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 off
- Harlan's two-part ripeness test, fitness and hardship, and why Toilet Goods Association came out the other way on the same day
- National Park Hospitality Association, and why a purely legal question still isn't automatically fit for review
- Why ripeness barely bites anymore, and the Court's open doubts about whether prudential ripeness survives
- APA section 704, Darby v. Cisneros, and why courts can't impose exhaustion requirements of their own
- Mandatory versus jurisdictional after Santos-Zacaria, and why the difference decides whether a missed deadline can ever be excused
- A disability claimant whose appeal goes in two weeks late, and what Smith v. Berryhill gets her
- Issue exhaustion for regulated companies, and why an objection nobody raised during the comment period may be forfeited
- McCarthy 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 exception
- Voluntary cessation: the Sacketts, FBI v. Fikre, and why a promise hedged with "based on currently available information" isn't enough
- Al Otro Lado, where keeping the case alive is what let the government win it
- Capable of repetition yet evading review, which started with railroad rates in 1911, and the same-party requirement people forget
- A detention case that died in September because a governor issued a pardon three days before a brief was due
- The 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 fails
Next week: the decisions courts say aren't reviewable at all, with Heckler v. Chaney and Patel v. Garland.
00:00:13 --> 00:00:15 Welcome to Administrative Remedies, because you
00:00:15 --> 00:00:17 can't fix what you don't understand. Brought
00:00:17 --> 00:00:19 to you in part by the University of Tulsa College
00:00:19 --> 00:00:22 of Law's Mabee Legal Information Center. I'm
00:00:22 --> 00:00:24 Gwendolyn Savitz, an associate professor here
00:00:24 --> 00:00:26 at TU. And I'm Marc Roark, a professor at the
00:00:26 --> 00:00:29 College of Law. We'll be breaking down complex
00:00:29 --> 00:00:31 doctrines with real-life analogies and examples
00:00:31 --> 00:00:33 to demystify the world of administrative law
00:00:33 --> 00:00:35 for everyone trying to understand how government
00:00:35 --> 00:00:38 actually works. Agencies are the main way the
00:00:38 --> 00:00:40 federal government gets things done. It's not
00:00:40 --> 00:00:42 through Congress, for reasons we'll be addressing
00:00:42 --> 00:00:47 over the course of the series. Marc, in 1963,
00:00:47 --> 00:00:49 the Food and Drug Administration issued a regulation.
00:00:49 --> 00:00:51 And this regulation required drug manufacturers
00:00:51 --> 00:00:54 to list a drug's generic name every time they
00:00:54 --> 00:00:56 used its brand name. They had to do it on labels,
00:00:56 --> 00:00:58 in advertisements, and in promotional materials.
00:00:59 --> 00:01:01 So every time you said Valium, you had to say
00:01:01 --> 00:01:04 diazepam. Every time. And the rule took effect
00:01:04 --> 00:01:07 immediately, the day it was published. That probably
00:01:07 --> 00:01:10 meant that they had to reprint everything that
00:01:10 --> 00:01:12 the industry then owned. Yes, they had to destroy
00:01:12 --> 00:01:15 everything they currently had. They needed to
00:01:15 --> 00:01:19 buy all new stuff, redo every ad. The FDA's lawyers
00:01:19 --> 00:01:21 told the court that immediate compliance was
00:01:21 --> 00:01:24 expected. Companies making more than 90% of
00:01:24 --> 00:01:26 the country's prescription drugs thought the
00:01:26 --> 00:01:29 every time requirement went beyond what Congress
00:01:29 --> 00:01:32 had authorized. So they had two choices. They
00:01:32 --> 00:01:34 could comply and pay for it. Or they could keep
00:01:34 --> 00:01:36 using their old labels and wait to be prosecuted
00:01:36 --> 00:01:39 for selling misbranded drugs. That really isn't
00:01:39 --> 00:01:42 a choice for a company whose whole business depends
00:01:42 --> 00:01:46 on doctors trusting in its products. So indicted
00:01:46 --> 00:01:48 for misbranding is not a headline a drug company
00:01:48 --> 00:01:52 survives lightly. Right. So they sued. And the
00:01:52 --> 00:01:54 district court agreed with them. The Third Circuit
00:01:54 --> 00:01:56 reversed. It said there was no case because the
00:01:56 --> 00:01:58 FDA hadn't done anything to anyone. And then
00:01:58 --> 00:02:01 in 1967, the Supreme Court sent it back. Abbott
00:02:01 --> 00:02:03 Laboratories v. Gardner. You can challenge a
00:02:03 --> 00:02:05 rule like this before the agency enforces it.
00:02:06 --> 00:02:09 So a pre-enforcement review. And the Court got
00:02:09 --> 00:02:13 there over a timing objection, which sounds a
00:02:13 --> 00:02:15 little like too early. Yes, it actually got there
00:02:15 --> 00:02:17 over two objections. The government's first argument
00:02:17 --> 00:02:19 was that the drug statute itself ruled out this
00:02:19 --> 00:02:22 kind of suit. It set up a special review procedure
00:02:22 --> 00:02:25 for certain other FDA regulations. But Harlan
00:02:25 --> 00:02:28 rejected that with a presumption that still is
00:02:28 --> 00:02:30 incredibly important in the field, that judicial
00:02:30 --> 00:02:33 review of final agency action isn't cut off unless
00:02:33 --> 00:02:34 there's some sort of persuasive reason to believe
00:02:34 --> 00:02:37 that Congress meant to cut it off. We need clear
00:02:37 --> 00:02:39 and convincing evidence. That was the language
00:02:39 --> 00:02:41 the court used. And the statute even had a clause
00:02:41 --> 00:02:43 saying that special remedies were in addition
00:02:43 --> 00:02:47 to any others. So a presumption in favor of review,
00:02:47 --> 00:02:50 that becomes the background rule for whether
00:02:50 --> 00:02:53 agency action can be reviewed at all. And the
00:02:53 --> 00:02:55 timing doctrines then ask whether this particular
00:02:55 --> 00:02:58 case can be heard now. Does that sound right?
00:02:58 --> 00:03:00 Yes. Timing is where a lot of administrative
00:03:00 --> 00:03:04 cases get decided. Too early is ripeness. Ripeness
00:03:04 --> 00:03:06 is largely the courts' own creation, although
00:03:06 --> 00:03:08 it has a constitutional core. Not yet through the
00:03:08 --> 00:03:11 agency's own process is exhaustion. Exhaustion
00:03:11 --> 00:03:13 can come from a statute or an agency rule or
00:03:13 --> 00:03:16 the courts. Too late, meaning the dispute is
00:03:16 --> 00:03:18 over, is mootness. And mootness comes from Article
00:03:18 --> 00:03:21 III. So where a timing rule comes from decides
00:03:21 --> 00:03:22 who has the power to forgive you for missing
00:03:22 --> 00:03:26 it. Let's go back to ripeness. Harlan gave the
00:03:26 --> 00:03:28 two-part test in Abbott Labs, and that's been
00:03:28 --> 00:03:31 the test ever since. He said the problem is best
00:03:31 --> 00:03:34 seen in a twofold aspect, requiring us to evaluate
00:03:34 --> 00:03:37 both the fitness of the issues for judicial decision
00:03:37 --> 00:03:39 and the hardship to the parties of withholding
00:03:39 --> 00:03:42 court consideration. Fitness came easily. Both
00:03:42 --> 00:03:45 sides agreed that this was purely legal. Did
00:03:45 --> 00:03:47 the statute authorize the every time rule? And
00:03:47 --> 00:03:49 the rule was final. It had gone through notice
00:03:49 --> 00:03:51 and comment and it was in force. Hardship came
00:03:51 --> 00:03:54 pretty easily, too. Harlan wrote that the rule
00:03:54 --> 00:03:56 put the companies in exactly the kind of dilemma
00:03:56 --> 00:03:58 that the Declaratory Judgment Act was designed
00:03:58 --> 00:04:00 to relieve. They could either choose to comply
00:04:00 --> 00:04:02 at great cost or they could risk serious criminal
00:04:02 --> 00:04:05 and civil penalties. The government must have
00:04:05 --> 00:04:08 said that expense alone isn't the hardship that
00:04:08 --> 00:04:11 can be enforced. Every regulation costs somebody
00:04:11 --> 00:04:14 some money. It did indeed say that. And the Court's
00:04:14 --> 00:04:17 answer was what kind of expense this was. So
00:04:17 --> 00:04:19 the rule was aimed at these companies specifically.
00:04:19 --> 00:04:21 It required immediate changes to how they ran
00:04:21 --> 00:04:24 their business every day. And noncompliance exposed
00:04:24 --> 00:04:27 them to strong sanctions. Harlan added that drug
00:04:27 --> 00:04:29 companies operate in a sensitive industry where
00:04:29 --> 00:04:32 public confidence matters. So forcing them to
00:04:32 --> 00:04:33 test the rule as defendants in an enforcement
00:04:33 --> 00:04:36 action could harm them severely and unnecessarily.
00:04:36 --> 00:04:38 So they couldn't really just wait and see what
00:04:38 --> 00:04:40 happened. Right. And you can see the difference
00:04:40 --> 00:04:42 clearly in Toilet Goods Association v. Gardner,
00:04:42 --> 00:04:45 which the Court decided the same day. So this
00:04:45 --> 00:04:48 was also the FDA. The FDA had issued a rule under
00:04:48 --> 00:04:50 the color additive amendments that said if a
00:04:50 --> 00:04:53 cosmetics company refused to give FDA inspectors
00:04:53 --> 00:04:55 free access to its facilities and its processes
00:04:55 --> 00:04:58 and its formulas, the Commissioner could suspend
00:04:58 --> 00:05:00 the company's certification for color additives.
00:05:00 --> 00:05:03 The industry said that FDA had no authority to
00:05:03 --> 00:05:05 demand that kind of access. The Court agreed
00:05:05 --> 00:05:08 the question was a legal one and it agreed the
00:05:08 --> 00:05:10 rule was final, but it still held the case wasn't
00:05:10 --> 00:05:13 ripe. So on what ground, if the question was
00:05:13 --> 00:05:16 legal and the rule was final? Here's what the
00:05:16 --> 00:05:18 Court said. At this juncture, we have no idea
00:05:18 --> 00:05:21 whether or when such an inspection will be ordered
00:05:21 --> 00:05:23 and what reasons the Commissioner will give to
00:05:23 --> 00:05:26 justify his order. The rule rested on FDA's general
00:05:26 --> 00:05:29 power to issue regulations for efficient enforcement,
00:05:29 --> 00:05:32 and whether that justified it depended on what
00:05:32 --> 00:05:35 enforcement problems FDA actually faced. This
00:05:35 --> 00:05:36 was something a court could judge better in a
00:05:36 --> 00:05:39 real inspection dispute. And the hardship here
00:05:39 --> 00:05:41 was light. Nobody had to change anything the
00:05:41 --> 00:05:43 day the rule was issued. A company that refused
00:05:43 --> 00:05:45 an inspector faced suspension of its certification,
00:05:45 --> 00:05:47 but it could challenge that when it occurred.
00:05:47 --> 00:05:50 So a purely legal question isn't automatically
00:05:50 --> 00:05:53 a fit. Right. It certainly helps, but by itself,
00:05:53 --> 00:05:56 it isn't enough. The modern example of a purely
00:05:56 --> 00:05:58 legal challenge that still wasn't ripe is National
00:05:58 --> 00:06:01 Park Hospitality Association. This is 2003, so
00:06:02 --> 00:06:04 more recent than the 1970s, but not super, super
00:06:04 --> 00:06:07 recent. Here, the Park Service had issued a regulation
00:06:07 --> 00:06:09 saying that a federal contract disputes statute
00:06:09 --> 00:06:12 didn't apply to concession contracts. Concessioners
00:06:12 --> 00:06:14 challenged it, and the Court said, wait, the
00:06:14 --> 00:06:16 regulation didn't command anyone to do anything,
00:06:16 --> 00:06:18 and a court would do a better job deciding the
00:06:18 --> 00:06:20 question with some sort of concrete contract
00:06:20 --> 00:06:23 dispute. Okay, first of all, I have to say, saying
00:06:23 --> 00:06:26 that 2003 just doesn't feel that recent really
00:06:26 --> 00:06:30 hurts me at my heart. I get that. Though most
00:06:30 --> 00:06:32 rule challenges now get filed the day the rule
00:06:32 --> 00:06:34 is published and nobody argues ripeness at all.
00:06:35 --> 00:06:37 Yeah, this is because Abbott won. So what ripeness
00:06:37 --> 00:06:40 used to screen out now is generally handled by
00:06:40 --> 00:06:43 standing or by finality. It can still affect
00:06:43 --> 00:06:45 interpretive rules or policy statements. But
00:06:45 --> 00:06:48 again, finality still kind of hits that too.
00:06:48 --> 00:06:50 So ripeness has two strands. One is constitutional.
00:06:51 --> 00:06:53 Here it overlaps with standing's requirement
00:06:53 --> 00:06:56 that the injury be actual or imminent. And no
00:06:56 --> 00:06:58 court can waive that. But the other is prudential.
00:06:58 --> 00:07:00 That's the court's own judgment about when to
00:07:00 --> 00:07:03 step in. The prudential strand is the part that's
00:07:03 --> 00:07:06 under pressure. So in 2014, the Court said that
00:07:06 --> 00:07:08 prudential ripeness sits in tension with a federal
00:07:08 --> 00:07:10 court's obligation to decide the cases within
00:07:10 --> 00:07:13 its jurisdiction and openly questioned how much
00:07:13 --> 00:07:16 of it survives. So the part the courts built
00:07:16 --> 00:07:18 is the part the courts are considering giving
00:07:18 --> 00:07:22 back now. So that's ripeness. There's not necessarily
00:07:22 --> 00:07:25 a ton left, but it still exists somewhat. Next,
00:07:25 --> 00:07:28 we go to exhaustion. Exhaustion asks whether
00:07:28 --> 00:07:30 you've used the agency's own process before going
00:07:30 --> 00:07:33 to court. And the reasons for it are real. The
00:07:33 --> 00:07:35 agency might fix its own mistakes. It builds
00:07:35 --> 00:07:37 a record that a court can review. It can apply
00:07:37 --> 00:07:40 its expertise first. And it won't be buried in
00:07:40 --> 00:07:41 lawsuits over something that it might have resolved
00:07:41 --> 00:07:44 on its own anyway. And this should feel common
00:07:44 --> 00:07:46 for lawyers, right? Private law has the same
00:07:46 --> 00:07:50 instinct. Plenty of homeowners' association declarations
00:07:50 --> 00:07:52 require mediation before anybody can sue over
00:07:52 --> 00:07:56 a covenant. So the lawsuit is supposed to be
00:07:56 --> 00:07:58 the last step, not the first step. And the first
00:07:58 --> 00:08:00 question with any sort of exhaustion requirement
00:08:00 --> 00:08:03 will be who wrote it. Was it Congress? Is this
00:08:03 --> 00:08:05 an agency rule? Was this the courts? For regular
00:08:05 --> 00:08:08 APA cases, Congress wrote the answer and it did
00:08:08 --> 00:08:11 it in section 704. Except as otherwise expressly
00:08:11 --> 00:08:14 required by statute, agency action otherwise
00:08:14 --> 00:08:17 final is final for the purposes of this section,
00:08:17 --> 00:08:20 whether or not there has been presented or determined
00:08:20 --> 00:08:24 an application for a declaratory order or for
00:08:24 --> 00:08:27 any form of reconsideration, or unless the agency
00:08:27 --> 00:08:30 otherwise requires by rule and provides that
00:08:30 --> 00:08:34 the action, meanwhile, is inoperative, for an
00:08:34 --> 00:08:38 appeal to superior agency authority. So that
00:08:38 --> 00:08:41 last clause is doing a lot of work. The agency
00:08:41 --> 00:08:44 can require an internal appeal, but only by rule
00:08:44 --> 00:08:46 and only if it holds the decision in suspense
00:08:46 --> 00:08:49 while the appeal is pending. Yes. So this is
00:08:49 --> 00:08:53 Darby v. Cisneros in 1993. A developer sanctioned
00:08:53 --> 00:08:55 by HUD went to court without taking an optional
00:08:55 --> 00:08:57 internal appeal. And the Court said that in an
00:08:57 --> 00:09:00 APA case, that's fine. Courts can't impose
00:09:00 --> 00:09:02 exhaustion requirements on their own. It has
00:09:02 --> 00:09:04 to come from a statute or from an agency rule
00:09:04 --> 00:09:07 that requires the appeal and makes the decision
00:09:07 --> 00:09:10 inoperative while the appeal is occurring. So
00:09:10 --> 00:09:12 the courts had built a general exhaustion doctrine,
00:09:12 --> 00:09:14 and the Court read the APA to take that tool
00:09:14 --> 00:09:17 away from the courts. And so then the harder
00:09:17 --> 00:09:20 cases are the statutory ones. Yes. And the key
00:09:20 --> 00:09:23 distinction there is between mandatory and jurisdictional.
00:09:23 --> 00:09:25 For a long time, statutory exhaustion requirements
00:09:25 --> 00:09:28 were described as jurisdictional. If you miss
00:09:28 --> 00:09:30 one, the court had no power to hear your case
00:09:30 --> 00:09:32 no matter what. But the Court's been moving away
00:09:32 --> 00:09:36 from that. In Santos-Zacaria v. Garland in 2023,
00:09:36 --> 00:09:38 it held that the immigration statute's exhaustion
00:09:38 --> 00:09:41 requirements were not jurisdictional. This was
00:09:41 --> 00:09:44 a claim-processing rule, and that mattered. A
00:09:44 --> 00:09:46 jurisdictional defect can be raised at any time
00:09:46 --> 00:09:48 and can never be excused. A claim-processing
00:09:48 --> 00:09:50 rule can be forfeited by the government if it
00:09:50 --> 00:09:52 doesn't raise it. And it might have exceptions,
00:09:52 --> 00:09:55 too. Okay. And the distinction only matters if
00:09:55 --> 00:09:59 somebody can actually use it. Put it on a person,
00:09:59 --> 00:10:01 for example. All right. So we'll take a Social
00:10:01 --> 00:10:04 Security disability claimant. Rosalind. She's 53.
00:10:05 --> 00:10:07 She's done 19 years as a hospital billing clerk.
00:10:07 --> 00:10:09 She's got a chronic nerve condition in both hands.
00:10:10 --> 00:10:12 The Social Security Act lets her into federal
00:10:12 --> 00:10:14 court only after a final decision is made after
00:10:14 --> 00:10:17 a hearing. And the Social Security Administration's
00:10:17 --> 00:10:19 regulations define the steps that you need to
00:10:19 --> 00:10:21 get there. You need an initial determination,
00:10:21 --> 00:10:23 a reconsideration, a hearing before an administrative
00:10:23 --> 00:10:26 law judge, and then Appeals Council review.
00:10:26 --> 00:10:28 So she's been denied each of the first three.
00:10:28 --> 00:10:31 That means up through the hearing before the
00:10:31 --> 00:10:33 administrative law judge. She has 60 days to
00:10:33 --> 00:10:35 ask the Appeals Council for review, but her lawyer's
00:10:35 --> 00:10:38 office has turnover. Her file slips. The request
00:10:38 --> 00:10:40 goes in two weeks late. And the Appeals Council
00:10:40 --> 00:10:44 dismisses it as untimely. So she's finished.
00:10:44 --> 00:10:47 That's what the lower court said about a similar
00:10:47 --> 00:10:49 person who was in almost exactly that position.
00:10:49 --> 00:10:52 But the Supreme Court unanimously disagreed in
00:10:52 --> 00:10:55 Smith v. Berryhill in 2019. There, an Appeals
00:10:55 --> 00:10:57 Council dismissed for untimeliness after an ALJ
00:10:57 --> 00:11:01 hearing, but the Court determined that that decision
00:11:01 --> 00:11:04 itself was a final decision a federal court could
00:11:04 --> 00:11:06 review. The court could ask whether the Social
00:11:06 --> 00:11:08 Security Administration was right to refuse to
00:11:08 --> 00:11:11 excuse the late filing, even though their own
00:11:11 --> 00:11:15 rules allow extensions for good cause. So Rosalind
00:11:15 --> 00:11:17 gets a judge to look at whether two weeks late
00:11:17 --> 00:11:20 should be forgiven. Right. She does. But the
00:11:20 --> 00:11:22 judge isn't starting fresh. The court asks whether
00:11:22 --> 00:11:24 the Social Security Administration abused its
00:11:24 --> 00:11:27 discretion in refusing to find good cause. And
00:11:27 --> 00:11:29 the agency's own rules give examples. Serious
00:11:29 --> 00:11:32 illness, if the records are destroyed, if there's
00:11:32 --> 00:11:34 misleading information from the agency itself.
00:11:34 --> 00:11:37 None of which is my lawyer's office lost track
00:11:37 --> 00:11:40 of my file. So that's the question the court
00:11:40 --> 00:11:43 would decide. And if she won, it wouldn't automatically
00:11:43 --> 00:11:45 mean she gets benefits. If the court decided
00:11:45 --> 00:11:47 that the Appeals Council should have heard the
00:11:47 --> 00:11:49 appeal, it would send it back to the Social Security
00:11:49 --> 00:11:52 Administration. So it would get her the review
00:11:52 --> 00:11:54 that she missed, but not necessarily more than
00:11:54 --> 00:11:57 that. Which is still more than the phrase jurisdictional
00:11:57 --> 00:11:59 would have left her. Yes. Social Security exhaustion
00:11:59 --> 00:12:03 has been softer for a while. Since Mathews v.
00:12:03 --> 00:12:04 Eldridge, the only thing that can't be waived
00:12:04 --> 00:12:07 is presenting the claim to the Social Security
00:12:07 --> 00:12:09 Administration in the first place. The rest
00:12:09 --> 00:12:11 can be excused, and the Court has excused it.
00:12:11 --> 00:12:13 It's also repeatedly refused to make claimants
00:12:13 --> 00:12:16 raise specific issues before the agency. And
00:12:16 --> 00:12:19 regulated companies, they don't go through an
00:12:19 --> 00:12:21 Appeals Council. Yeah, their version is different.
00:12:21 --> 00:12:24 When a company challenges a rule, some courts,
00:12:24 --> 00:12:26 like the D.C. Circuit, hold that an objection
00:12:26 --> 00:12:28 generally can't be raised in court unless someone
00:12:28 --> 00:12:31 raised it during the comment period. That doesn't
00:12:31 --> 00:12:32 necessarily have to be the company that's now
00:12:32 --> 00:12:35 suing. It just needs to be someone. That's issue
00:12:35 --> 00:12:37 exhaustion, and it rewards whoever had lawyers
00:12:37 --> 00:12:39 who were carefully watching the docket. Show
00:12:39 --> 00:12:41 up in court with an objection nobody put before
00:12:41 --> 00:12:43 the agency, and you may find it forfeited. Then
00:12:43 --> 00:12:46 statutory exhaustion isn't the wall I'd assumed.
00:12:47 --> 00:12:50 Where exactly is the wall? So the clearest is
00:12:50 --> 00:12:53 prisons. So McCarthy v. Madigan in 1992, a federal
00:12:53 --> 00:12:56 prisoner sued prison officials for money damages.
00:12:56 --> 00:12:58 The Bureau of Prisons had a grievance process,
00:12:58 --> 00:13:00 but it couldn't award money under that process.
00:13:00 --> 00:13:02 No statute required exhaustion, so the Court
00:13:02 --> 00:13:05 applied judge-made doctrine and identified circumstances
00:13:05 --> 00:13:08 where exhaustion shouldn't be required, like
00:13:08 --> 00:13:09 where it would unduly prejudice a later court
00:13:09 --> 00:13:12 action, or where the agency can't grant effective
00:13:12 --> 00:13:14 relief, or where the agency is biased or has
00:13:14 --> 00:13:16 already decided the answer. The prisoner didn't
00:13:16 --> 00:13:20 have to exhaust. And Congress answered. It did,
00:13:20 --> 00:13:22 four years later in the Prison Litigation Reform
00:13:22 --> 00:13:26 Act. Okay, and in that Act, Congress said no
00:13:26 --> 00:13:28 action shall be brought with respect to prison
00:13:28 --> 00:13:32 conditions under section 1983 of this title or
00:13:32 --> 00:13:34 any other federal law by a prisoner confined
00:13:34 --> 00:13:36 in any jail, prison or other correctional facility
00:13:36 --> 00:13:40 until such administrative remedies as are available
00:13:40 --> 00:13:43 are exhausted. So no action shall be brought
00:13:43 --> 00:13:47 is mandatory on its face. Yeah. And the Court
00:13:47 --> 00:13:51 has read it that way. In 2001, it held that prisoners
00:13:51 --> 00:13:53 must exhaust even when the grievance process
00:13:53 --> 00:13:55 can't give them the money they're seeking. That's
00:13:55 --> 00:13:58 the exact situation in which McCarthy had excused
00:13:58 --> 00:14:01 exhaustion. And then in 2006, it held that exhaustion
00:14:01 --> 00:14:03 means proper exhaustion, including deadlines.
00:14:04 --> 00:14:07 So here we had a California prisoner who filed
00:14:07 --> 00:14:09 his grievance roughly six months after the prison's
00:14:09 --> 00:14:12 deadline had passed, and it was rejected as untimely.
00:14:12 --> 00:14:15 That wasn't exhaustion, so his federal case couldn't
00:14:15 --> 00:14:18 go forward. Okay, that's the two weeks late scenario.
00:14:19 --> 00:14:22 Except it's a prisoner, not a disability claimant.
00:14:22 --> 00:14:25 Yes, and Ross v. Blake in 2016 shows how far
00:14:25 --> 00:14:28 that can go. So here we had a Maryland prisoner
00:14:28 --> 00:14:30 who was assaulted by a corrections officer. The
00:14:30 --> 00:14:32 state's internal investigations unit looked into
00:14:32 --> 00:14:34 it, and Blake assumed that was enough. He didn't
00:14:34 --> 00:14:37 think he also needed to file a grievance. That's
00:14:37 --> 00:14:39 a reasonable mistake. The state investigated
00:14:39 --> 00:14:42 the thing he would have complained about. Yeah,
00:14:42 --> 00:14:44 the Fourth Circuit thought so, too, but a unanimous
00:14:44 --> 00:14:47 Supreme Court reversed. Judges can't add special
00:14:47 --> 00:14:49 circumstances exceptions to the statute. The
00:14:49 --> 00:14:51 only exception is the one Congress wrote, the
00:14:51 --> 00:14:54 word available. And the Court said a process
00:14:54 --> 00:14:57 is unavailable if it's a dead end in practice,
00:14:57 --> 00:14:59 when it's so opaque no ordinary prisoner could
00:14:59 --> 00:15:02 use it, or when officials thwart prisoners through
00:15:02 --> 00:15:04 machination, misrepresentation or intimidation.
00:15:05 --> 00:15:09 So the difference between McCarthy and Ngo isn't
00:15:09 --> 00:15:12 that prisoner grievances got better. It's who
00:15:12 --> 00:15:15 wrote the rule. Yes. And there's one more doctrine
00:15:15 --> 00:15:17 we have to hit, and that's mootness. Article
00:15:17 --> 00:15:20 III requires a live controversy at every stage.
00:15:20 --> 00:15:23 Ordinarily, if the controversy ends, like the
00:15:23 --> 00:15:25 plaintiff got what they wanted, or the policy
00:15:25 --> 00:15:28 was withdrawn, or the person was released, the
00:15:28 --> 00:15:31 case ends too. Which means a defendant can sometimes
00:15:31 --> 00:15:34 end a case by changing its behavior. Or they
00:15:34 --> 00:15:36 can try. So if we go back to the Sacketts in
00:15:36 --> 00:15:39 2020, while their appeal is being briefed, EPA
00:15:39 --> 00:15:41 withdrew the compliance order that had previously
00:15:41 --> 00:15:44 issued to them in a two-paragraph letter and argued
00:15:44 --> 00:15:46 that the case was now over. But the Ninth Circuit
00:15:46 --> 00:15:48 said it wasn't because nothing in the letter
00:15:48 --> 00:15:51 bound a future EPA not to come back. And refusing
00:15:51 --> 00:15:53 to call it moot is the only reason there was
00:15:53 --> 00:15:57 ever a Sackett decision in 2023. Yes. So the
00:15:57 --> 00:15:59 Ninth Circuit ruled for the EPA on the merits
00:15:59 --> 00:16:02 anyway, but the case stayed alive. And then three
00:16:02 --> 00:16:04 years later, the Supreme Court decided that wetlands
00:16:04 --> 00:16:07 question in the Sacketts' favor. That's voluntary
00:16:07 --> 00:16:09 cessation. Stopping the conduct doesn't end the
00:16:09 --> 00:16:12 case unless you can show you won't start again.
00:16:13 --> 00:16:15 And the standard is demanding. The defendant
00:16:15 --> 00:16:17 has to make it absolutely clear the conduct could
00:16:17 --> 00:16:19 not reasonably be expected to recur. In 2024,
00:16:20 --> 00:16:22 Yonas Fikre, an American citizen, sued over
00:16:22 --> 00:16:25 his placement on the No Fly List. After he sued,
00:16:25 --> 00:16:27 the government took him off and then filed a
00:16:27 --> 00:16:29 declaration saying he wouldn't be put back on
00:16:29 --> 00:16:32 based on currently available information. A unanimous
00:16:32 --> 00:16:34 Court said that wasn't enough. The government
00:16:34 --> 00:16:36 bears a formidable burden, the same as any defendant,
00:16:36 --> 00:16:38 and a promise hedged with currently available
00:16:38 --> 00:16:42 information doesn't cover it. So based on currently
00:16:42 --> 00:16:44 available information, that isn't a promise.
00:16:44 --> 00:16:47 That's a lawyer leaving a door open, though I
00:16:47 --> 00:16:49 assume the government said it couldn't say more
00:16:49 --> 00:16:52 without revealing why he was listed on the No
00:16:52 --> 00:16:54 Fly List in the first place. It sure did.
00:16:54 --> 00:16:56 And the Court said only that classified material
00:16:56 --> 00:16:59 has to be handled carefully. And two justices
00:16:59 --> 00:17:01 wrote separately to add that the government can't
00:17:01 --> 00:17:03 be required to disclose it, that other evidence
00:17:03 --> 00:17:06 might be enough. But either way, the government
00:17:06 --> 00:17:09 hadn't made the showing here. Some courts give
00:17:09 --> 00:17:11 government defendants more benefit of the doubt,
00:17:12 --> 00:17:14 don't they? The Fifth Circuit does. Its precedent
00:17:14 --> 00:17:17 gives the government a lighter burden. It acts
00:17:17 --> 00:17:19 on a presumption of good faith, which is a little
00:17:19 --> 00:17:21 hard to square with Fikre. All of that predates
00:17:21 --> 00:17:24 this term, though, where the Court took up voluntary
00:17:24 --> 00:17:27 cessation again. This is Mullin v. Al Otro Lado,
00:17:28 --> 00:17:31 the metering case. Starting in 2016, border officials
00:17:31 --> 00:17:33 told asylum seekers arriving at the ports of
00:17:33 --> 00:17:36 entry to wait on the Mexican side. This practice
00:17:36 --> 00:17:39 was rescinded in 2021, but the challengers had
00:17:39 --> 00:17:41 already won a class-wide declaratory judgment
00:17:41 --> 00:17:44 that it was unlawful. So the policy is gone.
00:17:45 --> 00:17:47 I'd expect the government to be happy to let
00:17:47 --> 00:17:51 that lie. It was not. So in 2025, it asked the
00:17:51 --> 00:17:53 Supreme Court to review the judgment. And the
00:17:53 --> 00:17:55 Court held that the case wasn't moot, that the
00:17:55 --> 00:17:58 declaratory judgment still barred metering inside
00:17:58 --> 00:18:00 the Ninth Circuit, and the challengers hadn't
00:18:00 --> 00:18:01 shown it was absolutely clear the government
00:18:01 --> 00:18:04 wouldn't bring metering back. And then it ruled
00:18:04 --> 00:18:06 for the government on the merits, six to three.
00:18:06 --> 00:18:09 The rescission is what kept the case alive, and
00:18:09 --> 00:18:12 the case is what got the judgment erased. Yes,
00:18:12 --> 00:18:14 Jackson made this point in dissent. The government
00:18:14 --> 00:18:16 had said only it wanted to keep the option of
00:18:16 --> 00:18:18 reviving the practice. And on her account, the
00:18:18 --> 00:18:20 Court had given mixed signals about how likely
00:18:20 --> 00:18:24 recurrence had to be. That's an odd use of the
00:18:24 --> 00:18:26 rule, though. And then there's the exception
00:18:26 --> 00:18:29 every lawyer remembers from Roe, capable
00:18:29 --> 00:18:32 of repetition, yet evading review. This was
00:18:32 --> 00:18:36 actually originally an agency case from 1911,
00:18:37 --> 00:18:40 Southern Pacific Terminal v. ICC. Here, there
00:18:40 --> 00:18:42 were Commission orders that expired before anyone
00:18:42 --> 00:18:44 could get them reviewed. The modern test has
00:18:44 --> 00:18:46 two parts, that the action is too short to be
00:18:46 --> 00:18:48 fully litigated, and there's a reasonable expectation
00:18:48 --> 00:18:50 that the same complaining party will face it
00:18:50 --> 00:18:54 again. So the exception everyone associates with
00:18:54 --> 00:18:57 Roe started with railroad rates. Yes. These were
00:18:57 --> 00:18:59 short-lived orders and a slow docket. But the
00:18:59 --> 00:19:03 problem hasn't changed much. And same party is
00:19:03 --> 00:19:06 the part people forget. It's doing all the work.
00:19:06 --> 00:19:09 It isn't enough that the policy will hit somebody
00:19:09 --> 00:19:12 else next year. Yeah. So a pregnancy fits because
00:19:12 --> 00:19:14 people can definitely get pregnant again. Most
00:19:14 --> 00:19:17 detention cases don't because a released detainee
00:19:17 --> 00:19:19 usually can't show that that particular person
00:19:19 --> 00:19:22 will be detained again. Those survive through
00:19:22 --> 00:19:24 a different way, as class actions, where a claim
00:19:24 --> 00:19:27 that's inherently transitory can continue after
00:19:27 --> 00:19:29 the named plaintiff's own claim ends. Okay, so
00:19:29 --> 00:19:31 the cleanest illustration that I can think of
00:19:31 --> 00:19:35 just happened mere weeks ago. Yes, it happened
00:19:35 --> 00:19:38 in September. This was a case called Genalo v.
00:19:38 --> 00:19:41 Black. Two men were held under a mandatory detention
00:19:41 --> 00:19:43 provision. These were noncitizens, which included
00:19:43 --> 00:19:46 green card holders, who had been convicted of
00:19:46 --> 00:19:48 certain crimes and were being held without bond
00:19:48 --> 00:19:51 while removal proceedings ran. So the Second
00:19:51 --> 00:19:53 Circuit held their detention had gone on so long
00:19:53 --> 00:19:55 that due process entitled them to a bond hearing.
00:19:55 --> 00:19:57 And the government asked the Supreme Court to
00:19:57 --> 00:20:00 review that. OK, so mandatory means no bond hearing
00:20:00 --> 00:20:03 at all, which is the part people find hard to
00:20:03 --> 00:20:06 believe. You can be held on the strength of a
00:20:06 --> 00:20:09 conviction you already served the time for. Yeah.
00:20:09 --> 00:20:12 So one of them, Keisy G.M., was a lawful permanent
00:20:12 --> 00:20:15 resident with five U.S. citizen children. He
00:20:15 --> 00:20:17 spent 21 months in immigration detention without
00:20:17 --> 00:20:19 ever getting a hearing on whether he was a flight
00:20:19 --> 00:20:22 risk or a danger. And this was based on a 2015
00:20:22 --> 00:20:24 assault conviction that he'd already served all
00:20:24 --> 00:20:28 his time for. ICE released him in 2022 under
00:20:28 --> 00:20:30 a court order requiring custody reviews during
00:20:30 --> 00:20:32 the pandemic. And in that review, ICE itself
00:20:32 --> 00:20:34 concluded that he was neither a security risk
00:20:34 --> 00:20:37 nor a risk of absconding. So by the time this
00:20:37 --> 00:20:39 reached the Supreme Court, he wasn't even in
00:20:39 --> 00:20:42 custody. Right, which is why mootness was in
00:20:42 --> 00:20:44 the case from the start. The other man who'd
00:20:44 --> 00:20:45 been part of this case had already had a final
00:20:45 --> 00:20:47 removal order and he'd left the country for good.
00:20:48 --> 00:20:50 So his half was clearly moot. G.M. here was
00:20:50 --> 00:20:52 out, but he had a pardon application pending
00:20:52 --> 00:20:55 in New York. When the Court granted review on
00:20:55 --> 00:20:58 June 15th, it took only his case and told the
00:20:58 --> 00:21:00 parties to brief whether his claim was moot along
00:21:00 --> 00:21:02 with the merits. Argument was set for October
00:21:02 --> 00:21:06 13th, and then on August 28th the governor granted
00:21:06 --> 00:21:09 a pardon, one of six she issued that day. Here
00:21:09 --> 00:21:11 is the letter that was sent to the Court after
00:21:11 --> 00:21:13 that. Three calendar days before G.M.'s merits
00:21:13 --> 00:21:16 brief was due, however, the Governor of New York
00:21:16 --> 00:21:18 granted a full and unconditional pardon with
00:21:18 --> 00:21:20 respect to G.M.'s aggravated-felony conviction.
00:21:20 --> 00:21:23 And as a result of the pardon, G.M. would no longer
00:21:23 --> 00:21:26 be subject to such detention. The government
00:21:26 --> 00:21:28 therefore agrees with G.M. that his habeas case
00:21:28 --> 00:21:31 is now moot. So this case was granted. There
00:21:31 --> 00:21:34 were amicus briefs filed, argument on the calendar,
00:21:34 --> 00:21:37 and now gone. And not because the agency did
00:21:37 --> 00:21:39 anything. Right. And that's worth stressing.
00:21:40 --> 00:21:42 Here, the federal government didn't control the
00:21:42 --> 00:21:44 timing. A state governor did. The case was dismissed
00:21:44 --> 00:21:48 on September 11th. Why didn't capable of repetition
00:21:48 --> 00:21:51 save it? Mandatory detention isn't going away.
00:21:51 --> 00:21:54 So the issue here is with the same-party rule. It
00:21:54 --> 00:21:56 is. It's unquestionably the case that the issue
00:21:56 --> 00:21:58 is going to come up again, but it will come up
00:21:58 --> 00:22:01 for other detainees. Under the immigration statute,
00:22:01 --> 00:22:03 a full and unconditional pardon from a governor
00:22:03 --> 00:22:05 means that the conviction can no longer make
00:22:05 --> 00:22:07 this guy deportable. And if he isn't deportable
00:22:07 --> 00:22:10 on that ground, the mandatory detention provision
00:22:10 --> 00:22:12 doesn't reach him. So there was no reasonable
00:22:12 --> 00:22:14 expectation that he would face the same detention
00:22:14 --> 00:22:17 again. So the question survives, but his case
00:22:17 --> 00:22:21 doesn't. And the question itself, the government
00:22:21 --> 00:22:24 wanted it answered. And it still does. They've
00:22:24 --> 00:22:26 lost this vehicle, but the Solicitor General
00:22:26 --> 00:22:27 told the Court it would look for other cases
00:22:27 --> 00:22:30 to bring the same question back during this term.
00:22:30 --> 00:22:32 That's in the same letter that they sent. Okay,
00:22:32 --> 00:22:34 let's put the three of these things together.
00:22:35 --> 00:22:38 So, not too early, not before you've used the
00:22:38 --> 00:22:41 agency's process, but not after the dispute has
00:22:41 --> 00:22:45 ended. Right. And the size of this varies enormously.
00:22:45 --> 00:22:47 For a regulated industry challenging a final
00:22:47 --> 00:22:50 rule, Abbott made this really wide. For a benefits
00:22:50 --> 00:22:52 claim, it depends on deadlines inside a multi-
00:22:52 --> 00:22:55 step process. And for a detainee, the window
00:22:55 --> 00:22:58 can close with a release. OK, but these doctrines
00:22:58 --> 00:23:01 have real purposes. Ripeness keeps courts from
00:23:01 --> 00:23:04 deciding abstractions. Exhaustion lets agencies
00:23:04 --> 00:23:06 correct themselves. Mootness keeps courts from
00:23:06 --> 00:23:09 issuing advisory opinions. It's easy to make
00:23:09 --> 00:23:12 every timing rule sound like an agency trick.
00:23:12 --> 00:23:15 Right. And it certainly wouldn't be right to
00:23:15 --> 00:23:17 do that. We can see that in recent cases. So
00:23:17 --> 00:23:20 a pardon mooted Genalo, not the government. In Fikre,
00:23:20 --> 00:23:22 the government tried to moot a case and lost.
00:23:23 --> 00:23:25 And in Mullin, keeping a case alive is what let it
00:23:25 --> 00:23:27 win. But one advantage is real. The government
00:23:27 --> 00:23:30 is in these cases constantly. So when a vehicle
00:23:30 --> 00:23:32 fails, it goes looking for another one. And so
00:23:32 --> 00:23:37 Rosalind gets one case. Ngo got one grievance. Yes,
00:23:37 --> 00:23:39 that's the asymmetry. So it isn't the doctrine
00:23:39 --> 00:23:42 here. It's the docket. The imbalance isn't that
00:23:42 --> 00:23:44 doctrines are rigged. It's that one side plays
00:23:44 --> 00:23:47 the game every day. Which is why the drug companies
00:23:47 --> 00:23:50 in Abbott weren't really underdogs either. 90%
00:23:50 --> 00:23:52 of an industry with the best lawyers in the
00:23:52 --> 00:23:55 country suing together. They know the court.
00:23:55 --> 00:23:58 They know the terrain. They work this place every
00:23:58 --> 00:24:01 day. Yes. Abbott drastically expanded pre-enforcement
00:24:01 --> 00:24:03 review, and the first beneficiaries were a group
00:24:03 --> 00:24:06 that could afford to use it. The doctrine isn't
00:24:06 --> 00:24:08 formally limited to sophisticated parties, but
00:24:08 --> 00:24:10 the capacity to use it at the right moment isn't
00:24:10 --> 00:24:13 evenly distributed. And the other half of this
00:24:13 --> 00:24:16 is who can forgive a miss. When courts create
00:24:16 --> 00:24:18 a prudential rule, they generally keep more authority
00:24:18 --> 00:24:21 to recognize exceptions to it. The McCarthy factors,
00:24:21 --> 00:24:23 the prudential side of ripeness that the Court's
00:24:23 --> 00:24:25 now questioning. When Congress builds a rule,
00:24:25 --> 00:24:28 the statute's text controls the exceptions. That's
00:24:28 --> 00:24:31 Ross. Available is the only escape hatch in the
00:24:31 --> 00:24:33 prison statute because that's the only one that
00:24:33 --> 00:24:36 Congress wrote. But statutory doesn't mean unforgiving.
00:24:36 --> 00:24:40 You just told me Rosalind gets a hearing on her late
00:24:40 --> 00:24:42 filing. So statutory here doesn't have to mean
00:24:42 --> 00:24:46 jurisdictional. Santos-Zacaria made that explicit,
00:24:46 --> 00:24:48 and Smith shows the related point from another
00:24:48 --> 00:24:51 direction. A missed agency deadline doesn't lock
00:24:51 --> 00:24:53 the courthouse door because the statute made
00:24:53 --> 00:24:55 the agency's timeliness dismissal itself reviewable.
00:24:55 --> 00:24:58 Congress can write a mandatory rule without stripping
00:24:58 --> 00:25:00 courts of power, and Congress can write flexibility
00:25:00 --> 00:25:03 into the statute itself. The hardest limits of
00:25:03 --> 00:25:05 all aren't statutory, they're constitutional.
00:25:05 --> 00:25:08 If a case is truly moot, no court can revive
00:25:08 --> 00:25:10 it no matter who's asking. So the question isn't
00:25:10 --> 00:25:13 just whether a rule is hard or soft. It's whose
00:25:13 --> 00:25:15 rule is it, and therefore, who has the power
00:25:15 --> 00:25:18 to bend it? Yes. That's the question that can
00:25:18 --> 00:25:19 decide whether a court ever reaches the merits
00:25:19 --> 00:25:23 of your case. So we've now covered five threshold
00:25:23 --> 00:25:26 doctrines. We've talked about standing, finality,
00:25:26 --> 00:25:29 ripeness, exhaustion, and mootness. Next week,
00:25:29 --> 00:25:31 we talk about the last one, and it's a little
00:25:31 --> 00:25:34 different. This is the cases where courts say
00:25:34 --> 00:25:36 that some agency decisions are just completely
00:25:36 --> 00:25:39 outside judicial review altogether, or at least
00:25:39 --> 00:25:41 nearly so. This isn't not yet or not you. This
00:25:41 --> 00:25:44 is not reviewable at all. We're talking about
00:25:44 --> 00:25:46 Heckler v. Chaney and Patel v. Garland, the places
00:25:46 --> 00:25:48 where courts have said this really isn't ours
00:25:48 --> 00:25:51 to decide. And after that, what courts actually
00:25:51 --> 00:25:54 do once they've decided to engage. If they decide
00:25:54 --> 00:25:56 to engage. So that does it for today's episode
00:25:56 --> 00:25:58 of Administrative Remedies. Thank you for joining
00:25:58 --> 00:26:01 us today. Please, if you enjoy this podcast and
00:26:01 --> 00:26:04 enjoy this episode, give us a like on Spotify,
00:26:04 --> 00:26:07 iTunes, or whatever platform you're listening
00:26:07 --> 00:26:09 on. And be sure to tune in next time where we'll
00:26:09 --> 00:26:11 continue to dive into the contours of administrative
00:26:11 --> 00:26:14 law, because remember, you can't fix what you
00:26:14 --> 00:26:15 don't understand.

