Amending CEQ's Worst Case Analysis Rule: Towards Better Decisionmaking?

Kenneth L. Rosenbaum

National Environmental Policy Act (NEPA) litigation is often a tactical exercise. NEPA requires decisionmaking procedures that ensure the full and timely consideration of the environmental impacts of major federal actions. Though NEPA suits are usually waged over procedural points, preserving the integrity of NEPA's procedures is not necessarily the plaintiffs' first goal. Plaintiffs may bring NEPA suits as an indirect means to a more substantive end — to force the agency to abandon what they judge to be a poor proposed action. Such plaintiffs do not expect the court to order the agency to take a better action; they only hope to delay and win reexamination of the bad one.

The worst case analysis regulation,¹ which addresses the procedures for dealing with uncertainties about environmental impacts, has become a dramatic focus of a handful of such tactical actions. In particular, in the cases dealing with pesticide use on public lands in the Pacific Northwest, worst case litigation has been a stunningly effective tool for thwarting Forest Service and Bureau of Land Management (BLM) spraying plans.² Enforcement of the worst case analysis requirement has been an unpleasant surprise to these agencies; they and others fear the same kinds of surprises may lurk in other contexts.

In response, they have launched a tactical assault of their own. They have lobbied the Council on Environmental Quality (CEQ) to amend the regulation. At this writing, with the comment period drawing to a close on a proposed rule that eliminates the requirement,³ it seems their move will succeed. It remains to be debated, however, whether their actions serve the broadly held end of improving environmental decisionmaking, and whether in the end the new rule will have really improved their tactical position.

Towards Better Decisionmaking?

The purpose of the procedural requirements of NEPA is to promote informed, public decisionmaking on matters that affect the quality of the human environment.⁴ To that end, NEPA requires decisionmakers to prepare an environmental impact statement (EIS) on proposals that significantly affect the environment.⁵ An EIS must lay out the impacts of the action and consider alternatives. Almost always, the impacts of an action cannot be predicted exactly. Courts have long recognized that NEPA inquiries cannot be limited to what is known with certainty.⁶ The worst case regulation lays out how agencies must deal with uncertainty. The detailed requirements of the rule have been examined before in these pages.⁷ Very briefly, if an agency faces an uncertainty that is important to its decision, it must try to eliminate the uncertainty. If the uncertainty cannot be eliminated, the agency must explore its outer bounds by examining a "worst case" scenario.

The recent CEQ proposal would recast the obligation to explore uncertain impacts, limiting the inquiry to impacts suggested by credible scientific evidence. The proposed rule delineates that obligation as follows:

In preparing an environmental impact statement, the agency shall make reasonable efforts, in light of overall costs and state of the art, to obtain missing information which, in its judgment, is important to evaluating significant adverse impacts on the human environment that are reasonably foreseeable. If for the reasons stated above, the agency is unable to obtain this missing information, the agency shall include within the environmental impact statement (1) a statement that such information is missing, (2) a statement of the relevance of the missing information to evaluating significant adverse impacts on the human environment, (3) a summary of existing credible scientific evidence which is relevant to evaluating the significant adverse [15 ELR 10276] impacts on the human environment, and (4) the agency's evaluation of such evidence. "Reasonably foreseeable" includes impacts which have catastrophic consequences, even if their probability of occurrence is low, provided that they have credible scientific support, are not based on pure conjecture, and are within the rule of reason.⁸

The proposed rule leaves part of the current scheme intact, but makes some key changes. An agency will still have to disclose important data gaps and uncertainties and will still have to try to fill them. If it cannot, though, it need not enter into the speculation of a worst case analysis. It need only describe the relevance of missing information and evaluate credible scientific evidence concerning the impacts. The proposal would seem to grant the agency considerable power in deciding what impacts are important enough to consider at all.According to CEQ, "the Council has chosen to impose scientific credibility as the 'thresh-old' to trigger the requirements of the proposed regulation."⁹ This explanation suggests that an agency might safely ignore areas of uncertainty where, in its expert opinion, it judged there to be no credible evidence of significant effects.

The critics of the existing worst case rule see these changes as necessary to promote efficient, informed decisionmaking. Read literally, they claim, the directive to examine the "worst case" condemns agencies to pursue an endless series of "what ifs."¹⁰ And in the process, it forces agencies to pass out of the realm of scientific decisionmaking and consider possibilities that have no scientific basis.¹¹ They point as an example to Save Our Ecosystems v. Clark,¹² where the Nine Circuit ruled that the BLM must consider that any level of herbicide exposure, no matter how slight, may cause cancer even though the agency found no credible data suggesting that very low levels are dangerous. Under the proposed rule, the agency's duty to speculate would be delimited so that the agency could avoid such inquiries beyond the realm of accepted science.

The critics' arguments have two faults, and these raise questions about CEQ's decision to amend the rule and its resulting proposal. First, no court has read the existing regulation literally to require identification of the single worst case. It is possible to extract some bald, unqualified dicta from the cases that suggest the agencies have to blindly seek "the worst."¹³ However, these dicta must be read in conjunction with the courts' repeated calls to apply reason to define the scope of an agency's worst case obligation.¹⁴ The cases have not required the agencies to explore possibilities that the courts believed would have no bearing on the decision at hand.

Second, forcing the agency to look beyond the current science may be quite rational, particularly in situations where science tells us little one way or another about possible impacts. Suppose you are going on a trip to Chicago for a few days and have no way to find out the weather forecast for the city. Do you leave your umbrella at home because you have no credible scientific evidence that it will be raining there? Or do you hypothesize a rational worst case for Chicago in September and at least consider bringing your umbrella, "just in case." Suppose that we know that some forms of radiation have significant health effects, but that no one has really explored the health effects of the particular form of radiation that a proposed power line will emit, and to which a large number of people will incidentally be exposed. Suppose that to search for impacts would take years and millions of dollars. Can the agency simply ignore the possibility of harm? Or might the decisionmaker find it prudent to hypothesize a qualitative worst case and explore building health monitoring or other safeguards into the action, just in case.¹⁵

Unquestionably, there are times when the evidence is so thin that speculation is not worthwhile. But just as surely, there are times when lack of credible scientific evidence¹⁶ is no excuse not to explore a possibility. Indeed, in the above hypotheticals the credible scientific evidence test proposed by CEQ does not seem to go as far a reasonably prudent [15 ELR 10277] decisionmaker would go to serve the end of informed decisionmaking.

Tactical Improvement for the Agencies?

One of the reasons CEQ gave for proposing to amend the worst case rule is that "in the institutional context of litigation over EIS(s) the 'worst case' rule has proved counterproductive, . . . ."¹⁷ The proposed regulation seems designed to reduce the leverage of NEPA litigators in two ways. First, as discussed above, it sets a higher threshold for exploring uncertainty, thus increasing the likelihood that an agency will not need to undertake any sort of analysis or that any analysis done will be thorough enough to pass review. Second, the regulation subtly changes the standard or review that courts may invoke. Under the present rule, the question of how far an agency must explore uncertainty seems to be a mixed question of law and fact involving scientific evidence, common sense, and evaluation of what is "reasonable speculation." The Ninth Circuit, in Save Our Ecosystems, examined scientific evidence outside that administrative record and gave little deference to the BLM's conclusions on what case was worst.¹⁸ The proposed regulation, on the other hand, makes it clear that the choice of what uncertainty to weigh in the EIS depends on the agency's judgment of the scientific credibilityof the evidence concerning the uncertainty. This is the sort of expert fact finding that courts traditionally review under highly deferential standards.

So, will the proposed regulation reduce the tactical power plaintiffs have under the present rule? In the long run, probably not. First, the present rule is becoming "litigated out." A new law holds only so many surprises; after a few novel court rulings, the bounds of the law emerge and compliance becomes easier, often routine. Compliance with the worst case rule is approaching this state.¹⁹ True, after 15 years, the potential for interpretive litigation over NEPA as a whole has not been exhausted; new issues emerge from time to time. But many aspects of the statute are well settled, and worst case analysis has been heading in that direction.²⁰

Amending the regulation as proposed will open the door to a number of difficult issues in any future litigation on the rule. Courts will not only have issues concerning the meaning of the new rule to resolve, in the first case or two they might face complex issues about CEQ's rulemaking power. No court has squarely addressed whether CEQ draws its rulemaking power from NEPA²¹ or solely from Executive Orders;²² whether CEQ regulations are legislative, interpretive, or procedural;²³ whether the regulations represent the bounds of agency NEPA obligations or just the minima; and whether the worst case regulation is merely a codification of prior case law.²⁴ All of these questions could have some bearing on the larger question of whether CEQ may, by rulemaking, "overrule" judicial acceptance of the worst case analysis requirement.

Assuming these issues were to be settled in favor of CEQ, the courts will still face issues concerning the validity of the rule as promulgated. A plaintiff could argue, for example, that the rule conflicts with NEPA by purporting to allow a less rigorous exploration of uncertainty than the statute demands. Though the way the courts will ultimately resolve this issue can be debated, the fact remains the issue is there to be raised, perhaps repeatedly as the regulation is applied to different decisionmaking situations.

Likewise, if the regulation is amended, we may see litigation over the fine points of CEQ's rulemaking. To judge the merits of such claims before the rulemaking is complete would be premature. Nonetheless, by way of example, if CEQ does not elaborate the reasons for amending the rule presented in its proposal, a plaintiff might claim that CEQ has been arbitrary and capricious in its decision to amend the rule at all.²⁵

Finally, we will see challenges to the meaning of the rule itself and its application in particular situations. It is easy to envision, for example, litigation over the meaning of "credible scientific evidence."²⁶

[15 ELR 10278]

Conclusion

CEQ is in an unenviable spot. It has been under tremendous pressure from both proponents of the current regulation and proponents of change. It has chosen, tentatively, to opt for change. It has tried to give the agencies the bright line that they have demanded, while still trying to hold them to as thorough a standard of review as NEPA demands. Despite the Council's considerable efforts, the proposed regulation may not be the best course.

The Council's bright line, "credible scientific evidence," is not stringent enough to ensure rational decisionmaking in all cases of uncertain impacts. And the line itself may not prove to be so bright; the amendment likely will simply prolong the frustrating litigation over uncertainty.

It would be a difficult choice for CEQ to make, but perhaps its best option would be to withdraw the proposed rule. In its place CEQ could issue guidance on the existing rule.Thoughtful guidance could discourage litigation by clearing up some lingering questions about the rule. It could re-acknowledge that the requirement to explore that worst case is a figurative one and that it no more condemns the agency to explore every contingency than NEPA's requirement to consider alternatives condemns the agency to describe every alternative. It could explain that good decisionmaking sometimes requires the decisionmaker to grapple with problems that science can tell us next to nothing about, beyond alerting us that they might occur. Equally important, it could explain that when evidence and common sense suggest that a contingency is so remote or so minor as to be irrelevant to the decision, it need not be explored. Admittedly, a guidance could not work the proposed rule's subtle changes in the posture of review of decisions on uncertainty. However, as agencies learned the worst case "ropes," just as they learned the ropes of other NEPA requirements in the early seventies, the requirement would become less a tactical threat and more a means to informed decisionmaking.

1. 40 C.F.R. § 1502.22, ELR REG. 46023.

2. See, e.g., Save Our Ecosystems v. Clark, 747 F.2d 1240, 15 ELR 20035 (9th Cir. 1984); Southern Oregon Citizens Against Toxic Sprays, Inc. v. Clark, 720 F.2d 1475, 14 ELR 20061 (9th Cir. 1983), cert. denied, 105 S. Ct. 446 (1984).

3. The proposal is published at 50 Fed. Reg. 32234 (Aug. 9, 1985). The comment period will close on Sept. 23, 1985. CEQ began to consider revising or reinterpreting the worst case requirement over two years ago. See 48 Fed. Reg. 36486 (1983). The guidance initially proposed would have allowed agencies to ignore very low probability events, even if they might have great impacts. That proposal drew strong criticism, see, e.g., Yost, Don't Gut Worst Case Analysis, 13 ELR 10394 (1983), and was withdrawn, 49 Fed. Reg. 4803 (1984).

4. See 40 C.F.R. § 1500.1, ELR REG. 46015.

5. NEPA § 102(2)(C), 42 U.S.C. § 4332(2)(C), ELR STAT. 41010.

6. See, e.g., Scientists' Institute for Public Information, Inc. v. Atomic Energy Commission, 481 F.2d 1079, 1092, 3 ELR 20525, 20532 (D.C. Cir. 1973) (noting that NEPA requires agencies to engage in reasonable forecasting and speculation and to identify areas where effects are essentially unknown).

7. See Comment, Update: The NEPA Worst Case Analysis Regulation, 14 ELR 10267 (1984); Comment, CEQ's "Worst Case Analysis" Rule for EISs: "Reasonable" Speculation or Crystal Ball Inquiry?, 13 ELR 10069 (1983).

8. 50 Fed. Reg. at 32238.

9. Id. at 32237.

10. See, e.g., Ketcham, Worst Case Analysis: The Forest Service Position in PROCEEDINGS OF A SYMPOSIUM ON WORST CASE ANALYSIS 45, 46 (L. Garrett & W. Covington, N. Ariz. U., eds. 1985) ("Some courts have interpreted 'worst' to mean the worst imaginable situation. If I should try to describe the worst possible effect that I can think of, you could think of one that was worse still, and vice versa. It could provide a never-ending sequence which would not be helpful to decisionmaking."); 50 Fed Reg. 32234, 32236 (Aug. 9, 1985) (One of CEQ's reasons for amending the rule: "Many respondents to the Council's Advance Notice of Proposed Rulemaking pointed to the limitless nature of the inquiry established by this requirement; that is, one can always conjure up a worse 'worst case' by adding an additional variable to a hypothetical scenario.")

11. "The 'worst case analysis' requirement has been interpreted to require agencies to present a discussion of particular disastrous impact even when the agency believes that no credible scientific data has indicated that the particular impact could be caused by the proposed action. . . . The Council believes that pure conjecture, that is, a conjectural analysis, lacking a credible scientific basis is not useful to either the decisionmaker or the public. . . ." CEQ's preamble to the proposed amendment, 50 Fed. Reg. at 32236. (emphasis in original).

12. 747 F.2d 1240, 15 ELR 20035 (9th Cir. 1984).

13. E.g., "Plainly, the worst result that can occur as a result of proceeding in the face of uncertainty as to whether a herbicide causes cancer is that it does cause cancer." Save Our Ecosystems, 747 F.2d at 1246, 15 ELR at 20038 (quoting the trial court's opinion, emphasis in original).

14. E.g., "The agency must consider the range of worst possible effects. . . ." Save Our Ecosystems, 747 F.2d at 1244, 15 ELR at 20037. "The analysis is formulated on the basis of available information, using reasonable projections of the worst possible consequences. . . ." Id., quoting Forty Most Asked Questions Concerning CEQ's NEPA Regulations, 46 Fed. Reg. 18026, 18032 (Mar. 3, 1981). "The [agency] must at least consider information relevant to a 'significant' effect of a proposal, if that information is both 'important' to that decision and not based on 'unreasonable speculation.'" Save our Ecosystems, 747 F.2d at 1245, 15 ELR at 20038, quoting Sierra Club v. Sigler, 695 F.2d 957, 974, 13 ELR 20210, 20218 (5th Cir. 1983).

15. We sometimes need to remind ourselves that science is not as well-developed or powerful as we like to imagine. There are few circumstances where the decisionmaker need not resort to educated guesses to evaluate the available options. And these guesses always entail uncertainty. The problems of uncertainty grow as one passes from consideration of physical impacts, to health impacts, to ecological impacts. We often lack both the theory and the data to predict the latter.

16. As that term is used here. It would be possible to interpret the term very broadly, to cover, for example, my common knowledge, non-expert opinion that it rains sometimes in Chicago, or a biophysicist's common sense, expert, but thinly supported opinion that a particular form of radiation may cause harm. Whether the term encompasses such educated guesses will be a matter for the courts to decide. See infra, note 26. The more broadly the courts interpret the term, the closer the proposed regulation would resemble the present one.

17. 50 Fed. Reg. at 32236.

18. "The BLM argues that the analysis plaintiffs say in necessary would be pure guesswork because no credible data exist to support the proposition that cancer can occur at any dose. This contention is specious in light of the evidence presented by plaintiffs' experts and the holding of [Southern Oregon Citizens Against Toxic Sprays, Inc. v. Clark] that '[t]he agency may not omit the analysis only because it believes that the worst case is unlikely.' "Save Our Ecosystems, 747 F.2d at 1246, 15 ELR at 20038 (citation omitted).

19. Or so I have argued elsewhere, see Comment, Update: The NEPA Worst Case Analysis Regulation, 14 ELR 10267 (1984).

20. At least one industry lawyer involved in worst case litigation has argued that the best course for CEQ to take with the regulation is to let it become settled law. "To the extent that it is ignored or resisted by the agencies or amended by CEQ in response to pleas from agencies or industry, the [regulation] will generate litigation and uncertainty — both of which are anathema to industry." Bach, Worst Case Analysis: A Historical and Legal Perspective in PROCEENDINGS OF A SYMPOSIUM ON WORST CASE ANALYSIS 19, 28 n.24 (L. Garrett & W. Covington, N. Ariz. U., eds. 1985).

21. The Fifth Circuit suggested that it does in Sierra Club v. Sigler, 695 F.2d 957, 967, 13 ELR 20210, 20214 (5th Cir. 1983).

22. See Exec. Order 11514, 35 Fed. Reg. 4247 (Mar. 5, 1970), as amended by Exec. Order 11991, 42 Fed. Reg. 26967 (May 25, 1977), ELR REG. 45003.

23. See Comment, Update: The NEPA Worst Case Analysis Regulation, 14 ELR 10267, 10268 n.13 (1984).

24. Some courts have so suggested. See, e.g., Save Our Ecosystems, 747 F.2d at 1244, 15 ELR at 20037.

25. CEQ has presented five reasons for amending the rule: (1) "the limitless nature of the inquiry established by this requirement; that is, one can always conjure up a worse 'worst case' . . .," 50 Fed. Reg. at 32236; (2) that worst case analysis stands without link to established disciplines of coping with uncertainty, id.; (3) the problem of counterproductive litigation, id. at 32236; (4) that "analysis . . . lacking a credible scientific basis is not useful to either the decisionmaker or the public." id. (emphasis in original); and (5) that the requirement is inconsistent with the "rule of reason" guiding most judicial construction of NEPA, id. at 32236-37.

26. If CEQ adopts the proposed rule, it would do well to elaborate on the meaning of "credible scientific evidence." Does it mean only the raw data that a scientist would consider evidence? Apparently not, for in discussing the proposal, CEQ mentions "opinions" and "views" that the agency would have to disclose and evaluate. 50 Fed. Reg. at 32237. Does it mean the sort of expert testimony that a court might admit? Does it go further? Does it extend to comments submitted by scientists in response to the agency proposal, or is it limited to material in the scientific literature? Is "credible" a term with similarly legal meaning? If so, how can an agency find a view that conflicts with its own "credible"? And what standards may an agency use to determine whether evidence is credible? When is failure of science to generate direct evidence of impacts credible evidence that no impacts exist?