Natural Resource Damages From Oil Spills: A Comparison of the Ohio Decision and the Oil Pollution Act

Richard W. Dunford

Editors' Summary: The subject of natural resource damages continues to become ever more prominent. Congress' 1990 enactment of the Oil Pollution Act with a strong provision on causes of action for natural resource damages, combined with the 1991 settlement for damages from the Exxon Valdez oil spill worth hundreds of millions of dollars, promises to create even more attention in the future.

In this Article, the author, an economist, focuses on an issue not yet widely analyzed: the different results that occur from using different formulations to measure natural resource damages. In 1989, the D.C. Circuit ruled in Ohio v. United States Department of the Interior, 19 ELR 21099, that the cost of restoring damaged resources is the starting point in calculating the damage award. Congress took a similar approach in the Oil Pollution Act. Yet, on close examination the two standards are not the same, and in some cases the results can be dramatically different depending on which formulation is used.

The author illustrates the differences with four hypothetical scenarios, and draws conclusions about which scenarios are most likely to occur in practice.

Richard W. Dunford, Ph.D., is the Assistant Director of the Natural Resource Damage Assessment Program at the Research Triangle Institute in North Carolina. Dr. Dunford and his colleagues have worked on seven major oil spills in U.S. waters, including the Exxon Valdez oil spill in Alaska.

[22 ELR 10263]

The 1977 amendments to the Federal Water Pollution Control Act (FWPCA)¹ authorized certain government agencies to recover compensatory damages for injuries to natural resources resulting from oil spills. Specifically, the FWPCA provides that cleanup costs "shall include any costs or expenses incurred by [authorized governmental agencies] in the restoration or replacement of natural resources damaged or destroyed as a result of [such spills]."² The authorized governmental agencies "shall act on behalf of the public as trustee of the natural resources to recover for the costs of replacing or restoring such resources."³

In 1980 the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA)⁴ extended the liability of the businesses responsible for releases of hazardous substances to "damages for injury to, destruction of, or loss of natural resources, including the reasonable costs of assessing such injury, destruction, or loss resulting from such release."⁵ CERCLA (also known as the Superfund Act) required the promulgation of regulations for natural resource damage assessments (NRDA) for both oil spills (under the FWPCA) and hazardous substance releases.⁶

In August 1986 the U.S. Department of the Interior (DOI) promulgated the NRDA regulations.⁷ These regulations apply to natural resources "belonging to, managed by, held in trust by, appertaining to, or otherwise controlled by the United States …, any state or local government, any foreign government, any Indian tribe," or members of Indian tribes under certain conditions.⁸ Eligible trustees include federal natural resource management agencies, designated state agencies, and Indian tribes.⁹

Within three months of their promulgation, petitioners filed nine lawsuits contesting various aspects of the DOI regulations in the D.C. Circuit Court of Appeals. Two more lawsuits were filed in 1988. The petitioners included ten states, three nonprofit environmental organizations (led by the National Wildlife Federation), the Chemical Manufacturers Association, a public utility, and one manufacturing company. At the court's request the nine cases were consolidated, and the D.C. Circuit issued its rulings in Ohio v. United States Department of the Interior in July 1989.¹⁰

In August 1990 Congress passed comprehensive oil spill legislation, the Oil Pollution Act of 1990 (OPA).¹¹ The OPA [22 ELR 10264] creates liability for the following six types of damages resulting from oil spills:

* natural resource damages;

* losses resulting from injuries to real or personal property;

* loss of subsistence use of natural resources;

* losses of government revenues (such as taxes, royalties, rents, and fees) resulting from injuries to real property, personal property, or natural resources;

* losses of profits resulting from injuries to real property, personal property, or natural resources; and

* net costs of providing increased or additional public services during or after removal activities.¹²

Natural resource damages under the OPA are compensatory damages "for injury to, destruction of, loss of, or loss of use of natural resources, including the reasonable cost of assessing the damage, which shall be recoverable by a United States trustee, a State trustee, an Indian tribe trustee, or a foreign trustee."¹³

The specific measure of natural resource damages in the Ohio decision is similar, but not identical, to the measure specified in the OPA. The first section of this Article presents the measure of natural resource damages under the Ohio decision. The second section describes the measure of natural resource damages under the OPA, and compares this measure with the Ohio measure. The third section presents several examples that illustrate the similarities and differences in the two natural resource damage measures. The last section examines the practical significance of the differences in the two natural resource damage measures.

1989 Ohio Decision

In the Ohio decision the D.C. Circuit Court of Appeals held that natural resource damages should equal (1) the cost of restoring, rehabilitating, replacing, or acquiring the equivalent of the injured natural resources (restoration cost, for simplicity), allowing the diminution in the value of natural resources prior to restoration to be included in restoration cost, plus (2) the cost of assessing natural resource damages.¹⁴ However, the court acknowledged two possible exceptions: when restoration is infeasible or when restoration costs are "grossly disproportionate" to the diminution in the value of the natural resources.¹⁵ In a footnote, the court provided an example to illustrate the "grossly disproportionate" exception: if restoration costs are three times the diminution of value, then the latter could be the basis for natural resource damages.¹⁶

While restoration costs and damage assessment costs are self-explanatory, the meaning of the "diminution in the value of natural resources prior to restoration" requires some explanation. Natural resources may provide two types of services: use services and nonuse services. Use services involve some physical or visual contact between people and the natural resources. Examples of use services include recreation activities such as fishing, boating, birdwatching, hunting, hiking, and camping. Foregone recreational activities are the most common use services giving rise to natural resource damages from oil spills.

Nonuse services include services that resources provide to other resources and services that resources provide to people that do not require physical or visual contact between people and the resources. The services that some natural resources provide to other natural resources are mainly limited to the provision of food, nesting habitat, protection from predators, and similar biotic support services. A marsh providing nesting habitat for migratory waterfowl is an example of this type of service. Nonuse services can also arise from natural resources that provide well-being to some people through their mere existence, even though these people have no plans to ever use the resources. Examples of natural resources that may provide such existence services include unique geologic resources (such as the Grand Canyon) and rare, threatened, or endangered biologic resources (such as bald eagles).

The total value of a natural resource equals the value of the use services provided by the natural resource over time (i.e., use value) plus the value of any nonuse services over time (i.e., nonuse value). A diminution of total value is the sum of the reduction in use value and nonuse value, if any, attributable to natural resource injuries. For example, suppose a wetlands provides some recreation services and some nonuse services, such as food for migratory waterfowl. If the use value of the recreation services in perpetuity is $ 4 million while the value of the food for the migratory waterfowl in perpetuity is $ 1 million, the total value of the wetlands is $ 5 million (i.e., $ 4 million plus $ 1 million). Suppose an oil spill eliminates recreational uses of the wetlands and the food available for migratory waterfowl for a year. If recreational use is valued at $ 400,000 a year and the provision of food for migratory waterfowl is value at $ 100,000 a year, then the diminution in the value of the wetlands as a result of the oil spill is $ 500,000 (i.e., $ 400,000 plus $ 100,000).

OIL POLLUTION ACT OF 1990

Under the OPA, natural resource damages from oil spills equal (1) restoration costs, plus (2) the diminution in value of those natural resources pending restoration, plus (3) the reasonable cost of assessing those damages.¹⁷ The Conference Report on the OPA indicates that "diminution of value" refers to "the standard for measuring natural resource damages used in the recent D.C. Circuit Court decision."¹⁸ Thus, the "diminution of value" means the diminution in total value (i.e., foregone use and nonuse values, if any) prior to restoration. So, the OPA specifies essentially the same basic measure of natural resource damages as the Ohio decision.

While they have the same basic measure of natural resource damages, the OPA and the Ohio decision specify different exceptions to the basic measure. In particular, the [22 ELR 10265] OPA Conference Report states that trustees "shall give priority to efforts to restore, rehabilitate and replace damaged resources. The alternative of acquiring equivalent resources should be chosen only when the other alternatives are not possible, or when the cost of these alternatives would, in the judgment of the trustee, be grossly disproportionate to the value of the resources involved."¹⁹

The OPA mandates an alternate measure of natural resource damages once restoration costs exceed the total value of the natural resources that were injured by the oil spill. In contrast, the Ohio decision requires an alternate measure of natural resource damages once restoration costs exceed the diminution in the total value of the natural resources as a result of the spill. Unless an oil spill completely destroys natural resources for many years, the diminution in total value will be less (and perhaps much less) than the total value of the natural resources. Therefore, the "grossly disproportionate" exception in the Ohio decision will be reached at a lower dollar value than the "grossly disproportionate" exception in the OPA.

In addition to differences in when the alternate measure of natural resource damages is operative, the OPA and the Ohio decision have different alternate measures. As noted above, the alternate measure of natural resource damages under the Ohio decision is the diminution of total value under natural recovery plus damage assessment costs. In contrast, the alternate measure of natural resource damages under the OPA is not clearly specified. At a minimum the OPA alternate measure would include the cost of acquiring the equivalent natural resource and damage assessment costs. It may also include the diminution in the total value of the injured natural resource prior to the acquisition of the equivalent natural resource. Under either interpretation the OPA alternate measure will probably exceed the Ohio alternate measure in most cases.

In summary, the Ohio decision and the OPA have the same basic measure of natural resource damages, but they have different exceptions to the basic measure and different alternate measures once the exceptions are operative. In general, natural resource damages under the OPA will equal or exceed natural resource damages under the Ohio decision, as shown by the examples in the next section.

Examples

As explained above, the relationship of restoration costs to the diminution of the total value of the injured natural resource plays a crucial role in determining the magnitude of natural resource damages under both the Ohio decision and the OPA. There are four possible relationships between restoration costs and natural resource values.

A. Restoration costs are less than the diminution of total value.

B. Restoration costs are greater than the diminution of total value, but are not "grossly disproportionately" greater than the diminution of total value.

C. Restoration costs are "grossly disproportionately" greater than the diminution of total value, but are not "grossly disproportionately" greater than total value.

D. Restoration costs are "grossly disproportionately" greater than the total value.

Scenarios A through D in Table 1 illustrate these relationships between restoration costs and the foregone natural resource values. The top half of the table provides assumed values for the following:

* diminution of total value under natural recovery;

* restoration costs;

* diminution of total value under restoration;

* total value of the injured natural resources; and

* damage assessment cost.

Note that the assumed values for each of these elements of natural resource damages are the same for all four scenarios (A, B, C, and D), except for restoration costs (i.e., only restoration costs vary across the scenarios).²⁰ The bottom half of Table 1 shows the value of natural resource damages for each scenario under the Ohio decision and the OPA.

*5*Table 1. Comparison of Natural Resource Damages Under

*5*Two Liability Schemes for Four Hypothetical Spill Scenarios

*4*Spill Scenarios

A B C D

Assumed Values for:

Diminution of total value

under natural recovery 8 8 8 8

Restoration costs 4 10 24 36

Diminution of total value

under restoration 2 2 2 2

Total value of natural

resources 30 30 30 30

Damage assessment costs 1 1 1 1

Natural resource damages under:

Ohio decision 7 13 9 9

Oil Pollution Act 7 13 27 31

In Scenario A, restoration cost plus the diminution of total value under restoration is less than the diminution of the total value under natural recovery. Consequently, both liability schemes yield the same amount of natural resource damages (i.e., $ 7) for Scenario A. This value is based on the sum of the restoration cost ($ 4), the diminution of total value under restoration ($ 2), and damage assessment cost ($ 1).

Under Scenario B, restoration cost plus the diminution of total value under restoration exceeds the diminution of value under natural recovery, but not by a grossly disproportionate amount. In this scenario the Ohio decision and the OPA base natural resource damages on restoration cost ($ 10), plus the diminution of total value under restoration ($ 2), and damage assessment costs ($ 1). This approach, which also applies to Scenario A, results in natural resource damages of $ 13 for Scenario B.

Under Scenario C, restoration cost plus the diminution of total value under restoration are grossly disproportionately higher than the diminution of total value under natural recovery, based on the "three-times" example given by the court of appeals. Consequently, the Ohio decision results in natural resource damages of $ 9, which is the diminution of total value under natural recovery ($ 8) plus damage assessment cost ($ 1). However, restoration cost plus the diminution of total value [22 ELR 10266] under restoration (plus damage assessment cost) determines natural resource damages under the OPA, because the sum of these components is not grossly disproportionately larger than the total value of the natural resource. (In fact, restoration costs are less than the total value of injured natural resources in Scenario C.) Thus, natural resource damages for Scenario C under the OPA are $ 27 (i.e., $ 24 of restoration cost, plus $ 2 for the diminution in total value under restoration, plus $ 1 in damage assessment cost).

Under Scenario D, restoration cost exceeds the total value of the natural resources by a grossly disproportionate amount (by assumption). As in Scenario C, the Ohio decision bases natural resource damages on the sum of the diminution in total value under natural recovery ($ 8) plus damage assessment cost ($ 1). The OPA bases natural resource damages on the cost of acquiring the equivalent natural resources, which we assume equals the total value of injured resources prior to the oil spill (i.e., $ 30),²¹ plus damage assessment cost ($ 1). Natural resource damages under the OPA for Scenario D total $ 31.

In summary, the Ohio decision and the OPA result in the same amount of natural resource damages whenever restoration cost plus the diminution in the total value of the injured natural resources is less than grossly disproportionately larger than the diminution in the total value of the natural resources under natural recovery. Scenarios A and B depict this situation. However, if restoration cost plus the diminution in the total value of the injured natural resources is grossly disproportionately larger than the diminution in the total value of the natural resources under natural recovery (as in Scenarios C and D), then the OPA will result in a much larger amount of natural resource damages than the Ohio decision.

Significance

The practical significance of the difference in the measures of natural resource damages under the Ohio decision and the OPA depends on the following two factors:

* the provisions of regulations implementing the Ohio decision and the OPA; and

* the likelihood of actual oil spills having characteristics similar to the scenarios that result in larger natural resource damages under the OPA.

These factors are discussed below.

The federal government has not yet issued final regulations implementing either the Ohio decision or OPA. Consequently, the specific provisions of their "grossly disproportionate" exceptions and their alternate measures of natural resource damages will not be known until those regulations are promulgated. Therefore, the practical significance of the differences in the Ohio and OPA measures of natural resource damages is unclear at this time.

In April 1991, the DOI issued proposed revisions in its NRDA regulations pursuant to the Ohio decision.²² The proposed NRDA revisions define natural resource damages as the sum of the following:

* restoration costs;

* foregone compensable value; and

* damage assessment costs.²³

The proposed regulations define compensable value as

the amount of money required to compensate the public for the loss in services provided by the injured resources between the time of the discharge or release and the time the resources and the services those resources provided are fully restored to their baseline conditions.²⁴

Foregone compensable value would include losses of both use values and nonuse values, as required by the Ohio decision.²⁵ The proposed regulations define use values as the value of direct, public uses of natural resources, while nonuse values are defined as the difference between compensable value and use value.²⁶

While the basic measure of natural resource damages in the proposed regulations is consistent with the Ohio decision, the "grossly disproportionate" exception in the Ohio decision is not part of the proposed regulations. As an alternative, the proposed regulations list the following 10 factors that trustees should consider in selecting one or more restoration alternatives:

* technical feasibility;

* net benefits of the alternatives;

* cost-effectiveness of the alternatives;

* results of response actions;

* potential for additional injury resulting from the actions;

* natural recovery period;

* ability of the resource to recover with or without alternative actions;

* acquisition of equivalent land where restoration, rehabilitation, and/or other replacement of land is not possible;

* potential effects of the action on human health and safety; and

* consistency with applicable federal and state laws.²⁷

The proposed regulations do not provide any guidance on the relative importance of these factors, nor are trustees required to meet any of them. Trustees could possibly select a restoration alternative having grossly disproportionate costs under the proposed regulations. If the final regulations implementing the Ohio decision contain similar provisions, then natural resource damages under the Ohio decision for Scenarios C and D in the previous section will increase to the extent that trustees favor the restoration of injured natural resources regardless of the cost.

The OPA directs the Commerce Department to develop the new NRDA regulations pertaining to oil spills by August 1992.²⁸ If the Commerce Department ignores the "grossly [22 ELR 10267] disproportionate" exception in its regulations, as the DOI has done in its proposed regulations, then the differences in natural resource damages under the Ohio decision and the OPA may greatly diminish.

The second factor affecting the practical significance of the difference in natural resource damages under the Ohio decision and the OPA is the likelihood that situations similar to Scenarios C and D in the previous section, which result in larger natural resource damages under the OPA, will actually occur. If these scenarios are very unlikely, the Ohio decision and the OPA will yield the same amount of natural resource damages in most cases. While it is difficult to generalize, the effects of many oil spills are relatively short lived. In some cases restoration occurs naturally within a few months or years. In other cases only minimal restoration activities are warranted to enhance natural recovery. These situations are consistent with Scenarios A or B, in which natural resource damages were the same under the Ohio decision and the OPA.

However, in some cases the effects of an oil spill extend over several decades. If the injured resources have a relatively low total value and restoration efforts are very costly, then the difference in natural resource damages shown in Scenarios C and D may be apparent. In conclusion, situations that are consistent with Scenarios A and B (i.e., the no-difference scenarios) are probably most likely, but situations that are consistent with Scenarios C or D, which show much higher natural resource damages under the OPA, are possible.

1. 33 U.S.C. §§ 1251-1387, ELR STAT. FWPCA 001-071.

2. FWPCA § 311(f)(4), 33 U.S.C. § 1321(f)(4), ELR STAT. FWPCA 046.

3. FWPCA § 311(f)(5), 33 U.S.C. § 1321(f)(5), ELR STAT. FWPCA 046.

4. 42 U.S.C. §§ 9601-9675, ELR STAT. CERCLA 001-075.

5. CERCLA § 107(a)(4)(C), 42 U.S.C. § 9607(a)(4)(C), ELR STAT. CERCLA 025.

6. CERCLA § 301(c)(1), 42 U.S.C. § 9651(c)(1), ELR STAT. CERCLA 062.

7. 43 C.F.R. pt. 11 (1990). See Richard W. Dunford, Natural Resource Damages From Oil Spills, in INNOVATIONS IN ENVIRONMENTAL LAW (Tom Tietenberg ed., forthcoming) for a description of the NRDA process in the DOI regulations.

8. 43 C.F.R. § 11.14(z) (1990).

9. 43 C.F.R. § 11.14(rr) (1990).

10. 880 F.2d 432, 19 ELR 21099 (D.C. Cir. 1989). A related case decided the same day is Colorado v. U.S. Dep't of the Interior, 880 F.2d 481, 19 ELR 21127 (D.C. Cir. 1989). For more information on these decisions, see Erik D. Olson, Natural Resource Damages in the Wake of the Ohio and Colorado Decisions: Where Do We Go From Here?, 19 ELR 10551 (Dec. 1989) and Raymond J. Kopp et al., Natural Resource Damages: The Economics Have Shifted After Ohio v. United States Department of the Interior, 20 ELR 10127 (Apr. 1990).

11. Pub. L. No. 101-380, 104 Stat. 484. See Russell V. Randle, The Oil Pollution Act of 1990: Its Provisions, Intent, and Effects, 21 ELR 10119 (Mar. 1991) for a detailed examination of the Act. See also Walter B. Jones, Oil Spill Compensation and Liability Legislation: When Good Things Don't Happen to Good Bills, 19 ELR 10333 (Aug. 1989) for information on the legislative history of the Oil Pollution Act of 1990.

12. OPA § 1002(b)(2), 33 U.S.C. § 2702(b)(2), ELR STAT. OIL POLL. ACT 005.

13. OPA § 1002(b)(2)(A), 33 U.S.C. § 2702(b)(2)(A), ELR STAT. OIL POLL. ACT. 005.

14. 880 F.2d at 441-59, 19 ELR at 21102-113.

15. Id. at 443-44, 19 ELR at 21104.

16. Id. at 443 n.7, 19 ELR at 21104 n.7.

17. OPA § 1006(d), 33 U.S.C. § 2706(d), ELR STAT. OIL POLL. ACT 007.

18. H. REP. NO. 653, 101st Cong., 2d Sess. 108 (1990).

19. Id. (emphasis added).

20. For simplicity, the scenarios assume no diminution in total value prior to the assessment of natural resource damages.

21. In some cases the cost of acquiring equivalent resources will be less than the total value of the natural resources injured by the oil spill, because the market value of the former will exclude nonuse values. This would lower the measure of natural resource damages under the OPA, which would still exceed the measure of natural resource damages under the Ohio decision in most cases.

22. 56 Fed. Reg. 19752 (Apr. 29, 1991).

23. Id. at 19769.

24. Id. at 19772.

25. Id.

26. Id.

27. Id. at 19770-71.

28. OPA § 1006(e), 33 U.S.C. § 2706(e), ELR STAT. OIL POLL. ACT 007.