Advertise on podcast: Administrative Remedies

On a Wednesday morning in January, an immigration judge in Annandale, Virginia was running a hearing with a dozen attorneys appearing remotely for detained clients. Her supervisor opened the courtroom door and asked whether she was on the record. She called a recess. In the hallway, she was told she was off the detained docket immediately. She asked why, and he said he didn’t know. She asked whether she could finish the hearing, and he said no. Her replacement was already waiting. Karen Donoso Stevens had been an immigration judge for nearly a decade, and a Homeland Security prosecutor for a decade before that. She denied most of the asylum claims she heard and met her deadlines. Months earlier, on her account, a supervisor told her that unnamed higher-ups wanted four bond decisions she had already issued reversed, and that she was to vacate them herself, on her own motion, because the government would not be filing anything. Gwen and Marc set that account aside from the question of who’s telling the truth and look at the structure it happened inside. Almost everything in this story runs on powers nobody disputes the executive branch has. Immigration judges are appointed by the attorney general, work in the Justice Department, and have none of the protections administrative law judges nominally hold. Their dockets can be reassigned. The Board of Immigration Appeals can reverse them. The attorney general can refer a case to himself and decide it. There is one thing that structure is not supposed to permit, and it is the thing alleged here. In this episode: The regulation that calls immigration judges delegates of the attorney general and tells them to exercise independent judgment in the same breathWhy that isn’t a contradiction, and where the difference lies between exercising authority through the system and reaching into a pending caseMarcello v. Bonds, and the Court’s acceptance of adjudication inside the attorney general’s chain of authorityAccardi v. Shaughnessy: the attorney general denies himself the right to sidestep the board while his own regulations remain in forceWhy Accardi reaches well past immigration, and what it doesn’t stop the attorney general from doing tomorrowAccardi’s epilogue, where the doctrine survived and the man was still deportedThe scale: roughly 700 judges, about as many as there are federal district judgeships, a backlog past four million cases, and no appointed counselWhy a credibility finding is often the entire asylum caseMatter of Yajure Hurtado and what it changed about bond for people who entered without inspectionThe gap Donoso Stevens thought the board had left open, and whether disagreement about an undiscussed category is defianceCase completion quotas, performance reviews, and the cumulative pressure problem from last season, against something far more directA termination letter citing Article II, nine months before pension eligibility, a month after the highest available performance ratingThe Article I immigration court proposal, the case against it, and why the votes aren’t thereSmaller fixes, including the one that costs nobody any authority: put any direction about a pending case in writing, in the file Source: This episode discusses reporting by the Washington Post: “How a respected U.S. immigration judge lost her job under Trump,” October 25, 2026. https://www.washingtonpost.com/nation/2026/10/25/how-respected-us-immigration-judge-lost-her-job-under-trump/

n the early 1980s, prisoners on death row in Texas and Oklahoma petitioned the FDA. Their argument was statutory, not constitutional: the drugs their states planned to use for execution had been approved for ordinary medical purposes and never tested for killing anyone, they would be administered by untrained personnel, and that made them misbranded under the Food, Drug and Cosmetic Act. They asked the FDA to label the drugs, warn prison administrators, and seize supplies. The commissioner declined, saying enforcement was indicated only where there was a serious danger to public health or a blatant scheme to defraud, and that state lethal injection laws were neither. In 1985, the Supreme Court held that an agency’s decision not to enforce is presumed to be beyond judicial review. It never reached whether the FDA had authority over execution drugs at all. This closes out the threshold series, and it’s different from the five doctrines that came before it. Standing, finality, ripeness, exhaustion, and mootness all ask about who sued, how far the agency got, or when the case arrived. This one says the decision isn’t a court’s to review no matter who asks or when. It comes out of one sentence in APA section 701, with two exceptions inside it: one Congress writes into particular statutes, and one the courts built out of the phrase “no law to apply.” In this episode: Rehnquist’s three reasons in Heckler v. Chaney, and why the prosecutor analogy does most of the workJustice Marshall’s concurrence, that easy cases make bad law and the presumption was created out of whole clothWhat it takes to rebut the presumption: Dunlop v. Bachowski, mandatory verbs, and defined triggersThe abdication footnote, Adams v. Richardson, and the difference between prioritizing and announcing you’ll never enforceWhy the DACA rescission wasn’t unreviewable non-enforcement, and what that says about announcing versus quietly stoppingCook v. FDA, where the import statute supplied the guidelines the main opinion said could rebut the presumptionWebster v. Doe and the clear-statement floor under colorable constitutional claimsPreclusion by statute, implied preclusion in Block v. Community Nutrition, and the pullback in Bowen v. Michigan AcademyPatel v. Garland: a checked box on a Georgia driver’s license form, a factual finding resting on a mistake about Georgia law, and no court that can revisit itThe odd alignment in Patel, where the Solicitor General argued for the narrower bar and the Court had to appoint counsel to defend the judgmentWhat counts as a question of law after Wilkinson v. Garland, and the other direction in the visa revocation caseNoem v. Doe on temporary protected status, and the principle that subsidiary determinations merge into an unreviewable final actionA Tenth Circuit panel that vacated its own judgment to see whether it had been overtaken, looked again, and came out the same wayPolitical question as the constitutional cousin of “no law to apply,” and why it rarely decides an administrative caseWhether prosecutorial discretion scales, since declining to enforce a standard at all is a practical repeal for everyone it covered

In 1963, the FDA told drug manufacturers that every time they used a brand name, on a label, in an ad, anywhere, they had to print the generic name alongside it. The rule took effect the day it was published. Thirty-seven manufacturers, making more than 90 percent of the country's prescription drugs, thought it exceeded what Congress had authorized. Their options were to reprint everything they owned, or keep their labels and wait to be prosecuted for selling misbranded drugs. For companies whose business depends on physicians trusting them, that second option isn't really an option. Abbott Laboratories v. Gardner let them sue before anyone enforced anything, and it reshaped how rules get challenged. Gwen and Marc use it to open up the three doctrines that decide when an administrative case can be heard: ripeness for too early, exhaustion for not yet through the agency's own process, and mootness for too late. The doctrines come from different places, and that turns out to matter more than whether they sound strict. A rule the courts built, the courts can bend. A rule Congress wrote comes with only the exceptions Congress wrote into it. And a constitutional limit can't be forgiven by anyone. In this episode: Abbott Labs and the presumption that judicial review isn't cut off without clear and convincing evidence that Congress meant to cut it offHarlan's two-part ripeness test, fitness and hardship, and why Toilet Goods Association came out the other way on the same dayNational Park Hospitality Association, and why a purely legal question still isn't automatically fit for reviewWhy ripeness barely bites anymore, and the Court's open doubts about whether prudential ripeness survivesAPA section 704, Darby v. Cisneros, and why courts can't impose exhaustion requirements of their ownMandatory versus jurisdictional after Santos-Zacaria, and why the difference decides whether a missed deadline can ever be excusedA disability claimant whose appeal goes in two weeks late, and what Smith v. Berryhill gets herIssue exhaustion for regulated companies, and why an objection nobody raised during the comment period may be forfeitedMcCarthy v. Madigan, the Prison Litigation Reform Act, and Ross v. Blake: the clearest wall in the whole area, and the one word Congress left as an exceptionVoluntary cessation: the Sacketts, FBI v. Fikre, and why a promise hedged with "based on currently available information" isn't enoughAl Otro Lado, where keeping the case alive is what let the government win itCapable of repetition yet evading review, which started with railroad rates in 1911, and the same-party requirement people forgetA detention case that died in September because a governor issued a pardon three days before a brief was dueThe real asymmetry: not that the doctrines are rigged, but that one side is in these cases every day and goes looking for another vehicle when one failsNext week: the decisions courts say aren't reviewable at all, with Heckler v. Chaney and Patel v. Garland.

In 2012, the EEOC issued enforcement guidance on how employers should handle arrest and conviction records. It wasn't a regulation. It never went through notice and comment. It said an across-the-board exclusion of everyone with a record screens people out by race and national origin, and that employers generally can't justify it. Texas barred convicted felons from every job in its Department of Public Safety, and parts of Texas law forbid exactly the individualized assessment the guidance treated as the safe route. So Texas sued the EEOC before anyone could sue Texas. The government's first answer wasn't that the guidance was correct. It was that nobody could challenge it, because guidance isn't final agency action, and because the EEOC lacks authority to issue binding interpretations of Title VII in the first place. That argument bought almost six years and two trips to the Fifth Circuit. Gwen and Marc use the case to work through finality: the gate that stands between having an injury and having something to sue over. The test comes from Bennett v. Spear and asks two things, whether the agency is done and whether legal consequences flow from what it did. Agencies have a lot of control over how both questions get answered, because they decide what the document looks like long before any court applies the test. In this episode: APA section 704, and why preliminary steps can be reviewed only alongside a final actionBennett v. Spear, the Klamath biological opinion, and Scalia's point that a document labeled advice can still function as a commandWhy the final judgment rule is the same instinct but not the same problem, since a trial court can't decline to ever enter judgmentFDA warning letters, Holistic Candlers, and why real pressure isn't the same as legal consequencesSackett v. EPA: a compliance order, $75,000 a day under the government's reading, and the answer to "efficiency of regulation conquers all"Army Corps v. Hawkes, where a determination that ordered nobody to do anything was still final, because a definitive document cuts both waysThe 2019 executive order conceding that non-binding guidance carries an implicit threat of enforcement, while creating no right to challenge itTexas v. EEOC: sovereign interest in a state's own laws, safe harbors, and a document that binds agency staffWhy the Fifth Circuit held the same document can't be final for a private employer and non-final for a stateThe 2011 Dear Colleague letter on campus sexual assault: six years of nationwide effect, withdrawn before any court reached the meritsThe case against all of this, that guidance binds nobody and an employer who ignores it gives up nothing it had a right to doRescission as the mirror image, including the EEOC's June vote on the 1979 affirmative action guidelines and the section 713 good-faith defense that went with themNext week: ripeness, exhaustion, and mootness, three more ways of asking when, with the Sacketts again.

Alison Hoagland has lived on Capitol Hill since 1977, about two miles from the White House. She's an architectural historian, a professor emerita at Michigan Tech, the author of six books on American vernacular architecture, and a former senior historian at the Historic American Buildings Survey, the National Park Service program that documented the White House itself. She's been a member of the National Trust for Historic Preservation for more than forty years and now sits on its board. She swore in a declaration that she walks through Lafayette Square about once a month and will keep doing it. On August 31, in an unsigned 5–4 order, the Supreme Court held that the government is likely to show she has no injury a federal court can hear. The Trust had relied on her declaration and nobody else's, so with her out, two lower-court rulings on whether the president needed Congress's permission to demolish the East Wing go unreviewed, and the ballroom keeps going up. Gwen and Marc work through both halves of the order and the Chief Justice's dissent, then set the case beside Bost v. Illinois State Board of Elections, decided seven months earlier by the same Court, where a congressman who won his last race with 75 percent of the vote did have standing to challenge how ballots are counted. In this episode: The 1912 statute at the center of the case: no structure in federal public grounds in D.C. without express authority of Congress, and whether a maintenance appropriation supplies itWhy the litigation runs against the Park Service at all, and the government's shifting position on who was actually building the ballroomTwo parallel tracks, APA and ultra vires, and what happens when the decisionmaker isn't an agencyValley Forge, FDA v. Alliance for Hippocratic Medicine, and the line between an injury and a strong objectionLaidlaw, Summers, and Lujan: why aesthetic injury counts for a river or a species, and why the majority says a building is differentSierra Club v. Morton to Lujan to this order, and the argument that the bar keeps moving once a plaintiff clears itThe Chief Justice's dissent: the crocodile, Teddy Roosevelt's 1902 message to Congress, and the fact that nobody fishes in the White HouseThe equities half of the order, the security declarations, and the dissent's footnote quoting the government against itselfBost: candidates, the hundred-meter dash, Justice Jackson's dissent on generalized grievances, and Justice Barrett's narrower pocketbook routeWhy the doctrine lets two reasonable justices sort the same plaintiffs in opposite directionsWhat Congress can still do, and why timing may make that academicWhether the difference between losing a river and losing the view of a building is a principled distinction or just the same fact described twice is, more or less, the whole case. Tags: standing, Article III, National Trust for Historic Preservation, White House ballroom, Bost v. Illinois, Lujan, Sierra Club v. Morton, Laidlaw, aesthetic injury, emergency docket, administrative law, Supreme Court

In 1986, Joyce Kelly traveled to Egypt to see the habitat of the endangered Nile crocodile, and she swore she meant to go back. That affidavit was supposed to get Defenders of Wildlife into federal court to challenge the Interior Department's position that the Endangered Species Act stopped at the border. The district court ruled for Defenders, and the Eighth Circuit affirmed. Then in Lujan v. Defenders of Wildlife, the Supreme Court held that the courts never had the power to decide the question at all. She had no plane ticket and no date. Someday isn't imminent. This episode is about the first door: standing, the doctrine that decides who gets to ask the question. Gwen and Marc walk through the three elements every law student memorizes, which are injury in fact, causation, and redressability. They also trace where those elements came from. The Court calls them the "irreducible constitutional minimum," but they aren't in Article III. They were assembled mostly between 1970 and 1992, from Sierra Club v. Morton to Lujan. The doctrine is younger than many of the judges applying it, and it overrode an act of Congress that said "any person" may sue. We cover Simon v. Eastern Kentucky Welfare Rights Organization and the problem of harm that runs through a third party. We look at TransUnion v. Ramirez and the 6,300 people whose false terrorist-watchlist flags never left the company's files. We explain why a bare statutory violation isn't automatically a concrete injury, and what Summers v. Earth Island Institute means for procedural plaintiffs. On the administrative law layer, we cover APA § 702 and the zone-of-interests test, Bennett v. Spear, Lexmark's retirement of "prudential standing," associational standing under Hunt, and the special solicitude Massachusetts v. EPA extended to states. Property law turns out to have been sorting beneficiaries from enforcers for centuries. Then we put it all together. A factory told to install pollution controls has standing without arguing for it. The person downstream who got sick when EPA didn't enforce has a real injury and a much harder path in. None of these doctrines asks whether the agency was wrong. They sort on the shape of the harm: how concentrated it is, how traceable, and how reliably a court order would fix it. That sorting has a pattern. Loper Bright, West Virginia v. EPA, Sackett, and Jarkesy were all brought by regulated parties. Calling the doctrine rigged would be wrong. Calling it neutral is nearly as wrong. Plus: the White House East Wing litigation, where the Court stayed an injunction after finding a D.C. preservationist likely lacked standing. The Chief Justice dissented and went straight back to the crocodile. Next week: the ballroom case in full, and a January standing decision that went the other way. Then finality.

Two executive orders came from the same desk, and the government gave the same defense for both: the challenge came too early, because nothing had happened to anyone yet. The argument lost against the birthright citizenship order, and it has lost every time since. Against the mail voting order, the Supreme Court accepted it and held that the district court never had jurisdiction. The difference was the drafting. The birthright order commands agencies to reach a result: no department or agency shall issue citizenship documents. The mail voting order commands the Postal Service to run a rulemaking, and a rulemaking's outcome has to stay open. Gwen and Marc trace how that choice of wording decides which body of law tests a policy, when the clock starts, who can get into court, and what a loss actually costs. In this episode: Why the birthright order was ready for review immediately, while agency guidance under it would have arrived slowly, agency by agencyBiden v. Nebraska, where a presidential policy was carried out by the Secretary of Education, and why losing on the statute left the Higher Education Act as a second routeA "ladder" of losses: procedural, statutory, and constitutional, and why only the constitutional one takes a policy off the table for goodConstitutional avoidance: the interpretive canon, the separate habit of not deciding things, and why avoidance works poorly when a question isn't closeTrump v. Barbara: a 6–3 affirmance on two different grounds, and Justice Kavanaugh's statutory route, which would have left a door open for CongressThe second birthright order, enjoined before its implementing guidance existedExecutive Order 14399 on mail voting: deadlines attached to proposals, a Supreme Court stay for prematurity, a final rule issued three days before that ruling, and then the government arguing the challengers were too lateThe twist: the Postal Service is exempt from ordinary APA review, so the case the plaintiffs were told to wait for may not be available to themThe lesson isn't that one instrument protects a policy. Each one buys time and changes which question a court gets to first. The real drafting question is which loss you can live with. Recorded September 11, 2026. These cases are moving quickly, and developments after that date aren't reflected here. Hosted by Professor Gwendolyn Savitz and Dean Marc Roark of the University of Tulsa College of Law. Next week: standing, and who is allowed to challenge agency action at all. Tags/keywords: administrative law, executive orders, birthright citizenship, Trump v. Barbara, Biden v. Nebraska, constitutional avoidance, ripeness, standing, APA, Postal Service, mail voting, Supreme Court, law school

Loper Bright Enterprises sued the Secretary of Commerce, and the case had no trial, no depositions, and no discovery. It was decided on cross-motions for summary judgment, on a record the agency finished building before anyone filed a complaint. That's not an anomaly — that's what challenging agency action looks like. And the herring fishermen were lucky in a way most challengers aren't: they got a court to answer the question they came to ask. Season Three opens by taking apart the picture everyone carries around — agency acts, you object, a judge takes a look and fixes it or doesn't. Error correction. Gwen and Marc walk through why that description is true often enough to be believable and wrong often enough to matter. Underneath it sits a set of doctrines that decide whether the error is ever reached at all: standing, finality, ripeness, exhaustion, mootness, and the categories courts have placed outside review entirely. None of them asks whether the agency was wrong. We read 5 U.S.C. § 706 closely — Congress wrote the grounds. Congress did not write the intensity, and the intensity is where the cases are decided. Section 706 read identically the morning Chevron came down and the morning it was overruled. Also in this episode: why Bennett v. Spear gave three statutory words their content and why agency lawyers draft against it; the asymmetry that makes standing easy for a regulated company and hard for the person downwind; Patel v. Garland and how far a jurisdiction-stripping statute reaches; and where Gwen and Marc disagree about whether any of this is construction or something more deliberate. Plus Sackett, Lucia, and Jarkesy — courts opening doors, not closing them — and what distinguishes the cases that get in. Next week: executive orders, and why the choice between signing one and routing a policy through an agency determines which body of law ever gets to test it.

Fritz Miot has lived in California for fifteen years under Temporary Protected Status, working in an Alzheimer's research lab while managing a diabetes diagnosis that Haiti's collapsed healthcare system can't treat. On June 25, 2026, the Supreme Court told him a federal court isn't allowed to ask whether the government even followed its own rules before ending his status — not whether he loses on the merits, but whether anyone can look at all. In the second half of our emergency two-part season opener, Gwen and Marc dig into Mullin v. Doe: how "no judicial review of any determination" swallowed a mandatory consultation requirement, why the State Department told DHS one thing and the public another, and what Justice Thomas's solo concurrence is quietly setting up for a future case. Slaughter closed the check inside the agency. Mullin closes the one outside it.

For ninety years, Congress could build federal agencies designed to operate at arm's length from the president. Bipartisan commissions, staggered terms, for-cause removal — a whole architecture of institutional independence protected by Humphrey's Executor v. United States (1935). On June 29, 2026, in a 6-3 decision, the Supreme Court overruled Humphrey's by name and said it had been wrong the day it was decided. Rebecca Slaughter, the FTC commissioner Donald Trump fired in March 2025 for reasons that had nothing to do with her job performance, loses her case. So does the general category of "independent agency" as a distinct constitutional structure. In this episode, we walk through what Trump v. Slaughter actually held, why the Court's reasoning goes wider than its holding, and what happens next. We cover the "Decision of 1789" the majority uses as its historical anchor and Justice Sotomayor's response. We examine the stare decisis fight — seven prior Supreme Court cases had blessed the Humphrey's structure by name, including recent decisions the Court had just handed down. We explain the companion case, Trump v. Cook, decided the same morning, which saved the Federal Reserve on historical grounds while everyone else lost the same protection. And we get into the part the majority didn't quite say out loud: the opinion carefully carves out the Fed and Article I courts, but not administrative law judges. What that silence means for the MSPB, for ALJ tenure protection, and for every federal adjudication system in the country is the question the next Supreme Court term will start to answer. Justice Gorsuch's concurrence lays out what he calls the "ratchet effect" — Congress delegated enormous power to agencies on the premise that those agencies would be independent, and that premise is now gone. What Congress thought it bargained for is not what Congress got. Justice Barrett's dissent in Cook asks how history can support both a categorical rule ending independence everywhere and a carveout preserving it at one institution. The Court, she notes, does not say. Read the full explainer at remediespodcast.com/learn-trump-v-slaughter-explained.

A visitation room in a private detention facility in rural Louisiana. Cinderblock walls, fluorescent lights, two plastic chairs bolted to the floor. Elena is on one side. A lawyer who drove three hours from New Orleans is on the other. They have ninety minutes. Elena has eight months until the hearing that decides whether she gets asylum or gets sent back to Honduras. The evidence she needs to prove her case is in a town she can't safely call into, held by family members who might be at risk just from gathering it. In the second part of their immigration deep dive — and the final episode of Season 2 — Gwen and Marc move from the machinery to the person standing inside it: Why representation matters more in immigration than anywhere else in the federal system — represented asylum seekers win at several times the rate of unrepresented ones, and unrepresented detained respondents face denial rates around ninety percentThe Sixth Amendment gap: immigration removal is classified as civil, not criminal, so there's no right to appointed counsel. If you can't find a lawyer and the pro bono organizations are full, you represent yourself against a trained government attorney through an interpreterHow detention structurally degrades the case before the hearing happens — bond decided in a ten-minute video proceeding, evidence that can't be gathered from inside a facility, and the same docket clock running whether you're preparing from a lawyer's office or a monitored phone boothThe constitutional floor that most people assume exists and doesn't: plenary power from the 1889 Chinese Exclusion Case, Thuraissigiam's holding that someone apprehended twenty-five yards inside the border has essentially no due process right to judicial review, and a 2025 executive order testing whether that logic extends to anyone anywhere in the country who can't prove two years of presenceWhy you don't have to disagree with the doctrine to be troubled by the outcomes — strict or generous, the law is supposed to apply the same way to the same facts, and the data says it doesn'tNext season: judicial review. What happens when this finally gets to court — and why it's not the rescue mechanism people imagine.

Two asylum seekers cross the southern border six months apart. Same country, same persecution, same statute. A clerk neither of them has ever met routes one to the regular docket and the other to an expedited docket. One gets heard in eight months. The other is waiting until 2028. That single routing decision is now six years of difference. In the first of a two-part deep dive on immigration adjudication, Gwen and Marc walk through the system that will decide both of their cases: 3.2 million pending cases, 2.3 million of them asylum, divided among roughly 700 immigration judges — about 3,200 cases per judge, each requiring hours of fact-intensive testimony through interpreters about events in another countryThe particular social group doctrine — the single most contested category in asylum law, which has flipped four times in eleven years through Attorney General certification. Same statute, same words, opposite results depending on who holds the office when your case is heardHow the AG cleared 767,000 cases in 2025 — the highest single-year total in the system's history — through a mix of procedural exits, compressed dockets, and doctrinal changes that foreclosed categories of claims that were viable when applicants filedThe MSPB decision that stripped immigration judges of civil service protections, holding they're inferior officers removable at will — meaning the judge deciding whether to grant asylum on a borderline theory now knows she can be fired tomorrow with no notice and no appealThe denial rate climbing from roughly fifty percent in 2023 to eighty percent in February 2026, without a single statutory changeRosa's hearing is in 2028. She'll argue a social group theory the Attorney General has already foreclosed, with six-year-old evidence about a country she fled six years ago, in front of a judge who can be removed at will. The statute hasn't changed. Everything else has. Next episode: the person standing inside the machinery — detention, representation, expedited removal, and why the Constitution doesn't fix any of it.

Until June 2024, the Social Security Administration was denying disability claims on the grounds that applicants could work as pneumatic tube operators — a job that functionally disappeared decades ago. The agency's catalog of occupations hadn't been updated since 1991. Congress noticed. Courts complained. The agency issued two emergency messages. The fix removed 114 jobs nobody was citing anyway and added a documentation requirement to 13 jobs people were. The vocational expert who cited pneumatic tube operator last year can cite it again next year, as long as they say an extra sentence. That catalog isn't the problem. It's the symptom. In this episode, Gwen and Marc walk through what Social Security disability adjudication actually looks like when you run it two and a half million times a year: The five-step sequential evaluation that decides every claim — substantial gainful activity, severity, the listings, past work, and the step five question that wins or loses most cases: is there any other job in the national economy you could do?The grids — literal rows and columns where you plug in age, education, work history, and RFC, and the table tells you whether you're disabled. The Supreme Court blessed this trade in Heckler v. Campbell: uniformity over individualized judgment, explicitly because the alternative would "hinder needlessly an already overburdened agency"The age cliff: at fifty-four, the grids say not disabled. At fifty-five, same medical condition, same RFC, same education, same work history — disabled. The borderline rule that's supposed to soften this treats five months as too far awayHow every structural feature from the trilogy — the forty-six point judge-level variation, the weak internal review, the ex ante control mechanisms — runs through the credibility finding and the RFC, which determine where you land on the gridsThe Matthews problem at scale: the test counts administrative burden as a factor, and at two and a half million applications a year, that factor dominates the math. There's no doctrinal floor on how thin process can get once the system is big enoughPeople die waiting for hearings. The GAO and Social Security's own inspector general have documented it. At a nine-month average wait — an improvement from the fifteen months it was running in 2023 — that number is meaningful every year. That's not the system failing. That's the system working as designed, at volume.

In the final episode of a three-part series on how agencies actually produce outcomes, Gwen and Marc trace the mechanisms that did the work — all of them upstream of the hearing room and mostly invisible from outside: Case completion targets that measure speed but not thoroughness — and a Seventh Circuit concurrence warning that even well-intentioned production pressure "can alter the essential function of adjudication"Quality review rubrics that aren't published, shift with administrations, and can't be appealed — but feed into performance evaluations that have real career consequencesSocial Security Rulings that describe themselves as not having "the force and effect of law" and in the same sentence say they are "binding on all components" — binding policy issued without notice and commentThe 1984 case that saw it coming: Association of Administrative Law Judges v. Heckler, where the court found outcome-based targeting of ALJs "violated the spirit of the APA, if no specific provision thereof"Immigration as the amplified version: 700-case-per-year quotas for judges without APA independence protections, deciding cases where the stakes are deportationSomewhere out there is a claimant with the same diagnosis as someone approved three years ago by the same ALJ in the same office. That claimant gets denied — not because the case changed, but because the system around the judge changed. And nobody outside the agency can see why.

In 2018, Attorney General Jeff Sessions reached into a single immigration case, certified it to himself, and overruled the precedent that had let domestic violence survivors qualify for asylum. In 2021, Garland certified it back. In 2025, Bondy certified it away again. Three reversals on the same doctrine in seven years — no statute changed, no rulemaking happened. One signature each time. In this episode — the second in a three-part series on how agencies actually produce outcomes — Gwen and Marc move up a level from the hearing room and ask whether internal appellate review fixes the variation E13 documented. The textbook answer is yes: that's what review is for. The real answer is that "internal review" describes three very different machines, and none is primarily in the business of producing consistency. The APA baseline is permissive. 5 USC 557 gives the reviewing body "all the powers which it would have in making the initial decision" — including the power to redo credibility findings from a cold transcript. The ALJ who sat three feet from the witness gets reversed; the federal court then defers to the agency under substantial evidence review. The one person who actually saw the testimony is the one whose findings get displaced.Three agencies, three pullbacks. Social Security walks the default back via 20 CFR 404.970 — the Appeals Council acts only on legal error or want of substantial evidence. Immigration walks it back furthest via 8 CFR 1003.1 — IJ fact findings get clear-error review. The SEC operates closest to the default, where "all the powers" actually bites.Model 1 — Social Security: review mostly doesn't happen. The Appeals Council grants review in roughly 15-20% of cases. The rest get a one-paragraph denial. That denial opens the 60-day window to federal court, but reads to a lay claimant like a final rejection — because in ordinary English "denied" means you lost, while in admin-law English it means you've exhausted and the clock is running. Represented claimants know this. Unrepresented claimants miss the window and lose by default.Model 2 — Immigration: review is a policy instrument. The BIA designates decisions as precedential, binding every IJ in the country. Under 8 CFR 1003.1(h), the Attorney General can pull any case — pending or decided — and write a precedential opinion on it. Matter of A-R-C-G (2014), Matter of A-B- (2018), Garland's vacatur (2021), Matter of S-F-M (2025): the same doctrine swung four times across four administrations through a single referral mechanism. That isn't error correction. It's substantive policymaking through the review channel.Model 3 — Enforcement (SEC, FTC, NLRB): review is automatic and institutionally tilted. Every ALJ decision goes up. About three-quarters of SEC reversals favor the enforcement division — finding violations the ALJ didn't or raising penalties the ALJ imposed. The Commission supervises the enforcement staff, sets priorities, allocates their budget, and then sits as the appellate tribunal reviewing whether that same staff proved its case. No bad faith required; the incentives only run one way.Three failure modes, one conclusion: SSA fails by not reviewing; immigration fails by reviewing through a politically directed mechanism whose priorities shift with administrations; enforcement fails by reviewing through a body with an institutional stake in the result. Review corrects some errors and generates precedent where precedent gets designated. What it does not do — anywhere — is pull similar cases toward similar outcomes. Next episode: agencies don't shape outcomes by correcting decisions. They shape the conditions under which decisions get made.