General Electric Co. v. United States Department of CommerceGeneral Electric Co. v. United States Department of Commerce
Opinion for the Court filed by Circuit Judge TATEL.
Seventeen months after the oil tanker Exxon Valdez ran aground in Prince William Sound, spilling almost eleven million gallons of North Slope crude, Congress enacted the Oil Pollution Act of 1990 to make parties responsible for oil spills liable for damage to natural resources. In this case, we consider both procedural and substantive challenges to the final rule that the National Oceanic and Atmospheric Administration issued pursuant to the Act. Concluding that the final rule’s authorization for removal of residual oil suffers from a lack of reasoned decision-making, we- vacate this portion of the rule and remand to the agency for further consideration. With the agency’s consent, we also vacate and remand the final rule’s authorization for recovery of legal fees. In all other respects, we sustain the final rule.
I
Prior to the Oil Pollution Act of 1990, Pub.L. No. 101-380,104 Stat. 486 (codified at
OPA focuses specifically on oil discharges in the nation’s waterways and along its coastlines. Amending the Clean Water Act, section 4201(a) of OPA directs the President, who has since delegated his authority to the Environmental Protection Agency and the Coast Guard, to remove spilled oil.
To facilitate damage recovery, OPA directs the President, acting through NOAA, to “promulgate regulations for the assessment of natural resource damages ... resulting from a discharge of oil.”
Engaging in a six-year rulemaking process, which produced proposed rules in 1994, 59 Fed.Reg. 1062 (1994), and in 1995, 60 Fed.Reg. 39,804 (1995), NOAA promulgated its “final rule” governing trustee assessment of natural resource damages in 1996. Natural-Resource Damage Assessments, 61 Fed. Reg. 440-510 (1996) (adding
The final rule lays out a three-stage procedure for assessing injuries resulting from oil spills and for developing and implementing plans to restore damaged resources. Termed the “Preassessment Phase,” the first stage requires trustees to determine whether they hаve jurisdiction under OPA to pursue restoration activities and whether actions taken by other agencies have adequately addressed the injuries. This first stage also requires the trustee to collect and analyze pertinent data, prepare a notice of intent to conduct restoration planning activities, and open a publicly available administrative record.
The second stage, the “Restoration Planning Phase,” has two substages. The “injury assessment” substage requires the trustee to determine whether an injury has occurred, whether a “pathway” can be established between the discharged oil and the injury, and whether the injury resulted from the discharge.
In the third and final stage of the process, the “Restoration Implementation Phase,” the trustee presents a written demand for payment to the owner of the tanker or other party or parties responsible for the oil discharge. Id. § 990.62(a). . If the responsible party refuses to satisfy the demand within ninety days, or if the trustee and the responsible party cannot agree on an alternative figure, the trustee may sue the responsible party or seek an appropriation from the Oil Spill Liability Trust Fund. Id. § 990.64(a). In any suit filed by the trustee, its damage assessment is entitled to a rebuttable presumption, id. § 990.13, if the trustee can demonstrate that its assessment procedures are “capable of providing assessment information of use in determining the type and scale of restoration appropriate for a particular injury,” id. § 990.27(a)(1), that any additional cost of a “more complex procedure” reasonably relates to the expected increase in the quantity or quality of information, id. § 990.27(a)(2), and, most important, that its assessment procedures are “reliable and valid for the particular incident,” id. § 990.27(a)(3).
Pursuant to section 1017(a) of OPA, which allows interested persons to petition this court to review any OPA regulation within ninety days of its promulgation,-
II
We begin with section 990.13’s rebut-table presumption. Relying on our decision in Chemical Mfrs. Ass’n v. Department of Transp.,
Industry petitioners also argue that the rebuttable presumption gives trustees a “powerful advantage” in any subsequent litigation and that the agency must issue regulations that are sufficiently “sound and credible” to deserve the presumption. Petitioners, however, cite no authority for this latter proposition, nor do they explain why the final rule’s reliability and validity requirement would not satisfy their “sound and credible” standard. More important, it is not at all clear that the rebuttable presumption even gives trustees a “powerful advantage.” In the final rule’s preamble, NOAA “interprets” the rebuttable presumption as imposing upon responsible parties “the burdens of presenting alternative evidence on damages arid of persuading the fact finder that the damages presented by the trustees are not an appropriate measure of damages.”
We turn to industry petitioners’ specific challenges to the final rule.
Contingent Valuation and Passive Use Values
In valuing damage to natural • resources caused by oil discharges, trustees may consider two types of losses: active and passive. “Active-use” losses refer to the loss of actual use of a natural resource. An oil spill that contaminates a National Seashore, for example, causes an “active-use” loss for those unable to use the beach for swimming or fishing. Under the final rule, the trustee develops a plan to restore the beach to its original condition and, while restoration takes place, to provide alternative fishing, swimming and other active-use opportunities.
NOAA’s final rule also authorizés recovery of what are known as nonuse or “passive” losses, the value individuals place upon the existence of natural resources, even if they never plan to make active use of them. In the case of the National Seashore, for example, people who have never used the beach may nevertheless value its existence. To assess this value, researchers employ a survey technique known as “contingent valuation,” in which they create a hypothetical market and ask people — survey respondents — how much they would pay to preserve or protect a given resource. Averaging the responses, researchers then determine the value the public places on the resource. See Jeffrey C. Dobbins, Note, The Pain and Suffering of Environmental Loss: Using Contingent Valuation to Estimate Nonuse Damages, 43 Duke L.J. 879, 882 (1994).
Because contingent valuation is not without controversy, NOAA commissioned a special panel to study the technique and report on its appropriateness for assessing natural resource damage. See 58 Fed.Reg. 4601, 4602-14 (1993) (Appendix I — Report of the NOAA Panel on Contingеnt Valuation). After considering the critiques of contingent valuation, the panel, which included two Nobel laureates, concluded that if properly conducted under strict guidelines, the technique can convey useful and rehable information that “can produce estimates reliable enough to be the starting point of a judicial process of damage assessment.” Id. at 4610. Based on the panel’s report, NOAA’s first proposed rule explicitly authorized trustees to employ contingent valuation and provided detailed standards for using it. 59 Fed.Reg. at 1182-84. In its next proposed rule, however, and then again in its final rule, NOAA omitted all references to contingent valuation, instead authorizing trustees to choose whichever assessment techniques they wish, subject to
NOAA believes that the standards set forth in§ 990.27 are sufficient to allow trustees and responsible parties to determine the acceptability of a particular assessment procedure for a given incident. NOAA supports the use of all of the procedures discussed in’ Appendix B of the preamble as reliable and valid within the appropriate context and when performed in accordance with accepted professional practices. NOAA does not believe that the rule should set forth specific standards regarding the implementation of individual procedures, as it is not feasible to prescribe all valid uses of these procedures. The validity and reliability of procedures will depend on the circumstances of particular incidents____ Thus, NOAA believes that most of the comments received, which relate to potential prоblems with certain applications of these procedures, will be dealt with in the context of specific incidents.
Id. at 470.
Industry petitioners argue that NOAA acted arbitrarily and capriciously by “ignoring” the panel’s warning that contingent valuation studies must be conducted subject to stringent standards. Relying on our Kennecott decision, NOAA responds that this argument is not ripe for judicial review because the preamble’s reference to contingent valuation neither imposes a legal obligation upon petitioners nor has any immediate effect upon them. While this argument has force with respect to petitioners’ temporary loss argument, see discussion infra at 774, it has no applicability to their purely procedural challenge. The question we faced in Kennecott — whether the Interior Department exceeded its CERCLA authority by arguably authorizing recovery of certain damages through language in the regulation’s preamble — was not ripe because whеther Interior had intended to bind anyone with the preamble’s language was not at all clear. Until a trustee invoked the preamble to affect the outcome of a real dispute, we held, we lacked both the need and the factual basis for resolving the question. Kennecott, 88 F.3d at. 1222-23.
Industry petitioners’ purely procedural argument, that NOAA failed to consider relevant comments during the rulemaking process, is quite different. Unlike in Kennecott, where our treatment of the preamble issue would have benefited from a “concrete case,” id. at 1223, we now know everything we need to know about NOAA’s treatment of the administrative record. The issue petitioners present will never be more fit for review.
Industry petitioners’ .argument fails, however. NOAA ignored neither the panel’s comments nor the criticisms of contingent valuation that the panel considered. It simply gave trustees discretion to use contingent valuation, so long as the technique produces, as required by
Going beyond their procedural claim, industry petitioners argue that NOAA acted arbitrarily and capriciously by failing to bar contingent valuation altogether. Because this argument amounts to a facial challenge to the final rule that does not depend on the facts of a particular case, it is, like petitioners’ purely procedural argument, ripe for review. But also like that argument, it fails. We held in Ohio that the Interior Department had not acted arbitrarily or capriciously by authorizing CERCLA trustees to use contingent valuation. Ohio,
Industry petitioners next argue that NOAA acted arbitrarily by authorizing the recovery of passive-use values for temporary losses of natural resources. According to petitioners, passive-use losses occur only where resources are lost forever; temporary losses, they claim, can never give rise to passive-use losses. Although the administrative record lends support to NOAA’s contention that temporary losses can cause loss of passive-use values, e.g., id. at 4608 (CV panel concluding that “interim passive-use values are additive over time”), we agree with NOAA that this issue is not ripe for judicial review. The proper time to address the question will come if and when a trustee actually assesses damages for temporary losses in a particular case. Unlike industry petitioners’ purely procedural argument and their challenge to NOAA’s refusal to bar trustee use of contingent valuation, right now we lack both the factual record and the detailed findings needed to resolve their temporary loss argument.
Removal Authority
Section 990.53(b)(3)(i) of the final rule authorizes trustees to “[rjemove conditions that would prevent or limit the effectiveness of any restoration action (e.g., residual sources of contamination).”
Industry petitioners argue that because OPA delegates sole responsibility for oil removal to the President, NOAA exceeded its statutory authority by authorizing trustees to remove residual oil. In support of their argument, petitioners point out that OPA treats restoration and removal separately and for each contains distinct definitional; limitation, and liability provisions.
[cjonsult with the affected trustees designated under .section 2706 of this title on the appropriate removal action to be taken in connection with any discharge of oil. For the purposes of the National Contingency Plan, removal with respect to any .dischargе shall be considered completed when so determined by the President in consultation with the Governor or Governors, of the affected States. However, this determination shall not preclude additional removal actions under applicable State law.
We would ordinarily analyze NOAA’s interpretation of OPA under Chevron, asking first whether Congress spoke clearly to the issue or, if not, whether NOAA’s interpretation of the statute is permissible and thus entitled to- deference. Chevron,
NOAA argues that under the proposed rule, the trustee could have actually removed the oil, but we think this contention conflicts with the proposed rule’s plain language, which merely directs trustees to “consider” whether conditions exist that would make the primary restoration alternative impractical. By comparison, the final rule quite clearly authorizes trustees to “[rjemove conditions that would prevent or limit the effectiveness of any restoration action (e.g., residual sources of contamination).”
Not only has NOAA failed to explain this difference between the final rule, on the one hand, and its preamble and the proposed rule, on the other, but it has also not explained the interrelationship between trustees’ residual removal authority and the primary removal authority of EPA and the Coast Guard. If trustees are to have residual removal authority, clearly explaining how and under what circumstances they will exercise such authority is critical, particularly since OPA requires the President to consult with trustees during primary removal actions.
Monitoring Costs and Legal Fees
Section 990.30 of the final rule defines reasonable assessment costs to include both “monitoring and oversight costs” as well as “administrative, legal, and enforcement costs.”
Because OPA is silent on the question before us, we proceed to Chevron’s second step, where we have no doubt that NOAA acted reasonably by authorizing the inclusion of monitoring costs as part of restoration costs. According to NOAA, monitoring is an essential element of restoration:
NOAA believes that restoration monitoring costs are a recoverable component of natural resource damages. Monitoring is essential to ensure that restoration actions accomplish their intended goals and objectives and do not cause unanticipated harm to the environment or public health. In addition, monitoring is essential to determine whether the terms of restoration agreements have been met, upon which a release from liability is premised.
61 Fed.Reg. at 491. Not only does this conclusion seem eminently reasonable to us, but industry petitioners have suggested no reason for excluding monitoring costs from the cost of restoration.
We are equally unpersuaded by industry petitioners’ argument that, because monitoring takes place after assessment, NOAA improperly included monitoring costs as a component of assessment costs. In view of our conclusion that NOAA acted reasonably by including monitoring costs as a recoverable cost, the precise heading under which the trustee may recover monitoring costs seems insignificant, particularly since industry petitioners have given no reason why it makes any difference at all. ■
Relying on Key Tronic Corp. v. United States,
NOAA’s concession does not end this matter, however, because the parties continue to disagree about what other legal costs trustees may recover. Although both sides agree that trustees may recover assessment costs attributable to tasks that lawyers happen to perform but which others, such as engineers or private investigators, could have performed, they disagree about whether trustees máy recover costs stemming from legal work not directly in furtherance of litigation (e.g., pre-litigatión legal opinions, title searches) that only lawyers could have performed. In view of NOAA’s consent to vacatur of this portion of the rule, we decline to resolve this question, instead leaving it to NOAA to draw the precise line between recoverable and nonrecoverable legal costs in subsequent rulemaking.
Ill
Industry petitioners present several other challenges to the final rule which, in view of representations made by NOAA in its brief and during oral argument, have been resolved. In order to document the agency’s representations, we summarize each issue briefly and include relevant oral argument excerpts in the appendix to this opinion.
In their opening brief,- industry petitioners argued that section ,990.51 of the final rule, which they read to require the trustee to demonstrate only “injury,” “exposure,” and “pathway,” Brief for Industry Petitioners at 41, would allоw trustees to assess liability without evidence that the responsible party actually caused the oil discharge that did the damage, thus violating OPA’s causation re
The parties also now agree on the correct interpretation of
The parties’ final area of agreement relates to
This brings us finally to the arguments advanced by insurance petitioners. Lacking merit, they require but brief discussion.
To begin with, petitioners have no standing to argue that section 990.20(b), which allows trustees who have begun damage assessments under CERCLA to switch to the final rule, is impermissibly retroactive. They make no claim that any partiсular trustee has switched to the final rule, nor have they argued, as they could not at this time, that such a switch resulted in a trustee' imposing greater damages upon a responsible party insured by petitioners. Because petitioners have thus shown neither concrete nor imminent injury, see Louisiana Envtl. Action Network v. Browner,
Pointing to section 1002(b)(2)(A) of OPA, which makes responsible parties liable for “[djamages for injury to, destruction of, loss of, or loss of use of, natural resources,”
Insurance petitioners also challenge . the final rule because it makes no reference to OPA’s liability limits. As NOAA points out, however, nothing in the final rule affects a responsible party’s right to invoke the statute’s liability limits. If trustees need resources above and beyond statutory limitations to implement restoration plans, they may draw upon the Oil Spill Liability Trust Fund.
Insurance petitioners next argue that, by authorizing trustees to assess costs for restoration activities undertakеn after the date of the initial demand, the final rule violates responsible parties’ rights under OPA to seek contribution from other responsible parties. Their concern stems from section 1017(f)(3), which prohibits bringing contribution actions more than three years after either the date of judgment in any action under OPA for recovery of costs or damages, or the date of entry of a judicially approved settlement with respect to those costs.
Insurance petitioners’ argue finally that the final rule grants trustees “uncontrolled discretion.” But we think
V
We vacate
So ordered.
Excerpts from September 11, 1997 Oral Argument
Causation
THE COURT: Do I understand, do we understand that the parties are in agreement on this issue?
MS. MEDINA: I’m not sure I would characterize this “in agreement.” I guess I would say that we have determined—
THE COURT: I read your brief as conceding that Petitioner was dead right and pointed out your regulation was never intended to avoid causation.
MS. MEDINA: Absolutely. To prove legal causation, to prove liability. That is something for the Court to do. All the regulation requires is that we prove that the spill, the injury was derived from or resulted from the spill.
THE COURT: Wait a minute. Now I don’t understand. You mean is the Government’s — the trustee can come in and say look, there was a spill. I’ve shown it caused terrible problems. Now it’s up to you, your Hon- or, to determine who did it.... That is not correct, is it?
MS. MEDINA: That is not correct.
MS. MEDINA: We do have a responsible party on the other side of the table.
THE COURT: You have to show causation as part of your case.
MS. MEDINA: Yes, we do, absolutely, but we don’t have to do it as part of our rule, as a part of our process in assessing the damages.
THE COURT: That’s not — I think the Petitioner quite legitimately read the rule as suggesting that there was some effort to avoid the obligation to show causation and the Government comes in, as often happens on a direct challenge to rulemaking, says no, we didn’t intend that. That was— now, what is important and we have done this on a couple of occasions, your brief then agrees with Petitioner and says look — agrees in the sense that we never intended the rule to mean that.
MS. MEDINA: Right.
THE COURT: We mean the same thing you think it should mean.
MS. MEDINA: Right.
THE COURT: That now, whenever it’s reflected in our opinion, then binds the Agency, does it not?
MS. MEDINA: Yes, it does.
THE COURT: Your position is that estimate means calculate?
MS. MEDINA: Yes, Your Honor.
THE COURT: Explicitly measure means being very precise; estimate means a little general and calculate means both is that it?
MS. MEDINA: I think we used the words very much interchangeably. We didn’t intend there to be this huge distinction.
THE COURT: So I think Judge Silberman’s question of an earlier issue if we write an opinion and say we. understand the word “estimate” in the regulation to mean “calculate” that’s something the Agency — it would be consistent with your understanding of the issue?
MS. MEDINA: Yes, Your Honor.
THE COURT: Okay, that’s it. So calculate and measure are equivalent?
MS. MEDINA: Yes, Your Honor.
Simplified Procedures
THE COURT: [D]o I understand it works this way? A simplified method is a computer program of some kind that takes account of inputs, I assume, that come from the site of the oil spill. Is that right?
MS. MEDINA: Yes.
MS. MEDINA: Right.
THE COURT: Okay ... let’s assume that that formula, once you take account of the site specific things produces a figure of $100,000 worth of damage. All right?
MS. MEDINA: Right.
THE COURT: Okay, are there any circumstances under which that is the end of the process?
MS. MEDINA: I would think that we would still develop a plan.
THE COURT: No, ho, don’t tell me what you think. I want to know what the regulations say. Are there any circumstances under the regulation under which a trustee would send that $100,000 bill to the responsible party?
MS. MEDINA: No ...
THE COURT: Okay, let me go back to my intermediate question. Are there any circumstances under which the $100,000 calculation that flows from the computer model that’s based on sitе specific data would be the end of the process, that that’s the bill that you send to the responsible party?
MS. MEDINA: That’s not the end of the process. We then develop a plan ...
MR. TSAO: Then you need to develop and implement a plan to figure out how you’re going to compensate for that injury and as you asked before, what you end up deciding in terms of restoration plan could be greater or less than $100,000 that comes out of the simplified procedure. There’s no absolute requirement that it be the same____
MR. LEHNER: [T]o give you an example of how models work, in my prior life I actually was involved in mini-oil spills. You will have a circumstance,, for example, big oil spill where the trustees сontract with people, go out, find out the actual extent of the oil, how many birds are dead, what percentage of the grass in the wetlands is dead, et cetera, and then try to develop a restoration plan accordingly.
You can also use in a smaller spill a model to help you do that. You put in the inputs of the amount of oil of the geographic area and they will come up with estimates of the biological injury, the type of biological injuries and also using their economic data base, the approximate costs of those. The trustees caii then use the simplified assessment method, the model to develop a restoration plan.
MR. LEHNER: The assessment is stage 1 in the stage.
THE COURT: You have to go on to stage 3?
MR. LEHNER: Yes, stage 2 is letting the feds help out the states, so it’s really a two stage— assess and develop a plan and they all have to go to stage 2.
THE COURT: You develop a plan without a site specific activity?
MR. LEHNER: The rules provide for regional restoration рlans and the Petitioners don’t challenge those. You do need—
THE COURT: Wait a minute, counsel. I didn’t understand that at all. What do you mean?
MR. LEHNER: The rules — you can have a site specific plan or you can — the rules allow a trustee to use what are called regional plans
THE COURT: What is a regional plan?
MR. LEHNER: Let me give an example and explain it. In 1990, in the New York Harbor, we had six oil spills, six big oil spills in the first half of the year. The oil kept the same place again and again and some oil was a little further to the west and a little to the east, but it was all in the same region.
What the trustees recognized is that it was absurd to develop six entirely sepa-
So what the rules do allow in our view is you can use a model and then if there is a regional plan, perhaps you can use that, but you have to do — you do have to come up with a plan. That’s what the statute makes clear and that’s what the regs make clear....
THE COURT: Do you really have as much problem with that after you heard the Government’s position?
MR. BRUCE: Well, if the Government’s position were written into the rules the way the Government has articulated its position here in Court today, I don’t know what our position would have been, but that’s not what the rules say. The rules explicitly incorporate the Type A, DOI models. They allow other models as well, but they explicitly incorporate Type A DOI models ...
THE COURT: What’s so terrible about the DOI—
MR. BRUCE: No, no. I just want to tell you what they do. The DOI models and this is 61 Fed.Reg. at page 20562 describing those models that were actually published after the NOAA’s rule, they say “when trustees use a Type A procedure, that’s the model, they perform injury determination, quantification and damage determination through a computer model.” Keep that in mind, damage determina-
THE COURT: Oh, but in the regulation it’s only an assessment?
MR. BRUCE: •No, no. The Type A models are designed to put out a number that is the whole shooting match, the whole thing, and that’s what these rules do. That’s what—
THE COURT: If we don’t read them that way and the Agency certainly, the Government certainly hasn’t suggested that in their argument here, then you don’t have a problem. The fact that they’re using computers and starting with the simplified method — that can’t possibly be your objection____.. Your objection is if they were going to impose, come to Court, seek damages based on a computer without real hard evidence.
MR. BRUCE: Alone. That’s correct.
THE COURT: And without any development of a plan.
MR. BRUCE: Without anything else. That’s what the rule allows them to do. Now if they’ve abandoned that position, if they’ve walked away from that and somehow through the adjudicatory process they can do that and the Court so holds, we don’t have an argument.
Notes
Ms. Medina is counsel for NOAA. Mr. Tsao, an attorney with the Natural Resources Division of the Department of Justice, also appeared on behalf of NOAA. Mr. Lehner appeared for NRDC. Mr. Bruce appeared for industry petitioners,