Permitting Reform Bill: NEPA Remedies

Efforts to tailor remedies for NEPA violations address an important issue, but may have misfired

To me the most important change in environmental law from the permitting reform bill is the effort to restrict remedies for NEPA violations.  I’ve written before about why I think this is important work to do, and why it is important to do it well.  And there are good components to the proposal on the table (not just because it aligns some with an earlier suggestion of mine).

But after spending some time considering the provision, I’m increasingly concerned it is poorly drafted, and will not function well.  I also think it will have much less reach than both proponents and opponents might think.  Let me explain my reasoning.

First, the key language in the bill is here:

‘‘(5) REMEDY.—

‘‘(A) IN GENERAL.—Except as provided in subparagraph (B) and notwithstanding any other provision of law, if a court determines that an authorization does not comply with the requirements of this Act, the only remedy the court may order to redress that violation is to remand, without vacatur or issuance of an injunction, the applicable authorization to the agency. . .

‘‘(B) PRELIMINARY INJUNCTIONS.—

‘‘(i) IN GENERAL.—Notwithstanding subparagraph (A) and subject to clause (ii), a court may issue a preliminary in junction for the applicable authorization when—

‘(I) there is a substantial likelihood the authorization violates a Federal law other than—

‘‘(aa) this Act; or

‘‘(bb) division A of subtitle III of title 54, United States Code;

‘‘(II) necessary to prevent irreparable harm to the plaintiff

‘‘(III) the balance of equities favors the plaintiff; and

‘‘(IV) a preliminary injunction is in the public interest

The primary remedy here is, if there is a NEPA violation, the court does not stop the relevant agency action at issue, but just orders the agency to fix the NEPA violations.  That is a sharp change from the default principles under the APA, in which the standard remedy for a legal violation by an agency is to vacate the underlying decision.  I wrote before about why there are good arguments to flip that default.

But (as I recommended before) there is a carveout.  If the standards for a preliminary injunction are met, a court can stop the agency action.  That carveout is important – without it, agencies would arguably have no incentive to comply with NEPA.

Most of the requirements for a preliminary injunction are standard (requirements (II) through (IV).  But (I) is a new one.  It requires the court to find that there is a “substantial likelihood” that there is a violation of a Federal law in the proposed agency action, other than a violation of NEPA or the National Historic Preservation Act.

The first thing to be noted about this requirement is that there appears to be no requirement that the legal violation that could support the preliminary injunction has to be connected to the NEPA violation at issue in the case.  The legal violation could be any violation at all – no matter how trivial – of some federal law somewhere.  Moreover, the language does not appear to require that the plaintiff has to include the violation of that other federal law as a separate claim in the complaint, or that the violation has to be justiciable or judicially enforceable.  Presumably it just has to be noted somewhere in the complaint (perhaps in the request for remedies in the complaint).  Could a plaintiff point to a random violation of federal contracting law to support the preliminary injunction?  It appears so.

But I think more substantial (and more defensible) is another law that can be the basis for a preliminary injunction under this provision.  The Administrative Procedure Act.  Note that this would be the statute that the plaintiffs would be suing under in challenging an agency decision under NEPA, and the NEPA violation would also be a violation of the APA.  Does that mean that the preliminary injunction would always issue for any NEPA violation because it would also be a violation of the APA?

My inclination is that a court won’t go that far (even though the text of the Act might lead you there) – in part because there are debates about whether you can have “independent” violations of the APA separate from another statute.  If you cannot have such “independent” valuations, then really all you have is a NEPA violation plus the APA.  Moreover, if the only other statute besides the APA being violated is NEPA, that would seem to make the restrictions on preliminary injunctive relief superfluous, since you generally have to use the APA to sue under the NEPA.

That being said, I do think the APA is a key step here, in combination with other statutes besides NEPA.  Plaintiffs can allege that an agency decision is “arbitrary and capricious” under the APA (which would be a violation of the APA).  The agency’s failure to properly analyze a core issue relevant for its decision under another statute – such as through an unexamined or improperly examined NEPA analysis of that core issue – might be the basis of a court finding that the agency “entirely failed to consider an important aspect of the problem, [or] offered an explanation for its decision that runs counter to the evidence before the agency,” the standards set out by the Supreme Court in the State Farm case.  And for statutes with broad requirements to consider the environment or the public interest in agency decisionmaking, such as the National Forest Management Act or the Federal Lands Policy and Management Act, the NEPA analysis will often be quite relevant.  And remember, a plaintiff need only show a “substantial likelihood” of a violation of federal law – I’m not sure that’s even 50%.

I think this standard does mean that courts that wish to do so will feel empowered to brush off ticky-tack NEPA violations that seem tangential to the core questions before the agency, and deny preliminary injunctions.  But I do think that a court can look at the inadequacy of the NEPA analysis itself to determine whether that analysis could and should have been relevant to the substance of the agency decision, and if that inadequacy is severe enough, conclude that makes the agency’s substantive decision inconsistent with another federal law, justifying a preliminary injunction.  (I also think, as I said before, that most courts that find a NEPA violation will find a way to a remedy, and most courts that would not find a remedy here just won’t find a NEPA violation in the first place.)  If that is the case, then this approach does seem to at least partially address the concern that small NEPA violations were being used to block large projects, even where those violations were not relatively important considering the environmental stakes at issue.

As an aside, I think this is a clunky – and less effective – way to address the problems I had identified earlier.  As I noted then, one approach would be to conclude there should be injunctive relief if the inadequacy of the NEPA analysis was relevant to whether there was a potential violation of another environmental law (including state or local law).  That standard would actually be narrower than the current bill text (it requires relevance).  It also would be clearer that the key question is whether the inadequate NEPA analysis implicated a possible violation of law,  that greater analysis would determine whether such a violation existed and what to do about it, and that given the possible violation of some other environmental law, we have good evidence the inadequate NEPA analysis matters.  I also think that it would be better to allow courts to be able to rely solely on irreparable harm that is significant even if there is not some other violation of federal law – in other words, violations of law would only be one factor that would weigh in favor of injunctive relief.

There is another problem with the proposal, one that is more practical.  Another provision of the bill would require all NEPA lawsuits to be filed in the first instance in the court of appeals.  I appreciate that as an effort to reduce litigation timeframes.  And if we were just doing pure legal review based on an administrative record, that would make sense.

But the remedies findings that the bill creates are not just pure legal review on an administrative record.  Whether there will be irreparable harm is a factual question.  And whether there is a “substantial likelihood” of a violation of another law is not just a legal question – it will require some factual evidence or support as well.  District courts can hold hearings with witnesses to resolve these questions.  Courts of appeal (generally) do not.  Will they have to?  Will all of these instead just be based on paper filings – claims or assertions in the complaints, answers, affidavits, motions for summary judgment, and whatever is relevant in the administrative record?  This could be . . . a mess.  My guess is that the drafters did not consider the implications of putting these two reforms together.

Again, my prior proposal was to have an administrative process that could compile this information – but that in turn requires such a process, including public comment on the relevant environmental document.  To me that seems a fair trade – if an agency thinks the decision in question is important enough to warrant limiting remedies, at least give the public a chance to speak up as to why they might disagree, and have the ability to place relevant information in the record.  That would obviate the need for hearings at the court of appeals level.

Finally, I also want to note that the bill does not indicate what power a court has to enforce a remand order.  Does the bill just allow the court to ask “pretty please” if an agency does not comply with the order to redo the environmental review in a certain period of time, correcting certain flaws?  Ifthe agency responds in a timely manner with a new review, but that review is inadequate, the bill says that a challenge to the revised document is treated as a new claim.  Does that mean the only option for addressing a recalcitrant agency that produces consistently inadequate review is just constant remands, as the decision continues to get built despite noncompliance with NEPA?  Or do courts retain their inherent authority to enforce the remand order – perhaps through fines, contempt, or stop work orders on the project?

NEPA, permitting reform, remedies