with whom The Chief Justice joins and with whom Justice Powell and Justice Rehnquist join as to Parts I, II, and III, dissenting.
The Court acknowledges that we noted probable jurisdiction in this case “[bjecause of the importance of the ques*354tion whether a monetary remedy in inverse condemnation is constitutionally required in appropriate cases involving regulatory takings,” ante, at 348, but avoids this issue by holding that the antecedent question — whether appellant adequately stated a takings claim — should be answered in the negative. I disagree. The factual allegations that we must consider, when the opinion below is correctly read, do state a takings claim and therefore present the remedial question that we have thrice before sought to resolve. See Williamson County Regional Planning Comm’n v. Hamilton Bank, 473 U. S. 172 (1985); San Diego Gas & Electric Co. v. San Diego, 450 U. S. 621 (1981); Agins v. Tiburon, 447 U. S. 255 (1980).
HH
The Court recognizes that “the complaint alleged that appellant was deprived of all beneficial use of its property,” ante, at 352, n. 8, but concludes:
“The California Court of Appeal, whose opinion on matters of local law and local pleading we must respect, cf. Agins v. Tiburon, 447 U. S. 255, 259-260, n. 5 (1980), apparently rejected what the Superior Court labeled a ‘conclusionary’ allegation of futility, and explained that appellant could seek an administrative application of the Yolo County General Plan and Zoning Ordinances to its property which, for aught that appears, would allow development to proceed.” Ibid.
The Court thus ignores the allegations in the complaint that the effect of the county’s denial of appellant’s subdivision application in conjunction with the reasons behind that denial and other actions taken by the appellees has been to deprive the appellant of any use of its property “which is not (a) economically infeasible, (b) prohibited by law, or (c) prevented by actions taken by [the appellees].” Fourth Amended Complaint, App. 46. The Court also disregards appellant’s allegation that the actions of the appellees demonstrate “That ANY APPLICATION FOR A ZONE CHANGE, VARIANCE OR OTHER *355relief would be futile.” Id., at 58. Instead, the Court focuses on the denial of the particular subdivision application alone.
In my view, the Court does not fairly read the record and the opinion below. Appellant’s initial complaint filed in Superior Court alleged that although the property was zoned for residential use by the county it was designated an “Agricultural Preserve or Reserve” by the city. The complaint further alleged that even though the property lay in the county outside of the city’s boundaries, the county implemented city policy relegating the land to agricultural uses. See Complaint of Oct. 13, 1977, pp. 9-12. Finally, the complaint asserted that the property was agriculturally impaired and could not economically be used for agricultural purposes. See id., at 5, 16.
In sustaining the appellees’ demurrer to this complaint, the Superior Court accepted as true the allegation that the property had effectively been rezoned agricultural but noted that there was no allegation that the property could not be used for a variety of nonagricultural purposes explicitly allowed in agricultural zones under the county and city codes. See Order of Mar. 30, 1978, in No. 36655 (Cal. Super. Ct., Yolo County), pp. 6-8. .The conclusion was that “[t]he failure to allege the property in question useless for other permitted purposes in an agricultural zone over and above an agricultural use renders the [inverse condemnation cause of action] demurrable.” Id., at 8.
In the Fourth Amended Complaint, the complaint that formed the basis for the judgment below, appellant responded to this earlier ruling by specifically alleging that the property was not suitable for the other uses permitted in an agricultural zone and by asserting facts in support of this allegation. See App. 52-58. The Superior Court, however, indicated that it found these allegations “conclusionary,” although it did not rely on this determination in sustaining the demurrer to the complaint, relying instead on the California Supreme Court’s general ruling in Agins v. City of Tiburon, *35624 Cal. 3d 266, 272-277, 598 P. 2d 25, 28-31 (1979), aff’d on other grounds, 447 U. S. 255 (1980), that no inverse condemnation action may be brought as a result of a land use regulation.1 The crucial fact here is that the Superior Court denominated “conclusionary” only those allegations in the complaint that rejected the possibility of feasible nonagri-cultural uses of the property that would be consistent with agricultural zoning. It did not categorize as “conclusionary” appellant’s allegations that all economically beneficial residential uses were foreclosed by the appellees’ actions.
In reviewing the Superior Court’s ruling on the demurrer to the Fourth Amended Complaint, the California Court of Appeal first noted that it would not consider whether the complaint was barred by the failure to exhaust administrative remedies or by res judicata. App. 125-126. It then summarized the allegations of the complaint, including the allegations that the property was not suitable for agricultural use or any of the other uses permitted in the county code and that it was suitable for residential use but that the county and city had acted to prevent this use entirely. Id., at 127-129. The Court of Appeal also noted that appellant had alleged that “[a]ny application for a zone change, variance or other relief would be futile.” Id., at 129. Nowhere did the court state that as a matter of California demurrer law it was rejecting any of these allegations as not properly pleaded. And nowhere did it refer to the Superior Court’s statement that the allegations as to the infeasibility of the nonagricul-tural uses that would be consistent with agricultural zoning might not be properly pleaded.
*357Having recited all of these allegations without indicating that it was rejecting any of them, the Court of Appeal first held that no cause of action was stated in inverse condemnation. This holding, it noted, was compelled by the California Supreme Court’s ruling in Agins that there is no such remedy for takings alleged to result from land use regulation. Id., at 130-132. See Agins, 24 Cal. 3d, at 272-277, 598 P. 2d, at 28-31.
In the alternative, however, the Court of Appeal found that even if such a cause of action were available, appellant had not stated a takings claim. The court concluded that “[p]ared to their essence, the allegations are that [appellant] purchased property for residential development, the property is zoned for residential development, [appellant] submitted an application for approval of development of the property into 159 residential units, and, in part at the urging of the City, the County denied approval of the application.” App. 132. The court then observed that this situation was “not unlike” that in Agins, in which a zoning ordinance that restricted a landowner of five acres to building a maximum of five residences on his property was found not to constitute a taking since on its face the ordinance still allowed that level of development, which was a reasonable use of the property. See Agins, 447 U. S., at 262-263; Agins, 24 Cal. 3d, at 277, 598 P. 2d, at 31. Citing Agins, the Court of Appeal than determined that appellant had not stated a takings cause of action because appellees’ refusal to allow the intensive development requested by appellant “does not preclude less intensive, but still valuable development.” App. 133.
In my view, given the absence of any expression of disapproval by the Court of Appeal of any of the appellant’s allegations summarized in its opinion and given the fact that the Superior Court had not labeled appellant’s allegations of futility “conclusionary,” there is no reason to read into this last statement by the Court of Appeal a state-law ruling that the allegations of futility were not well pleaded. Instead, the *358Court of Appeal’s focus on what it termed the essence of appellant’s complaint together with its conclusion that with respect to these essential allegations this case was analytically the same as Agins imply that it believed that as a matter of federal takings law certain allegations controlled in terms of determining if a takings claim had been stated. Specifically, in concluding that the allegations of futility were not material and in determining that under Agins further application must be made before a takings claim could be stated, the Court of Appeal held that no takings cause of action had been stated because no reapplication had been made, even if further application would be useless.
II
Whethera regulatory taking has occurred is an inquiry that cannot be completed until a final decision is made as to how the allegedly confiscatory regulations apply to the pertinent property. Williamson County Regional Planning Comm’n, 473 U. S., at 190-191. Thus, in Penn Central Transportation Co. v. New York City, 438 U. S. 104, 136-137 (1978), and in Agins, supra, at 262-263, we considered for takings purposes only the actual regulatory decision that had been made by the governmental decisionmaker; we declined to speculate as to further restrictions that might be imposed. In Hodel v. Virginia Surface Mining & Reclamation Assn., Inc., 452 U. S. 264, 293-297 (1981), we refused to consider a takings claim based on general regulatory provisions that had not yet been applied to specific properties and that were susceptible of administrative exemption. Most recently, in Williamson County Regional Planning Comm’n, the Court determined that the denial of a particular use for a property did not constitute a final decision where variance procedures were available that “[left] open the possibility that [the landowner] may develop the subdivision according to its plat after obtaining the variances.” 473 U. S., at 193-194.
*359These holdings recognize that a regulatory takings determination is closely tied to the facts of a particular case and that there is often an ongoing process by which the relevant regulatory decisions are made. Given these characteristics of a regulatory taking, the final decision requirement is necessary to ensure that “the initial decisionmaker has arrived at a definitive position on the issue that inflicts an actual, concrete injury.” Id., at 193. Nothing in our cases, however, suggests that the decisionmaker’s definitive position may be determined only from explicit denials of property-owner applications for development. Nor do these cases suggest that repeated applications and denials are necessary to pinpoint that position.
Moreover, I see no reason for importing such a requirement into the “final decision” analysis. A decisionmaker’s definitive position may sometimes be determined by factors other than its actual decision on the issue in question. For example, if a landowner applies to develop its land in a relatively intensive manner that is consistent with the applicable zoning requirements and if the governmental body denies that application, explaining that all development will be barred under its interpretation of the zoning ordinance, I would find that a final decision barring all development has been made — even though the landowner did not apply for a less intensive development. Although a landowner must pursue reasonably available avenues that might allow relief, it need not, I believe, take patently fruitless measures.
The Court of Appeal’s reliance on Agins in this case was therefore misplaced. Appellant alleged not simply that its application had been denied but that the overall effect of (1) that denial, (2) the reasons given for the denial, and (3) other actions taken by appellees to prevent appellant from ever being able to meet county development requirements was that appellant’s property had been taken. When the Court of Appeal purported to reduce appellant’s claim to its essence, it ignored a critical distinction between Agins, in *360which there was no indication that upon application the property owner would not be allowed to develop his property in some economically beneficial manner, and the factual situation here, in which further application would allegedly be futile. In this case, I believe that appellant sufficiently alleged a final decision denying it all reasonable economically beneficial use of its property.2
Ill
Assuming that appellant adequately alleged a final decision, the next question is whether a takings cause of action was stated by the allegations in the complaint. Discerning the answer to this question involves two separate inquiries: Whether a land use regulation restricting the use of property may ever amount to a taking and, if the answer to this first inquiry is affirmative, whether the allegations here are sufficient to state a takings claim.
As to the first question, our cases have long indicated that police-power regulations may rise to the level of a taking if the restrictions they impose are sufficiently severe. See, e. g., Agins, 447 U. S., at 260; PruneYard Shopping Center v. Robins, 447 U. S. 74, 83 (1980); Kaiser Aetna v. United States, 444 U. S. 164, 174-175 (1979); Andrus v. Allard, 444 U. S. 51, 65-66 (1979); Penn Central, supra, at 130-131, 138, n. 36; United States v. Central Eureka Mining Co., 357 U. S. 155, 168 (1958); Pennsylvania Coal Co. v. Mahon, 260 U. S. 393, 415-416 (1922). Nevertheless, the California Supreme Court in Agins concluded:
“[A] landowner alleging that a zoning ordinance has deprived him of substantially all use of his land may attempt through declaratory relief or mandamus to invalidate the ordinance as excessive regulation in violation of the Fifth Amendment to the United States Constitu*361tion. ... He may not, however, elect to sue in inverse condemnation and thereby transmute an excessive use of the police power into a lawful taking for which compensation in eminent domain must be paid.” 24 Cal. 3d, at 273, 598 P. 2d, at 28.3
In addition to being inconsistent with this Court’s statements, this view, as Justice Brennan explained in his dissent in San Diego Gas, ignores the fact that
“[pjolice power regulations such as zoning ordinances and other land-use restrictions can destroy the use and enjoyment of property in order to promote the public good just as effectively as formal condemnation or physical invasion of property. From the property owner’s point of view, it may matter little whether his land is condemned or flooded, or whether it is restricted by regulation to use in its natural state, if the effect in both cases is to deprive him of all beneficial use of it. From the government’s point of view, the benefits flowing to the public from preservation of open space through regulation may be equally great as from creating a wildlife refuge through formal condemnation or increasing electricity production through a dam project that floods private property. ... It is only logical, then, that government action other than acquisition of title, occupancy, or physical invasion can be a ‘taking,’ and therefore a defacto exercise of the power of eminent domain, where the effects completely deprive the owner of all or most *362of his interest in the property.” 450 U. S., at 652-653 (citations and footnotes omitted).
I agree that land use restrictions may constitute a taking under the Constitution.
This resolution of the general question brings me to the more specific question whether the allegations in the complaint here were sufficient to state a takings claim. Here, appellant alleged the existence of a final decision denying it all economically beneficial use of its property. It also alleged that it had paid “good and valuable consideration,” App. 43, for the property. Factual allegations of interference with reasonable investment-backed expectations and denial of all economically feasible use of the property are certainly sufficient allegations of a regulatory taking to state a cause of action. See, e. g., Penn Central, 438 U. S., at 136-138. Consequently, I would hold that appellant adequately alleged a taking.
IV
The final question presented is whether a State can limit to declaratory and injunctive relief the remedies available to a person whose property has been taken by regulation or whether the State must pay compensation for the interim period between the time that the government first “took” the property and the time that the “taking” is rescinded by amendment of the regulation.4 On this question, I am again in substantial agreement with Justice Brennan’s discussion in San Diego Gas. See 450 U. S., at 653-660. Even where a property owner is deprived of its property only temporarily, if that deprivation amounts to a taking the Constitu*363tion requires that just compensation be paid. If the governmental body that has taken the property decides to rescind the taking by amending the regulation, that does not reverse the fact that the property owner has been deprived of its property in the interim. “[I]t is only fair that the public bear the cost of benefits received during the interim period between application of the regulation and the government entity’s rescission of it.” Id., at 656-657. Thus, just compensation must be available for the period between the time of the final decision taking the property and the time that that decision is rescinded.
I recognize that such a constitutional rule admits of problems of administration that are by no means insignificant. Aside from the problems that the Court has already addressed in some measure regarding the determination of when a taking shall be deemed to have occurred, there are questions of valuation and of procedure. As to the latter, the Constitution requires no particular procedures, although as the Court today notes, “[a] property owner is of course not required to resort to piecemeal litigation or otherwise unfair procedures in order to obtain this determination.” Ante, at 350, n. 7. As to the former, the issue of what constitutes just compensation in this context is a particularly meaty one, which merits substantial reflection and analysis. Nevertheless, these unsettled questions should not deter us from acting to protect constitutional requirements in this type of case. Consequently, I would vacate the judgment below and remand for further proceedings not inconsistent with the views I have expressed.
V
In sum, I believe that the Court of Appeal’s decision is most properly read as taking as true all of the allegations in the complaint, including the allegations of futility, and as rejecting those allegations as insufficient as a matter of substantive takings law. At the very least, the Court’s reading of the opinion below, however plausible, is not the only sen*364sible reading of that opinion. Given this arguable ambiguity, I would not, as the Court does, withdraw from appellant all chance of relief at this stage. That is, if the Court of Appeal in fact did reach its judgment by the reasoning I have summarized rather than as the Court hypothesizes, appellant should not be precluded from seeking relief on the facts currently alleged in the complaint. I would at least vacate the judgment below and remand for explanation by the Court of Appeal as to the precise basis for its judgment.
The Superior Court also sustained the demurrer on the ground that appellant had failed to exhaust administrative and judicial remedies; that the county’s denial of appellant’s subdivision application was res judicata not subject to collateral attack in the Superior Court; and that no taking in the form of actual “invasion or appropriation of a cognizably valuable property right” had been alleged. App. Ill, 116.
1 emphasize that the futility of further application would have to be proved at trial for appellant to prevail here on the merits. I address only the question whether appellant’s allegations of futility are sufficient support for assuming that a final decision has been made.
Although the California Supreme Court’s ruling in Agins rests on the rationale that excessive land use regulation simply cannot constitute a lawful taking, the Court of Appeal in this case seemed to proceed on the assumption that such regulation could constitute a taking but that no inverse condemnation remedy for such a taking would be available. See App. 131. My discussion here follows the reasoning given by the California Supreme Court in Agins rather than the somewhat inexact summary of that reasoning given by the Court of Appeal below.
1 assume here that the normal action by the governmental entity following a determination that a particular regulation constitutes a taking wall be to rescind the regulation. I believe that this is a permissible course of action, limiting liability for the taking to the interim period. See San Diego Gas & Electric Co. v. San Diego, 450 U. S. 621, 658 (1981) (BRENNAN, J., dissenting). Of course, the governmental entity could actually condemn the property and pay permanent compensation for it.