The Administrative Record in NEPA Litigation



In the vast majority of cases hearing petitions for review of final agency action under the Administrative Procedure Act (APA), the “record” amassed by the agency en route to its final action is the exclusive record by which the court will judge the validity of that action.  This rule is usually bedrock in the court/agency relationship.  Deviations from it should be rare.  See, e.g., WildWest Institute v. Bull, 547 F.3d 1162 (9th Cir. 2008); Citizens for Alternatives to Radioactive Dumping v. U.S. Dept. of Energy, 485 F.3d 1091 (10th Cir. 2007).

NEPA litigation is something else, though. Oddly enough, courts have often affirmed the bedrock rule just as they make an exception for NEPA.  Deviations have taken two principal forms.  The first—what I will call a standard response—is where the court remands the matter back to the agency so that the agency can ‘correct’ an administrative record the court believes is incomplete.  See, e.g., Klickitat County v. Columbia River Gorge Commission, 770 F. Supp. 1419, 1424 (E.D. Wash. 1991); Cronin v. USDA, 919 F.2d 439, 444 (7th Cir. 1990).  There’s nothing out of the ordinary here except perhaps the frequency at which this happens in NEPA litigation.

The second, more extraordinary response is for the reviewing court, typically a district court, to simply add the proffered evidence to the “record” being “reviewed” along with the final agency action itself.  It could be new expert testimony, new scientific papers or other published matter, or anything really.  This is an odd response in my view. Although that doesn’t necessarily make it wrong, litigants and courts in NEPA cases would do well to take care here.

Amending a Record After the Fact?

The Supreme Court has periodically allowed that, under exceptional circumstances, district courts may have to supplement the administrative record in order to discharge their responsibilities to petitioners seeking review.  See, e.g., Camp v. Pitts, 411 U.S. 138, 142-43 (1973).

But this is and should be rare.  For one reason, the APA provides that, in cases where they’re authorized, districts courts may hold a hearing de novo and decide whether the agency action is “unwarranted by the facts” as the court records them. See 5 U.S.C. § 706(2)(F).  The caveat there is the thing, though: in cases where they are authorized to hold a trial de novo.  Such authorizations are rare.  The Freedom of Information Act provides such authority.  Precious few other statutes do, though.

For another reason, adding information to a “record” after the fact is almost certain to render the subject decision “arbitrary and capricious” under the APA (§ 706(2)(A)).  It’s elementary: certainly a judgment that doesn’t take account of all the evidence is arbitrary in some sense.  The scare quotes are the point here: if a record is kept and grounds a decision when made but that record is later supplemented with after-acquired information, then there really are two different records in hand.  And it basically ignores the essence of a record in legal proceedings to amend one after the fact.

The Suffolk Doctrine

The final and most anomalous category, created by a famous NEPA saga involving oil drilling off the Atlantic coast in County of Suffolk v. Secretary, 562 F.2d 1368 (2d Cir. 1977), dynamites the bedrock.  It does so on the theory that NEPA is special.  The Suffolk court grounded its analysis in NEPA’s central dependence on alternatives analysis (a good EIS should weigh and compare several alternative actions and their environmental consequences).  Charitably interpreted, the court argued that challenging an EIS’s adequacy is like attacking the record itself and that decision-makers who don’t sincerely weigh real alternatives often tip it off by how little they’ve tried to learn about those alternatives.

Remedying such agency malfeasance by ignoring the court/agency relationship and bedrock principles like record review, though, does less to rectify a perceived deviation from NEPA than it does to destabilize institutional settlements and NEPA litigation itself within the federal system.  Indeed, a further holding in Suffolk that petitioners who fail to submit their evidence to an agency during a NEPA procedure cannot thereafter be barred from bringing it forward in a review proceeding was itself dynamited less than a year later by Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519 (1978).  According to Vermont Yankee, such petitioners absolutely can lose their right to put their evidence into the “record” if they fail to bring it forward during the agency’s NEPA proceeding.

None of this is to say that exceptions to the exclusive record rule are always wrong.  But agencies often scope their fact-finding as they do knowing full well that they aren’t constructing the world’s most inclusive, most exhaustive, or most encyclopedic collection of “available” information.  They do so not because they are ignorant but rather on the theory that information costs time and money to gather, substantiate, sort, and interpret.  If a court is going to second guess those judgments, it should probably keep to the beaten path of remanding with instructions for further proceedings. Having to remand for a do-over should give enough pause to that court for it to be sure the relief is justified.