with whom Justice Sotomayor joins, concurring in part and concurring in the judgment.
The Court’s analysis of the principles that control ownership of the land in question, and of the rights of petitioner’s members as adjacent owners, is correct in my view, leading to my joining Parts I, IV, and V of the Court’s opinion. As Justice Breyer observes, however, this case does not re*734quire the Court to determine whether, or when, a judicial decision determining the rights of property owners can violate the Takings Clause of the Fifth Amendment of the United States Constitution. This separate opinion notes certain difficulties that should be considered before accepting the theory that a judicial decision that eliminates an “established property right,” ante, at 726, constitutes a violation of the Takings Clause.
The Takings Clause is an essential part of the constitutional structure, for it protects private property from expropriation without just compensation; and the right to own and hold property is necessary to the exercise and preservation of freedom. The right to retain property without the fact or even the threat of that sort of expropriation is, of course, applicable to the States under the Due Process Clause of the Fourteenth Amendment. Chicago, B. & Q. R. Co. v. Chicago, 166 U. S. 226, 239 (1897).
The right of the property owner is subject, however, to the rule that the government does have power to take property for a public use, provided that it pays just compensation. See First English Evangelical Lutheran Church of Glendale v. County of Los Angeles, 482 U. S. 304, 314-315 (1987). This is a vast governmental power. And typically, legislative bodies grant substantial discretion to executive officers to decide what property can be taken for authorized projects and uses. As a result, if an authorized executive agency or official decides that Blackacre is the right place for a fire station or Greenacre is the best spot for a freeway interchange, then the weight and authority of the State are used to take the property, even against the wishes of the owner, who must be satisfied with just compensation.
In the exercise of their duty to protect the fisc, both the legislative and executive branches monitor, or should monitor, the exercise of this substantial power. Those branches are accountable in their political capacity for the proper discharge of this obligation.
*735To enable officials to better exercise this great power in a responsible way, some States allow their officials to take a second look after property has been condemned and a jury returns a verdict setting the amount of just compensation. See, e. g., Cal. Civ. Proc. Code Ann. § 1268.510 (West 2007). If the condemning authority, usually acting through the executive, deems the compensation too high to pay for the project, it can decide not to take the property at all. The landowner is reimbursed for certain costs and expenses of litigation and the property remains in his or her hands. See, e. g., § 1268.610(a).
This is just one aspect of the exercise of the power to select what property to condemn and the responsibility to ensure that the taking makes financial sense from the State’s point of view. And, as a matter of custom and practice, these are matters for the political branches — the legislature and the executive — not the courts. See First English, supra, at 321 (“[T]he decision to exercise the power of eminent domain is a legislative function”).
If a judicial decision, as opposed to an act of the executive or the legislature, eliminates an established property right, the judgment could be set aside as a deprivation of property without due process of law. The Due Process Clause, in both its substantive and procedural aspects, is a central limitation upon the exercise of judicial power. And this Court has long recognized that property regulations can be invalidated under the Due Process Clause. See, e.g., Lingle v. Chevron U. S. A. Inc., 544 U. S. 528, 542 (2005); Goldblatt v. Hempstead, 369 U. S. 590, 591, 592-593 (1962); Demorest v. City Bank Farmers Trust Co., 321 U. S. 36, 42-43 (1944); Broad River Power Co. v. South Carolina ex rel. Daniel, 281 U. S. 537, 539, 540-541 (1930); Washington ex rel. Seattle Title Trust Co. v. Roberge, 278 U. S. 116, 121 (1928); Nectow v. Cambridge, 277 U. S. 183, 188 (1928); Village of Euclid v. Ambler Realty Co., 272 U. S. 365, 395 (1926); see also Pennsylvania Coal Co. v. Mahon, 260 U. S. 393, 413 (1922) (there *736must be limits on government’s ability to diminish property values by regulation “or the contract and due process clauses are gone”). It is thus natural to read the Due Process Clause as limiting the power of courts to eliminate or change established property rights.
The Takings Clause also protects property rights, and it “operates as a conditional limitation, permitting the government to do what it wants so long as it pays the charge.” Eastern Enterprises v. Apfel, 524 U. S. 498, 545 (1998) (Kennedy, J., concurring in judgment and dissenting in part). Unlike the Due Process Clause, therefore, the Takings Clause implicitly recognizes a governmental power while placing limits upon that power. Thus, if the Court were to hold that a judicial taking exists, it would presuppose that a judicial decision eliminating established property rights is “otherwise constitutional” so long as the State compensates the aggrieved property owners. Ibid. There is no clear authority for this proposition.
When courts act without direction from the executive or legislature, they may not have the power to eliminate established property rights by judicial decision. “Given that the constitutionality” of a judicial decision altering property rights “appears to turn on the legitimacy” of whether the court’s judgment eliminates or changes established property rights “rather than on the availability of compensation, . . . the more appropriate constitutional analysis arises under general due process principles rather than under the Takings Clause.” Ibid. Courts, unlike the executive or legislature, are not designed to make policy decisions about “the need for, and likely effectiveness of, regulatory actions.” Lingle, supra, at 545. State courts generally operate under a common-law tradition that allows for incremental modifications to property law, but “this tradition cannot justify a carte blanch judicial authority to change property definitions wholly free of constitutional limitation.” Walston, The Con*737stitution and Property: Due Process, Regulatory Takings, and Judicial Takings, 2001 Utah L. Rev. 379, 435.
. The Court would be on strong footing in ruling that a judicial decision that eliminates or substantially changes established property rights, which are a legitimate expectation of the owner, is “arbitrary or irrational” under the Due Process Clause. Lingle, 544 U. S., at 542; see id., at 548-549 (Kennedy, J., concurring); see also Perry v. Sindermann, 408 U. S. 593, 601 (1972) (“‘[Pjroperty’” interests protected by the Due Process Clauses are those “that are secured by ‘existing rules or understandings’ ” (quoting Board of Regents of State Colleges v. Roth, 408 U. S. 564, 577 (1972))). Thus, without a judicial takings doctrine, the Due Process Clause would likely prevent a State from doing “by judicial decree what the Takings Clause forbids it to do by legislative fiat.” Ante, at 714. The objection that a due process claim might involve close questions concerning whether a judicial decree extends beyond what owners might have expected is not a sound argument; for the same close questions would arise with respect to whether a judicial decision is a taking. See Apfel, supra, at 541 (opinion of Kennedy, J.) (“Cases attempting to decide when a regulation becomes a taking are among the most litigated and perplexing in current law”); Penn Central Transp. Co. v. New York City, 438 U. S. 104, 123 (1978) (“The question of what constitutes a ‘taking’ for purposes of the Fifth Amendment has proved to be a problem of considerable difficulty”).
To announce that courts too can effect a taking when they decide cases involving property rights would raise certain difficult questions. Since this case does not require those questions to be addressed, in my respectful view, the Court should not reach beyond the necessities of the case to announce a sweeping rule that court decisions can be takings, as that phrase is used in the Takings Clause. The evident reason for recognizing a judicial takings doctrine would be *738to constrain the power of the judicial branch. Of course, the judiciary must respect private ownership. But were this Court to say that judicial decisions become takings when they overreach, this might give more power to courts, not less.
Consider the instance of litigation between two property owners to determine which one bears the liability and costs when a tree that stands on one property extends its roots in a way that damages adjacent property. See, e. g., Fancher v. Fagella, 274 Va. 549, 650 S. E. 2d 519 (2007). If a court deems that, in light of increasing urbanization, the former rule for allocation of these costs should be changed, thus shifting the rights of the owners, it may well increase the value of one property and decrease the value of the other. This might be the type of incremental modification under state common law that does not violate due process, as owners may reasonably expect or anticipate courts to make certain changes in property law. The usual due process constraint is that courts cannot abandon settled principles. See, e. g., Rogers v. Tennessee, 532 U. S. 451, 457 (2001) (citing Bouie v. City of Columbia, 378 U. S. 347, 354 (1964)); Apfel, supra, at 548-549 (opinion of Kennedy, J.); see also Perry, supra, at 601; Roth, supra, at 577.
But if the state court were deemed to be exercising the power to take property, that constraint would be removed. Because the State would be bound to pay owners for takings caused by a judicial decision, it is conceivable that some judges might decide that enacting a sweeping new rule to adjust the rights of property owners in the context of changing social needs is a good idea. Knowing that the resulting ruling would be a taking, the courts could go ahead with their project, free from constraints that would otherwise confine their power. The resulting judgment as between the property owners likely could not be set aside by some later enactment. See Plaut v. Spendthrift Farm, Inc., 514 U. S. 211, 217 (1995) (leaving open whether legislation re*739opening final judgments violates Due Process Clause). And if the litigation were a class action to decide, for instance, whether there are public rights of access that diminish the rights of private ownership, a State might find itself obligated to pay a substantial judgment for the judicial ruling. Even if the legislature were to subsequently rescind the judicial decision by statute, the State would still have to pay just compensation for the temporary taking that occurred from the time of the judicial decision to the time of the statutory fix. See First English, 482 U. S., at 321.
The idea, then, that a judicial takings doctrine would constrain judges might just well have the opposite effect. It would give judges new power and new assurance that changes in property rights that are beneficial, or thought to be so, are fair and proper because just compensation will be paid. The judiciary historically has not had the right or responsibility to say what property should or should not be taken.
Indeed, it is unclear whether the Takings Clause was understood, as a historical matter, to apply to judicial decisions. The Framers most likely viewed this Clause as applying only to physical appropriation pursuant to the power of eminent domain. See Lucas v. South Carolina Coastal Council, 505 U. S. 1003, 1028, n. 15 (1992). And it appears these physical appropriations were traditionally made by legislatures. See 3 J. Story, Commentaries on the Constitution of the United States §1784, p. 661 (1833). Courts, on the other hand, lacked the power of eminent domain. See 1 W. Blackstone, Commentaries 135 (W. Lewis ed. 1897). The Court’s Takings Clause jurisprudence has expanded beyond the Framers’ understanding, as it now applies to certain regulations that are not physical appropriations. See Lucas, supra, at 1014 (citing Mahon, 260 U. S. 393). But the Court should consider with care the decision to extend the Takings Clause in a manner that might be inconsistent with historical practice.
*740There are two additional practical considerations that the Court would need to address before recognizing judicial takings. First, it may be unclear in certain situations how a party should properly raise a judicial takings claim. “[I]t is important to separate out two judicial actions — the decision to change current property rules in a way that would constitute a taking, and the decision to require compensation.” Thompson, Judicial Takings, 76 Va. L. Rev. 1449, 1515 (1990). In some contexts, these issues could arise separately. For instance, assume that a state-court opinion explicitly holds that it is changing state property law, or that it asserts that it is not changing the law but there is no “fair or substantial basis” for this statement. Broad River, 281 U. S., at 540. (Most of these cases may arise in the latter posture, like inverse condemnation claims where the State says it is not taking property and pays no compensation.) Call this Case A. The only issue in Case A was determining the substance of state property law. It is doubtful that parties would raise a judicial takings claim on appeal, or in a petition for a writ of certiorari, in Case A, as the issue would not have been litigated below. Rather, the party may file a separate lawsuit — Case B — arguing that a taking occurred in light of the change in property law made by Case A. After all, until the state court in Case A changes the law, the party will not know if his or her property rights will have been eliminated. So res judicata probably would not bar the party from litigating the takings issue in Case B.
Second, it is unclear what remedy a reviewing court could enter after finding a judicial taking. It appears under our precedents that a party who suffers a taking is only entitled to damages, not equitable relief: The Court has said that “[ejquitable relief is not available to enjoin an alleged taking of private property for a public use ... when a suit for compensation can be brought against the sovereign subsequent to the taking,” Ruckelshaus v. Monsanto Co., 467 U. S. 986, 1016 (1984), and the Court subsequently held that the Tak*741ings Clause requires the availability of a suit for compensation against the States, First English, supra, at 321-322. It makes perfect sense that the remedy for a Takings Clause violation is only damages, as the Clause “does not proscribe the taking of property; it proscribes taking without just compensation.” Williamson County Regional Planning Comm’n v. Hamilton Bank of Johnson City, 473 U. S. 172, 194 (1985).
It is thus questionable whether reviewing courts could invalidate judicial decisions deemed to be judicial takings; they may only be able to order just compensation. In the posture discussed above where Case A changes the law and Case B addresses whether that change is a taking, it is not clear how the Court, in Case B, could invalidate the holding of Case A. If a single case were to properly address both a state court’s change in the law and whether the change was a taking, the Court might be able to give the state court a choice on how to proceed if there were a judicial taking. The Court might be able to remand and let the state court determine whether it wants to insist on changing its property law and paying just compensation or to rescind its holding that changed the law. Cf. First English, 482 U. S., at 321 (“Once a court determines that a taking has occurred, the government retains the whole range of options already available — amendment of the regulation, withdrawal of the invalidated regulation, or exercise of eminent domain”). But that decision would rest with the state court, not this Court; so the state court could still force the State to pay just compensation. And even if the state court decided to rescind its decision that changed the law, a temporary taking would have occurred in the interim. See ibid.
These difficult issues are some of the reasons why the Court should not reach beyond the necessities of the case to recognize a judicial takings doctrine. It is not wise, from an institutional standpoint, to reach out and decide questions that have not been discussed at much length by courts and *742commentators. This Court's dicta in Williamson County, supra, at 194-197, regarding when regulatory takings claims become ripe, explains why federal courts have not been able to provide much analysis on the issue of judicial takings. See San Remo Hotel, L. P. v. City and County of San Francisco, 545 U. S. 323, 351 (2005) (Rehnquist, C. J., concurring in judgment) (“ Williamson County’s state-litigation rule has created some real anomalies, justifying our revisiting the issue”). Until Williamson County is reconsidered, litigants will have to press most of their judicial takings claims before state courts, which are “presumptively competent... to adjudicate claims arising under the laws of the United States.” Tafflin v. Levitt, 493 U. S. 455, 458 (1990). If and when future cases show that the usual principles, including constitutional principles that constrain the judiciary like due process, are somehow inadequate to protect property owners, then the question whether a judicial decision can effect a taking would be properly presented. In the meantime, it seems appropriate to recognize that the substantial power to decide whose property to take and when to take it should be conceived of as a power vested in the political branches and subject to political control.