Worst Case Analysis: The Final Chapter?

Vicki O'Meara Masterman

Editors' Summary: This Term, the Supreme Court will decide a case construing NEPA for the first time since 1983. The Court must decide, among other issues, whether NEPA requires federal agencies to include worst case analyses in EISs. The Court will review two Ninth Circuit cases ordering agencies to perform worst case analyses even though the Council on Environmental Quality (CEQ) removed the requirement from its NEPA regulations in 1986. The Ninth Circuit held that the worst case requirement, which required agencies to analyze the worst possible consequences of a proposed action when faced with incomplete or unavailable information, is a requirement of prior NEPA case law and thus survives CEQ's decision to delete it from its regulations. The author sets out the regulatory history leading to CEQ's decision to rescind the requirement and discusses the case law interpreting the requirement. She argues that the worst case analysis is not a requirement of NEPA or NEPA case law. The author concludes that mandatory worst case analysis would likely prevent federal agencies from using other risk evaluation methods and would actually diminish the importance of environmental considerations in agency decisionmaking.

Ms. Masterman is an associate with the law firm of Jones, Day, Reavis & Pogue in Chicago, Illinois. She has formerly served as the Acting Deputy General Counsel of the U.S. Environmental Protection Agency, a White House Fellow, and an Assistant to the Army General Counsel.

[19 ELR 10026]

This Term, the Supreme Court will decide whether the National Environmental Policy Act (NEPA)¹ requires federal agencies to include worst case analyses in environmental impact statements (EISs). Worst case analysis is the identification and consideration of the worst potential consequences of a future action. Until amending its NEPA regulations in 1986, the Council on Environmental Quality (CEQ) required agencies to perform worst case analyses when they lacked information on the significant environmental effects of proposed federal actions. Though CEQ has rescinded its worst case analysis requirement, the U.S. Court of Appeals for the Ninth Circuit has continued to impose the requirement on federal agencies, tracing the requirement to perform worst case analyses to NEPA case law preceding CEQ's regulation.

In response to petitions by the Solicitor General, the Supreme Court has granted certiorari to two Ninth Circuit cases that required agencies to perform worst case analyses after the CEQ had revoked the worst case rule. In Methow Valley Citizens Council v. Regional Forester² and Oregon Natural Resources Council v. Marsh,³ the Ninth Circuit Court of Appeals explicitly ignored CEQ's revocation of the worst case rule and, citing NEPA case law, required the Forest Service and the Army Corps of Engineers, respectively, to perform worst case analyses. In his petitions seeking writs of certiorari for these decisions, the Solicitor General argued that the Ninth Circuit erred in concluding that NEPA requires agencies to perform worst case analyses whenever they are unable to obtain information on which to base reasoned decisions.⁴

[19 ELR 10027]

Whatever the wisdom or the value of worst case analysis, it is not a requirement of NEPA or of NEPA common law. NEPA requires federal agencies to disclose and to consider the environmental effects of proposed actions about which they lack information, yet it prescribes no specific method to accomplish these purposes. Furthermore, while the NEPA cases upon which the Ninth Circuit has relied consider reasonable speculation to be an inherent part of the NEPA process, these cases do not even mention worst case analysis, let alone suggest that it is a requirement of NEPA.

This Article first sets out the regulatory history of the worst case analysis requirement and explains the current CEQ regulation on incomplete or unavailable information.⁵ The Article then discusses the court decisions interpreting the worst case rule and focuses in particular on the NEPA cases cited by the Ninth Circuit as supporting its holdings in Methow Valley and Oregon Natural Resources.⁶ Finally, the Article addresses the importance of the Court's decision both with respect to the treatment of incomplete information in EISs and to the CEQ's power to guide federal NEPA compliance.⁷

CEQ's Worst Case Analysis Regulation

The worst case analysis requirement was a component of CEQ's first regulation addressing the problem of evaluating environmental impacts about which there is only limited information.⁸ This 1978 regulation, entitled "[i]ncomplete or unavailable information," required federal agencies confronting incomplete information about environmental effects to follow certain procedures including, in some circumstances, a worst case analysis.⁹ In 1986, after three years of public deliberation, the CEQ revised this regulation.¹⁰ The revised rule retains all the requirements of the previous rule, except that it replaces worst case analysis with the requirements to summarize the existing credible scientific evidence and to evaluate the impacts by "theoretical approaches or research methods generally accepted in the scientific community."

CEQ's rule on incomplete or unavailable information is, and has been since its promulgation in 1978, a part of the CEQ's EIS regulations.¹¹ The rule on incomplete or unavailable information applies when agencies lack information about the significant adverse direct and indirect environmental effects they are evaluating in EISs for proposed actions. Agencies may lack information about the magnitude of an impact or about the likelihood of an impact's occurrence. The rule on incomplete or unavailable information does not guide the identification of the significant adverse direct and indirect effects of proposed actions, but rather applies when agencies identify such impacts about which they lack information.¹²

The Old Rule

CEQ's 1978 regulation on incomplete or unavailable information required agencies assessing adverse impacts about which there was only limited information to: disclose the lack of information, obtain the information if possible, weigh the need for the action against the risk of uncertainty about its effects, and perform worst case analyses if they decided to proceed with the action. The rule contained two subsections, with the worst case analysis requirement appearing in the last sentence of the second subsection.¹³

Specifically, the regulation first required agencies encountering "gaps in relevant information or scientific uncertainty" to "make clear that such information is lacking or that uncertainty exists." The rule then required agencies to obtain the missing information if it was "essential to a reasoned choice among alternatives" and if the "overall costs of obtaining it [were] not exorbitant." If the overall costs of obtaining "essential" information were exorbitant, or if the means to obtain "important" information were unknown, the rule required agencies to weigh the need for the proposed actions against the "risk and severity of possible adverse impacts were the action[s] to proceed in the face of uncertainty." Finally, if the agency decided to proceed with the proposed action in the face of uncertainty, the rule required the agency to "include a worst case analysis and an indication of the probability or improbability of its occurrence."

In its 1981 guidance memorandum (the "Forty Most Asked Questions"), CEQ explained the worst case analysis rule as requiring "reasonable projections of the worst possible consequences of a proposed action."¹⁴ The guidance contained the explanatory example of an agency preparing an EIS for a proposed cooling water facility where the agency lacked information on the number of juvenile fish threatened by entrainment in the facility. In [19 ELR 10028] this case, CEQ stated, "the responsible agency must disclose and consider the possibility of the loss of the commercial or sport fishery." CEQ's guidance summarized the worst case analysis rule as requiring agencies to consider the worst possible consequences of a spectrum of events from those with low probability and catastrophic effect to those with high probability and less drastic impact.

The Problems with the Old Rule

Beginning in 1983 and continuing through 1986, CEQ published a series of rulemaking notices on the worst case analysis requirement. In these notices, CEQ explained the problems with the worst case analysis requirement, proposed guidance on worst case analysis (which it later withdrew), solicited public responses to five specific questions on worst case analysis, and finally rescinded the worst case requirement as part of the amendments to the overall rule on incomplete or unavailable information.

On August 11, 1983, CEQ published for public comment a proposed guidance memorandum on worst case analysis.¹⁵ As a result of the comments it received on this proposal, CEQ withdrew it on February 8, 1984.¹⁶ In the 1983 proposed guidance, CEQ stated that additional guidance on the worst case rule was needed because the requirement had been "subject to a wide variety of conflicting interpretations by both federal agencies and reviewing courts." Specifically, CEQ was "concerned that the worst case analysis requirements [were] being read to require federal agencies to conduct such analyses of potential effects that may well be highly remote or unlikely."¹⁷

To resolve these issues, the proposed guidance would have directed agencies to perform worst case analyses for those environmental effects that cross "an initial threshold of probability." The proposed guidance defined this threshold of probability by reference to CEQ's regulatory definition of the term "effects." Under CEQ's definition of "effects," agencies need only consider the "reasonably foreseeable" effects of proposed federal actions in EISs.¹⁸ Thus, CEQ reasoned that the worst case analysis requirement applied to a reasonably foreseeable environmental effect of a proposed federal action when there was a lack of information about the "consequences of that effect."¹⁹

Instead of resolving the problems with the worst case analysis rule, however, the 1983 proposed guidance aggravated the controversy by improperly equating the principle of reasonable foreseeability with a threshold of probability. Of particular concern to commenters was the proposal's statement that "effects with an extremely low probability of occurrence … would not meet the threshold of 'reasonable foreseeability.'"²⁰ This statement would have affected the identification of all effects in EISs regardless of whether the worst case analysis requirement applied.

As CEQ later acknowledged, reasonably foreseeable environmental effects may include those with a low probability of occurrence. Indeed, courts have found impacts with a low probability of occurrence to require consideration under NEPA as reasonably foreseeable effects when such effects would be severe.²¹ Probability is certainly an important aspect of reasonable foreseeability, but the terms are not synonymous.

The 1983 proposed guidance unintentionally highlighted one of the major problems with worst case analysis in the NEPA context — its inconsistency with the NEPA common law "rule of reason." Courts interpreting NEPA have developed a rule of reason to judge the sufficiency of agencies' consideration of the effects of proposed actions and of the alternatives to these actions in EISs.²² The judicial decisions applying the rule of reason to determine whether an agency has adequately considered the effects of a proposed action in an EIS accord with CEQ's regulatory definition of the term "effects" in requiring agencies to consider only the reasonably foreseeable effects of their proposed actions.²³

It is difficult to reconcile worst case analysis with the NEPA rule of reason. Even assuming it is possible to identify the worst potential consequence of a proposed federal action, this consequence may or may not be within the range of reasonably foreseeable effects. For instance, the worst potential consequence of a proposed action may be based on a lengthy series of purely conjectural assumptions. In such a case, the worst potential consequence of the proposed action is possible, yet it is so hypothetical as to be outside of the range of reasonably foreseeable effects.

Though not explicitly acknowledging the conflict, the Ninth Circuit has construed the worst case analysis requirement to override both the rule of reason and the CEQ definition of "effects." In Southern Oregon Citizens Against Toxic Sprays, Inc. v. Clark, the court determined that an agency's conclusion that an effect was not reasonably foreseeable did not relieve the agency of the duty to perform a worst case analysis of the effect.²⁴ Moreover, in Save Our Ecosystems v. Clark, the Ninth Circuit required [19 ELR 10029] an agency's worst case analysis to consider a "spectrum of possible events" regardless of whether these events were reasonably foreseeable.²⁵

After recognizing the problems with the 1983 proposed guidance and withdrawing the guidance, CEQ published an advance notice of proposed rulemaking on December 31, 1984.²⁶ In the advance notice, CEQ stated it had received a private petition to amend the worst case rule as well as numerous requests from federal agencies to initiate a rulemaking. CEQ requested public responses to five questions related to worst case analysis. The questions focused on how the NEPA rules should address the problem of limited information about the environmental impacts of major federal actions.²⁷

On August 9, 1985, after reviewing the responses to the questions in the advance notice, CEQ announced its decision to amend the regulation on incomplete or unavailable information. CEQ stated that it had found the worst case analysis requirement to be an unsatisfactory approach to the analysis of potential consequences in the face of missing information.²⁸ CEQ summarized the numerous negative comments it had received about the worst case analysis requirement as including: (1) the limitless nature of the task of conjuring the worst possible case; (2) the lack of expert support for worst case analysis in the growing field of risk analysis; (3) the inability of federal agencies to perform useful worst case analyses; (4) the minimal value of fanciful worst case analyses to federal decisionmakers who must balance a full range of proven competing interests; (5) the counterproductive allocation of federal resources caused by agencies attempting to comply with the worst case analysis rule at the expense of other potentially useful environmental studies; and (6) the inconsistency of the worst case analysis rule with the NEPA common law rule of reason, the judicial standard courts have used to interpret NEPA's requirements in the context of uncertainty.²⁹

CEQ also cited the expansive interpretation given the worst case rule by the Ninth Circuit as a basis for its decision to amend the rule. CEQ's proposed rule discussed the appellate court's holding in Save Our Ecosystems v. Clark³⁰ as requiring agencies to develop purely conjectural worst case scenarios that lack a credible basis and that are not useful to decisionmakers. CEQ described the decision as requiring the Bureau of Land Management (BLM or the Bureau) to assume its use of herbicides would cause cancer in humans despite the Bureau's finding that there was no credible scientific data suggesting the herbicides caused cancer at any dose level. CEQ stated that it viewed such decisions as inconsistent with NEPA's rule of reason.³¹

The New Rule

On April 25, 1986, CEQ published its final amended rule on incomplete or unavailable information. The rule replaced the worst case analysis requirement with the requirement to evaluate effects by means of accepted scientific or theoretical methods.³² The revised rule retains all the requirements of the former rule except for worst case analysis.³³ Accordingly, the new rule requires agencies confronting missing information about a reasonably foreseeable effect to disclose the fact that they lack relevant information and to obtain the information if it is essential and if the overall costs of obtaining it are not exorbitant. If the agency cannot obtain the missing information because the overall costs are exorbitant or because the means to obtain it are unknown, the new rule requires agencies to explain the relevance of the missing information, summarize the existing credible scientific evidence, and evaluate the impacts by means of theoretical approaches or research methods generally accepted in the scientific community. The final rule also makes clear that reasonably foreseeable effects may include those with a low probability of occurrence.

In the preamble to the final rule, CEQ clearly repudiates the worst case analysis rule. CEQ concluded that the worst case requirement was an unproductive and ineffective means to implement NEPA in the face of incomplete or unavailable information. Further, CEQ stated that it clearly had the authority to revoke the worst case rule in a manner binding on federal agencies. CEQ explained that the worst case requirement was an innovation of CEQ. Since CEQ had developed the worst case requirement, it had authority to rescind the requirement through rulemaking.³⁴

The new requirement to evaluate effects about which [19 ELR 10030] there is only limited information according to accepted scientific or theoretical methods allows for, but does not require, the use of risk assessment techniques. CEQ explained that many commenters favored risk assessment as a substitute for worst case analysis. CEQ decided not to constrain agencies to a particular methodology for addressing uncertainty, but rather to allow agencies to employ a variety of commonly accepted methods, such as literature searches and peer reviews. The CEQ also declined to define "credible scientific evidence" on the grounds that it did not want to narrow the range of possible approaches to evaluating effects in the face of uncertainty.

In its final rule, CEQ states that early NEPA case law supports its decision to replace the worst case analysis requirement with the requirement to evaluate impacts by accepted scientific and theoretical methods. CEQ's preamble states that the NEPA rulings preceding its worst case regulation construed NEPA to require reasonable forecasting and reasonable speculation, but did not mandate application of worst case analysis or any other particular technique.³⁵

The Judicial Decisions on Worst Case Analysis

Though the Ninth Circuit was not the first appellate court to apply the worst case analysis requirement,³⁶ it has been the dominant force behind the worst case rule in two important respects. First, in tracing the worst case analysis requirement to early NEPA case law, the Ninth Circuit has raised the issue now before the Supreme Court — whether worst case analysis is a requirement of NEPA itself. Second, the court has enforced the worst case requirement to its full literal extent. The court has enjoined federal actions when agencies failed to include worst case analyses in EISs (and even in environmental assessments), when agencies prepared insufficient worst case analyses, and when agencies failed to perform the original research required by the worst case analysis rule.³⁷

In finding that early NEPA case law requires worst case analysis, the Ninth Circuit relied on the first appellate court decision interpreting the worst case requirement, the Fifth Circuit's decision in Sierra Club v. Sigler.³⁸ In this case, the Fifth Circuit required the Army Corps of Engineers to perform a worst case analysis for the construction of a deepwater port in Galveston, Texas. The Solicitor General stated, in his petitions for writs of certiorari in Methow Valley and Oregon Natural Resources, that the Ninth Circuit is mistaken in its reliance on Sigler. The Solicitor General notes the Sigler court found worst case analysis to be consistent with NEPA, but not required by it.³⁹

Interestingly, all of the judicial and administrative authorities that have taken a position on the origin of the worst case analysis requirement — the Fifth and Ninth Circuit courts, the Solicitor General, and CEQ — rely on the same early NEPA cases to support their respective views. In particular, these authorities all find support in the 1973 decision of the District of Columbia Court of Appeals in Scientists' Institute for Public Information v. Atomic Energy Commission (SIPI).⁴⁰ Though the specific holding in SIPI is not directly relevant to worst case analysis, the court's opinion discusses at length the reasonable forecasting and speculation implicity required by NEPA.

The Ninth Circuit Cases

The Ninth Circuit's 1983 decision in Southern Oregon Citizens Against Toxic Sprays, Inc. v. Clark (SOCATS), was the court's first application of the worst case analysis requirement.⁴¹ In SOCATS, the court held that the BLM had to perform a worst case analysis for its herbicide spraying program in Oregon forests in light of the uncertainty regarding the safe dosage levels of the herbicides. Rejecting the Bureau's argument that it need not posit and analyze a worst case unless it is a probable consequence of a proposed action, the court held that CEQ's regulation [19 ELR 10031] on incomplete or unavailable information required "separate consideration of (1) the worst case and (2) 'the probability or improbability of its occurrence.'"⁴²

Significantly, the SOCATS opinion explains that CEQ's "worst case analysis regulation codifies prior NEPA case law." The court does not include any additional discussion of this finding. The court merely cites the holding in Sierra Club v. Sigler as support for this statement without explaining how Sigler supports this conclusion.⁴³

It is important to note that the SOCATS decision required the BLM to include a worst case analysis in an environmental assessment, despite the clear placement of the worst case analysis requirement in CEQ's EIS rules.⁴⁴ The court limited its holding by describing the Bureau's environmental assessment as supplementing its earlier programmatic EIS on its spraying program. Yet the court also reasoned that applying the worst case requirement to the environmental assessment in this case furthered the purposes of NEPA. Such rationale could support extending the worst case requirement to environmental assessments regardless of whether they supplement preexisting EISs.

In its 1984 decision in Save Our Ecosystems v. Clark,⁴⁵ the Ninth Circuit found the worst case analysis performed by the BLM in response to SOCATS to be inadequate. The court also enjoined the Forest Service's similar spraying program on the grounds of an inadequate worst case analysis. Though the Environmental Protection Agency (EPA) had approved the herbicides being used by the BLM and the Forest Service in their spraying programs, no conclusive data existed regarding the safe dosage levels of the herbicides. Given the lack of a proven safe exposure level, the court required the agencies to assume in their worst case analyses that the herbicides cause cancer. The court held that the CEQ regulation first required agencies to analyze the worst case without regard to whether credible data exist to support the worst case assumption, and then required agencies to analyze the likelihood that the worst case will occur.

The court again interpreted the early NEPA case law as requiring agencies to perform worst case analyses when confronting uncertainty. Specifically, Save Our Ecosystems explains that the worst case rule is "a codification of prior case law that required analysis of the costs of proceeding without more and better information."⁴⁶ The decision cites SOCATS as support for this proposition yet offers no additional analysis.

Another important aspect of the Save Our Ecosystems ruling is its requirement for agencies to perform original research if no adequate data exist on the effects of its actions.⁴⁷ The court found the worst case rule's requirement that agencies obtain missing information if the overall costs are not exorbitant requires agencies to conduct independent research. The court rejected the Forest Service's argument that the court cannot force it to do research under NEPA. This holding is particularly interesting in this case where the agencies that the court was requiring to perform original research are not the agencies to which Congress has delegated authority or responsibility for analyzing the health effects of herbicides; this is the responsibility of EPA.

The two Ninth Circuit cases now before the Supreme Court for review continue the full enforcement of the worst case analysis requirement. In Methow Valley Citizens Council v. Regional Forester,⁴⁸ the court held that the Forest Service must perform a worst case analysis for its designation of a forest area for use as a major ski resort. Focusing on the effect of the ski resort on the migratory mule deer herd in the area, the court held that if the Forest Service lacked information about the effect on mule deer, it would have to conduct an analysis of the worst case — presumably the loss of the herd.

In Oregon Natural Resources Council v. Marsh,⁴⁹ the appellate court held that the Army Corps of Engineers must either perform a worst case analysis or conduct further research on the effect of its proposed construction of the Elk Creek Dam project on the turbidity of an environmentally valuable river into which the Elk Creek runs. According to the court, exposing uncertainty is not sufficient for NEPA compliance.

In Methow Valley and Oregon Natural Resources, the Ninth Circuit explicitly rejected the argument that CEQ's repeal of its worst case analysis rule rescinds the worst case analysis requirement. The court offers little basis for this conclusion in either opinion. In both cases, the court merely asserted in a footnote that CEQ's repeal of the regulation does not rescind the worst case analysis requirement, since the regulation was a codification of prior NEPA case law. The Methow Valley decision cites Oregon Natural Resources, SOCATS, and Save Our Ecosystems to support this conclusion.⁵⁰ The Oregon Natural Resources decision cites Save Our Ecosystems in support of the finding that early NEPA case law requires a worst case analysis.⁵¹

The Sigler Decision

The only outside authority the Ninth Circuit relied on in finding worst case analysis to be a requirement of NEPA common law is the Fifth Circuit's decision in Sierra Club v. Sigler. The court in Sigler required the Army Corps of Engineers to include in its EIS for a deepwater port in Galveston Bay a worst case analysis of the environmental damage that could result from a total cargo loss by a supertanker in the Bay. Though there was a very low probability of a total cargo loss, the court determined that the remoteness of the impact did not render it unreasonably speculative since all parties agreed it could occur.⁵²

Before applying the worst case rule to the proposed deepwater port, the court considered the Corps' argument that it did not have to perform a worst case analysis because the worst case requirement was beyond the "statutory [19 ELR 10032] minima" of NEPA.⁵³ The court upheld the worst case analysis requirement as consistent with the language, legislative history, and common law of NEPA. Accordingly, the court required the Corps to include in its EIS an analysis of a worst case oil spill from a supertanker using the proposed deepwater port.

It is difficult to see how the Sigler decision supports the Ninth Circuit cases finding worst case analysis to be a requirement of NEPA case law. In examining whether the worst case requirement was beyond the statutory minima of NEPA, the court in Sigler first examined the language of NEPA itself. Although the court found "some language in NEPA which may be said to endorse generally the concept of a worst case analysis," the court concluded that NEPA's "literal language does not require a worst case analysis."⁵⁴

The Sigler court then turned to NEPA's legislative history to see whether it supported CEQ's worst case analysis requirement. The court found that the legislative history illustrated congressional awareness of the limited information available about the environmental effects of federal actions. However, the court noted that Congress only required agencies to disclose fully the environmental consequences of their actions, "apparently leaving to the courts and the CEQ the determination of the extent of the mandate when information is missing."⁵⁵

Finding NEPA silent on the issue of uncertainty and finding little assistance in NEPA's legislative history, the court in Sigler analyzed the judicial decisions interpreting NEPA in the context of incomplete information. The court noted in particular the application of NEPA's rule of reason in the context of a lack of information about environmental impacts. Specifically, agencies need not decline to undertake any actions in the face of uncertainty, but must engage in reasonable forecasting and speculation in their assessments of the environmental effects of proposed actions.

After discussing NEPA's common law rule of reason, the court stated that "CEQ's worst case analysis regulation merely codifies these judicially created principles."⁵⁶ This statement is most likely the basis for the citation to Sigler in the Ninth Circuit SOCATS decision. However, this statement refers to the court cases that have held that NEPA requires federal decisionmakers to evaluate potential environmental consequences in the face of uncertainty. The Sigler opinion does not state that NEPA case law requires agencies to perform worst case analyses specifically. Rather, the court characterizes the worst case analysis technique as a CEQ "innovation" that has ample support in the case law.⁵⁷

The Early NEPA Case Law

In the decisions preceding CEQ's worst case rule, courts acknowledged both the difficulty of evaluating environmental effects in the face of uncertainty and the implicit requirement in NEPA for agencies to engage in reasonable forecasting and speculation about such effects. The Ninth Circuit court has concluded that these early NEPA cases require agencies to analyze the worst possible consequences of proposed actions. On the other hand, CEQ states that these cases support its revised rule on incomplete and unavailable information, which does not include a worst case analysis requirement.

The decision of the District of Columbia Court of Appeals in SIPI is most often cited for the proposition that reasonable forecasting and speculation are implicit in NEPA.⁵⁸ In SIPI, the court required the Atomic Energy Commission to prepare an EIS for its liquid metal fast breeder program notwithstanding the Commission's argument that the program was at the research and development stage. Though the case did not discuss worst case analysis, the court addressed compliance with NEPA in the face of uncertainty about the environmental effects of proposed actions and required agencies to engagein reasonable forecasting in their assessments of environmental impacts.

Although the Ninth Circuit cited SIPI in support of its holding that the worst case analysis rule codified early NEPA case law, the holding in SIPI actually supports CEQ's revised rule on incomplete or unavailable information. The court in SIPI noted that "NEPA requires predictions, but not prophecy, and impact statements ought not to be modeled upon the works of Jules Verne or H.G. Wells."⁵⁹ By definition, the worst case analysis rule enforced by the Ninth Circuit requires agencies to conjure up the worst imaginable consequence of a proposed action regardless of whether it is reasonable to consider such an effect.

The NEPA common law rule of reason also supports CEQ's revision of the regulation on incomplete or unavailable information. In applying the rule of reason to judge the sufficiency of EISs,⁶⁰ courts avoid requiring agencies to go to "unreasonable extremes" in assessing the environmental effects of their proposed actions.⁶¹ CEQ replaced the worst case analysis requirement with the requirement to study effects by means of accepted scientific and theoretical methods in order to bring the rule on incomplete or unavailable information back in line with the judicial standard of reason.

Conclusion: The Importance of the Supreme Court's Rulings

The Supreme Court's decisions in Methow Valley and in Oregon Natural Resources will largely direct the methods by which federal agencies will make decisions on major future projects about which there is some uncertainty. The Court's holdings will also affect CEQ's authority to guide federal compliance with NEPA.

[19 ELR 10033]

If the Court affirms the Ninth Circuit holdings on worst case analysis, then federal agencies will probably not utilize other techniques, such as risk analysis, to assess the effects of their proposed actions in the face of missing information. This is because they will devote their limited resources to developing defensible worst case analyses. Moreover, without a legislative amendment to NEPA, agencies will not be free to substitute other risk evaluation methods that may be developed in the future for the required worst case analysis.⁶²

Mandatory worst case analysis would most likely have the surprising effect of diminishing the importance of environmental considerations in federal decisions. Worst case scenarios will be pure conjecture in many cases. Federal decisionmakers could justifiably decide not to rely heavily on such guesswork in deciding whether to proceed with proposed actions.

With respect to CEQ's authority to implement NEPA, the forthcoming Supreme Court decisions will either affirm or severely restrict CEQ's power.⁶³ The Court has held that CEQ's regulations are entitled to substantial deference, even when the rules conflict with earlier CEQ guidelines.⁶⁴ If the Court determines that CEQ does not have the power to rescind the worst case analysis requirement, the Court would substantially reduce the weight given CEQ's regulations by federal agencies and by the courts. Moreover, if the Court were to hold that NEPA common law requires worst case analysis, the Court would be allowing courts to draft the implementing regulations to NEPA and to substitute their judgment for that of CEQ.

Though the Ninth Circuit traces the worst case requirement to early NEPA case law, the Solicitor General is correct in articulating the issue as whether NEPA itself requires worst case analysis. Since neither NEPA's language nor its legislative history reveals how Congress intended agencies to evaluate environmental effects in the face of uncertainty, it is far-fetched to suggest that NEPA directs agencies to conduct worst case analyses when assessing environmental effects about which little is known.

1. 42 U.S.C. § 4321-4370a, ELR Stat. NEPA 001-012.

2. 833 F.2d 810, 18 ELR 20163 (9th Cir. 1987), cert. granted sub nom. Robertson v. Methow Valley Citizens Council, 56 U.S.L.W. 3879 (U.S. June 27, 1988) (No. 87-1703). The appellate court found the Forest Service's EIS on the designation of forest property for use as a downhill ski area to be inadequate, in part because the agency did not perform a worst case analysis of the effect of the ski area on a migratory mule deer herd. The Forest Service had determined that mitigation measures would minimize the impact of the ski area on the mule deer herd. Yet, since the Forest Service had not fully developed its mitigation plan or assessed the effectiveness of the mitigation measures, the court held the agency had to perform a worst case analysis of the effects of the ski area on the herd.

3. 820 F.2d 1051 (9th Cir. 1987), as amended, 832 F.2d 1489, 18 ELR 20321 (9th Cir. 1987), cert. granted sub nom. Marsh v. Oregon Natural Resources Council, 56 U.S.L.W. 3879 (U.S. June 27, 1988) (No. 87-1704). The appellate court found the Army Corps of Engineers' EIS for construction of the Elk Creek Dam to be insufficient on the grounds that the Corps failed to prepare a worst case analysis of the effect of the dam on the turbidity of the river into which Elk Creek runs.

4. The Court has consolidated these two cases for review. These cases raise other issues not discussed in this Article. In both cases, the Court will decide whether NEPA requires federal agencies to include fully developed mitigation plans in EISs. Also in Methow Valley, the Court will decide whether the Forest Service adequately considered alternatives to the proposed action; and in Oregon Natural Resources the Court will decide whether the Corps of Engineers adequately considered new information acquired after completion of the original EIS.

5. See infra text accompanying notes 8-35.

6. See infra text accompanying notes 36-61.

7. See infra text accompanying notes 62-64.

8. Formerly codified at 40 C.F.R. § 1502.22.

9. The former CEQ regulation on incomplete or unavailable information provided:

When an agency is evaluating significant adverse effects on the human environment in an environmental impact statement and there are gaps in relevant information or scientific uncertainty, the agency shall always make clear that such information is lacking or that uncertainty exists.

(a) If the information relevant to adverse impacts is essential to a reasoned choice among alternatives and is not known and the overall costs of obtaining it are not exorbitant, the agency shall include the information in the environmental impact statement.

(b) If (1) the information relevant to adverse impacts is essential to a reasoned choice among alternatives and is not known and the overall costs of obtaining it are exorbitant or (2) the information is relevant to adverse impacts, is important to the decision and the means to obtain it are not known (e.g., the means for obtaining it are beyond the state of the art) the agency shall weigh the need for the action against the risk and severity of possible adverse impacts were the action to proceed in the face of uncertainty. If the agency proceeds, it shall include a worst case analysis and an indication of the probability or improbability of its occurrence.

Formerly codified at 40 C.F.R. § 1502.22.

10. 51 Fed. Reg. 15618 (Apr. 25, 1986).

11. 40 C.F.R. §§ 1501.1-.25 (1987).

12. However, the worst case analysis requirement has been applied to expand the range of environmental effects under consideration. See infra text accompanying notes 22-25.

13. See supra note 9.

14. Forty Most Asked Questions Concerning CEQ's National Environmental Policy Act Regulations, 46 Fed. Reg. 18026, 18032, ELR ADMIN. MATERIALS 35020 (Mar. 23, 1981).

15. 48 Fed. Reg. 36486 (Aug. 11, 1983).

16. 49 Fed. Reg. 4803 (Feb. 8, 1984).

17. 48 Fed. Reg. at 36487.

18. 40 C.F.R. § 1508.8(b) (1987).

19. 48 Fed. Reg. at 36487.

20. Id.

21. City of New York v. Department of Transportation, 715 F.2d 732, 746 n.14, 13 ELR 20823, 20829 n.14 (2d Cir. 1983) (an agency must estimate "both the consequences that might occur and the probability of their occurrence…. The fact that effects are only a possibility does not insulate the proposed action from consideration under NEPA, but it does accord an agency some latitude in determining whether the risk is sufficient to require preparation of an EIS"); Carolina Environmental Study Group, Inc. v. United States, 510 F.2d 796, 799, 5 ELR 20181, 20182 (D.C. Cir. 1976) ("[i]t is entirely proper, and necessary, to consider the probabilities as well as the consequences of certain occurrences in ascertaining their environmental impact. There is a point at which the probability of an occurrence may be so low as to render it almost totally unworthy of consideration…. Recognition of the minimal probability of an event is not equatable with nonrecognition of its consequences").

22. National Helium Corporation v. Morton, 486 F.2d 995, 1002, 4 ELR 20041, 20043 (10th Cir. 1973) (the rule of reason is the appropriate standard where the sufficiency of the EIS is being tested); Sierra Club v. Morton, 379 F. Supp. 1254, 4 ELR 20690 (D. Colo. 1974) (rule of reason is the standard to judge the sufficiency of the EIS).

23. See supra note 18. See also Swain v. Brinegar, 542 F.2d 364, 6 ELR 20609 (7th Cir. 1976) (EIS need not review all possible environmental effects of a project, but must review those that are reasonably foreseeable); Arkansas Power and Light Co. v. Federal Power Commission, 517 F.2d 1223 (D.C. Cir. 1975) (agency is required to make good faith effort to describe the reasonably foreseeable environmental impacts).

24. 720 F.2d 1475, 1478-79, 14 ELR 20061, 20062 (9th Cir. 1983). The court characterizes the agency's finding as being that the effect was not "probable or reasonably likely to occur" rather than not reasonably foreseeable. See infra text accompanying notes 41-44.

25. 747 F.2d 1240, 15 ELR 20035 (9th Cir. 1984). See infra text accompanying notes 45-47.

26. 49 Fed. Reg. 50744 (Dec. 31, 1984).

27. The five questions were:

1. Under what circumstances and to what extent must a federal agency engage in forecasting or speculation when confronted with scientific uncertainty or gaps in information concerning the environmental effects of a proposed action?

2. How can an analysis be structured to present reasonable forecasting in the face of scientific uncertainty or information gaps about the effects of proposed action to provide more useful and understandable information for decisionmakers and other interested parties?

3. Does the type of analysis called for in 40 C.F.R. § 1502.22 require federal agencies to go beyond the "rule of reason," as traditionally expressed in judicial decisions interpreting NEPA?

4. Should a threshold standard be established which would trigger the preparation of the type of analysis identified in response to question one, such as a threshold of severe consequences, a threshold of probability or a threshold of scientific credibility?

5. Is the term "worst case" appropriate for this type of analysis? If so, how should it be defined? If not, what is the most appropriate term for this type of analysis, and how should it be defined?

49 Fed. Reg. at 50744.

28. 50 Fed. Reg. 32234, 32236 (Aug. 9, 1985).

29. Id. at 32236-37.

30. 747 F.2d 1240, 15 ELR 20035 (9th Cir. 1984).

31. 50 Fed. Reg. at 32236.

32. 51 Fed. Reg. 15618 (Apr. 25, 1986).

33. Though the rule also states it "eliminates the requirement for agencies to 'weigh the need for the action against the risk and severity of possible adverse impacts were the action to proceed in the face of uncertainty' while in the process of preparing an EIS," the rule later notes this weighing of risks and benefits is "properly done after completion of the entire NEPA process in the decision on whether to proceed with the project." Id. at 15621.

34. Id. at 15625. The United States Senate recently passed a bill that would specifically authorize CEQ to issue regulations under NEPA. S. 1792, 100th Cong., 2d Sess., 134 CONG. REC. S16346 (daily ed. Oct. 14, 1988). Currently, CEQ is authorized to issue regulations under Exec. Order 11514, § 2(g), 3 C.F.R. § 902 (1966-70), as amended by Exec. Order 11991, 3 C.F.R. § 123 (1987), ELR ADMIN. MATERIALS 45001, which also requires federal agencies to comply with these CEQ regulations.

35. The preamble states:

The regulation also requires that analysis of impacts in the face of unavailable information be grounded in the "rule of reason." The "rule of reason" is basically a judicial device to ensure that common sense and reason are not lost in the rubric of regulation. The rule of reason has been cited in numerous NEPA cases for the proposition that, "An EIS need not discuss remote and highly speculative consequences…. This is consistent with the (CEQ) Council on Environmental Quality Guidelines and the frequently expressed view that adequacy of the content of the EIS should be determined through use of a rule of reason."

51 Fed. Reg. at 15621, ELR ADMIN. MATERIALS 35040 (citing Trout Unlimited v. Morton, 509 F.2d 1276, 1283, 5 ELR 20151, 20154 (9th Cir. 1974)).

36. The first appellate court to interpret the rule was the Fifth Circuit in Sierra Club v. Sigler, 695 F.2d 957, 13 ELR 20210 (5th Cir. 1983); see infra text accompanying notes 52-57.

37. See infra text accompanying notes 41-51.

38. 695 F.2d 957, 13 ELR 20210.

39. Petition of the Solicitor General of the United States for a Writ of Certiorari to the United States Court of Appeals for the Ninth Circuit at 15, Marsh v. Oregon Natural Resources Council (filed Apr. 14, 1988)(No. 87-1704); Petition of the Solicitor General of the United States for a Writ of Certiorari to the United States Court of Appeals for the Ninth Circuit at 19, Robertson v. Methow Valley Citizens Council (filed Apr. 14, 1988) (No. 87-1703).

40. 481 F.2d 1079, 3 ELR 20525 (D.C. Cir. 1973). CEQ's final rule refers to this decision as "the seminal case which applied the rule of reason to the problem of unavailable information." 51 Fed. Reg. at 15621.

41. 720 F.2d 1475, 14 ELR 20061 (9th Cir. 1983). Other Ninth Circuit cases involving worst case analysis issues include: Northwest Coalition for Alternatives to Pesticides (NCAP) v. Lyng, 844 F.2d 588, 18 ELR 20738 (9th Cir. 1988) (upholding BLM worst case analysis for weed elimination program under previous CEQ rule); Animal Defense Council v. Hodel, 840 F.2d 1432, 18 ELR 20497 (9th Cir. 1988) (Bureau of Reclamation need not perform worst case analysis of health effect of chlorination of Central Arizona Project water); Oregon Environmental Council v. Kunzman, 817 F.2d 484, 17 ELR 20756 (9th Cir. 1987) (worst case analysis for gypsy moth spraying program satisfied readability standard); National Wildlife Federation v. United States Forest Service, 801 F.2d 360, 17 ELR 20344 (9th Cir. 1986), vacating in part, 592 F. Supp. 931, 14 ELR 20755 (D. Or. 1984) (vacating requirement to perform worst case analysis for each timber sale as part of larger settlement for entire timber sales program); San Luis Obispo Mothers for Peace v. Nuclear Regulatory Commission, 799 F.2d 1268, 17 ELR 20011 (9th Cir. 1986) (overruling NRC operating licenses on alternate grounds but refering to the lack of a worst case analysis as a deficiency in the NEPA document that should be cured on remand); Friends of Endangered Species v. Jantzen, 760 F.2d 976, 15 ELR 20455 (9th Cir. 1985) (upholding Fish and Wildlife Service determination that no worst case analysis was required for permit for taking of endangered species); Village of False Pass v. Clark, 733 F.2d 605, 14 ELR 20398 (9th Cir. 1984) (worst case analysis must be performed at the last decision point that would allow the agency to avoid the worst case); Save Lake Washington v. Frank, 641 F.2d 1330, 11 ELR 20645 (9th Cir. 1981) (upholding agency's worst case analysis performed for construction of docking facilities); and Friends of the Earth v. Hall, 693 F. Supp. 904 (W.D. Wash. 1988) (district court notes worst case analysis requirement is in force in the Ninth Circuit notwithstanding CEQ's regulatory amendment).

42. 720 F.2d at 1479, 14 ELR at 20062 (quoting Sierra Club v. Sigler, 695 F.2d 957, 13 ELR 20210 (5th Cir. 1983)).

43. Id. at 1478, 14 ELR at 20061.

44. Id. at 1480, 14 ELR at 20062-63.

45. 747 F.2d 1240, 15 ELR at 20035 (9th Cir. 1984).

46. Id. at 1244, 15 ELR at 20037.

47. Id. at 1247, 15 ELR at 20039.

48. 833 F.2d 810, 18 ELR 20163; see supra note 2.

49. 832 F.2d 1489, 18 ELR 20321; see supra note 3.

50. 833 F.2d at 817 n.11, 18 ELR at 20166 n.11.

51. 832 F.2d at 1497 n.8, 18 ELR at 20324 n.8.

52. 695 F.2d at 974, 13 ELR at 20217. See also McChesney, CEQ's "Worst Case Analysis" Rule for EISs: "Reasonable" Speculation or Crystal Ball Inquiry?, 13 ELR 10069 (Mar. 1983).

53. 695 F.2d at 969, 13 ELR at 20215.

54. Id.

55. Id. at 970 n.9, 13 ELR at 20215 n.9.

56. Id. at 971, 13 ELR at 20216.

57. Id. 972, 13 ELR at 20217. Indeed, the court described the worst case rule as precisely the type of regulatory innovation due substantial judicial deference pursuant to the Supreme Court's holding in Andrus v. Sierra Club, 442 U.S. 347, 9 ELR 20390 (1979) (courts and agencies must accord CEQ's regulations substantial deference).

58. 481 F.2d at 1092, 3 ELR at 20531.

59. Id. at 1093, 3 ELR at 20532.

60. See supra note 22.

61. National Resources Defense Council, Inc. v. Morton, 458 F.2d 827, 838, 2 ELR 20029, 20034 (D.C. Cir. 1972) (so long as agencies take a hard look at environmental effects, a court will not impose unreasonable extremes).

62. The federal government is continuing to improve regulatory application of risk assessment techniques. Risk assessment is the use of the factual base to define the health effects of exposure of individuals or populations to hazardous situations. Risk management is the process of weighing policy alternatives and selecting the most appropriate regulatory action, integrating the results of risk assessment with engineering data and with social, economic, and political concerns to reach a decision. See NATIONAL RESEARCH COUNCIL, RISK ASSESSMENT IN THE FEDERAL GOVERNMENT: MANAGING THE PROCESS 3 (1983).

63. The Court's decision may affect the passage of the Senate bill discussed supra at note 34. Should the Court hold that the worst case analysis requirement survives CEQ's revocation of the requirement, then passage of the Senate bill that would give CEQ legislative authorization to promulgate NEPA regulations would be of great importance for this and future NEPA controversies.

64. Andrus v. Sierra Club, 442 U.S. 347, 358, 9 ELR 20390, 20393 (1978). See also the discussion of CEQ regulations in the context of legislative versus interpretive rules in Rosenbaum, Update: The NEPA Worst Case Analysis Regulation, 14 ELR 10267, 10268 n.13 (July 1984).