Administrative Law Attack Outline: How to Build One

Your master outline explains doctrine. Your attack outline routes you through it under time pressure. The unit of a master outline is a rule statement; the unit of an attack outline is a question with branches — "Is this a rule or an order?" "If a rule, is it legislative or interpretive?" You should never be reading prose off an attack outline during an exam. You should be reading prompts.

Administrative law rewards this format more than almost any other 1L or 2L course, because agency-action fact patterns are procedurally layered. A single hypo can raise standing, finality, notice-and-comment adequacy, arbitrary-and-capricious review, and a removal-power challenge. Students lose points by spotting the merits issue and skipping the four gates in front of it. The attack outline's job is to stop that.

Target four to eight pages. If it runs longer, you have copied your master outline. Compress by deleting explanations and keeping triggers: a case name earns its spot only if the name alone reminds you of the rule.

Use the litigation sequence, not the syllabus sequence. Most administrative law courses teach constitutional structure first (nondelegation, appointments, removal) because it is conceptually foundational, but courts reach it last or not at all. Your outline should mirror the court.

The five branches, in order:

Write this branch as a checklist you run every time, because on an exam at least one gate will be closed or contestable. For standing, chain Lujan's three elements to the fact pattern: injury in fact that is concrete and particularized, causation, redressability, plus the procedural-injury and special-solicitude wrinkles from Massachusetts v. EPA and the tightened causation analysis in FDA v. Alliance for Hippocratic Medicine (2024). Add TransUnion for bare statutory violations and Summers for organizational standing.

Then run the APA gates. Final agency action under Bennett v. Spear requires consummation of decisionmaking plus legal consequences; Sackett v. EPA and U.S. Army Corps v. Hawkes show how broadly "legal consequences" reads. Exhaustion of intra-agency appeals is not required unless a statute or rule mandates it and makes the initial decision inoperative — Darby v. Cisneros. Ripeness runs through Abbott Laboratories' fitness-and-hardship test. Add Heckler v. Chaney's presumption that non-enforcement is committed to agency discretion, rebuttable by statutory guidelines, and Webster v. Doe's constitutional-claim carve-out.

Include a line on timing: Corner Post v. Board of Governors (2024) holds the six-year default limitations period in 28 U.S.C. § 2401(a) runs from when the plaintiff is injured, not when the rule issues. That is a live exam issue for challenges to old rules by new entities.

The controlling fork is rulemaking versus adjudication, and Londoner v. Denver and Bi-Metallic Investment Co. v. State Board of Equalization give you the intuition: individualized adjudicative facts get a hearing, general policy does not. Then ask whether the organic statute triggers formal process — the phrase "on the record after opportunity for an agency hearing" is the tell for §§ 556-557.

For informal rulemaking under § 553, list the four failure modes an exam will test: inadequate notice of the rule's terms or subjects; failure to disclose the technical data the agency relied on; a final rule that is not a logical outgrowth of the proposal (Long Island Care at Home v. Coke); and a statement of basis and purpose that does not respond to significant comments. Ohio v. EPA (2024) is the recent example of a rule falling because the agency did not respond to a significant comment. Anchor the whole section with Vermont Yankee: courts may not impose procedures beyond the APA, the organic statute, the Constitution, or the agency's own rules.

Add the exemptions branch — interpretive rules, general statements of policy, and good cause — and the rule from Perez v. Mortgage Bankers Ass'n that an agency may revise an interpretive rule without notice and comment. Include a short due process node: Goldberg v. Kelly for pre-termination hearings on welfare benefits, Mathews v. Eldridge's three-factor balancing for everything else, Withrow v. Larkin on combined investigative and adjudicative functions, and the Accardi principle that an agency must follow its own regulations.

Rewrite this section entirely if your materials predate June 2024. Chevron is overruled. Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024), directs courts to exercise independent judgment on questions of law, using traditional tools of statutory construction, while respecting statutes that genuinely confer discretion on the agency — for example, statutes that use terms like "appropriate" or "reasonable" or that expressly authorize the agency to define a term. Your outline should ask: does the statute delegate discretionary authority, and if so, did the agency stay within the boundaries of that delegation and act reasonably?

Keep Skidmore v. Swift & Co. as the residual: an agency interpretation gets weight to the extent it has the power to persuade, based on thoroughness, validity of reasoning, consistency, and the agency's expertise. Keep Kisor v. Wilkie for an agency's reading of its own ambiguous regulation — genuine ambiguity after exhausting interpretive tools, a reasonable reading, and the reading must be the agency's authoritative and considered position implicating its substantive expertise, plus no unfair surprise. Keep the major questions doctrine from West Virginia v. EPA for assertions of vast economic and political significance requiring clear congressional authorization. Add a note that Loper Bright preserved stare decisis for prior holdings that had applied Chevron.

For arbitrary-and-capricious review, use Motor Vehicle Manufacturers Ass'n v. State Farm as the spine: the agency must examine relevant data and articulate a rational connection between facts and choice; it fails if it relies on impermissible factors, ignores an important aspect of the problem, offers an explanation contrary to the evidence, or is implausible. Layer on Encino Motorcars for policy changes requiring acknowledgment of the change and reliance interests, Department of Commerce v. New York for pretextual reasoning, DHS v. Regents for failing to consider alternatives and reliance, and Citizens to Preserve Overton Park v. Volpe for record-based review. Close with SEC v. Chenery Corp. (Chenery I): the action stands or falls on the rationale the agency gave.

This branch is short but must be complete, because it is easy to award points on and easy to skip. Nondelegation: J.W. Hampton's intelligible principle, effectively unchanged by Whitman v. American Trucking Ass'ns, with a note that the Gundy v. United States plurality drew four votes for a narrower approach. Legislative veto: INS v. Chadha.

Appointments: Buckley v. Valeo on who counts as an officer, Edmond v. United States and United States v. Arthrex on the principal/inferior line turning on supervision and directability, and Lucia v. SEC holding SEC ALJs are inferior officers. Removal: Humphrey's Executor for multimember bodies, Morrison v. Olson for inferior officers, Free Enterprise Fund for dual for-cause layers, Seila Law for single-director agencies with substantial executive power, and Collins v. Yellen on remedy — an unconstitutional removal restriction does not automatically void the officer's actions.

Add agency adjudication and Article III: SEC v. Jarkesy (2024) holds that when the SEC seeks civil penalties for securities fraud, the Seventh Amendment requires a jury trial in an Article III court, because the claim is legal in nature and the public rights exception does not apply. Add Axon Enterprise v. FTC for the proposition that structural constitutional challenges may bypass the agency's review scheme and go straight to district court — that one belongs in your reviewability branch too, cross-referenced.

Run three old fact patterns using only the attack outline, timed. Every place you had to stop and think about which branch to enter is a structural defect — fix the routing, not the content. Every place you needed a rule the outline did not carry is a content gap. Two or three iterations is usually enough.

Format matters. Closed-book means the outline is a memorization scaffold, so use mnemonics and keep it to four pages. Open-book means it should have a one-page index on top with page numbers, so you can jump. If your professor allows only handwritten notes or limits pages, build to the constraint from the start rather than compressing at the end.

On citation form: in exam prose, case name and year is enough — "State Farm (1983)" — unless your professor says otherwise. If the course requires a memo or brief, use full Bluebook forms: statutes as 5 U.S.C. § 706(2)(A), regulations as 40 C.F.R. § 52.21, proposed and final rules as 89 Fed. Reg. 1234 (Jan. 5, 2024) (to be codified at 40 C.F.R. pt. 52), and cases as Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024). Do not cite a page number you are not sure of; cite the case by name and year instead.

Finally, ask your professor which framing they want for deference. Post-Loper Bright, courses differ on how much Chevron history to preserve, and some professors still test Chevron's two steps as background to explain what changed. That is a genuine school-to-school variation, and it drives whether your deference section is one page or three.