NEPA and Remedies

Trying to solve the difficult challenge of reforming remedies for NEPA violations

Earlier this year, the Institute for Progress (a pro-permitting reform organization) published a series of proposals for changing remedies for agency violations of NEPA.  There are good reasons to think about the problem.  If any NEPA violation can stop an entire project, that can incentivize efforts to fly-speck environmental review documents to strategically stop projects – whether they are beneficial to the environment or not.  While this approach can benefit opponents of oil and gas leasing or other fossil fuel infrastructure, it also benefits opponents of transmission lines, wind farms, and solar and storage projects – an issue I’ve covered before.

The focus on remedies also makes sense – it can be a simpler, broader solution to the overall problem, rather than trying to tinker with the underlying legal structure of NEPA.  That of course has a flip side – any changes may be blunt tools that may have counterproductive outcomes.

Remedies are particularly complicated because there are three components.  First, a plaintiff might initially seek a preliminary injunction from a court to stop a project while the litigation proceeds.  After litigation is resolved, if the plaintiff is successful, there are two separate remedial paths.  One is a permanent injunction – which can be applied both against the agency in question, but also against third parties (for instance, permittees or proponents).  The other is vacatur (or not) of the underlying agency action that triggered NEPA in the first place.  This is a remedy that comes from the Administrative Procedure Act (APA).

A couple of decades ago, at least some federal circuits generally presumed that injunctive relief was appropriate for a plaintiff who was likely to succeed (for a preliminary injunction) or had succeeded (for a permanent injunction) on a NEPA claim, because environmental harm was presumed to be irreparable (and thus worth protection by an injunction).  The Supreme Court, in a series of cases, made clear that courts could not apply such a blanket presumption.  Environmental harms must be weighed against the costs and benefits of granting an injunction, and while they may well be irreparable, plaintiffs must show that harm is irreparable.

However, that still leaves vacatur under the APA, which has long been a contested issue.  Section 706 of the APA states that courts shall “hold unlawful and set aside” agency action that a court finds violates the APA.  (The APA is the primary means by which challenges under NEPA would be brought by plaintiffs.)  Does that language mean that the only remedy, if a court finds a violation of the APA, is to vacate the underlying agency decision (in the case of NEPA, the agency decision about what NEPA review to do and what substantive decision to make after the NEPA review is complete)?  Or do courts have the ability to remand the decision back to the agency without vacatur?  The question is unsettled, with the circuits divided on the issue.

But if a court concludes that it must vacate the underlying agency decision, note that this would bypass the balancing test for permanent injunctions – the agency decision must be vacated regardless of any balancing of harms.  That raises the concerns I flagged at the beginning of this post about the possibility of litigation finding relatively minor NEPA flaws that are used to stop entire projects, even if no permanent injunction is issued.

Even the courts that do allow remand without vacatur apply a test that is ill-suited to questions about remedies for failure to do adequate environmental review.  It’s a test that looks at how badly the agency got the underlying legal issue wrong (resulting in a finding that the agency lost under the APA), and how disruptive vacating the agency decision would be.  Notably, nothing in this balancing test considers the benefits of vacating the agency decision (for instance, advancing environmental goals).  And they are fairly wide-open factors, allowing for a wide range of outcomes.

As I noted in my prior blog post, the problem here is that moving remedies away from “always” vacate will generally result in a high degree of variability in outcomes.  One could move the dial to the other extreme – “never” vacate.  But, as the report notes, that would be the effective repeal of NEPA as agencies would face no consequences for violations.  And so the report (correctly to my mind), rejects that approach.

Instead, the report focuses on two alternative approaches to remedies for NEPA – in either case, the court would conduct a specific, unified analysis to determine whether to impose vacatur and/or injunctive relief, thus eliminating the current separation between injunctive relief and APA vacatur analysis.  To my mind, a unified approach that focuses on the standards traditionally used in injunctive relief makes a lot of sense, especially if it allows for explicit consideration of environmental harms.

The first option proposed is that injunctive relief or vacatur would only be applied by a court where there are extraordinary circumstances (such as major environmental harms) or blatant violations of NEPA.  This has some promise – but creates risks as well.  If courts interpret “extraordinary” very narrowly, lots of preventable environmental harm might occur.  In addition, if courts interpret “extraordinary” as “infrequent”, then only the most blatant violations or harms get enjoined.  But if agencies consistently make worse and worse decisions, then the standard becomes a moving target that protects less and less.

The second proposal requires plaintiffs to show that the error in the NEPA analysis could plausibly have resulted in a change in the agency’s decision that would obviate or mitigate the environmental harm the plaintiffs are complaining of.

Both of these proposals are intriguing, but both raise a number of complications:

• Any proposal must avoid “magic words” that one side or another can just say to get their desired result. Could a plaintiff just say (in briefing or in the administrative process) that agency decision would have been different, or that there is serious environmental harm – that is an easy standard to meet (since it is likely always a plausible hypothetical that the agency decision could change in some way, or the plaintiff’s claim of serious environmental harm can just stand on its own).  Reciprocally, the agency could always argue, since NEPA is a procedural statute, that the agency decision could have been the same regardless of the NEPA analysis, or it could argue that the harm will be minimal.
• The whole point of NEPA is to provide analysis to determine whether environmental harm is serious (or not). But the plaintiff’s very claim is that the analysis is inadequate – and if the analysis were adequate, the agency can and should make a different decision.  How are courts supposed to assess those claims as to the seriousness of the environmental impacts and their relevance for the agency decision, if the environmental review itself has not done so, short of conducting their own trial (which would presumably be worse both on delay and expertise grounds).  Relatedly, there is often wide dispute over what counts as an environmental harm.  Should courts consider that at the first instance as well?
• So ideally, we would have some sort of administrative process in which these issues would first be raised by plaintiffs – arguing that the harm from the inadequate analysis would be serious, and providing supporting evidence, and/or the agency decision would be different – with agencies having the ability to respond. That material would be in the administrative record, facilitating deferential and speedier judicial review.
• But that requires a public comment process – something that the Trump Administration has been eliminating for many NEPA processes.

One possible way of resolving these tensions is to allow courts to only grant remand without vacatur or injunctive relief for NEPA violations if agencies have provided a public comment period, and plaintiffs failed to provide in that comment period support for the possibility of serious environmental harm connected to the inadequate analysis and/or support for the possibility that the agency decision would be different if the analysis was adequately undertaken.  (Note that the bar for showing whether serious environmental harm could occur should not be set too high, since after all the analysis of whether that harm might occur is contested!)  Agencies could respond to those arguments in the administrative process, and the court could consider the record evidence and agency response in determining remedy.  In other words, there is a trade: agencies can get the possibility of avoiding vacatur, but only if they provide public comment.  (One can imagine agencies making strategic choices about when public comment is warranted.)

This still leaves the uncertainty of a vague standard for granting injunctive relief or vacatur.  As noted above, I’m a little skeptical of “extraordinary circumstances” as it may set the bar too high.  But there might be other possible options.  One would be that if plaintiffs can show the possibility of violations of other environmental standards connected to the inadequate analysis, that would weigh in favor of injunctive relief or vacatur.  Alternatively, the finding of “irreparable” harm used in the preliminary injunctive relief context could provide some more guidance to courts.  Or a finding by the court that there is higher “certainty” that the environmental review, if adequately performed, would identify significant environmental impacts.

The report also notes the possibility of bonds – allowing agencies and project proponents to put up bonds to allow them to initiate a project despite NEPA violations, with the bond covering the costs of mitigating any significant environmental impacts identified in the subsequent analysis.  There are a couple of issues with this approach.  First, it may be difficult to quantify what mitigation might require up front (the history of reclamation bonds in the mining and oil and gas industries is not a pretty one).  Second, some harms may not be feasibly mitigated (think driving a species to extinction).  But a variant on this approach might reduce these concerns: Allow agencies and project proponents to take on an enforceable commitment to mitigate any subsequently identified environmental impacts, so long as those impacts are known to be feasibly mitigated (i.e., not irreparable).  This reduces the concern about quantifying the cost of mitigation (if the cost is higher than expected, the agency and the proponent are still on the hook for it) and allows courts to conclude that mitigation is not appropriate for the particular impacts at issue.

Overall, I think the IFP report is a nice effort at a hard problem.  I hope these suggestions are productive in moving the conversation forward.

administrative law, APA, NEPA, permitting reform, remedies