A Critical Analysis of the Seven County Opinion

The court’s opinion ignores basic principles of law and statutory text in order to achieve a policy goal.

This is the second in a series of blog posts about the Court’s Seven County opinion.  In our first post, we summarized the key points from the opinion.  Here, we provide our assessment of the Court’s analysis.

The Court’s analysis, especially in Part II.B, where it sets specific limits on the scope of NEPA, relies primarily on two steps.  First, it begins with the proposition that the “proposed action” is the basis for NEPA review.  Such a starting point is sensible, given the text of NEPA itself, which mandates review of the proposed action.  43 USC 4332(2)(C).  Slip op. at 16.

From there, the Court argues that a separate project “breaks the chain of proximate causation between the project at hand and the environmental effects of the separate project.”  Slip op. at 17.  As we’ve noted in our prior posts and our online article, the Court’s conclusion here does not follow from tort law principles of causation, which is the most relevant common law area of causation.  Indeed, subsequent actions that are foreseeable have long been considered within the scope of what is proximate cause in tort.  (For instance, a landlord who refuses to repair the lock on a tenant’s door may be liable for a subsequent break-in.) Thus, the Court must have some reason for a narrower approach to causation in NEPA than it does in common law.

That analysis is lacking in the Court’s opinion, which is limited to a single paragraph on causation.  Slip op. at 16-17.  The Court simply states that effects of subsequent actions “are not relevant to the agency’s decisionmaking process or that it is reasonable to hold the agency responsible for those effects.”  Slip op. at 17.  That is true, the Court claims, even if those effects are “factually foreseeable.”  Id.  Both of those points seem wrong to us.  It is not clear to us why an agency, when thinking about the environmental impacts of an action it is considering to take, shouldn’t consider the foreseeable effects of follow-on decisions.  The point about responsibility is even odder.  Tort law holds actors responsible for their actions through damages – and yet tort law considers subsequent actions within the scope of proximate cause.  NEPA does not impose any liability on agencies, other than a mandate to conduct thorough review – given the lower stakes of NEPA, one might argue we should be less concerned about whether we are holding the agency “responsible” for additional causal steps.

The Court does say “[a] relatively modest infrastructure project should not be turned into a scapegoat for everything that ensues from upstream oil drilling to downstream refinery emissions.” Slip op. at 19.  Courts may be reluctant to place huge liability based on a relatively minor action, but this is not an ironclad rule, and what is “modest” is a matter of context. Flipping the latch on a gate is a seemingly minor action. But if someone unlatches a gate and lets a bull escape into a crowded fairground, wouldn’t they be responsible for the ensuring harm — even if some of it is delayed or happens at a distance?  Opening a gate is a “relatively modest” action, but that doesn’t excuse the actor in question from liability for the ensuing results.  The key question should be whether the results are something that a reasonable person would take into account.

It is not explicit in the Court’s reasoning, but this may be where the Court’s emphasis on the merely procedural element of NEPA in Part II.A is driving its analysis.  If NEPA is only procedural, then perhaps it is not worth time and effort for courts to force agencies to look too far down the causal chain in their NEPA review.  But such a conclusion is of course in tension with the language in NEPA that the point of the statute is to force agencies to consider the “reasonably foreseeable” effects of their actions – which presumably includes what other actors might reasonably foreseeably do based on the proposed action. And it’s in tension, to say the least, with the important environmental policies that Congress expressed in NEPA in 1970 and left unchanged in 2023. It’s also a bit odd for the Court to dismiss something as “merely procedural” given the amount of attention that the Court itself gives to procedural issues in judicial proceedings, administrative law, and criminal law.

The other limit the Court draws on the scope of NEPA is based on whether an agency has “regulatory authority” over subsequent projects.  (This is another example of sloppiness in the opinion — the Court first seems to say that the agency doesn’t have to consider separate projects and then that it might matter whether the separate project is regulated by another agency.) Here the Court argues that the NEPA provision that requires consultation with other agencies about the effects of a proposed action does not mean that NEPA requires consideration of other agency actions in NEPA.  Instead the Court argues that the consultation provision goes to effects of a proposed action that might fall within the scope of another agency’s jurisdiction: The Court uses an example of consultation with the Forest Service when the railroad would go through national forest lands as an example of consultation.  This may, in fact, be one of the reasons the Court did not adopt the broader position of the petitioners, which was that any effects of a proposed action that fell within the jurisdictional space of another agency were outside the scope of NEPA.  The Court’s position that NEPA only excludes actions that are outside the jurisdictional space of the agency is more consistent with the NEPA consultation provision, and less constraining of NEPA scope.

Nonetheless, the Court’s decisions on scope, whether for separate projects later in space or time, or for projects under the regulatory authority of a different agency, are not tethered to the language of NEPA or to earlier Supreme Court precedent.  As noted above, the causation language is inconsistent with general principles of proximate causation (and thus the “reasonably foreseeable” language in the 2023 amendments).  The Court does not cite any statutory language in its discussions of causation and of limiting the scope of NEPA to exclude projects under the jurisdictional scope of another agency.  This despite the Court’s claim that its analysis is based on “NEPA’s textual focus,” slip op. at 19.

Indeed, for a Court that supposedly in Loper Bright emphasizes the “best reading” of statutes, that has disclaimed that statutory interpretation involves policy, and that has generally emphasized textualism in its analysis, this is a remarkably atextual, policy-based opinion.  One might even say that the Court is legislating…

There is one, potentially significant, caveat to the Court’s restriction of the scope of NEPA analysis, as noted in our prior post.  There may be circumstances, the Court concedes, where “other projects may be interrelated and close in time and place to the project at hand” such that there “is a single project within the authority of the agency in question.”  As an example of such a situation, the Court cites Robertson v. Methow County, 490 U.S. 332 (1989), in which the Court considered the adequacy of NEPA review for a ski project on federal land.  In Robertson, the Court explicitly noted, and endorsed, NEPA review of the follow-on residential development that would occur from the ski area.

The problem is, as the Court in Robertson itself noted, the residential development that would have occurred would not have been on federal land, but on private land, under state and local regulation.  Indeed, that was the very reason the Robertson Court gave for its ruling in the decision.  So, we’re left with a caveat that inaccurately describes the very example the Court relies upon.

This is not just evidence of the general sloppiness of the opinion, but also we think, a hint of confusion to come.  The Court here appears to be giving leeway for courts to conclude that what on the surface are framed as separate projects are really one and the same project, such that the true scope of the project for NEPA review purposes includes them all.  And in some ways, the Court has to make such a caveat.  Otherwise, agencies could engage in rampant “segmentation”  — the term used for the division of a larger project into smaller bits to avoid NEPA review, such as dividing a highway project into multiple phases, even though the project really is one larger project.  And indeed, this concern drove a range of concepts in the prior NEPA regulations, such as connected actions, to address situations where agencies might break up a larger project to sidestep NEPA.

But of course, that leaves courts to some extent back where they were before this case, with difficult questions about the scope of NEPA review.  And indeed, the inconsistency of the Court’s use of the Robertson case raises real questions about the bright-line nature of the Court’s exclusion of projects under another agency’s jurisdiction.  If the projects are closely related enough, even if they are under different agency jurisdictions, do they get considered together in NEPA?

We don’t want to overemphasize the uncertainty here.  The Court emphasizes deference by courts as to agency decisions on this question.  Slip op. at 20.  And as we’ll discuss in the next post, there are a range of indirect effects that the Court’s ruling may put off the table for NEPA purposes.  But there will still be lots of room for dispute, and varying outcomes across courts.

NEPA