OPINION OF THE COURT
ADAMS, Circuit Judge.
Various owners and operators of bituminous coal mines brought suit pursuant to 42 U.S.C. § 1983 (1982) challenging the constitutionality of state statutes and regulations governing the mining of coal in Pennsylvania. The district court granted summary judgment for defendants, holding that the state program violated neither the takings clause nor the contract clause of the Constitution, and was not an invalid exercise of the power of eminent domain. We agree with the district court, and accordingly affirm.
I
A.
Plaintiffs1 sought in an action in the district court to have several provisions of [710]*710the Pennsylvania Bituminous Mine Subsidence and Land Conservation Act, 52 Pa. Cons.Stat.Ann. § 1406.1 et seq. (Purdon Supp.1984-85) (Subsidence Act), and its implementing regulations promulgated by the state Department of Environmental Resources (DER), declared unconstitutional. Named as defendants were those officials of the DER responsible for administering the Act.2
In February 1984, following pretrial proceedings designed to narrow the factual and legal issues, the district court denied plaintiffs’ motion for summary judgment, granted the DER partial summary judgment and dismissed the complaint. As a consequence of post-trial motions the matter was reopened and upon joint request of the parties, the district court certified several issues for appeal and stayed further proceedings in the district court.3 We accepted jurisdiction pursuant to 28 U.S.C. § 1292(b) (1982).4
B.
Plaintiffs operate underground bituminous coal mines in Western Pennsylvania, generally employing two different mining methods. The “room and pillar” method consists of a two-step process. During the first step, as coal is removed from the mine, blocks of coal known as pillars are left in place in a pre-planned pattern to support the strata overlying the coal seam being mined. Second, when the primary mining is substantially completed in a particular area, the coal pillars are systematically removed as the mining operation retreats. Operators seek to remove as many coal pillars as possible, consistent with safety.
The second method, known as the “long-wall panel” method, also has two phases. In the first, as mining proceeds coal pillars are left in place on either side of a longwall panel of coal-laden earth. Once a series of large panels has been laid out, a piece of equipment known as a longwall miner is installed. It advances continuously through the panels removing coal.
For purposes of this litigation, the mineralogical fact of most significance is that subsidence — the lowering of the strata overlying a coal mine because of underground coal extraction — eventually results from both forms of coal mining. App. at 45. A substantial amount of the coal reserves in the mines operated by plaintiffs lies beneath buildings and other features located on land whose surface is owned by others.5 Predominantly during the period 1890-1920, plaintiffs purchased from land owners the rights to mine and remove this coal. They also regularly secured waivers of the surface owners’ rights to collect damages for harm to the surface or surface structures caused by subsidence.
C.
In 1966, Pennsylvania passed the Subsidence Act which, with later amendments and implementing regulations, creates a comprehensive arrangement governing bi[711]*711tuminous coal mining in Pennsylvania. The present state program is in part a response to the federal Surface Mining Control and Reclamation Act of 1979, 30 U.S.C. §§ 1201 et seq. (1982 and Supp. 1983). The federal act provides that a state may assume primary control over mine reclamation activities within its borders by adopting a regulatory scheme at least as stringent as the minimum guidelines set forth in the federal laws. Pennsylvania’s program was eventually approved under this scheme. See 30 C.F.R. §§ 938.1-.20 (1983).
An original purpose of the state law was to preserve land in Pennsylvania, thus promoting the health, safety, and welfare of the people of the Commonwealth. An additional purpose was the preservation of a tax base for certain municipalities in order to enhance the Commonwealth’s economic welfare. 52 Pa.Cons.Stat.Ann. § 1406.3.
Plaintiffs challenge four aspects of the Subsidence Act. Section 4 and various Subsidence Regulations implementing both section 4 and section 5 require coal mine operators to leave a certain amount of coal in the ground for support under specified types of surface structures. These provisions apply notwithstanding an operator’s ownership of mineral rights or support rights. It is alleged that these provisions abridge the takings clause of the Constitution, U.S. Const, amend. V (“nor shall private property be taken for public use, without just compensation”), and specifically the holding of the Supreme Court in Pennsylvania Coal Co. ¶. Mahon, 260 U.S. 393, 43 S.Ct. 158, 67 L.Ed. 322 (1922).
Also challenged by the plaintiffs is the validity of section 6 of the Subsidence Act, which requires mine operators to pay compensation for subsidence damage to various structures, notwithstanding the existence of a damage waiver executed by the surface owner. In this regard, claims are raised under the contract clause, U.S. Const, art. I, § 10, cl. 1, as well as the takings clause of the Constitution.
Similarly plaintiffs urge that a regulation requiring mine operators to repair any damage to surface land caused by subsidence to the extent technically feasible, 25 Pa.Admin.Code § 89.147(b) (Shepard’s 1985), violates the contract clause. Finally, section 15 of the Subsidence Act, which gives surface owners a right to purchase underlying coal for support notwithstanding a waiver of the right of support, is claimed to be an invalid exercise of the power of eminent domain in violation of the Fourteenth Amendment. The district court rejected all four of the plaintiffs’ asserted grounds for relief. We address each contention in turn.
II.
Section 4 of the Subsidence Act, 52 Pa. Cons.Stat.Ann. § 1406.4, prohibits mine operators from mining “bituminous coal so as to cause damage as a result of the caving-in, collapse or subsidence of the following surface structures in place on April 27, 1966” in the proximity of a mine:
(1) Any public building or any noncommercial structure customarily used by the public, including but not being limited to churches, schools, hospitals, and municipal utilities or municipal public service operations.
(2) Any dwelling used for human habitation; and
(3) Any cemetery or public burial ground; unless the current owner of the structure consents and the resulting damage is fully repaired or compensated.
Pursuant to the legislation, § 89.145 of the DER regulations repeats the section 4 list of protected structures, and adds:
(4) perennial streams and impoundments with a storage volume exceeding 20 acre feet;
(5) acquifers which serve as a significant source of water supply to any public water system; and
(6) coal refuse deposited under requirements of Chapter 90 (relating to coal refuse disposal).
25 Pa.Admin.Code § 89.145 (Shepard’s 1985).
[712]*712Section 5(e) of the Act requires that, in order to prevent subsidence causing material damage, mine operators must adopt technologically and economically feasible measures to maximize mine stability and maintain the value and reasonably forseeable uses of surface land. Accompanying regulations specify that, at a minimum, the measures adopted must consist of the following:
(1) For each structure or feature to be protected, the operator shall provide a support area in which the amount of extraction is limited to 50%, leaving for support pillars of coal of a size and in a pattern which maximize bearing strength and are approved by the Department. No mining is permitted beneath a structure where the depth of overburden is less than 100 feet.
25 Pa.Admin.Code § 89.146(b)(1) (Shepard’s 1985).
Plaintiffs assert that these support requirements transgress the takings clause, relying particularly on Pennsylvania Coal Co. v. Mahon, 260 U.S. 393, 43 S.Ct. 158, 67 L.Ed. 322 (1922).
The Constitution guarantees that private property shall not be taken for public use without just compensation. The Supreme Court, however, has not adopted a single precise formula for determining when a challenged government action is properly characterized as a taking. See, e.g., Kaiser Aetna v. United States, 444 U.S. 164, 175, 100 S.Ct. 383, 390, 62 L.Ed.2d 332 (1979); Penn Central Transportation Co. v. New York City, 438 U.S. 104, 124, 98 S.Ct. 2646, 2659, 57 L.Ed.2d 631 (1978). The inquiry turns on the particular circumstances of each case, United States v. Central Eureka Mining Co., 357 U.S. 155, 168, 78 S.Ct. 1097, 1104, 2 L.Ed.2d 1228 (1958), with special emphasis on the character of the government action and the degree of interference with the private property. Pruneyard Shopping Center v. Robins, 447 U.S. 74, 82-84, 100 S.Ct. 2035, 2041-2042, 34 L.Ed.2d 741 (1980). While the takings analysis thus involves an “essentially ad hoc, factual inquiry,” Penn Central, 438 U.S. at 124, 98 S.Ct. at 2659, several general propositions emerge from the Supreme Court’s decisions in this area.
Respecting the character of the government action, the Court has stated that a “taking may more readily be found when the interference with property can be characterized as a physical invasion by government, ... than when interference arises from some public program adjusting the benefits and burdens of economic life to promote the common good.” Id. The concept of governmental invasion or occupation of property now encompasses certain government acts causing a direct and sustained interference with the use of property. Thus, in Kaiser Aetna, the owner of a private pond in Hawaii had dug an inlet through a barrier beach to form a private marina with access to the bay. The government claimed that the lagoon had become part of the navigational waterways and must be open to the public. The Supreme Court held that the imposition of this navigational servitude was an “actual physical invasion” and thus a taking which required compensation. 444 U.S. at 179-80, 100 S.Ct. at 392-93. A more traditional physical occupation was present in Loretta v. Teleprompter Manhattan CATV Corp., 458 U.S. 419, 102 S.Ct. 3164, 73 L.Ed.2d 868 (1982). There the Court found a New York statute forcing a landlord to permit installation of cable facilities on private property to be a taking.
A taking may also occur despite the absence of any form of physical occupation. Pursuant to its police powers the state may regulate the uses of property to promote the public good. In instances in which a state “reasonably concluded that the ‘health, safety, morals, or general welfare’ would be promoted by prohibiting particular contemplated uses of land,” the Supreme Court has upheld regulations that “destroyed or adversely affected recognized real property interests.” Penn Central, 438 U.S. at 125, 98 S.Ct. at 2659; [713]*713Nectow v. Cambridge, 277 U.S. 183, 188, 48 S.Ct. 447, 448, 72 L.Ed. 842 (1928).
Nevertheless, when “regulation reaches a certain magnitude” there may be a taking requiring compensation. E.g., Mahon, 260 U.S. at 413, 43 S.Ct. at 159; see Penn Central, 438 U.S. at 127, 98 S.Ct. at 2660. For example, in Nashville, Chattanooga & St. Louis Ry. v. Walters, 294 U.S. 405, 55 S.Ct. 486, 79 L.Ed. 949 (1935), the Court invalidated a local regulation requiring a railroad to bear half the cost of separating railroad tracks from the level of the highway. It found the purpose of the requirement was merely to facilitate highway travel, and that it was unreasonable to impose such a cost on the railroad, a private party. Id. at 421-31, 55 S.Ct. at 491-96.
With regard to the prong of analysis dealing with the degree of interference with the property, courts have applied a number of considerations, each developed under varying factual circumstances. At one time the concept of diminution of value was considered the dominant method of analysis. See Sax, Takings, Private Property and Public Rights, 81 Yale L.J. 149, 151 (1971). While it is clear that a diminution of value standing alone is not enough to establish a taking, see, e.g., Penn Central, 438 U.S. at 131, 98 S.Ct. at 2662, it has remained a viable part of the Court’s “degree of interference” inquiry. See Note, Regulation Without Just Compensation, 69 Geo.L.J. 1083, 1105 (1981). Where there is no diminution in value, it is likely that there will be no taking. In Goldblatt v. Hempstead, 369 U.S. 590, 82 S.Ct. 987, 8 L.Ed.2d 130 (1962), for example, the plaintiff — owner of a gravel pit whose right to excavate below the water table was prohibited — did not establish a taking because he failed to present evidence that the prohibition diminished the present value of the lot on which the pit was located. There is no set point at which a diminution in value is so great that a taking occurs. A diminution of 50% in the value of a chicken farm caused by low-flying government planes was sufficient to constitute a taking in United States v. Causby, 328 U.S. 256, 66 S.Ct. 1062, 90 L.Ed. 1206 (1946), while a 75% diminution in the value of a developer’s land caused by a zoning change did not create a taking in Village of Euclid v. Ambler Realty Co., 272 U.S. 365, 47 S.Ct. 114, 71 L.Ed. 303 (1926).
Some commentators have suggested that the diminution of value concept is most useful when the challenged government action is more narrowly focused, as in Causby. See Note, supra, 69 Geo.L.J. at 1109. In any event, it seems clear that the extent of diminution is but “one fact for consideration” in determining whether governmental action constitutes a taking. See Mahon, 260 U.S. at 413, 43 S.Ct. at 159. The Supreme Court has emphasized that “[gjovernment 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.” Penn Central, 438 U.S. at 124, 98 S.Ct. at 2659; Mahon, 260 U.S. at 413, 43 S.Ct. at 159.
Another element the courts have pointed to is the extent to which the regulation interferes with reasonable, distinct, investment-backed expectations. Penn Central, 438 U.S. at 124, 98 S.Ct. at 2659; see Goldblatt, 369 U.S. at 594, 82 S.Ct. at 990. In Penn Central the Court focused on the impact of a city regulation, rejecting the proposition that a taking is established simply by showing the denial of “the ability to exploit a property interest that [the plaintiffs] heretofore had believed was available.” 438 U.S. at 130, 98 S.Ct. at 2662. While the plaintiffs in that case were not permitted to use their ownership of air rights to build an office tower, the Court believed that other reasonable beneficial uses of the site remained. There was thus an insufficient interference with investment-backed expectations. Id. at 138, 98 S.Ct. at 2666.
Similarly, in Andrus v. Allard, 444 U.S. 51, 100 S.Ct. 318, 62 L.Ed.2d 210 (1979), the Court confronted a challenge to two federal statutes that prohibited the sale of por[714]*714tions of endangered birds, thereby preventing the sale, but not other uses, of artifacts containing eagle feathers. The Court rejected the takings claim, noting that the denial of one traditional property right does not always amount to a taking where alternative beneficial uses remain. Id. at 66.
Appellants urge that we need not undertake an analysis applying the above principles relating to takings because this case is factually on all fours with Mahon. In Ma-hon the plaintiffs owned the surface of certain land in Pennsylvania on which their dwelling was located. The defendant, the Pennsylvania Coal Co., owned the coal underlying the land and had secured the right to remove that coal as well as a waiver of all claims from subsidence damage caused by the company’s mining. 260 U.S. at 412, 43 S.Ct. at 159. Relying on a state law known as the Kohler Act, the plaintiffs sought to enjoin the mining operations to the extent the mining would damage their home. The defense was that the Kohler Act was an unconstitutional violation of the takings clause. That Act, passed by the state legislature in 1921, forbade the mining of anthracite coal in such a way as to cause the subsidence of certain classes of buildings. It thus closely resembled parts of the Subsidence Act challenged in this appeal.
The similarity between the two acts suggests that the Mahon case may be particularly relevant to our inquiry. Nevertheless, for several reasons Mahon does not dispose of this appeal. The gravamen of the company’s complaint in Mahon was that the Kohler Act merely protected the rights of a small group of private parties at the expense of other private parties, and was not legislation intended to promote a broad public interest pursuant to the police power, 260 U.S. at 394-404, 43 S.Ct. at 159; see McGinley and Barrett, “Pennsylvania Coal Co. v. Mahon Revisited: Is the Federal Surface Mining Act a Valid Exercise of the Police Power or an Unconstitutional Taking?,” 16 Tulsa L.J. 418, 430 (1981). Justice Holmes, writing for a majority of the Court, agreed with the company’s characterization of the Act’s purpose. He noted, in focusing on the nature of the government action, “This is the case of a single private house.... A source of damage to such a house is not a public nuisance even if similar damage is inflicted on others in different places. The damage is not common or public.” 260 U.S. at 413, 43 S.Ct. at 159. Furthermore, “the extent of the public interest [was] shown by the statute to be limited” since the Kohler Act did not “apply to land when the surface [was] owned by the owner of the coal.” Id. at 414, 43 S.Ct. at 159.
The Kohler Act was defended as legislation designed to protect personal safety, but the Court found the Act to be over-broad in this respect. Id.; see also McGinley and Barrett, supra, 16 Tulsa L.J. at 412. The goal of public safety could have been provided for by giving notice to homeowners of intended mining activities. The Court in this way distinguished Plymouth Coal Co. v. Pennsylvania, 232 U.S. 531, 34 S.Ct. 359, 58 L.Ed. 713 (1914), in which it had upheld legislation requiring pillars of coal to be left in the ground on the basis that those requirements were necessary “for the safety of employees.” 260 U.S. at 415, 43 S.Ct. at 160.
Not only was the Kohler Act found to lack substantial police power justifications, but it apparently rendered the removal of coal “commercially impracticable.” Id. at 414, 43 S.Ct. at 160; Penn Central, 438 U.S. at 127, 98 S.Ct. 2661; Note, supra, 69 Geo.L.J. at 1108. With so little evidence of justification for such a sweeping interference with the defendants’ rights, see Comment, Mine Subsidence Legislation in Pennsylvania and the Developing Concept of the Police Power, 27 U.Pitts.L.Rev. 835, 850 (1966), the Court concluded the state law had “gone too far” and consequently was a taking.
In contrast, the Subsidence Act at issue here does seek to prevent common or public damage. The legislature found that the [715]*715Subsidence Act was necessary to serve the public interest in preserving the land. Damage from mine subsidence was found seriously to have impeded land development throughout the Commonwealth and to have eroded the tax base of numerous municipalities. 52 Pa.Cons.Stat.Ann. § 1406.3. Moreover, unlike under the Kohler Act, no surface owner is excluded from coverage, app. at 445, indicating that the Subsidence Act was intended to serve the public purpose of preserving surface estates and structures regardless of whether mine operators or other persons own them.
As the above legislative findings demonstrate, the Subsidence Act is designed to protect the environment of the Commonwealth, its economic future, and its well-being. Id. While the Supreme Court stated in Mahon that notice might be adequate to promote the limited safety purposes of the Kohler Act, the legislature here found that such notice had not sufficed to promote the varied public concerns at issue. Id. Plaintiffs offer no evidence to show that this legislative determination was erroneous.
While we reject the suggestion that Ma-hon has been overruled sub silentio, see Andrus, 444 U.S. at 65-67, 100 S.Ct. at 326-328; Kaiser Aetna, 444 U.S. at 174, 100 S.Ct. at 389; Penn Central, 438 U.S. at 136-37, 98 S.Ct. at 2665, we conclude that because of the significant differences in the respective statutes’ scopes and purposes, Mahon is not dispositive of the takings issue in this case.
D.
Because Mahon is distinguishable, we must proceed to apply the takings analysis to the situation present here. The cases suggest that there are three relevant fields of inquiry: first, whether the government’s action entails a physical invasion of the plaintiff’s property; second, the extent to which it results in a diminution in the value of plaintiff’s property; and third, the degree to which it interferes with plaintiff’s reasonable, distinct, investment-backed expectations. See, e.g., Andrus, 444 U.S. at 65-66, 100 S.Ct. at 326-327; Penn Central, 438 U.S. at 124, 98 S.Ct. at 2659.
We first note that the legislation here does not authorize any actual physical invasion of the coal companies’ property but rather regulates the manner in which the property may be used. Thus, the support requirements are more properly viewed as part of a public program adjusting the benefits and burdens of economic life to promote the common good and to prevent certain serious and identifiable harm to the general public. As previously observed, a government act of this character is less likely to be considered a taking. E.g., Penn Central, 438 U.S. at 124, 98 S.Ct. at 2659. The legislative purposes underlying the public program — protection of the environment, public safety, and economic welfare of the state — are classic grounds for the use of the police power.
The question of the degree of interference is complicated by the fact that the property in question is divided into multiple estates under state law. Pennsylvania property law recognizes surface, support, and mineral estates. See Captline v. County of Allegheny, 74 Pa.Super. 85, 459 A.2d 1298 (1983), cert. denied, — U.S. ---, 104 S.Ct. 1679, 80 L.Ed.2d 154 (1984). The support estate consists of the right to remove the strata of coal and earth that undergird the surface or to leave those layers intact to support the surface and prevent subsidence. These two uses cannot co-exist and, depending upon the purposes of the owner of the support estate, one use or the other must be chosen. If the owner is a mine operator, the support estate is used to exploit the mineral estate. When the right of support is held, by the surface owner, its use is to support that surface and prevent subsidence. Thus, although Pennsylvania law does recognize the support estate as a “separate” property interest, id., it cannot be used profitably by one who does not also possess either the mineral estate or the surface estate. See Montgomery, The Development of the Right of Subjacent Support and the [716]*716“Third Estate in Pennsylvania,” 25 Temple L.Q. 1, 21 (1951).
To focus upon the support estate separately when assessing the diminution of the value of plaintiffs’ property caused by the Subsidence Act therefore would serve little purpose. The support estate is more properly viewed as only one “strand” in the plaintiff’s “bundle” of property rights, which also includes the mineral estate. As the Court stated in Andrus, “[t]he destruction of one ‘strand’ of the bundle is not a taking, because the aggregate must be viewed in its entirety.” 444 U.S. at 65, 100 S.Ct. at 326. The use to which the mine operators wish to put the support estate is forbidden. However, because the plaintiffs still possess valuable mineral rights that enable them profitably to mine coal, subject only to the Subsidence Act's requirement that they prevent subsidence, their entire “bundle” of property rights has not been destroyed. That the statute and regulations challenged prevent the most profitable use of their property is not in itself dispositive. Id.; Central Eureka Mining Co., 357 U.S. at 168, 78 S.Ct. at 1104. This case thus does not present the situation confronted by the Court in Mahon, in which it was found that the Kohler Act prevented the mining companies from profitably exercising their right to mine coal. 260 U.S. at 414-15, 43 S.Ct. at 159-60.6
With respect to interference with reasonable investment-backed expectations, the support requirements do not appear to work so substantial an interference as to result in a taking. In Penn Central, the Court applied this test to the divided property context. It read Mahon as stating that a statute may “frustrate distinct investment-backed expectations” when it has “nearly the same effect as the complete destruction of rights [a party] had reserved from the owners of the surface land.” Penn Central, 438 U.S. at 127, 98 S.Ct. at 2661. The Court thus stressed that the coal owners in Mahon had expressly contracted with individual surface owners for the waiver of any claim for damages due to subsidence. It was reasonable for them to have distinct expectations, grounded upon those waivers, to be free to mine coal without liability for damage caused to the surface owners’ estates. The Kohler Act thwarted those expectations by shifting the burden contractually imposed upon the surface owners to the mine operators. In so doing the Act appeared only to affect private interests.
In the present appeal, however, the statute at issue is clearly designed to serve broad and legitimate public interests. The ownership of the support estate does not afford a mine operator a reasonable expectation to profit at the expense of the public at large, but only at the expense of the surface owner with whom it contracted. Likewise, ownership of the surface estate does not give a landowner the right to waive away rights of support if the legislature deems them necessary for the public good. Just as coal mining activities may [717]*717be restricted when they intrude upon the interests of adjacent owners, or the holders of water rights, police power may be invoked to protect the public. Thus, a property owner’s reasonable expectations as to the possible uses of his property are always circumscribed by the limitations on its use that may be imposed by the state in the public interest. “A prohibition simply upon the use of property for purposes that are declared, by valid legislation, to be injurious to the health, morals, or safety of the community, cannot in any just sense, be deemed a taking or an appropriation of property for the public benefit.” Goldblatt, 369 U.S. at 593, 82 S.Ct. at 989 (quoting Mugler v. Kansas, 123 U.S. 623, 668-69, 8 S.Ct. 273, 300-01, 31 L.Ed. 205 (1887)). The support requirements thus cannot be said to interfere with reasonable investment-backed expectations.
In light of the character of the governmental action and the degree of interference present, we cannot say that the district court erred in concluding that sections 4 and 5 do not violate the takings clause.
Ill
Section 6 of the Act provides that if “the removal of coal or other mining operations ... causes damage to structures set forth in section 4” the operator must “submit evidence that such damage has been repaired or that all claims arising therefrom have been satisfied____” 52 Pa.Cons.Stat. Ann. § 1406.6. This duty is apparently imposed even though a mine operator has left coal in place beneath that structure or holds a release or waiver of damage from the surface owner.
Plaintiffs argue that this provision violates the contract clause of the Constitution.7 We believe that the district court properly rejected this claim. Two levels of scrutiny are employed to evaluate challenges under the contract clause: one applies when economic or social legislation affects a wholly private contract, and the other when the state itself is a party to the contract. See Energy Reserves Group, Inc. v. Kansas Power and Light Co., 459 U.S. 400, 103 S.Ct. 697, 74 L.Ed.2d 569 (1983). Under either level of scrutiny, the threshold inquiry is whether the state law has substantially impaired a contractual relationship. See, e.g., Allied Structural Steel Co. v. Spanraus, 438 U.S. 234, 244, 98 S.Ct. 2716, 2722, 57 L.Ed.2d 727 (1978). If the state regulation does constitute a substantial impairment, the second step is to determine whether there is a significant and legitimate public purpose behind the regulation, such as the remedying of broad and general social or economic problems. E.g., Troy Ltd. v. Renna, 727 F.2d 287, 297 (3d Cir.1984); see also Energy Reserve, 459 U.S. at 411, 103 S.Ct. at 704. Once a legitimate public purpose has been identified, the court must address whether the adjustment of the parties' rights and responsibilities is based upon reasonable conditions and is of a character appropriate to the legislature’s public purpose. E.g., United States Trust Co. v. New Jersey, 431 U.S. 1, 22, 97 S.Ct. 1505, 1517, 52 L.Ed.2d 92 (1977).
It is at this third stage that the analysis shifts for contracts involving the government. Unless the state itself is a contracting party the courts should defer to the legislative judgment as to the reasonableness of the particular measure, as is customary in reviewing economic and social regulations in other contexts. See United States Trust, 431 U.S. at 22-23, 97 S.Ct. at 1517-18. If the state is a party, however, the court must inquire whether a less drastic modification would be sufficient and whether the legislation was reasonable in light of changed circumstances. Id. at 30-32, 97 S.Ct. at 1521-23; see also Troy, 727 F.2d at 296.
The district court held, and the parties agree, that although section 6 does operate as a substantial impairment of con[718]*718tractual relationships, the state has shown that the regulation serves a legitimate public purpose. With regard to the contracts between private parties, however, the plaintiffs assert that the district court did not exercise sufficient scrutiny at the third step of the test, i.e., whether the legislation was based upon reasonable conditions and of a character appropriate to the public purposes behind the Act. We disagree. The legislature expressly found that subsidence damage devastated many surface structures and thus endangered the health, safety, and economic welfare of the Commonwealth and its people. The legislature concluded, not without basis, that property owners who received damage awards would repair their property, thus preserving the state’s limited tax base and land available for development. In the alternative, the operator might repair the damage, accomplishing the same end. As we noted in Troy, in “assessing the reasonableness of these purposes, we are admonished not to substitute our views for those of the legislature. Because the state is not itself a contracting party, in this sphere of economic and social regulation we ‘properly defer to legislative judgment as to the necessity and reasonableness of’ the Act.” 727 F.2d at 298 (quoting Energy Reserves, 459 U.S. at 413, 103 S.Ct. at 706). The district court did not err in rejecting this contract claim.8
IV.
Section 89.147(b) of the DER regulations provides:
When underground coal mining activities reduce the value or the reasonably foreseeable uses of surface land, the operator shall restore the land to the extent technologically and economically feasible to its premining condition.
25 Pa.Admin.Code § 89.147(b). Plaintiffs urge that this section violates the contract clause by imposing upon them the obligation to restore the surface estate even in situations in which they have contracted for the right to remove all coal, including support coal.
The companies concede that the public has a legitimate interest in the restoration of land damaged by subsidence, but argue that § 89.147 is not reasonably tied to that permissible interest. They also acknowledge that the provision “actually furthers, in some sense, a public purpose.” Brief for Appellants at 47. Still, they insist that the imposition of a duty to restore land where the damage was unavoidable and nonnegligent and the surface owner has waived his right to damages is unjustifiable. The legislative finding that the restoration of damaged land is central to the future economic well-being of the Commonwealth is entitled to substantial deference. See Energy Reserves, 459 U.S. at 412-13, 103 S.Ct. at 705-06. Especially in light of the fact that § 89.147 does not impose an absolute obligation, but is conditioned on both economic and technological feasibility, we cannot say that the district court erred [719]*719in concluding that the restoration requirements are reasonable conditions of a character appropriate to the public purposes. See Troy, 727 F.2d at 297-98.
V.
The mine operators finally allege that section 15 of the Subsidence Act is an invalid exercise of the power of eminent domain because it serves no “public use.” That portion of the statute permits the owners of structures not otherwise protected by the Act to purchase, for “just compensation,” the “amount of the coal necessary to be left in place for surface support.” 52 Pa.Cons.Stat.Ann. § 1406.15.
Our disposition of this issue is guided primarily by the Supreme Court’s recent decision in Hawaii Housing Authority v. Midkiff, 467 U.S. 229, 104 S.Ct. 2321, 81 L.Ed.2d 186 (1984), which was handed down after the district court’s ruling in this case. Midkiff dealt with a challenge to a state law that created a land condemnation arrangement whereby title in real property was taken from lessors and transferred to lessees in order to reduce the social and economic evils of a land oligopoly traceable to the early high chiefs of the Hawaiian Islands. The landowners claimed that the statute violated the Constitution’s public use clause, U.S. Const, amend. V (“private property [shall not] be taken for public use, without just compensation”).
The Court explained that the “public use” requirement of the Constitution is coterminous with the scope of a sovereign’s police powers. Thus, in reviewing a legislature’s judgment of what constitutes a public use, the scope of review by the courts is “an extremely narrow one.” 104 S.Ct. at 2329 (quoting Berman v. Parker, 348 U.S. 26, 75 S.Ct. 98, 99 L.Ed.2d 27 (1954)). Courts are not to substitute their judgment for a legislature’s judgment as to what constitutes a public use “unless the use be palpably without reasonable foundation.” Id. (quoting United States v. Gettysburg Electric Ry. Co., 160 U.S. 668, 680, 16 S.Ct. 427, 429, 40 L.Ed. 576 (1896)). Under that standard of review the Supreme Court had “no trouble concluding that the Hawaii Act is constitutional.” 104 S.Ct. at 2330.
In the present appeal the mine operators recognize the obstacle that Midkiff places in their path. Nevertheless they press their claim, arguing that an important distinction, and the fatal flaw of section 15, is that there is no requirement in the Act that the surface owner demonstrate that the coal being purchased at a just price is absolutely necessary to avoid subsidence damage. We find no merit in this argument. In Midkiff the Court refused to “second guess” the legislature’s presumption that because a number of persons had delcared they were willing but unable to buy lots at fair prices the land market was malfunctioning. Id. Similarly, we are unwilling to find wholly irrational the Pennsylvania legislature’s presumption that only those surface owners that needed coal for support of the surface would avail themselves of the right to purchase such support coal.9 Under Midkiff the question is not whether the state plan is perfect, or the best possible scheme, or even likely to achieve its intended goals. The constitutional requirement is satisfied if the state legislature rationally could have believed that the Act would promote its objective. Id.; see also Western & Southern Life Insurance Co. v. State Board of Equalization, 451 U.S. 648, 671-72, 101 S.Ct. 2070, 2084-85, 68 L.Ed.2d 574 (1981). We cannot say the legislature’s conclusion was irrational.
VI.
The Commonwealth has plenary power to engage in regulation that adjusts economic benefits and burdens for the common good. See, e.g., Penn Central, 438 U.S. at 124-25, 98 S.Ct. at 2659-60; Hadacheck, 239 U.S. 394 at 410, 36 S.Ct. 143, 145, 60 L.Ed. 348. We conclude that in this [720]*720case the Commonwealth has exercised that power without breaching constitutional boundaries. For the foregoing reasons, the district court’s grant of summary judgment will be affirmed.