Institutional Controls at Superfund Sites

David F. Coursen

David F. Coursen has been an attorney for the U.S. Environmental Protection Agency's Office of General Counsel in Washington, D.C., for the last five years. Previously, he was a staff attorney for the U.S. Court of Appeals for the Fifth Circuit. He graduated from the University of Oregon Law School. Mr. Coursen most recently was the author of Lucas v. South Carolina Coastal Council: Indirection in the Evolution of Takings Law, 22 ELR 10778. The views expressed in this Article are solely the author's and do not represent the views of EPA.

[23 ELR 10279]

Institutional controls (ICs) are restrictions on the use of land. In the context of the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA or Superfund),¹ they are used to reduce the dangers to the public from releases or threatened releases of hazardous substances or dangerous pollutants or contaminants. ICs are used frequently as part of the overall strategy for cleaning up a release, instead of or in addition to active response measures, such as treatment or disposal. ICs, such as restrictive zoning, may operate broadly by preventing particular activities or classes of activities at or near a site. ICs may also be narrow, specific restrictions, such as fencing off a particular area and prohibiting entry, or restricting the use of contaminated groundwater.

While CERCLA does not expressly provide for ICs, they are consistent with CERCLA's purpose. CERCLA recognizes that a remedial action may "result] [ in … hazardous substances … remaining at [a] site."² In addition, it states that relocation of affected persons or facilities may be an appropriate part of a remedy if "such relocation is more cost-effective than and environmentally preferable to the transportation, storage, treatment, destruction, or secure disposition offsite of hazardous substances."³

The National Oil and Hazardous Substances Pollution Contingency Plan (NCP),⁴ the regulation that governs CERCLA's implementation, establishes a limited role for ICs: "The use of institutional controls shall not substitute for active response measures … [that actually reduce, minimize, or eliminate contamination] as the sole remedy unless such active measures are determined not to be practicable, based on the balancing of trade-offs among alternatives that is conducted during the selection of remedy."⁵ However, in revising the NCP in 1990, the U.S. Environmental Protection Agency (EPA) provided that where active remediation is not practicable, ICs may be "the only means available to provide protection of human health."⁶ EPA elaborated further, stating that where ICs are the sole remedy, "special precautions must be made to ensure that the controls are reliable."⁷ ICs may also be used as "a necessary supplement when some waste is left in place, as it is in most response actions."⁸

The NCP does not discuss or identify the precautions needed to ensure the reliability of ICs. It addresses this matter solely by noting that, in appropriate cases, EPA cannot provide remedial action unless a state assures "that institutional controls implemented as part of the remedial action are in place, reliable, and will remain in place after initiation of operation and maintenance."⁹

The NCP's provisions relating to ICs were challenged in Ohio v. U.S. Environmental Protection Agency.¹⁰ Plaintiffs in the case argued, among other things, that ICs may not be selected on the basis of cost considerations, and that ICs may never be the sole remedy at a site. As of this date, no decision has been issued on any issue regarding ICs in that case. Any such decision could affect the availability or usefulness of ICs in particular cases.

Limitations of ICs

There are various ways that an IC may be said to "fail." First, an IC may operate precisely as intended and still not reduce human exposure sufficiently to provide the requisite level of protection to human health and the environment. For example, a control that depends on restricting public access to a site will fail if the contamination migrates beyond the restricted area. This type of failure is most likely to occur as a result of inadequate design or faulty analysis.

An IC may also fail because of its inherent limitations, [23 ELR 10280] as when restrictive zoning that limits human use of or exposure to a property is altered or eliminated. This is mostsignificant for controls which, by their nature, are subject to change or elimination without the approval of persons responsible for conducting or monitoring the response.

Finally, and most obviously, an IC will not succeed if it is not fully and effectively implemented for the necessary time period. This may occur, for example, when restrictions on land use are erratically enforced or simply ignored. This may be problematic if a control, to be effective, must be vigorously implemented for an extended period. As a practical matter, it is difficult to predict how effective implementation of an IC will be years or decades into the future.

The fact that ICs have such limitations points up a qualitative distinction between ICs and more active response measures. Once the contamination is gone from a site, the site can no longer threaten public health, regardless of how it is used. By contrast, the effectiveness of an IC depends on how the property is used. Thus, it is critical to understand how particular ICs function, what their limitations are, and what measures can be taken to maximize their effectiveness.

Types of ICs

There are two distinct types of ICs, which may be characterized as governmental and proprietary controls. Governmental ICs involve a government using its sovereign powers to impose restrictions on citizens or sites under its jurisdiction. Proprietary ICs involve property owners using their rights as owners to control the use of, or access to, their property. The two types of controls differ significantly in regard to scope, reliability, and implementation.

Governmental Controls

Governmental ICs typically involve "restrictions in respect of the use and occupation of private lands."¹¹ Generally, state and local governments impose and enforce such restrictions under their police powers.¹² Among the more common governmental ICs are water and well use advisories and restrictions, well drilling prohibitions, building permit requirements, zoning, and other land use restrictions.

Governmental ICs are not typically implemented by EPA. Indeed, the NCP suggests "that EPA may not have the authority to implement institutional controls at a site."¹³ This reflects the fact that governmental ICs most often involve the exercise of typically local police power authorities, such as zoning and land use management. However, the NCP should not be read to suggest that EPA lacks the authority to establish ICs. On the contrary, EPA has express authority, in response to an "imminent and substantial endangerment," to issue orders or take other appropriate actions "as may be necessary to protect public health and the environment."¹⁴ An order issued under this authority may effectively direct a third party to establish and maintain an IC. In addition, such an order, to the extent it restricts or prohibits certain land uses, may itself function as an IC, at least with respect to the party to whom it is directed. Nevertheless, governmental ICs are most often implemented by local governments.

Governmental ICs are often established legislatively,¹⁵ for example, by enactment of a local ordinance. The same legislative mechanism, however, may also be used to alter or eliminate an IC, for example, by repealing the ordinance.¹⁶ Thus, where waste is left in place at a site, a governmental IC might take the form of a zoning restriction that limits uses of the site; a subsequent change in zoning, such as from open space or industrial use to residential use, could effectively reduce or eliminate the control and, thus, cause it to fail.

A fundamental characteristic of many types of governmental ICs is that they can be altered through the legislative process. In a democracy it is desirable to develop and utilize decisionmaking processes that are responsive to the needs and preferences of the public. However, such responsiveness may be environmentally undesirable if it erodes or weakens the stability and reliability needed to ensure the continuing protectiveness of an IC. Indeed, the most reliable way of preventing the alteration and, thus, failure of a governmental control might be to restrict the decisionmaking authority of future local governments. Even if this were possible, it might not be lawful, prudent, practical, or desirable.

Ultimately, this suggests that the legislative process may not be well-suited to providing the optimal stability for the continued, long-term effectiveness of a governmental IC. Thus, caution should be used before relying on governmental ICs as integral parts of remedial strategies, particularly in circumstances where the failure of the control might have serious consequences.

One way to address this limitation of governmental ICs is to obtain contractual assurances concerning the continued effectiveness of a control. For example, EPA cannot provide remedial action at a site unless the state where the site is located assures EPA that it will operate and maintain a remedy.¹⁷ In many cases, the continued enforcement of an IC is characterized as an aspect of the effective operation and maintenance (O&M) of a site.¹⁸ Thus, a state providing the required assurance, typically through a Superfund cooperative agreement or Superfund state contract,¹⁹ may be contractually obligated to EPA to continue the controls to the extent it has the authority to do so. If the remedy fails, EPA may be able to pursue a breach of contract claim against the state.

However, the practical value of ensuring the retention of a governmental IC through such an agreement may be limited. Where legislation imposes such a control, there may be limitations on a state government's ability to limit the exercise of legislative authority of future governments. In addition, states may have delegated to local governments the types of police powers exercised in implementing and enforcing ICs.²⁰ However, often local governments are not [23 ELR 10281] parties to the types of agreements discussed above, and are not required to assure effective O&M of a site.²¹ If a state has made an assurance, but lacksthe legal authority to prevent the local government from actions that might lead to the failure of the IC, such as a change in zoning regulations, an assurance from a state may be of limited value.

As a practical matter, the most effective way of promoting the reliability of governmental ICs may be to promote community understanding of, support for, and commitment to, the IC. When a community recognizes the importance and appropriateness of an IC, the political process is likely to be used to protect, rather than to alter, the IC. This applies both to governmental ICs that prevent individuals from taking actions that increase public exposure to contamination, and to ICs that prevent individuals from taking actions that increase their own exposure, such as water use restrictions. The effectiveness of such restrictions ultimately depends on voluntary compliance, which in turn depends on the affected individuals' belief that compliance is necessary for their protection and well-being.

Proprietary Controls

Where governmental authority is not the basis for an IC, property ownership often provides the authority. For example, a consent agreement in which a party agrees to ensure that a particular IC will remain in place will only be reliable to the extent the party has the power to enforce the restrictions in the IC. If the party is not a sovereign, the most likely source of power to control the use of private property rests with the ownership interests in that property.²²

With a proprietary IC, a party owning sufficient property rights restricts the uses that can be made of that property. This may be done, for example, by the owner erecting a fence or otherwise prohibiting entry. It can also be done by creating an easement or taking some affirmative legal action to impose servitudes on private lands to prevent or restrict uses.²³ Proprietary ICs can be established by a party possessing sufficient rights in a property to prevent an inappropriate use, and who formally agrees to exercise those rights and prevent the use. Ideally, a proprietary IC should be tailored to impose the needed restrictions on the property, while permitting all appropriate uses.

The most common legal basis for a proprietary control is the ownership of some estate in land.²⁴ Fee simple absolute, which comprises full ownership of land and encompasses all the rights, privileges, powers, and immunities connected to that land, is comprehensive enough to support an IC. Fee owners can generally impose any lawful restrictions on the uses of their property.²⁵ Other, lesser possessory estates may also encompass rights sufficient to impose the needed restrictions.²⁶

Nonpossessory interests, or servitudes, may also provide the basis for proprietary ICs.²⁷ The most common nonpossessory interest is an easement. Easements are well-recognized at common law, and are sufficiently flexible to be crafted to give the holder precisely the rights needed to restrict the use of the property as required.²⁸ A covenant, running with the land, that restricts the use as needed, may also be adequate if some party has the ability and willingness to enforce it.²⁹

Another alternative is a reverter clause in a deed, by which the property reverts to a former owner or some other party if it is ever used in a prohibited way.³⁰ Yet another option would be the creation of an irrevocable trust to hold the interest and ensure that the property is not used in a prohibited way. Ultimately, any interest (preferably recordable) that encompasses rights and control over the property sufficient to enforce a use restriction can be adequate.

The identity of the party with the power and responsibility for enforcing the IC may prove critical to whether the control will function effectively. EPA may hold the relevant interests or enforce the relevant rights,³¹ but a state or local government is more likely to do so.³² It may also be possible and desirable for a nongovernmental party to enforce the restriction, for example, under the terms of a settlement agreement.³³

A local community group, motivated by a desire to ensure adequate environmental protection of an area, might also be an appropriate entity to assume authority and responsibility for maintaining an IC. In considering this possibility, it is critical to examine relevant factors affecting the long-term viability of the group, including its past history, likely longevity, its resources for taking legal actions to address violations of the control, and its ability to take other action that might be appropriate.

The party responsible for implementing or overseeing an IC must possess the legal authority to manage use of the property in the desired ways. Certain instruments, such as deed notices and deed restrictions, may not reliably ensure such authority. For example, a deed notice may have limited utility if it does no more than refer to the existence of an IC without providing any means to enforce it. A facially enforceable deed restriction may be of limited value if the ability to enforce ceases once the property passes out of privity, since conveyance may trigger a "failure" of the control.

An assessment of the reliability of an IC should include a determination of the effects of the possible conveyance of the property or of the interest needed to enforce the IC. [23 ELR 10282] Ideally, the party with the power to enforce an IC should have an ongoing contractual duty to do so. Thus, if the property is conveyed and the use restriction is violated, the party with the enforcement authority would be in breach of the terms of the agreement.

Controls on Government Property

ICs may be particularly useful where the property at issue is owned by a governmental entity. For example, some of the largest and most seriously contaminated Superfund sites are on federal lands. Creative proprietary ICs at these sites can provide a cost-effective method of ensuring adequate protection to the public.

One example of a governmental proprietary IC is at the Rocky Mountain Arsenal Superfund site near Denver, Colorado, one of the most toxic sites in the nation.³⁴ Under the Rocky Mountain Arsenal National Wildlife Refuge Act of 1992 (the Act), legislation was enacted to create a national wildlife refuge on that site.³⁵ Under the Act, management responsibilities for the site are being transferred from the Department of the Army to the Department of the Interior.³⁶ Any conveyance of property from the refuge, such as for highway use, is subject to a variety of perpetual restrictions, including a ban on residential or industrial use and the use of groundwater located under the property as a source of drinking water.³⁷ The Act is intended to ensure that the site will be cleaned up to meet CERCLA's standards, while upholding the purposes for which the refuge was established.³⁸ At the same time, transforming the site into a refuge effectively limits future uses and precludes intensive development.

This appears to be a hybrid of governmental and proprietary ICs. The Act creating the refuge constitutes the governmental control, but the viability of the control depends on the fact that the government owns the property.

The Threat of Liability

As a practical matter, it is somewhat misleading to focus solely on the existence and effectiveness of formal means to enforce a restriction. To the extent that failure of a control entails consequences that can be characterized as a CERCLA release, the owner or operator of the property may be liable for all costs of responding to the release.³⁹ While the prospect of such liability may not directly preclude inappropriate uses of property, it may effectively deter such uses by clarifying their potential legal consequences. Moreover, any record of an attempt to put a control in place, including the presence of a use restriction or notice in a deed, is likely to make it difficult for a party to maintain an innocent landowner defense to CERCLA's liability.⁴⁰

Takings

Because ICs restrict uses of property, in extreme cases the implementation of an IC may effectively prevent any meaningful use of a property. This could give rise to a claim that the control effectively "takes" the property, entitling the owner to just compensation under the Fifth Amendment to the U.S. Constitution. Under the Fifth Amendment, it is entirely proper for government to take property, so long as it provides just compensation to the owner.⁴¹

The risk of takings arising from the imposition of ICs, however, in many cases may be largely theoretical. An IC, after all, is imposed to restrict human exposure to hazardous substances that remain on site. To establish that such a control worked a taking, mandating compensation, the owner would have to demonstrate both a vested property right to maintain the property in its existing condition,⁴² and that, absent the control, the property could be put to some economically beneficial use.⁴³ In many cases, this would be tantamount to a showing that the IC is not needed, at least in its existing form.

Further, to recover just compensation, the property owner would also have to prevent the government from establishing that the owner is deriving a benefit from the IC, or from the response action itself, that offsets any burdens imposed on the property.⁴⁴ Absent such a demonstration, it could be difficult to establish that the owner was being forced "alone to bear public burdens which, in all fairness, should be borne by the public as a whole."⁴⁵

Finally, if a property owner were able to establish a claim for just compensation, as an owner of the site of a release, he or she might be subject to liability for all costs of responding to the release.⁴⁶ Presumably, any just compensation payments would constitute necessary costs of the response, and thus the government could ultimately recover them from the property owner. This incongruous situation should not arise, however, because any party legally responsible for cleanup costs may have difficulty passing the "fairness" test articulated above, which is the overarching principle for determining the compensability of government actions.

Conclusion

Certain types of proprietary ICs can be extremely reliable when implemented properly. The effectiveness of a proprietary IC depends on at least three key elements. Whether it provides sufficient rights to ensure that uses of the underlying property are restricted in the required ways, the willingness of the party holding the rights to [23 ELR 10283] exercise them, and the likely effects of a conveyance of the property on those rights.

Conversely, governmental ICs, to the extent that they are expressed in laws, policies, or regulations, may be subject to change. To maximize the effectiveness of this type of IC, it may be useful to attempt to develop some formal agreement in which the government possessing police power over the activity to be regulated assumes responsibility for implementing and maintaining the control. The most effective means of safeguarding a governmental IC may be public support for, and understanding the need for, the control.

1. 42 U.S.C. §§ 9601-9675, ELR STAT. CERCLA 001-075. CERCLA was amended in 1986 by the Superfund Amendments and Reauthorization Act, Pub. L. No. 99-499, 100 Stat. 1613. Since then, CERCLA and Superfund have been commonly used interchangeably.

2. 42 U.S.C. § 9621(c), ELR STAT. CERCLA 051.

3. Id. § 9601(24), ELR STAT. CERCLA 009 (defining remedial action).

4. 40 C.F.R. pt. 300 (1992). The NCP, promulgated by EPA under CERCLA § 105(a)(8)(B), establishes operating rules for Superfund cleanups.

5. 40 C.F.R. § 300.430(a)(1)(iii)(D) (1992).

6. 55 Fed. Reg. 8666, 8706 (1990).

7. Id. at 8706.

8. Id.

9. 40 C.F.R. § 300.510(c)(1) (1992). See also 42 U.S.C. § 9604(c)(3), ELR STAT. CERCLA 013 (describing criteria for continuing obligations over response actions).

10. Nos. 86-1096 et al. (D.C. Cir. filed Feb. 10, 1986).

11. Village of Euclid v. Ambler Realty Co., 272 U.S. 365, 386 (1926).

12. See, e.g., 1 ROBERT M. ANDERSON, AMERICAN LAW OF ZONING § 2.01, at 28 (2d ed. 1976) [hereinafter ANDERSON].

13. 55 Fed. Reg. 8706 (1990).

14. 42 U.S.C. § 9606(a), ELR STAT. CERCLA 024.

15. See Village of Euclid, 272 U.S. at 388; see also ANDERSON, supra note 12, § 2.01.

16. ANDERSON, supra note 12, § 4.26.

17. 42 U.S.C. § 9604(c)(3), ELR STAT. CERCLA 013.

18. See, e.g., 40 C.F.R. § 300.510(c)(1).

19. See 40 C.F.R. §§ 35.6105(b), -.6805(i)(1).

20. ANDERSON, supra note 12, § 2.01.

21. See, e.g., 42 U.S.C. § 9604(c)(3)(C)(ii), ELR STAT. CERCLA 013 (requiring states to pay share of costs when a political subdivision operates a facility).

22. A government that owns property may also, in its proprietary capacity, establish and implement such a control.

23. See Lucas v. South Carolina Coastal Council, 112 S. Ct. 2886, 2895, 22 ELR 21104, 21109 (1992).

24. See ROGER A. CUNNINGHAM ET AL., THE LAW OF PROPERTY, §§ 1.2, 2.1. (1984 ed.) [hereinafter CUNNINGHAM].

25. Id. § 2.2.

26. See, e.g., id. § 2.16 (life estates) and § 2.17 (leaseholds).

27. See, e.g., id. § 8.1.

28. See, e.g., Lucas, 112 S. Ct. at 2895, 22 ELR at 21109.

29. CUNNINGHAM, supra note 24, § 8.13.17.

30. Id. § 3.3.

31. See 42 U.S.C. § 9604(j)(1)-(2), ELR STAT. CERCLA 021 (authorizing EPA to acquire an "interest in real property" to conduct remedial action, and precluding use of this authority unless the state in which the interest is located assures that it will accept transfer of interest "following completion of the remedial action").

32. See 55 Fed. Reg. 8665, 8778 (1990) (preamble to the NCP, discussing state obligations to assure that the operation and maintenance of a facility, which may include an assurance that an IC is reliable, will remain in place).

33. See 42 U.S.C. § 9622, ELR STAT. CERCLA 054.

34. See 138 CONG. REC. S14093 (daily ed. Sept. 18, 1992) (statement of Sen. Brown describing severity of contamination at site).

35. Pub. L. No. 102-402, 106 Stat. 1061 (Oct. 9, 1992).

36. Id. § 2.

37. Id. § 5(d).

38. See 138 CONG. REC. S14093 (daily ed. Sept. 18, 1992) (statements of sponsors of the Act).

39. 42 U.S.C. § 9607, ELR STAT. CERCLA 024.

40. See id. § 9607(b)(3), ELR STAT. CERCLA 025; see also id. § 9622(g)(1).

41. See First English Evangelical Lutheran Church of Glendale v. County of Los Angeles, 482 U.S. 304, 315, 17 ELR 20787, 20790 (1987).

42. See Lucas, 112 S. Ct. at 2899-902, 22 ELR at 21111-21112.

43. See, e.g., Agins v. City of Tiburon, 447 U.S. 255, 260, 10 ELR 20361, 20363 (1980).

44. See, e.g., Pennsylvania Coal Co. v. Mahon, 260 U.S. 393, 415 (1922); Monongahela Navigation Co. v. United States, 148 U.S. 312, 325 (1893).

45. Armstrong v. United States, 364 U.S. 40, 49 (1960).

46. 42 U.S.C. § 9607, ELR STAT. CERCLA 024.