The Freedom of Information Act (FOIA) is codified with the Administrative Procedure Act (APA) as part of Title 5 of the U.S. Code, and FOIA law is uncontroversially considered a subset of administrative law.  At the same time, FOIA is unique, with its own judicial review provision and standards, as well as rather unconventional litigation practice.  Some administrative law practitioners might be surprised to learn the Supreme Court’s groundbreaking decision in Loper Bright Enterprises v. Raimondo, and its overruling of Chevron deference, could be relevant to FOIA.  For that reason, and to mark last month’s celebration of Sunshine Week, I would like to explore how Loper Bright will impact FOIA jurisprudence, specifically with respect to judicial review of agency use of Exemption 3.¹1. The author has served as lead counsel for the fishermen in Loper Bright since its filing in February 2020.  Most recently, he argued the case on remand before the D.C. Circuit.

For those who are unfamiliar, every year in mid-March, Sunshine Week affords advocates of open government the opportunity not only to highlight their work but to celebrate the principles of transparency underscored by such figures as Presidents James Madison and John Adams.  Adams, in particular, wrote highly of the importance of what today we might call government transparency:

[L]iberty cannot be preserved without a general knowledge among the people, who have a right from the frame of their nature to knowledge . . . but besides this they have a right, an indisputable, unalienable, indefeasible divine right to that most dreaded, and envied kind of knowledge . . . the characters and conduct of their rulers.

The spirit expressed in this passage has long inspired my own practice and the work of my organization, Americans for Prosperity Foundation, particularly in its efforts to organize a center-right transparency coalition, “Right on Transparency.”  That coalition holds fast to the notion that the fight for free markets, limited government, and individual liberty goes hand-in-hand with comprehensive transparency reforms.  Seeing as those of us who are interested in—or even celebrate—doctrinal developments like the end of Chevron deference might also have an interest in open government, I believe it fitting to highlight the intersection of Loper Bright and FOIA.

I. FOIA and Chevron Deference in the Pre-Loper Era

To start, a quick refresher: Under the methodological framework for resolving disputed interpretations of law that the Supreme Court designed in Chevron v. National Resources Defense Council, Article III courts were once expected, first, to examine “whether Congress ha[d] directly spoken to the precise question at issue.”  Judges, in other words, would start by asking if a law was “ambiguous.”  If the “intent of Congress” was “clear,” that would be the “end of the matter.”  But if a court determined “Congress ha[d] not directly addressed the precise question at issue”—that is, if the law at issue was “ambiguous” or “silent”—it would instead consider whether the agency’s position was “based on a permissible construction of the statute.”  This Chevron two-step required judges to defer to agencies’ legal interpretations of “ambiguous” or “silent” statutes, so long as they were “reasonable,” notwithstanding the APA’s directive in Section 706 that reviewing courts “decide all relevant questions of law.”

Despite its tremendous impact on administrative law generally, the Supreme Court’s Chevron doctrine was never quite relevant—with one exception, discussed below—for judicial review under FOIA.  There are several reasons why that was so.

First, FOIA provides, as a matter of statutory text, that courts are to adopt a de novo standard of review for disputes related to the withholding of an agency record, see 5 U.S.C. § 552(a)(4)(B), or the denial of a fee waiver.  See id. § 552(a)(4)(A)(vii).  While fee-waiver review is limited to the record that was before an agency, de novo review of law and facts is the theoretical standard for other types of challenges.  Since all FOIA fights can be conceptualized as disputes over fees or the improper withholding of records, there could arguably be only one standard of review—de novo.  But that is not how courts previously understood Section 706 of the APA.  Hence the forty-year existence of Chevron deference.

Second, and perhaps more obviously, even if Congress had designed FOIA to rely on the general standard of review found in Section 706, there would have been strong doctrinal reasons for maintaining de novo review pre-Loper Bright.  Consider the Supreme Court’s instruction in United States v. Mead that Chevron deference is only appropriate when “Congress [has] delegated authority to [an] agency generally to make rules carrying the force of law.”  This preliminary inquiry—Chevron “Step Zero”—drew on language found in Chevron itself, which instructed that deference was only appropriate if an agency not only had “great expertise” but was “charged with responsibility for administering” the statute at issue.

FOIA, of course, is a law that applies to all agencies, but Congress has never tasked any one of those agencies with its government-wide administration, at least as to all its aspects.  To be sure, the Department of Justice’s Office of Information Policy (OIP) has assumed de facto responsibility for providing guidance to agencies on various FOIA issues.  And the National Archives and Records Administration’s Office of Government Information Services (OGIS) has authority to “review policies and procedures,” “review [agency] compliance,” and identify ways of improving FOIA administration.  See 5 U.S.C. § 552(h)(2).  But neither OIP nor OGIS has statutory authority to regulate agency FOIA programs in any binding way.  Thus, agency interpretations of the FOIA could never pass “Step Zero” and trigger the Chevron framework.  The D.C. Circuit confirmed as much in Tax Analysts v. Internal Revenue Service: “The meaning of FOIA should be the same no matter which agency is asked to produce its records.  One agency’s interpretation . . . is therefore no more deserving of judicial respect than the interpretation of any other agency.”

Remarkably, this was the approach throughout the Chevron era, even when the government had a colorable argument for at least some aspects of FOIA having been assigned to one agency for regulatory implementation.  Consider Section 552(a)(4)(A)(i), with my emphasis added:

In order to carry out the provisions of [the FOIA], each agency shall promulgate regulations, pursuant to notice and receipt of public comment, specifying the schedule of fees applicable to the processing of requests under this section and establishing procedures and guidelines for determining when such fees should be waived or reduced.  Such schedule shall conform to the guidelines which shall be promulgated, pursuant to notice and receipt of public comment, by the Director of the Office of Management and Budget and which shall provide for a uniform schedule of fees for all agencies.

Under this provision, Congress gave the Office of Management and Budget (OMB) the duty to develop fee guidelines that would be binding across the entire government, that is, to which all agency FOIA regulations would need to “conform.”  OMB published its guidelines in 1987 and has only updated them once, in 2020, after litigation by the Cause of Action Institute to compel a rulemaking.²2. The author served as lead counsel in this case.  See generally Cause of Action Inst. v. White House Office of Mgmt. & Budget, No. 19-2784 (D.D.C. filed Sept. 17, 2019). (Notably, OMB went so far as to argue the fee guidelines did not constitute a “rule.”)  Despite the possible argument for Chevron deference in the context of such definitional fee issues, the government repeatedly took the view that Chevron was inapplicable.  In one of the more recent fee-category fights that helped shape the updated guidelines, the Department of Justice entirely omitted any argument “that the OMB Guideline[s] [were] entitled to Chevron deference.”  And the D.C. Circuit has reiterated the inapplicability of Chevron in the fee space, regardless of whether it went unaddressed by an agency.

II. The Special Case of Exemption 3

Yet as I’ve written elsewhere, “things were always a bit tricky with Exemption 3,” which cross-references withholding statutes that exist outside of Section 552.  Those statutes either “require[] that the matters [contained in a record] be withheld from the public in such a manner as to leave no discretion on the issue,” or “establish[] particular criteria for withholding or refer[] to particular types of matters to be withheld.”

The reference to provisions of law extrinsic to the statutory framework of FOIA presents a fascinating interpretative question: If a FOIA case involves a fight over the proper use of Exemption 3—or, more specifically, a dispute over the proper use of an “Exemption 3” statute—is (or, rather, was) there ever justification for deference under the Chevron framework? 

Professor Bernard Bell wrote an excellent piece on this blog several years ago addressing the issue:

While Exemption 3 statutes are often construed in the context of FOIA claims, they exist independently of FOIA.  Moreover, such statutes contain their own standards of disclosure, and often their own procedures governing withholding or release of information.  Those procedures may be both self-contained and distinct from FOIA.  Often, unlike FOIA, they address documents or information maintained by a particular agency, rather than by a broad cross-section of agencies. . . .

Exemption 3 statutes are typically separate legislation to be reconciled with FOIA, not necessarily incorporated into it.  Exemption 3 operates in two ways.  It functions as a savings clause for legislation regarding the disclosure of government information that predated FOIA or the relevant amendments to Exemption 3.  But perhaps more importantly, it provides a stable background assumption against which Congress can legislate; it specifies that future legislation satisfying Exemption 3’s standards is not subject to FOIA statute and its processes.  This reconciliation of FOIA with separate statutes clearly accords Exemption 3 statutes, and the standards and procedures set forth therein, primacy over FOIA.  In enacting Exemption 3 statues, Congress has consciously balanced the interest in disclosure against the need to maintain the confidentiality of the particular information covered, and often made the agency responsible for compliance with those specialized statutory provisions.  The requirement that Congress expressly reference Exemption 3 in the text of an Exemption 3 statute provides an additional assurance that Congress consciously makes the decision that FOIA’s general transparency standards are inappropriate with respect to the particular type of information or files exempted.

Setting aside Professor Bell’s reference to the requirement for Congress, as a result of the OPEN FOIA Act of 2009, to “specifically cite[]” to Exemption 3 when devising new withholding statutes—a requirement that may be an unconstitutional entrenchment—the doctrinal “split” on Exemption 3 must not be understated.  There already was (and remains) an arguable tension with respect to FOIA’s de novo review standard vis-à-vis application of Exemption 3 statutes, which outside the context of a FOIA claim would normally implicate the APA’s “arbitrary and capricious” review standard, as is the norm in “reverse FOIA” suits.  The more pressing disagreement among jurists for present purposes, however, involved whether an agency’s understanding of the meaning of the terms of a withholding statute—or its scope—could receive Chevron deference, precisely because the disputed statutory terms exist “outside” of FOIA.

On this question, there is a split in the circuits, as noted by Professor Bell.  On the one hand, the First Circuit in Aronson v. Internal Revenue Service decided that because FOIA, as a result of Exemption 3, “does not apply” to matters exempted by another statute, the default administrative law doctrines, including Chevron, must apply.  973 F.2d 962, 965–66 (1st Cir. 1992).  “[I]f courts were to use principles of FOIA de novo review to decide how to interpret or to apply the exceptions” in those non-FOIA provisions, it “would risk distorting Congress’s very different aims in enacting the two different statutes.”  Thus, once it is clear a recognized Exemption 3 statute applies, and “the information requested at least arguably falls” within its scope, “de novo review normally ends” and “review must take place under more deferential, administrative law standards.”³3. See generally A. Michael’s Piano, Inc. v. Fed. Trade Comm’n, 18 F.3d 138, 144 (2d Cir. 1994) (discussing the circuit split as it stood in 1994). The D.C. Circuit has adopted this same approach, albeit somewhat inconsistently.

The Ninth Circuit took a different tack in Long v. Internal Revenue Service, 742 F.2d 1173 (9th Cir. 1984), and Carlson v. U.S. Postal Service, 504 F.3d 1123 (9th Cir. 2007), where it viewed the relationship between Exemption 3 and various withholding statutes through an “incorporationist” lens, to paraphrase Bell.  If an agency seeks to use a withholding statute for purposes of applying FOIA, as opposed to applying the statute in a non-FOIA context, a court is required to “proceed de novo,” to ensure the agency does “not interpret the exemption too broadly.”  That approach arguably heeds the Supreme Court’s direction that FOIA exemptions be construed narrowly.

Professor Bell appears to have once endorsed the First Circuit approach, arguing “[c]ourts should adopt a general presumption that Chevron deference applies to agencies’ interpretation of the limitations on disclosure of information in Exemption 3 statutes,” at least unless those statutes “do no more than replicate FOIA requirements,” for example, by “preclud[ing] disclosure [of] trade secrets or commercial or financial information that is privileged or confidential,” which would fall under Exemption 4.

III. The Impact of Loper Bright on the Exemption 3 Interpretative Question

“Chevron is overruled.”  With these historic words, Loper Bright eliminated Chevron deference.  The resolution of all legal questions—including the meaning of withholding statutes used with the FOIA—should now be resolved de novo.  Whether the use of extrinsic withholding statutes in the FOIA context is best understood as a matter of “reconciliation,” which might technically trigger review under Section 706, or “incorporation,” under Section 552, the same method for judicial decision-making applies across the board.

When a requester disputes that a record is protected by Exemption 3, a court is obliged to provide its independent, best judgment about the scope and application of the underlying withholding provision.  Of course, there may be instances where “respect” for the agency’s position is appropriate.  And Congress could, in theory, design a withholding statute to grant an agency express discretion to set the bounds of its reach.  But even then, the general rule is de novo review, and the legal limits of the scope of the statute under review remain with the court to determine. 

There have been at least 19 published decisions involving Exemption 3 since the Supreme Court released its Loper Bright decision.  Most of these cases involve non-interpretative issues like waiver under the public-domain or official-domain doctrines, the use of a Glomar response to refuse to admit or deny the existence of records, whether an agency followed correct procedures, or whether it provided adequate justification for applying a withholding statute.

Three cases, however, have touched on the scope of an Exemption 3 statute.  Unfortunately, none of them mentions Loper Bright.  Most notably, the Second Circuit in Bloomberg LP v. U.S. Postal Service interpreted the meaning of “commercial” in the Postal Reorganization Act’s provision prohibiting disclosure of commercial information “which under good business practice would not be publicly disclosed.”  39 U.S.C. § 410(c)(2).  Because the law at issue lacked any relevant statutory definition, the Bloomberg court—after citing the Ninth Circuit’s opinion in Carlson—turned to ordinary meaning and contemporary dictionary definitions.  That is precisely the sort of judicially driven, textualist approach anticipated by the Loper Bright majority.  The Bloomberg court also considered the separate meaning of “good business practice” by looking to an earlier decision in a sister circuit that similarly refused to afford deference to an agency’s understanding of the phrase but instead examined common meaning.  The other two cases from the past year—Jimenez v. Department of Homeland Security and Appeal, Inc. v. Department of Justice’s Office of Justice Programs—were hardly different in relying on text-focused approaches to determining the meaning of the withholding statutes in question without deference to the agency’s views.

As Exemption 3 cases work their way through the courts, especially in those jurisdictions that were previously amenable to Chevron deference, we should hopefully see a repudiation of deference and an explicit embrace of the Loper Bright interpretative paradigm.  This will eventually resolve the existing split discussed above. One final, somewhat unrelated note.  There are many kinds of informal “deference” apart from Chevron that have long existed in FOIA jurisprudence, which requesters find deeply troubling given the statute’s promise of de novo review.  Those doctrines—as Professor Margaret Kwoka and others have observed—undermine the level playing field promised by de novo review.  Indeed, Professor Kwoka goes so far as to call de novo review under the FOIA a “myth.”  The unfortunate reality—certainly in my own experience and the experience of many of my fellow FOIA litigators—is that courts afford significant deference to agency determinations, legal or factual, especially as they are set out in agency affidavits. Professor Bell, for example, points to the Supreme Court’s decision in Central Intelligence Agency v. Sims, which involved Exemptions 1 and 3. Although the Sims court did not expressly depend on Chevron, it explained the agency’s position was “worthy of great deference given the magnitude of the national security interests and potential risks at stake.” In fact, FOIA itself only mentions according “substantial weight” to an agency’s position, as presented in its supporting declarations, regarding the “technical feasibility” of partial redactions or the “reproducibility” of records in the form and format sought by a requester. 5 U.S.C. § 552(a)(4)(B). Again, as I have explained elsewhere:

Consider, for example, the language of “substantial weight” in Exemption 1 litigation, or the “good faith” and “regularity” presumptions in all cases.  There’s also a radical uniqueness to FOIA and FOIA litigation that ends up giving de facto deference to the government.  We might look to the lack of an administrative record, the rarity of discovery, the multiple rounds of summary judgment so that agencies can get another bite at the apple, etc.  Finally, the D.C. Circuit has decided to afford deference to an agency’s interpretation of its own regulations that provide for additional grounds for expedited processing beyond those set out in the statute.  See, e.g., Al-Fayed v. CIA, 254 F.3d 300, 307 n.7 (D.C. Cir. 2001). (That, of course, implicates Auer deference, which is still good doctrine, albeit limited by Kisor v. Wilkie.)  Loper Bright and the broader movement to reform the administrative state haven’t yet touched these kinds of concerns.

Perhaps we can hope the atmospheric impacts of Loper Bright will eventually carry over to the broader FOIA context and provide fodder for requesters to argue that courts ought to take the plain text of FOIA, and its promise of de novo review, more seriously. On this point, just two weeks ago a district court declined to adopt a deferential standard—“either arbitrary-and-capricious review . . . or Auer deference”—and instead undertook de novo review of U.S. Customs & Border Protection’s understanding of its own regulations governing the scope of the “business submitter” consultation required by Executive Order 12600, and its inclusion of such consultation in the calculation of a fee estimate. If “[t]he interpretation of the laws is the proper and peculiar province of the courts,” as suggested by Federalist 78, then that principle should apply with equal force to all legal questions under FOIA.

Ryan P. Mulvey is policy counsel at Americans for Prosperity Foundation and counsel at Cause of Action Institute.  He is lead counsel for the Cause of Action Institute team litigating Loper Bright.  Ryan’s practice focuses, in large part, on FOIA.  He is currently President of the American Society of Access Professionals (a 501(c)(3) organization that provides training on the FOIA), helps run FOIAadvisor.com, and is a member of the 2024-2026 term of the National Archives and Records Administration’s Federal FOIA Advisory Committee.