skip navigation

110 F.3d 150, reversed and remanded.
[ O'Connor ]
[ Thomas ]
[ Stevens ]
[ Breyer ]
[ Kennedy ]
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version

Opinion of Kennedy, J.


No. 97—42



[June 25, 1998]

Justice Kennedy, concurring in the judgment and dissenting in part.

The plurality’s careful assessment of the history and purpose of the statute in question demonstrates the necessity to hold it arbitrary and beyond the legitimate authority of the Government to enact. In my view, which is in full accord with many of the plurality’s conclusions, the relevant portions of the Coal Industry Retiree Health Benefit Act of 1992 (Coal Act), 26 U.S.C. § 9701 et seq. (1994 ed. and Supp. II), must be invalidated as contrary to essential due process principles, without regard to the Takings Clause of the Fifth Amendment. I concur in the judgment holding the Coal Act unconstitutional but disagree with the plurality’s Takings Clause analysis, which, it is submitted, is incorrect and quite unnecessary for decision of the case. I must record my respectful dissent on this issue.


The final Clause of the Fifth Amendment states:

“nor shall private property be taken for public use, without just compensation.” U.S. Const., Amdt. 5.

The provision is known as the Takings Clause. The concept of a taking under the Clause has become a term of art, and my discussion begins here.

Our cases do not support the plurality’s conclusion that the Coal Act takes property. The Coal Act imposes a staggering financial burden on the petitioner, Eastern Enterprises, but it regulates the former mine owner without regard to property. It does not operate upon or alter an identified property interest, and it is not applicable to or measured by a property interest. The Coal Act does not appropriate, transfer, or encumber an estate in land (e.g., a lien on a particular piece of property), a valuable interest in an intangible (e.g., intellectual property), or even a bank account or accrued interest. The law simply imposes an obligation to perform an act, the payment of benefits. The statute is indifferent as to how the regulated entity elects to comply or the property it uses to do so. To the extent it affects property interests, it does so in a manner similar to many laws; but until today, none were thought to constitute takings. To call this sort of governmental action a taking as a matter of constitutional interpretation is both imprecise and, with all due respect, unwise.

As the role of Government expanded, our experience taught that a strict line between a taking and a regulation is difficult to discern or to maintain. This led the Court in Pennsylvania Coal Co. v. Mahon, 260 U.S. 393 (1922), to try to span the two concepts when specific property was subjected to what the owner alleged to be excessive regulation. “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.” Id., at 415. The quoted sentence is, of course, the genesis of the so-called regulatory takings doctrine. See Lucas v. South Carolina Coastal Council, 505 U.S. 1003, 1014 (1992) (“Prior to Justice Holmes’s exposition in Pennsylvania Coal Co. v. Mahon, it was generally thought that the Takings Clause reached only a ‘direct appropriation’ of property or the functional equivalent of a ‘practical ouster of [the owner’s] possession’ ”) (citations omitted). Without denigrating the importance the regulatory taking concept has assumed in our law, it is fair to say it has proven difficult to explain in theory and to implement in practice. Cases attempting to decide when a regulation becomes a taking are among the most litigated and perplexing in current law. See 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”); Kaiser Aetna v. United States, 444 U.S. 164, 175 (1979) (the regulatory tak-
ing question requires an “essentially ad hoc, factual

Until today, however, one constant limitation has been that in all of the cases where the regulatory taking analysis has been employed, a specific property right or interest has been at stake. After the decision in Pennsylvania Coal Co. v. Mahon, supra, we confronted cases where specific and identified properties or property rights were alleged to come within the regulatory takings prohibition: air rights for high-rise buildings, Penn Central, supra; zoning on parcels of real property, e.g., MacDonald, Sommer & Frates v. Yolo County, 477 U.S. 340 (1986), Agins v. City of Tiburon, 447 U.S. 255 (1980); trade secrets, Ruckelshaus v. Monsanto Co., 467 U.S. 986 (1984); right of access to property, e.g., PruneYard Shopping Center v. Robins, 447 U.S. 74 (1980); Kaiser Aetna, supra; right to affix on structures, Loretto v. Teleprompter Manhattan CATV Corp., 458 U.S. 419 (1982); right to transfer property by devise or intestacy, e.g., Hodel v. Irving, 481 U.S. 704 (1987); creation of an easement, Dolan v. City of Tigard, 512 U.S. 374 (1994); Nollan v. California Coastal Comm’n, 483 U.S. 825 (1987); right to build or improve, Lucas, supra; liens on real property, Armstrong v. United States, 364 U.S. 40 (1960); right to mine coal, Keystone Bituminous Coal Assn. v. DeBenedictis, 480 U.S. 470 (1987); right to sell personal property, Andrus v. Allard, 444 U.S. 51 (1979); and the right to extract mineral deposits, Goldblatt v. Hempstead, 369 U.S. 590 (1962); United States v. Central Eureka Mining Co., 357 U.S. 155 (1958). The regulations in the cited cases were challenged as being so excessive as to destroy, or take, a specific property interest. The plurality’s opinion disregards this requirement and, by removing this constant characteristic from takings analysis, would expand an already difficult and uncertain rule to a vast category of cases not deemed, in our law, to implicate the Takings Clause.

The difficulties in determining whether there is a taking or a regulation even where a property right or interest is identified ought to counsel against extending the regulatory takings doctrine to cases lacking this specificity. The existence of at least this outer boundary for application of the regulatory takings rule provides some necessary predictability for governmental entities. Our definition of a taking, after all, is binding on all of the States as well as the Federal Government. The plurality opinion would throw one of the most difficult and litigated areas of the law into confusion, subjecting States and municipalities to the potential of new and unforeseen claims in vast amounts. The existing category of cases involving specific property interests ought not to be obliterated by extending regulatory takings analysis to the amorphous class of cases embraced by the plurality’s opinion today.

True, the burden imposed by the Coal Act may be just as great if the Government had appropriated one of Eastern’s plants, but the mechanism by which the Government injures Eastern is so unlike the act of taking specific property that it is incongruous to call the Coal Act a taking, even as that concept has been expanded by the regulatory takings principle. In the terminology of our regulatory takings analysis, the character of the governmental action renders the Coal Act not a taking of property. While the usual taking occurs when the Government physically acquires property for itself, e.g., Chicago, B. & Q. R. Co. v. Chicago, 166 U.S. 226 (1897), our regulatory takings analysis recognizes a taking may occur when property is not appropriated by the Government or is transferred to other private parties. See, e.g., United States v. Security Industrial Bank, 459 U.S. 70, 78 (1982) (“[O]ur cases show that takings analysis is not necessarily limited to out-
right acquisitions by the government for itself”); Loretto, supra (transfer of physical space from landlords to cable companies).

As the range of governmental conduct subjected to takings analysis has expanded, however, we have been careful not to lose sight of the importance of identifying the property allegedly taken, lest all governmental action be subjected to examination under the constitutional prohibition against taking without just compensation, with the attendant potential for money damages. We have asked how the challenged governmental action is implemented with particular emphasis on the extent to which a specific property right is affected. See id., at 432 (physical invasion “is a government action of such a unique character that it is a taking without regard to other factors”); Hodel, supra, at 715—716 (declaring a law, which otherwise would not be a taking because of its insignificant economic impact, a taking because the character of the governmental action destroyed the right to pass property to one’s heirs, a right which “has been part of the Anglo-American legal system since feudal times”); Penn Central, supra, at 124 (“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”) (citation omitted). The Coal Act neither targets a specific property interest nor depends upon any particular property for the operation of its statutory mechanisms. The liability imposed on Eastern no doubt will reduce its net worth and its total value, but this can be said of any law which has an adverse economic effect.

The circumstance that the statute does not take money for the Government but instead makes it payable to third persons is not a factor I rely upon to show the lack of a taking. While there are instances where the Government’s self-enrichment may make it all the more evident a taking has occurred, e.g., Webb’s Fabulous Pharmacies, Inc. v. Beckwith, 449 U.S. 155 (1980); United States v. Causby, 328 U.S. 256 (1946), the Government ought not to have the capacity to give itself immunity from a takings claim by the device of requiring the transfer of property from one private owner directly to another. Cf. Hawaii Housing Authority v. Midkiff, 467 U.S. 229 (1984). At the same time, the Government’s imposition of an obligation between private parties, or destruction of an existing obligation, must relate to a specific property interest to implicate the Takings Clause. For example, in United States v. Security Industrial Bank, we confronted a statute which was alleged to destroy an existing creditor’s lien in certain chattels to the benefit of the debtor. We acknowledged that, given the nature of the property interest at stake, which resembled a contractual obligation, the takings challenge “fits but awkwardly into the analytic framework” of our regulatory takings analysis. 459 U.S., at 75. We decided the analysis could apply because the property interest was a “traditional property interes[t],” though in the end the statute was found inapplicable to the lien at issue. In so holding, we relied on Louisville Joint Stock Land Bank v. Radford, 295 U.S. 555 (1935), which invalidated the Frazier-Lemke Act, because it interfered with mortgages on farms and thus worked a “ ‘taking of substantive rights in specific property acquired by the Bank prior to the Act.’ ” 459 U.S., at 77 (quoting Radford, supra, at 590, 601). Unlike the statutes at issue in Security Industrial Bank and Radford, the Coal Act does not affect an obligation relating to a specific property interest.

If the plurality is adopting its novel and expansive concept of a taking in order to avoid making a normative judgment about the Coal Act, it fails in the attempt; for it must make the normative judgment in all events. See, e.g., ante, at 35 (“[T]he governmental action implicates fundamental principles of fairness”). The imprecision of our regulatory takings doctrine does open the door to normative considerations about the wisdom of government decisions. See, e.g., Agins v. City of Tiburon, 447 U.S., at 260 (zoning constitutes a taking if it does not “substantially advance legitimate state interests”). This sort of analysis is in uneasy tension with our basic understanding of the Takings Clause, which has not been understood to be a substantive or absolute limit on the Government’s power to act. The Clause operates as a conditional limitation, permitting the Government to do what it wants so long as it pays the charge. The Clause presupposes what the Government intends to do is otherwise constitutional:

“As its language indicates, and as the Court has frequently noted, [the Takings Clause] does not prohibit the taking of private property, but instead places a condition on the exercise of that power. This basic understanding of the Amendment makes clear that it is designed not to limit the governmental interference with property rights per se, but rather to secure compensation in the event of otherwise proper interference amounting to a taking.” First English Evangelical Lutheran Church of Glendale v. County of Los Angeles, 482 U.S. 304, 314—315 (1987) (emphasis and citations omitted).

Given that the constitutionality of the Coal Act appears to turn on the legitimacy of Congress’ judgment rather than on the availability of compensation, see ante, at 19 (“[I]n a case such as this one, it cannot be said that monetary relief against the Government is an available remedy”), the more appropriate constitutional analysis arises under general due process principles rather than under the Takings Clause.

It should be acknowledged that there are passages in some of our cases on the imposition of retroactive liability for an employer’s withdrawal from a pension plan which might give some support to the plurality’s discussion of the Takings Clause. See Connolly v. Pension Benefit Guaranty Corporation, 475 U.S. 211, 223 (1986); Concrete Pipe & Products of Cal., Inc. v. Construction Laborers Pension Trust for Southern Cal., 508 U.S. 602, 641 (1993). In Connolly, the Court said the definition of a taking was not controlled by “any set formula,” but was dependent “on ad hoc, factual inquiries into the circumstances of each particular case.” 475 U.S., at 224. The Court then applied the three-factor regulatory takings analysis set forth in Penn Central, which examines the economic impact of the regulation, the extent to which it interferes with investment-backed expectations, and the character of the governmental action. 475 U.S., at 225. This analysis did not result in a finding of a taking. The Court, moreover, prefaced the entire takings discussion with the admonition it would be surprising to discover that there had been a taking in the instance where a due process attack had been rejected. See id., at 223; see also Concrete Pipe, supra, at 641 (“Given that [the] due process arguments are unavailing, ‘it would be surprising indeed to discover’ the challenged statute nonetheless violating the Takings Clause”) (quoting Connolly, supra, at 223). At best, Connolly is equivocal on the question whether we should apply the regulatory takings analysis to instances like the one now before us. My reading of Connolly, and Concrete Pipe, is that we should proceed first to general due process principles, reserving takings analysis for cases where the governmental action is otherwise permissible. See Connolly, supra, at 224 (“[H]ere, the United States has taken nothing for its own use, and only has nullified a contractual provision limiting liability by imposing an additional obligation that is otherwise within the power of Congress to impose”); see also Duke Power Co. v. Carolina Environmental Study Group, Inc., 438 U.S. 59, 94, n. 39 (1978) (upholding on due process grounds the Price-Anderson Act, 42 U.S.C. § 2210 (1970 ed., Supp. V), which placed a cap on civil liability for nuclear accidents, but declining to address petitioner’s request that the Act be declared a taking because compensation would be available under the Tucker Act, 28 U.S.C. § 1491(a)(1) (1976 ed.)). These authorities confirm my view that the case is controlled not by the Takings Clause but by well-settled due process principles respecting retroactive laws.

Given my view that the takings analysis is inapplicable in this case, it is unnecessary to comment upon the plurality’s effort to resolve a jurisdictional question despite little briefing by the parties on a point which has divided the Courts of Appeals.


When the constitutionality of the Coal Act is tested under the Due Process Clause, it must be invalidated. Accepted principles forbidding retroactive legislation of this type are sufficient to dispose of the case.

Although we have been hesitant to subject economic legislation to due process scrutiny as a general matter, the Court has given careful consideration to due process challenges to legislation with retroactive effects. As today’s plurality opinion notes, for centuries our law has harbored a singular distrust of retroactive statutes. Ante, at 31. In the words of Chancellor Kent, “A retroactive statute would partake in its character of the mischiefs of an ex post facto law … ; and in every other case relating to contracts or property, it would be against every sound principle.” 1 J. Kent, Commentaries on American Law *455; see also ibid. (rule against retroactive application of statutes to be “founded not only in English law, but on the principles of general jurisprudence”). Justice Story reached a similar conclusion: “Retrospective laws are, indeed, generally unjust; and, as has been forcibly said, neither accord with sound legislation nor with the fundamental principles of the social compact.” 2 J. Story, Commentaries on the Constitution §1398 (1833).

The Court’s due process jurisprudence reflects this distrust. For example, in Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 15 (1976), the Court held due process requires an inquiry into whether in enacting the retroactive law the legislature acted in an arbitrary and irrational way. Even though prospective economic legislation carries with it the presumption of constitutionality, “[i]t does not follow … that what Congress can legislate prospectively it can legislate retrospectively. The retrospective aspects of [economic] legislation, as well as the prospective aspects, must meet the test of due process, and the justifications for the latter may not suffice for the former.” Id., at 16—17. We have repeated this formulation in numerous recent decisions and given serious consideration to retroactivity-based due process challenges, all without questioning the validity of the underlying due process principle. United States v. Carlton, 512 U.S. 26, 31 (1994); Concrete Pipe, supra, at 636—641; General Motors Corp. v. Romein, 503 U.S. 181, 191 (1992); United States v. Sperry Corp., 493 U.S. 52, 64 (1989); United States v. Hemme, 476 U.S. 558, 567—572 (1986); Pension Benefit Guaranty Corporation v. R. A. Gray & Co., 467 U.S. 717, 729—730 (1984). These decisions treat due process challenges based on the retroactive character of the
statutes in question as serious and meritorious, thus confirming the vitality of our legal tradition’s disfavor of retroactive economic legislation. Indeed, it is no accident that the primary retroactivity precedents upon which today’s plurality opinion relies in its takings analysis were grounded in due process. Ante, at 22—26 (citing Turner Elkhorn, R. A. Gray, and Concrete Pipe).

These cases reflect our recognition that retroactive lawmaking is a particular concern for the courts because of the legislative “tempt[ation] to use retroactive legislation as a means of retribution against unpopular groups or individuals.” Landgraf v. USI Film Products, 511 U.S. 244, 266 (1994); see also Hochman, The Supreme Court and the Constitutionality of Retroactive Legislation, 73 Harv. L. Rev. 692, 693 (1960) (a retroactive law “may be passed with an exact knowledge of who will benefit from it”). If retroactive laws change the legal consequences of transactions long closed, the change can destroy the reasonable certainty and security which are the very objects of property ownership. As a consequence, due process protection for property must be understood to incorporate our settled tradition against retroactive laws of great severity. Groups targeted by retroactive laws, were they to be denied all protection, would have a justified fear that a government once formed to protect expectations now can destroy them. Both stability of investment and confidence in the constitutional system, then, are secured by due process restrictions against severe retroactive legislation.

The case before us represents one of the rare instances where the Legislature has exceeded the limits imposed by due process. The plurality opinion demonstrates in convincing fashion that the remedy created by the Coal Act bears no legitimate relation to the interest which the Government asserts in support of the statute. Ante, at 27—34. In our tradition, the degree of retroactive effect is a significant determinant in the constitutionality of a statute. United States v. Carlton, supra, at 32; United States v. Darusmont, 449 U.S. 292, 296—297 (1981) (per curiam); see also Dunbar v. Boston & P. R. Corp., 181 Mass. 383, 386, 63 N. E. 916, 917 (1902) (Holmes, C. J.). As the plurality explains today, in creating liability for events which occurred 35 years ago the Coal Act has a retroactive effect of unprecedented scope. Ante, at 30.

While we have upheld the imposition of liability on former employers based on past employment relationships, the statutes at issue were remedial, designed to impose an “actual, measurable cost of [the employer’s] business” which the employer had been able to avoid in the past. Turner Elkhorn, supra, at 19; accord, Concrete Pipe, supra, at 638; Romein, supra, at 191—192; R. A. Gray, supra, at 733—734. As Chancellor Kent noted, “[s]uch statutes have been held valid when clearly just and reasonable, and conducive to the general welfare, even though they might operate in a degree upon existing rights.” 1 Kent, supra, at *455—*456. The Coal Act, however, does not serve this purpose. Eastern was once in the coal business and employed many of the beneficiaries, but it was not responsible for their expectation of lifetime health benefits or for the perilous financial condition of the 1950 and 1974 Plans which put the benefits in jeopardy. As the plurality opinion discusses in detail, the expectation was created by promises and agreements made long after Eastern left the coal business. Eastern was not responsible for the resulting chaos in the funding mechanism caused by other coal companies leaving the framework of the National Bituminous Coal Wage Agreement. Ante, at 31—33. This case is far outside the bounds of retroactivity permissible under our law.

Finding a due process violation in this case is consistent with the principle that “under the deferential standard of review applied in substantive due process challenges to economic legislation there is no need for mathematical precision in the fit between justification and means.” Concrete Pipe, 508 U.S., at 639 (citing Turner Elkhorn, 428 U.S., at 19). Statutes may be invalidated on due process grounds only under the most egregious of circumstances. This case represents one of the rare instances in which even such a permissive standard has been violated.

Application of the Coal Act to Eastern would violate the proper bounds of settled due process principles, and I concur in the plurality’s conclusion that the judgment of the Court of Appeals must be reversed.


About us

Send email