regulation, but must probe into its content and other considerations. We thus conclude that Congress' abrogation of what was an explicit and material term of their mortgage contracts is simply not a change the Model Plaintiffs should have expected.

Beyond assuming the very thing in issue here, the trial court relied on two decisions that we conclude actually do not support the court's holding. The first case, Parkridge Investors Ltd. Partnership v. Farmers Home Administration, 13 F.3d 1192 (8th Cir. 1994), does involve some factual similarities: the plaintiff had received a loan from the Farmers Home Administration ("FmHA") and, in spite of contrary terms, its right to terminate the low-income rental use restrictions of its loan agreement through prepayment of the loans was deferred because of the passage of ELIHPA. The United States Court of Appeals for the Eighth Circuit held that the plaintiff could have foreseen that Congress might so amend the law and, therefore, the short-term deferral of prepayment did not effect a compensable taking. Id. at 1198. Any contrary expectation, according to the court, was simply not reasonable. Id. at 1199.

In relying on this decision, however, the trial court ignored critical dissimilarities. First, in Parkridge, the plaintiff's mortgage loan was directly from the FmHA, not from a private lender under a separate contract. Id. at 1195. Under ELIHPA and LIHPRHA, in this other program, the FmHA was not prohibited from accepting prepayment; rather the FmHA could "accept prepayment only at the end of an intricate, six-month-long (or longer) procedure," and the party desiring to prepay could be forced to sell the property instead if a willing non-profit or public agency offered to buy the property at market value. Id. While this six-month delay undoubtedly caused some inconvenience to the Parkridge plaintiff, the economic effect was minor and short-term. Here, the effect was major and the deferral was for up to twenty years. Second, the plaintiff in Parkridge was seeking enforcement of its prepayment rights rather than damages for its injury. As the court emphasized, the plaintiff had no right to ask for anything more than fair value of what was being taken — that is the only remedy offered by the Fifth Amendment. Moreover, the possibility of being forced to sell would actually offer ample compensation, rather than causing injury as the plaintiff suggested. Id. at 1199. The Parkridge plaintiff's claim for specific performance of the contract terms was misplaced since a claim for damages was possible and the Fifth Amendment does not constrain Congress' prerogative to pass legislation; it only requires compensation. Id. at 1200.

The trial court's reliance on Federal Housing Administration v. Darlington, Inc., 358 U.S. 84, 91, 79 S.Ct. 141, 3 L.Ed.2d 132 (1958), was also misplaced. First, the court's quotation that "[t]hose who do business in [a] regulated field cannot object if the legislative scheme is buttressed by subsequent amendments to achieve the legislative end," was, in its original context, referring to legislation that merely clarified the originally-intended meaning of an existing statute, not, as in this case, legislative amendments that fundamentally changed the scheme legislated previously. Id. at 86-87, 91, 79 S.Ct. 141. The government argues that the Supreme Court's citation to Darlington in Connolly v. Pension Benefit Guaranty Corp., 475 U.S. 211, 106 S.Ct. 1018, 89 L.Ed.2d 166 (1986), shows that the reliance by the trial court on the quoted language from Darlington was appropriate. Connolly, 475 U.S. at 227, 106 S.Ct. 1018. Connolly was concerned with the question whether amendments to the Employee Retirement Income Security Act of 1974 ("ERISA") imposing additional financial obligations upon employers terminating pension plans, interfered with reasonable investment-backed expectations. Id. In Connolly, the Supreme Court partially relied on the language in Darlington to conclude that the ERISA amendments did not. Id.

We disagree with the government and the trial court with respect to the implications of Darlington and Connolly. The reference to Darlington in Connolly was actually consistent with our conclusion that the Model Plaintiffs had reasonable, investment-backed expectations. Connolly gave, as the reason plaintiffs should have expected the particular amendments to ERISA, the pre-existing legislation addressing termination. ERISA, enacted in 1974, was "designed to ensure that employees and their beneficiaries would not be deprived of anticipated retirement benefits by the termination of pension plans before sufficient funds have been accumulated in the plans. . . . Congress wanted to guarantee that if a worker has been promised a defined pension benefit upon retirement — and if he has fulfilled whatever conditions are required to obtain a vested benefit — he will actually receive it." Connolly, 475 U.S. at 214, 106 S.Ct. 1018 (citation and internal punctuation omitted, emphasis added). Thus, the very problem addressed by the initial legislation was the one further addressed by the amendment that the plaintiff argued constituted a taking. Here, no pre-existing legislation suggested that prepayment was limited in any way or even that Congress would consider the termination of individual developers' participation in the programs a problem. As already explained, Congress was concerned with a housing shortage, but the developers already participating in the housing programs were not the only ones in a position to alleviate the problem. Indeed, there was every reason to think that Congress would merely induce new parties to participate in the programs. In addition, Connolly was a vastly different case as ERISA created a compulsory program, while the housing programs were completely voluntary.

Second, in Darlington, as here, mortgage insurance was granted in return for agreeing to restrictions to rent to tenants approved by the Federal Housing Administration (veterans and their families were the intended beneficiaries of the program). 358 U.S. at 86, 79 S.Ct. 141. But the Darlington plaintiff claimed a right to rent to different tenants while still participating in the program. Id. In other words, the plaintiff claimed the right to ignore restrictions on the very question — which tenants would program participants have to accept — that the program was expressly designed to regulate. We have already agreed that the Model Plaintiffs in this case should reasonably have expected changes to exactly what tenants they must accept, the rental rates, and similar details relating to housing. Such restrictions were the very purpose of the regulatory programs here, just as in Darlington. Unlike the Darlington plaintiff, the Model Plaintiffs are not, however, challenging that type of restriction. Rather, the provisions they are challenging, those abrogating the prepayment provisions of the mortgage contracts, are not central to the social purpose of the program. Not only is the duration of participation in this below-market business venture fundamental, but the prepayment rights provided, from all objective analyses, the crucial economic incentive here. And they were separate from the social goals. The assumption by these plaintiffs that the prepayment provisions would remain in effect is, thus, reasonable, in contrast to the Darlington plaintiff.

In sum, in light of the unchallenged findings of fact made by the trial court in Cienega III, we must conclude that the Model Plaintiffs did in fact expect to retain the right to prepay and exit the housing programs after twenty years. Our analysis of the circumstances surrounding their participation, the indications given to them by the written agreements — the explicit right to twentieth-year prepayment in the mortgage contracts and the regulations mandating that HUD grant twentieth-year prepayment — and the ultimately unpersuasive nature of all of the contrary arguments presented to us by the government, moreover, lead us to conclude as a matter of law that their expectations were reasonable.

The trial court thus erred in accepting the government's assertion that no expectation of the continuation of the prepayment term in a housing program participant's mortgage contract could be reasonable in light of the applicable regulation. "[K]nowledge of the Government's role in their chosen business venture," Alexander Investment, 51 Fed. Cl. at 110, we hold, is not enough to destroy the reasonableness of the expectation of this right to prepay that was expressly granted both by contract and by regulation when the Owners entered the programs. Where a trial court erred in granting summary judgment, ordinarily we would have to remand the case. Here, however, the reasonableness of the Model Plaintiffs' expectations can be confidently adjudicated without a remand because of the very specific factual findings following a trial on a contract claim about the two housing programs, the text of the relevant documents, regulations, and legislation, and the role of HUD in approving and endorsing the mortgage contracts and the riders therefor which contain the express prepayment right. We, therefore, hold that ELIHPA and LIHPRHA frustrated the Model Plaintiffs' reasonable investment-backed expectations that they would be entitled to prepay. We thus hold that all three of the Penn Central factors individually and together establish that the Model Plaintiffs are entitled to compensation without need for further fact-finding.

CONCLUSION

We vacate the trial court's sua sponte grant of judgment in favor of the government. Contrary to the trial court, we hold that all of the Owners had vested property interests under the Fifth Amendment in their contractual and regulatory rights to post-twentieth-year prepayment and under real property law to repossess. These property interests were expressly and deliberately abrogated by ELIHPA and LIHPRHA. Unlike the trial court we see no basis in takings case law for holding that as a matter of law ELIHPA and LIHPRHA could not effect a compensable taking under Penn Central. In addition, because whether or not a taking has occurred is a question of law based on factual underpinnings, Bass Enters., 133 F.3d at 895, and because we have findings of fact by the trial court adequate for a Penn Central analysis, we can and do conclude that the Model Plaintiffs suffered a compensable, temporary, regulatory taking. As to these plaintiffs, we reverse. We further direct that the original damages judgment entered in Cienega III, 38 Fed. Cl. 64, be reinstated in the amount awarded therein for each of the four Model Plaintiffs. In Chancellor Manor v. United States, 331 F.3d 891 (Fed. Cir. 2003), also issued today, we hold that for plaintiffs for whom this court has no findings of fact, it is not possible to complete a Penn Central analysis. Thus for the thirty-eight non-model plaintiffs, because the record before us is not developed at all as to, for example, economic impact, we remand for the trial court to allow the parties to develop an appropriate record and to rule on liability, and if liability is found, also on damages.

On remand, the court may of course consider any materials presented by either party in furtherance of its case that are relevant under regulatory takings case law. As we explain in Chancellor, the trial court may consider a broad array of relevant public and private documents in assessing expectations for prepayment by participants in the sections 221(d)(3) and 236 programs. Chancellor, 331 F.3d at 906. Specifically, we suggest in Chancellor that relevant evidence may be found in "contemporaneous publications" including "private placement memoranda" as well as "the legislative history of the various statutes . . . with respect to other federally-subsidized housing programs." Id. (emphasis added). We also explain that the trial court may undertake a review of "past actions taken" with respect to those other programs, meaning actions taken by HUD to implement those statutes through regulations and agency actions or by Congress through statutory amendments. Id. Such documents may reveal relevant evidence. That is not to say, however, that the trial court must weigh every document conceivably related to any aspect of federal housing law, or that we are making an advance determination of what evidence is relevant. In fact, the relevance of "other federally-subsidized housing programs" is not self-evident. Nor is the relevance of their "legislative history." And without knowing the contents of documents the government may produce on remand, it cannot be predicted whether or to what extent each one may be relevant. In saying in remand instructions, then, that the "court should consider" them, we do not prejudge their relevance. See id.

In Chancellor we do note that " Commonwealth Edison requires that the state of the law as a whole be considered in assessing whether investment-backed expectations are reasonable." Id. (citation omitted). But in Commonwealth Edison while there were contracts with the federal agency offering uranium enrichment services and also various applicable statutes and regulations, none contained guarantees of freedom from assessment of the power companies using the services for clean-up costs. Therefore, a broad search into the "regulatory environment" was necessary. In addition, the Comprehensive Environmental Response, Compensation and Liability Act, 42 U.S.C. § 9601, et seq. ("CERCLA"), already imposed retroactive liability on anyone who caused the creation of pollution, including nuclear waste pollution. Here, by contrast, no party has cited a prior statute keeping owners in the programs after the specified prepayment periods. The extant regulation and contract with the lenders specifically supported the expectations of these Owners to prepay at twenty years. Further, regulatory takings cases based on contracts containing key guarantees later negated by Congress may be fundamentally different from those involving only the generalized "regulatory environment" seen in earlier statutes, regulations, agency policies and practices, and industry understandings, relied upon in Commonwealth Edison, 271 F.3d at 1348-54. Thus, the need for as broad a review of documents in these housing cases remains to be determined. For example, "contemporaneous publications" is a very broad class of documents. Presumably not every document of this broad class will be relevant under Penn Central to claims by sections 221(d)(3) and 236 participants who sought to exit the programs after twenty years.

In deciding whether to rely on particular evidence, the trial court should also remain mindful of our holding that it was reasonable for the Model Plaintiffs not to have expected a material contract term and regulatory provision to be negated just because the federally-subsidized low income housing industry was admittedly highly-regulated. The high level of regulation for the housing programs may have less effect in this type of case than in some others because here the regulations themselves established the program participants' right to prepay. Nor should the Model Plaintiffs reasonably have expected divestment of their prepayment right simply because at the time of the enactments a shortage of low-income housing was projected that would have been exacerbated if these owners exercised their right to prepay and were not replaced by new participants, as Congress had other alternatives which it had already utilized in the past. On the other hand, if on remand for the non-model plaintiffs the government can produce documents from HUD to those plaintiffs, or other evidence that similarly negates the reasonableness of the expectation based on a material contract term, that would present a different case.

In addition, when assessing any other federal housing statutory provisions, the trial court should only consider those provisions that are somehow related to the sections 221(d)(3) or 236 programs in the late 1970s. While conceivably other statutory provisions or even some parts of their legislative histories may be relevant, the court must consider them in light of the parallel contractual agreements and HUD regulations then in effect governing the rights and expectations of the participants in the two programs. It is even possible that no additional legislative or regulatory materials or contemporaneous publications will be relevant.

In sum, the trial court should rely on the statements of law in this opinion and those in the Chancellor opinion to guide its conclusions about the relevance of particular documents to a Penn Central analysis.

VACATED-IN-PART, REVERSED-IN-PART, and REMANDED.

COSTS

Costs to be awarded to the appellants.