Natural Resource Damages: The Economics Have Shifted After Ohio v. United States Department of the Interior

Raymond J. Kopp, Paul R. Portney, and V. Kerry Smith

Raymond J. Kopp, Ph.D., is Senior Fellow and Director of the Quality of the Environment Division, Resources for the Future. Paul R. Portney, Ph.D., is Senior Fellow and Vice President of Resources for the Future. V. Kerry Smith, Ph.D., is University Distinguished Professor at North Carolina State University, and a University Fellow at Resources for the Future.

[20 ELR 10127]

Litigation under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA)¹ is now a booming business. Most of the lawsuits seek to recover cleanup costs. But attention could soon shift to another class of cases: post-cleanup liability for residual injury to natural resources. These natural resource damage cases may eventually become as prevalent and as costly as cleanup cases.

In 1989, the District of Columbia Circuit Court of Appeals changed the rules of the game when it handed down its decision in Ohio v. United States Department of the Interior.² From an economist's viewpoint, this decision has greatly widened the scope of admissible damages, changed the relative bargaining positions of plaintiffs and defendants in settlement proceedings, and altered defendants' optimal litigation strategies.

The Court's Decision

Overview

CERCLA § 301(c)³ directs the President to promulgate regulations for assessing damages for injury to, destruction of, or loss of natural resources resulting from releases of hazardous substances or oil.⁴ The President in turn delegated this responsibility to the Department of the Interior, which published final rules governing natural resource damage assessments.⁵

Shortly after rules were issued, 10 states, three environmental organizations, a chemical trade association, a manufacturing company, and an electric utility petitioned for judicial review of the rules. Arguments were presented before the District of Columbia Circuit Court of Appeals on February 22, 1989, and the court issued its opinion on July 14, 1989.⁶

The opinion remanded the regulations to the Interior Department for reissuance. Five issues raised by the challenge to the rules involved choices of economic values and policies.⁷ While the court upheld the Interior Department on two issues, it overturned three choices that dramatically change the playing field for conflicts between plaintiffs and defendants in natural resource damage cases.

First, the court struck down the "lesser-of" rule, prefering restoration of the damaged resources to mere compensation for lost values. Second, the court overruled the Interior Department's relegation of so-called nonuse values to second-class status. Finally, the court rejected the Interior Department's "hierarchy of assessment methods." Two other issues — the discount rate and use of "contingent valuation" techniques — are also important economically. On these, the court sustained the Interior Department. Each of these five issues is analyzed in more detail below.

The Lesser-of Rule

The Interior Department had required that natural resource damages be measured by either restoration or replacement costs on the one hand, or certain lost values on the other [20 ELR 10128] hand, whichever is less. The Interior Department argued that CERCLA provides for a common law measure of damages and that such a measure is also economically efficient when interpreted as the lesser of restoration or lost use value.

The court was not persuaded by these arguments, however. It struck down the lesser-of damage concept, concluding that Congress expressed a "distinct preference"⁸ in CERCLA for restoring damaged natural resources rather than simple compensation for lost use values defined in narrow terms. The court observed that common law interpretations of damage do not bind the natural resource damage provisions of CERCLA.

Had the court not "waffled" a bit by introducing the notion of "grossly disproportionate cost" in its decision, one could infer that all future damage cases would revolve solely around estimates of restoration costs. But the court did waffle, noting that its decision is not "that DOI [the Department of the Interior] may not establish some class of cases where other considerations — i.e., infeasibility of restoration or grossly disproportionate cost to use value — warrant a different standard."⁹

One naturally wonders how large the gap must be between restoration cost on the one hand and lost values on the other to call forth this "different standard." To economists, something has value only to the extent that it provides satisfaction to people. That is, things do not have innate value independent of people. To the extent that people value wildlife and natural wilderness areas — as they so obviously do — these things will show up in economic accounting. If individuals' preferences change such that they lose all interest in the natural world, however, no value would be attributed to these things. While one might argue with this approach, these arguments usually reflect the intensity of the preferences of those advocating other standards.

The Interior Department argued that its lesser-of rule made sense from an economic standpoint, since one would not want to restore a resource if the cost of doing so exceeded the value society placed on it. But, the court found that the congressional preference for restoration was motivated not by a hostility to this reasoning per se, but rather by skepticism regarding human ability to measure a resource's true value. The court might have found the lesser-of rule to be consistent with congressional intent if it weighed restoration cost against the "true value of the resource."

The distinction here is between "use values" and "nonuse values," both of which are components of total value. Generally, a use value is one that involves a close physical connection between an individual and a resource. For instance, swimming, fishing, and boating provide use values of a river. Camping, hiking, and birdwatching are use values in a forest or wilderness area. Nonuse values, on the other hand, do not depend on such in situ activities. For instance, most people attach a value to the mere knowledge that the Grand Canyon exists, a value independent of their visiting it. Similarly, one may attribute a nonuse value to the preservation of certain species perhaps because of a belief that one's descendents would derive benefits from its existence.

While both kinds of values can be difficult to measure, in general nonuse values are much harder to measure than are use values.

Concern about measuring the true value of natural assets has been expressed by economists as well as legislators and judges. While writing on preservation versus development of unique natural environments in 1975, for example, a path-breaking economics text proposed that analysts ask the following question: How much would the intangible (or currently unmeasurable) value of preservation need to be for society to be indifferent about choosing between preservation and the development of natural resources?¹⁰ In other words, if the economic benefits of developing a wilderness area (for skiing, say) were $ 10 million, and if the recreation and other quantifiable benefits of leaving the area in its natural state were $ 3 million, decisions about development should focus on whether the nonquantifiable benefits were worth at least $ 7 million. The answer could well be that society should avoid developing it for skiing and leave it alone. This approach acknowledged the difficulty in defining and measuring these intangible values by specifying the minimum level of intangible values rather than trying to measure the actual level.

Since that early effort in 1975, economists have made great progress in developing techniques for measuring the values people place of nonmarketed resources. But the Interior Department's rules rejected this progress made in the quantification of nonuse value and forced attention solely on use value. In so doing, a large portion of the value of many unique natural resources — the nonuse or intrinsic value — was specifically excluded from consideration.

The most interesting economic issue raised by the court's opinion on the lesser-of rule concerns the size of the gap between full restoration or replacement cost and lost use value. How big must this gap be before restoration cost is viewed as being unreasonable? From an economic perspective, a modified lesser-of rule is appropriate. This modified rule would compare total lost value (i.e., both use and all categories of nonuse value) to restoration or replacement cost. This modified rule is efficient in the sense that damages cannot exceed the total value society places on the natural resource.

The modified lesser-of rule proposed here involves comparing restoration or replacement cost with total lost value, where "total" refers not only to the sum of use and nonuse values, but also to the aggregation of these values over all generations, present and future. Most economists believe that future generations will place a relatively higher value on the services provided by natural resources than does the present generation. If true, losses from natural resource damage grow over time and may tip the scales more toward full restoration. Even if future generations value the services of natural assets the same as the present generation does, the aggregate loss in value obviously will be larger than if the focus was exclusively on the loss suffered by the present generation alone.

The implication of this approach for redrafting the Interior Department's damage assessment regulations is straightforward. If restoration cost is less than the diminution in use value narrowly defined, the value of the natural [20 ELR 10129] resource damage should equal restoration cost. If restoration cost exceeds lost use value alone, but is less than the total loss in value, the damage should equal restoration cost. Otherwise, the damage should equal the sum of lost use and nonuse value. This rule gives operational meaning to the court's interpretation of the Interior Department's authority to "establish some class of cases where other considerations — i.e., infeasibility of restoration or grossly disproportionate cost to use value — warrant a different standard."

A final remark on the court's lesser-of ruling concerns the pace of restoration. The House Merchant Marine and Fisheries Committee report accompanying the 1986 amendments to CERCLA states that the lost use value from the time of release to the time of restoration shall be a component of the damage award.¹¹ The length of this time period, however, is determined by the pace of restoration, and the pace has a bearing on the cost. For example, the injury caused by releases of some substances may be mitigated in large part by natural activities; if one is willing to wait for these processes to have their effect, the cost of restoration may be quite low. On the other hand, if one wishes to speed the process by direct human intervention, the cost rises.

If onetakes CERCLA's emphasis on cost-effective restoration strategies to heart, "nature's way" may in many instances be preferable to direct intervention. However, as restoration time grows, so too do lost values. The court and the Interior Department are silent on the speed of restoration and also on who has the authority to determine the time path.

From an economic perspective, an explicit balancing would be desirable. An economic rule would suggest that the greater the lost use values per unit of time, the more rapid should restoration proceed.

When considering the speed of restoration in the context of lost total value, optimal decision rules for restoration speed become more complex. It seems reasonable to assume that an approximately linear relationship exists between lost use value and time to restoration. In other words, the loss in use value per unit of time may be approximately constant, say, so many dollars per month or year. On the other hand, it may be unreasonable to assume that lost nonuse values are approximately constant over time. Since nonuse values are derived from a complex appreciation of natural assets, injury to such assets might cause immediate and large losses in nonuse value that diminish rapidly over time.¹² If lost nonuse values compose the bulk of lost total value and lost nonuse value has this "sudden burst" time profile, efficient restoration might proceed at a pace far slower than would be the case if lost nonuse value had a linear time profile.

"Committed Use" Requirement

In its regulations the Interior Department defined "committed use" as "either: a current public use; or a planned public use of a natural resource for which there is a documented legal, administrative, budgetary, or financial commitment established before the discharge of oil or release of hazardous substance is detected."¹³ This definition limited consideration of nonuse values. Separately under the regulations, nonuse values could be considered only if use values could not be measured.¹⁴ By requiring current or documented plans for use of the damaged resource, the rule precluded consideration of nonuse values as sources of damages — by their very definition, such values do not involve use. The Interior Department's regulations did allow nonuse values to be considered if some use values could not be measured.

While the court's opinion upholds the committed use requirement, it relegates it to the status of a simple mechanism for avoiding highly speculative future uses. Now, the committed use requirement only excludes a tendency for damages to reflect the "what might have been" category of use. For example, it is easy to imagine conjectural descriptions of booming recreational facilities after the fact, when there were no plans or documented signs of this process prior to any release. Under the court's interpretation, these conjectures do not represent compensable damages.

Hierarchy of Assessment Methods

In its original regulations, the Interior Department prescribed a hierarchy of valuation methods in which market-based techniques were preferred to nonmarket approaches, and more importance was attached to use values than to nonuse values. The D.C. Court of Appeals found both of these preferences inconsistent with congressional intent in CERCLA.

Certainly, from an economic perspective the court's decision to reject the methods hierarchy is laudable. Even if markets for natural resources existed, it would be hard to argue that observed market prices reflect social value, if only for the public goods nature of the vast majority of these assets. Similarly, a broad definition of the values to be considered seems a reasonable policy choice to which economists would not object.

The implication of this ruling for damage assessments seems clear. Nonuse values would now appear to have equal standing with conventional use values and must be included in damage awards. This may lead to significantly larger awards than those based exclusively on the Interior Department's earlier narrow definition of use values. The expansion could be dramatic. Two economists recently studied the relative size of nonuse to use values for water quality improvements in rivers such as the Potomac and Monongahela. Even though these rivers are not "unique" resources for which one might expect large nonuse values, they found that nonuse values were as much as 80 percent of use values.¹⁵

[20 ELR 10130]

The 10 Percent Discount Rate

Discounting involves converting the stream of future monetary values an asset provides into a single present value. Since 1972, the Office of Management and Budget has mandated that the discount rate used in federal benefit-cost analysis and project evaluation should be 10 percent in real terms. The Interior Department adopted the 10 percent discount rate for determining the present value of natural resource damages occurring in future years. Although there has been great debate within the economics profession about the appropriate discount rate,¹⁶ most economists now agree that the rate should reflect an individual's willingness to trade off present for future consumption. There is also general agreement that the rate is less than 10 percent in real terms. That is, a 10 percent rate understates the true damage from a long-lasting injury to natural resources.

The court avoided these technical economic issues and upheld the 10 percent discount rate on entirely different grounds. The court reasoned that CERCLA requires a lump-sum payment now to cover future costs of restoration. Thus, the court reasoned, a current-period damage award placed in an interest-bearing account can be smaller than the simple sum of the future years' expenditures, because it will earn interest in the meantime. The rate at which future costs should be discounted must obviously include the expected rate of inflation and must be based on market rates of return. Given this interpretation of CERCLA, the court found no grounds upon which to challenge the 10 percent discount.

However, the court failed to appreciate the unique role of discounting in establishing the value of an environmental asset. This oversight means that litigants will continue to disagree about the appropriate rate to use when capitalizing forward the past stream of lost values and discounting back those future values lost during restoration or lost in perpetuity if restoration is infeasible.

To understand the role of discounting one must understand the principle of damage valuation employed by economists. Economists view natural resources as assets providing a flow of services to humans. Humans place value on this service flow and in turn on the natural asset. Injury to the asset diminishes the quantity or quality of these service flows and thus lowers the level of human utility provided by the services. This causes a loss in use or nonuse value, or both.

Given this framework, one can think of natural resource damage as the aggregate loss in human utility among all individuals and over time. Aggregation over time requires discounting future losses in utility by each individual's rate of time preference. This rate of time preference may bear no relation to the mandated 10 percent rate. Moreover, damage calculations over long periods of time are very sensitive to the rate selected, so the seemingly arbitrary 10 percent rate could introduce a serious error.

There is one final point to make regarding the case where restoration is deemed infeasible, or where the cost of restoration or replacement fails to meet the suggested modified lesser-of rule. In this case, lost values continue forever. Any discount rate, no matter how small, will produce a finite current period damage for this infinite stream of lost values. However, choosing not to discount the loss in utility suffered by future generations (an approach often argued on ethical grounds) will result in a damage assessment of infinity and suggest that restoration should proceed "at all cost."

Contingent Valuation

One aspect of the Interior Department's regulations — the acceptance of contingent valuation techniques — has the effect of expanding damage awards, and the court sustained the regulations here. Contingent valuation is the use of surveys to ask a wide sample of citizens how much they would pay for a resource in a hypothetical market.¹⁷ Economists now generally agree that at least in some circumstances properly designed contingent valuation surveys will yield valid results.

The importance of this issue cannot be overemphasized. Using contingent valuation to measure nonuse values means that someone in Florida who never intends to go near Valdez, Alaska, can still suffer a measurable loss as a result of the Exxon Valdez oil spill there. This greatly expands the possible number of people experiencing damage as a result of an accidental release. Since contingent valuation techniques are the only way known to measure these nonuse damages, they will play a key role in determining the magnitude of awards in certain cases.

Deciding whether a contingent valuation study is well-designed is not a mechanistic decision. It involves subjective decisions by experienced economists. Toward this end, it seems reasonable for plaintiffs anddefendants to set up peer review panels of experts, not to testify, but rather to comment critically on the design and execution of the studies.

Conclusions

The court's ruling will most likely lead to a substantial redrafting of the regulations, dramatically alter their character, and have significant impacts on current and future natural resource damage cases. The emphasis on restoration, the elevation of nonuse values to an equal footing with those that arise from use of the resources in question, and the confirmation of contingent valuation methods will form the basis for the character of the revised rules.

Moreover, the court's implicit suggestion that the Interior Department develop rules to govern situations where there is a big gap between restoration costs and diminution in total value has introduced considerable uncertainty into the range of outcomes possible in current damage cases. For example, until new rules are promulgated, lower courts may follow the D.C. Circuit's suggestion and define for themselves circumstances in which restoration cost is viewed as grossly disproportionate to lost value. Alternatively, courts may choose to follow the "congressional preference" for restoration, which the D.C. Circuit recognized, and award damages based on restoration cost without considering lost value.

With respect to the redrafting of the damage assessment [20 ELR 10131] regulations, the Interior Department should look to well-established principles of applied welfare economics for resolution of the problem of "grossly disproportionate cost of restoration to lost value." Nonetheless, revised rules adopting these or other justifications are unlikely to be proposed in the near term. Thus, uncertainty, postponement, and a preference (on the part of defendants) for protracted cases may be some of the undesirable by-products of the D.C. Circuit's ruling.

On related matters, the court failed to address the substance of other economic issues in damage assessments, such as the discount rate.

How is all of this likely to play out in the immediate future? It should come as no surprise that the position of trustees in negotiating out-of-court settlements has been considerably strengthened, since, in many cases, restoration cost can be expected to exceed the diminution in lost value (both use and nonuse). This may have been part of the court's purpose, tilting pretrial settlements closer to actual restoration of injured resources.

But, the court's language regarding "grossly disproportionate cost" will also have an impact. Because lost value will serve to define the floor for future damage awards, economic measurement of the diminution in value will continue to have an important place in litigation. In fact, the outcome of many cases may well turn on the gap between restoration and lost value. While plaintiffs in current and past cases have sought to measure lost value, future plaintiffs may refrain from doing so if they believe restoration cost will exceed it. In contrast, defendants — who so far have been content to attack plaintiffs' estimates of lost value — may now have to undertake their own valuation studies to support claims that restoration is "grossly disproportionate to use value."

1. 42 U.S.C. §§ 9601-9675, ELR STAT. CERCLA 001-075.

2. 880 F.2d 432, 19 ELR 21099 (D.C. Cir. 1989).

3. 42 U.S.C. § 9651(c), ELR STAT. CERCLA 062.

4. Two types of rules were to be promulgated: so-called type A rules are to apply to "simplified" assessments and are to define a set of standard procedures for valuing the damage done by minor, short-duration releases of hazardous waste or oil; type B rules, on the other hand, are to apply in situations requiring individualized assessments and are to specify protocols for selecting the best procedures for estimating damages.

The literature on natural resource damages increasingly distinguishes between "damages" and "injury." Damages are the monetary quantification of a physical injury.

5. 51 Fed. Reg. 27674 (Aug. 1, 1986) (type B rules); 52 Fed. Reg. 9042 (Mar. 20, 1987) (type A rules).

6. Ohio v. United States Department of the Interior, 880 F.2d 432, 19 ELR 21099 (D.C. Cir. 1989). The Ohio opinion considered the type B rules. A companion case incorporated much of the reasoning of the Ohio opinion in considering the type A rules. Colorado v. United States Department of the Interior, 880 F.2d 481, 19 ELR 21127 (D.C. Cir. 1989).

7. Some issues raised in the litigation did not involve economics. For a complete legal analysis of the decision, see Olson, Natural Resource Damages in the Wake of the Ohio and Colorado Decisions: Where Do We Go From Here? 19 ELR 10551 (Dec. 1989).

8. 880 F.2d at 459, 19 ELR at 21113.

9. Id.

10. J. KRUTILLA & A. FISHER, THE ECONOMICS OF NATURAL ENVIRONMENTS: STUDIES IN THE VALUATION OF COMMODITY AND AMENITY RESOURCES (1975).

11. H.R. REP. No. 253 (IV), 99th Cong., 1st Sess. 50 (1985).

12. Such a relationship is purely conjectural for now, but could be tested empirically by economists using contingent valuation methods.

13. 43 C.F.R. § 11.14(h) (1989).

14. 43 C.F.R. § 11.83(b) (1989).

15. Fisher & Raucher, Intrinsic Benefits of Improved Water Quality: Conceptual and Empirical Perspectives, in ADVANCES IN APPLIED MICRO ECONOMICS (V. Smith & A. Witte, eds. 1984). There are reasons for caution in interpreting the studies summarized by Fisher and Raucher because they generally focus on the relative size of monetary value increases from uncertainty in comparison with definite or certain use. Even so, these uncertainty components probably include nonuse values.

16. See, e.g., R. LIND, DISCOUNTING FOR TIME AND RISK IN ENERGY POLICY (1982).

17. See generally R. MITCHELL & R. CARSON, USING SURVEYS TO VALUE PUBLIC GOODS (1989).