Federal courts issue hundreds of decisions in FOIA cases every year. Most do not break new legal ground or attract significant media attention. As 2022 gets under way, the legal eagles of FOIA Advisor -- Allan Blutstein (AB) and Ryan Mulvey (RM) -- look back at 2021 and identify the decisions that stood out to them (in no particular order).
(1) U.S. Fish & Wildlife Serv. v. Sierra Club (U.S., Mar. 4, 2021) -- in a 7-2 decision, ruling that the agency properly relied on Exemption 5’s deliberative process privilege to withhold draft biological opinions that were prepared by lower-level staff and never approved by decisionmakers or sent to EPA under the Endangered Species Act.
AB: This was only the third FOIA decision issued by the U.S. Supreme Court since 2011, so readers will not be surprised to see it on our list. The Court’s earlier two FOIA decisions—Argus Leader (2019) and Milner (2011)— were far more consequential, however, and I suspect the media would have ignored this case if it had not been Justice Barrett’s first majority opinion. In sum, I agree with the Office of Information Policy’s guidance in which it advised that the decision “did not introduce new factors or considerations that would govern the application of Exemption 5,” but underscored two principles: (1) “[a] record is not ‘final’ for purposes of the deliberative process privilege simply because it is the last version and nothing else follows it”; and (2) “[a] record is final not because it causes an agency to change course, but because it is treated by the agency as a final decision with legal effect.”
RM: My initial impression last spring was that this case would have a fairly limited impact, given its unique facts and the nature of the biological opinions at issue. I’ve grown more pessimistic, especially as I’ve seen how DOJ attorneys in my own practice are trying to use the case. There are several aspects of the Court’s reasoning that, over time, could seriously change the deliberative-process privilege landscape. Now, I have no quarrel with the notion that a “final” record is one that “communicates a policy on which the agency has settled.” And that no doubt requires attending to how the agency treats that record. I’m not sure, however, what it means to consider “real operative effect” in light of “legal, not practical consequences.” In the context of biological opinions, it is reasonable to think of “finality” as you would under Bennett v. Spear. But agencies makes lots of decisions that never end up determining rights or obligations, or which never result in “legal” consequences, strictly speaking. So OIP’s reference to “legal effect,” and Justice Barret’s repeated citation to Bennet is concerning. If the pre-decisional aspect of a record turns on whether its communicates a policy with actual legal consequences, we could witness a massive expansion of what is considered privileged.
(2) Cause of Action Inst. v. DOJ (D.C. Cir., June 1, 2021 ) -- reversing district court’s decision and concluding that: (1) DOJ improperly segmented one large electronic file into separate records and withheld portions as non-responsive; (2) plaintiff had standing to challenge agency’s practice and policy of segmenting records, but issue was unripe for adjudication.
RM: As a disclaimer, I argued this case. I’m honored you would choose to include it in our list of consequential cases, Allan! Unsurprisingly, I agree it’s an important decision, and a consequential follow-up to the Circuit’s 2016 decision in American Immigration Lawyers Ass’n v. Executive Office for Immigration Review. The heart of the case deals with the definition of a “record,” and the importance of understanding that foundational term before turning to the question of agency control or the application of exemptions. In the wake of AILA, and with the end of scoping, many agencies turned from withholding portions of records as “Non-Responsive,” to arguing those same portions were distinct non-responsive “records.” Even though the Circuit didn’t settle on an objective definition and didn’t reach the policy-and-practice claim that challenged the legality of OIP’s guidance on defining a “record” under the FOIA, I still think it took a step in the right direction by focusing on how records are maintained and treated prior to the submission of a FOIA request. If an agency maintains a record as unitary whole, it shouldn’t be able to break it up when processing for disclosure, at least without requester consent. In case anyone is interested, I explored the import of the case in greater detail a while back at the Yale Journal on Regulation’s Notice & Comment blog.
I’d only add that the court’s decision on ripeness is bizarre. I don’t understand how ripeness is relevant to a policy-and-practice claim, and I don’t see how Judge Edward’s concern about “contingent future events” squares with his discussion of standing and the demonstration of likely future application of a challenged policy or practice. Taken to its logical extension, Cause of Action Institute could gut Payne, especially as far as informal policies and practices are concerned. In the end, the practical effect may be that OIP guidance is immune from judicial review, unless there is a clear adoption of the guidance within an agency.
AB: No one outside the government knows this case better than you and your colleagues, Ryan, so I will address a different aspect of it, namely the perplexing way this litigation managed to occur. Once the requester filed an administrative appeal disputing the “non-responsive” withholdings, DOJ could have—and I would argue should have—treated the appeal as a new request and processed the disputed records. Instead of mooting the matter as such, DOJ took the requester’s bait and risked its “distinct records” policy and practice in court. Given the lopsided oral argument (kudos, Ryan), DOJ must have been thrilled with the mixed verdict here.
(3) Rojas v. FAA (9th Cir., Mar. 2, 2021) (en banc) -- holding in most relevant part that the “consultant corollary” applied to documents prepared for agency by outside consulting firm and affirming district court’s decision that two of three disputed documents were protected by Exemption 5’s attorney work-product privilege; Jobe v. NTSB (5th Cir., June 17, 2021) -- reversing and remanding district court’s decision and holding that “outside parties solicited by the NTSB qualify as ‘consultants’ under Exemption 5’s consultant corollary; rejecting district court’s view that technical personnel employed by aircraft manufacturers and operators have too much “self-interest” in outcome of NTSB investigations to be regarded as consultants;
AB: The government thwarted attempts to upend its longstanding use of the consultant corollary and it topped off those wins in early 2022 when the U.S. Supreme Court denied petitions for each case. Nicely done, government litigators. Requesters are running out of circuits in which to litigate this issue. Perhaps they’ll try to persuade lawmakers to do their bidding instead.
RM: As a policy matter, I have no real quarrel with the consultant corollary per se, although I find the textualist analysis offered by the dissenters in the Ninth and Fifth Circuits to be persuasive. Whatever ambiguity there may be in the term “intra-agency,” I don’t think the plain meaning calls to mind the kinds of consultants implicated in Rojas, let alone Jobe. (Side note: I think Jobe could have been reversed, and perhaps should have been reversed, simply by applying Klamath and without deciding the validity of the consultant corollary.) Every time I read footnote two of Justice Scalia’s dissent in Department of Justice v. Julian, with its purposivist betrayal of the statutory text, I just shake my head. But here’s a hot take: my real heartburn is caused by the long-standing D.C. Circuit caselaw undergirding the consultant corollary—e.g., Soucie v. David and Ryan v. Department of Justice—and how it has been used to expand the meaning of Exemption 5’s threshold to cover inter-branch communications between agencies and the President or Congress.
(4) Cause of Action Inst. v. OMB (D.C. Cir., Aug. 20, 2021) -- affirming district court’s decision that Internet browsing histories of OMB and USDA officials did not qualify as “agency records,” because “the agencies’ retention and access policies for browsing histories, along with the fact that they did not use any of the officials’ browsing histories” indicated that the agencies did not control the requested records.
RM: Another case I argued…I promise I didn’t add it to our list! As interested as I was to read OMB Director Mulvaney’s Internet browsing history, I always saw our fight as touching on something much more impactful and theoretical. In my mind, this case was really about proper application of Burka, and whether its third factor—namely, actual use—is dispositive to the control inquiry, especially in the context of agency-created records. We were trying to get some clarity on the proper import of Judicial Watch, Inc. v. Federal Housing Finance Agency. Alas, we were unsuccessful. The silver lining, however, was the Court’s clarification that “[t]he text and structure of FOIA . . . make clear that whether the requested materials are ‘agency records’ goes to the merits of the dispute . . . rather than the court’s jurisdictional power to hear the case.” This is something requesters have had to fight with the government about for a number of years; hopefully, DOJ will stop trying to dismiss novel FOIA suits under Rule 12(b)(1)! One final note: although this case involved requests denied by OMB and USDA, there was a third request directed to Amtrak that ended up not making it into the Complaint. Why? Amtrak gave us their top officials’ browsing histories. And guess what? They really like looking at pictures of trains!
AB: Government employees are breathing slightly easier now. Had the court ruled in appellant’s favor, FOIA offices would have been flooded by these pesky requests. Granted, agencies would have been able to invoke applicable exemptions (Exemptions 5 and 6 come to mind), and employees would have learned to delete their histories, browse in private mode, or use personal devices to do their Cyber Monday shopping. I agree with you about the silver lining. The D.C. Circuit had not squarely addressed that particular jurisdictional/merits question, though the answer seemed clear enough given the court’s similar 2009 decision in Citizens for Responsibility & Ethics in Wash. v. Office of Admin.
(5) Nat. Res. Def. Council v. EPA (2nd Cir., Nov. 29, 2021) -- reversing in part and vacating in part district court’s decision and holding that: (1) “messaging documents”— i.e., records relating to agency’s decision about how to communicate its policies to people outside the agency—merit protection under Exemption 5’s deliberative process privilege unless they reflect “merely descriptive discussions”; and (2) “briefing documents”—i.e., records “created to brief senior agency staff about various topics”—could qualify under the deliberative process privilege even if they did not relate to a specific decision facing the agency.”
AB: My reaction to this decision can be summed up in one word that was used frequently by John Madden, the recently departed Hall of Fame football coach and broadcaster: “Boom!” The D.C. Circuit has long rejected the Exemption 5 arguments shilled by plaintiff here; it was satisfying to see the Second Circuit finally lower the boom. Unfortunately, requesters can still find refuge in the Ninth Circuit, which clings to a shortsighted notion about messaging records. See, e.g., Nat'l Pub. Radio v. U.S. Int'l Dev. Fin. Corp. (C.D. Cal., Nov. 21, 2021).
RM: It was only a matter of time. I tend to think the deliberative-process privilege is most appropriate when applied to records that reflect substantive policy decision-making. But I recognize it’s hard to come up with a categorical rule, and I also acknowledge there could be real harm in revealing some internal discussions about “messaging.” I suppose my view is colored by agencies trying to withhold records to avoid political embarrassment, rather than to protect employee candor and the integrity of their decision-making processes. At this point, any real effort to limit the use of the deliberative-process privilege is going to require congressional intervention and a reworking of Exemption 5. Alternatively, a robust “foreseeable harm” standard could be an effective check on over-redaction. The D.C. Circuit nudged things in favor of the requesters in that regard this past year in Reporters Committee for Freedom of the Press v. Federal Bureau of Investigation.
(6) NY Legal Assistance Grp. v. Bd. Immigration Appeals (2nd Cir., Feb. 5, 2021) -- in a 2-1 decision, vacating and remanding district court’s decision that agency was not required to affirmative publish its non-precedential opinions pursuant to 5 U.S.C. § 552(a)(2).
RM: There were some interesting circuit splits that materialized or deepened in 2021. The Ninth Circuit, for example, parted ways with the Second and took a different approach to the Tiahrt Rider, the OPEN FOIA Act’s Exemption 3 reforms, and the question of legislative entrenchment. Ctr. for Investigative Reporting v. DOJ, 14 F.4th 916 (9th Cir. Sept. 23, 2021); Everytown for Gun Safety Support Fund v. Bureau of Alcohol, Tobacco & Firearms, 984 F.3d 30 (2d Cir. 2020). We also saw the Second Circuit join with the Ninth to reject the D.C. Circuit’s extremely limited provision of relief for requesters bringing claims under 552(a)(2)—otherwise known as the FOIA’s “reading room” provision. See CREW v. DOJ, 846 F.3d 1235 (D.C. Cir. 2017). Frankly, I’ve always found the CREW court’s narrow reading of the FOIA’s remedial provision to be quite weak. I’m glad to see the Second Circuit willing, in theory, to force agencies to comply with their proactive disclosure obligations.
AB: Unlike you, I was not “glad” to see this Second Circuit decision. Rather, I agree with the dissent’s conclusion that “[t]his state of affairs makes little sense.” Congress ought to fix this mess instead of leaving it for the Supreme Court to resolve, but I certainly won’t hold my breath. My sense is that FOIA lobbyists are pursuing several grander proposals, such as a (terrible) across-the-board public interest balancing test and a (reasonable by comparison) reinstatement of Exemption 4’s substantial competitive harm test.