We have another short administrative law episode, analyzing the Supreme Court's decision about e-cigarettes in FDA v. Wages and White Lion. But first we field some listener pushback about facial challenges in administrative law, and discuss the shadow docket ruling, and ensuing fallout, in Noem v. Abrego Garcia.
We have another short administrative law episode, analyzing the Supreme Court's decision about e-cigarettes in FDA v. Wages and White Lion. But first we field some listener pushback about facial challenges in administrative law, and discuss the shadow docket ruling, and ensuing fallout, in Noem v. Abrego Garcia.
[Divided Argument theme]
Will Baude: Welcome to Divided Argument, an unscheduled, unpredictable Supreme Court podcast. I'm Will Baude.
Dan Epps: And I'm Dan Epps. Back in the studio, Will, after what I'd like to think was a successful live show at Harvard. I like doing those occasionally with you so we can reconnect in person.
Will Baude: Yeah.
Dan Epps: Although you said you were concerned with the recording, that it didn't quite capture the laughs we got from the audience.
Will Baude: Yeah. That's one thing I've noticed is that I think sometimes our editing is a little aggressive or maybe my memory is a little optimistic about how much people are--
Dan Epps: Well, I think our mics are just very good. And the mics are designed to pick up close sounds and not pick up atmospheric sounds.
Will Baude: Maybe we need an atmospheric mic to pick up the room.
Dan Epps: Yeah, it couldn't hurt.
Will Baude: Sometimes the weird energy, I think, in the show comes from the fact that, “Oh, at this point, somebody was laughing, but you can't hear that anymore,” so it just sounds like we're shouting into the void.
Dan Epps: [laughs]That's possible. So, we'll see if people still keep complaining to you about me being nervous when we'll watch it.
Will Baude: One thing I like about the live show, I was just realizing, is when we did a live show, you play the intro music in the room right beforehand, which gets me in the mood. And we don't do that here.
Dan Epps: We can just play that for ourselves.
Will Baude: Yeah, I might have to start doing that.
Dan Epps: Yeah. I mean, it always is a little awkward just in the room. I'm there putting on the PowerPoint, and we all just sit there staring while the music plays.
Will Baude: I smile and bop along.
Dan Epps: Yeah. No, I mean, I think it's important. I think it's not a live show of the podcast if we don't do the music.
Will Baude: Yeah.
Dan Epps: Okay. So, as always, there's some feedback. Do you want to do 30 or 40 minutes on Supreme Court ethics? [Will laughs] We haven't really had any news on it, but I feel like that's a segment we haven't returned to in a while.
Will Baude: It's funny how people's interest in delegitimizing the Supreme Court seems to have changed in the past couple months. [crosstalk]
Dan Epps: Yeah. I'm thinking about writing something about that on the blog, blog.dividedargument.com, which I think is very successful so far. My wife Danielle basically refuses to listen to the show. She says my voice sounds too weird to her relative to what my voice sounds in real life. But she likes the blog and she was like, “Oh, I really like those things to read every week that you post.” I'm like, “Well, I don't post those. Will post those.” And she's like, “Oh, I assume that you're contributing to those.” I'm like, [laughs] “No. Will is just doing that. But I agree that they are very good and useful.” So, check it out. As the semester has been busy, I've not been super active, but I plan to pick up the pace a little bit in the coming weeks and months.
Will Baude: Yeah, I think it's going well. Hopefully we'll keep attracting more readers to that end. We got several emails about our last in studio episode on Vanderstock, the Ghost Guns case, complaining about our views about administrative law [laughs] or the various things we said. I don't think I can do all the complaints justice but-- [crosstalk]
Dan Epps: Nor do you want to--[crosstalk]
Will Baude: Oh well.
Dan Epps: -broadcast the complaints too much.
Will Baude: Oh, complaints are good. But my colleague, one of our Bigelow Fellows, Brian Lipshutz, emailed to complain that my discussion of facial APA challenges overlooked, as did the court and the dissents, he says the text of the APA review provisions. A reviewing court isn't deciding whether the plaintiff gets an injunction or declaratory judgment. It's deciding whether the agency action is not in accordance with law or in excess of statutory jurisdiction, authority or limitations. And if it is, then the remedy is to hold unlawful and set aside the order subject to separability. So, he says from his view, the textual approach to the APA suggests you can't just use the facial challenges model. The question is just, is the regulation not in accordance with law?
Dan Epps: So, that means the decision was wrong?
Will Baude: I think so. Now I think this does get into a bigger debate that we've talked about on the show before about what is the action that is being reviewed in these cases. This goes to the question about the legitimacy of vacatur of a rule. Is the agency action in question a rule and the thing you're supposed to do to vacate it? Or is the agency action a particular thing the agency's doing in the world? And so, the thing you're supposed to do is to set aside the individual action, the John Harrison position.
Dan Epps: So, the particular enforcement action.
Will Baude: Right. So, it reopens the vacatur can of worms. But maybe one point is that you can't necessarily have a view about the facial versus as applied question without having a view about the remedy structure of the APA more generally. And then Emily Bremer from Notre Dame also wrote in to complain about my endorsement of the accepted narrative that the Supreme Court only allowed facial challenges to administrative rules a few decades ago. She points to a case, CBS v. United States from 1942 before the APA, that she says is codified by the APA and shows that this kind of remedy is allowed and argues that's part of the fundamental point of a legislative rule is to resolve with binding legal effect questions that will regularly arise individual cases.
The whole point is to eliminate the need and legal obligation on the part of the agency to address those questions in subsequent administrative proceedings. In other words, having as applied challenges to a rule defeats the whole point for having a legislative rule. And so, this is bad policy and very inefficient.
Dan Epps: So, sorry, I'm not following which way that cuts. So, does that mean that she also thinks it was wrong?
Will Baude: Yeah.
Dan Epps: So, then what does that look like? What does the scope of that review look like? Do you have to imagine every possible scenario that the rule could apply to? And if there's one theoretical one, then--[crosstalk]
Will Baude: Maybe we should get some of these people on the show. I take it you could still have a doctrine of severability. You could still say you judge the rule, you conclude that the rule is not in accordance with law, and maybe you can still sever parts of the rule and leave the other parts of the rule in place, or set aside, break the rule into multiple pieces and set aside one or the other, I take it.
Dan Epps: But you'd have to go through. It's like, “Well, what if it applied to this or what if it applied to that?” You know, just coming up with all sorts of hypotheticals, would you have to consider every possible hypothetical? And if they don't consider that hypothetical, does that mean the rule stands? Or does someone get to come back and say you didn't consider this hypothetical?
Will Baude: I think they don't get to come back. But I think this gets into the general question about the APA and agency authority more generally. In a world where agencies have a decent amount of running room and maybe a little bit of now overruled Chevron deference, it's easier to see how you're like, “Well, look, if there are a couple of cases where this doesn't make sense, the rule is still valid because we give the agency some discretion.” In a world where we're not giving agencies any discretion, it seems like it's just going to be much higher stakes. You know how this works. If the rule doesn't match the statute a tiny, tiny bit, is that enough to say the whole rule is not in accordance with law? I guess so, under this theory, but I don't know if that makes sense to me.
Dan Epps: Yeah. So, why does the court miss this?
Will Baude: Well, I mean, one theory is just the court was looking for a way out of this case, and this was a minimalist dodge. Another theory would be they don't agree. They don't mind if we change the administrative. They don't really like legislative rules. So, they don't mind if we rob legislative rules of the main force they're supposed to have. That would be sabotaging the administrative state eventually. But one possibility is that the court doesn't mind doing that. Another possibility is the court just doesn't really understand administrative law that well, like us, apparently.
Dan Epps: But we're trying to give the generalist approach.
Will Baude: Yeah. I used to litigate several tax cases when I was working at the firm, despite not being a tax expert. And I used to joke that my comparative advantage was that I did not understand tax law any better than your average Fifth Circuit Law Clerk. And so, I'd be good at translating these tax arguments to something that the court might actually understand. Not entirely joke.
Dan Epps: Yeah, I mean, that seems not clearly wrong.
Will Baude: Yeah. So, we may not understand administrative law, or at least you should take our views with a grain of salt. Although I'm not convinced, we're wrong.
Dan Epps: Yeah, I mean, I'm not convinced that we're wrong in trying to channel what the court is thinking. Maybe, I don't know.
Will Baude: Yeah. As you said, I think there are going to be about 85 articles about this issue coming out in the next year or two. So, I will read those and put the best ones on the things to read feed of blog.dividedargument.com.
Dan Epps: Do you actually plan on reading all 85?
Will Baude: Well, how do you define read?
Dan Epps: [laughs]Read the abstracts, I suppose.
Will Baude: I see things online, I look at the abstract, and then I decide whether it's worth downloading it and putting it in the “readme” folder on my iPad. And then at various points I blow through the “readme” folder. And sometimes I read those things thoroughly. And sometimes I read them less thoroughly and conclude they're not worth my time. So, things have kind of--[crosstalk]
Dan Epps: There's two stages of cuts. There's the iPad cut and then there's the “Do you read it once it's on the iPad” cut.
Will Baude: Yeah. Well, there's several stages before it's like the “Do I even click on a link to the article?”
Dan Epps: Okay.
Will Baude: [laughs] Then I read the abstract and decide whether to even download the article. [chuckles] Then once the article's on the iPad, do I actually read the whole thing or do I glance through it, notice that the footnotes are bad, and then get rid of it or something.
Dan Epps: Okay. I feel like you should announce publicly what your choice is, what your assessment is. You'd be like, “This is a no click article.” Do you remember when you used to call everything self-recommending?
Will Baude: Yeah.
Dan Epps: That was like one of your catchphrases.
Will Baude: Well, I stole it from Tyler Cowen. It's not mine.
Dan Epps: But I feel like you don't do that as much.
Will Baude: Is that true?
Dan Epps: I don't know. Well, I mean I think really you don't post as much.
Will Baude: Yeah. Like on Twitter.
Dan Epps: Yeah, on anything you post on--[crosstalk]
Will Baude: I post on our blog. I've blogged my blog--
Dan Epps: No, you post longer form things but you don't post the one line you have, self-recommending.
Will Baude: Well, now that I have the blog to regularly recommend things to read, I don't feel as much of a need to do like the self-recommending tweets.
Dan Epps: Yeah, I mean self-recommending is also self-defeating because you are recommending it.
Will Baude: Well, what self-recommending means is the nature of the project and or the author means this is inherently going to be good. Here is a new article by Caleb Nelson on any topic, self-recommending. But as you start to call some people and some topics self-recommending and not others, people start to get insulted about like who you think is inherently good and who you don't. And people can get insulted. And you may not know this, but I actually don't like insulting people.
Dan Epps: Yeah. So, we've already talked about the blog. Your choice to another's choice to create the blog. Paul Horwitz plugged the blog a little bit on Prawf'sBlawg blog a few weeks ago. I liked what he said. It now includes some of the finest former Volokh Conspiracy bloggers, it requires immediate bookmarking. I don't know their reasons for moving over there and quite agree with those reasons. So, we don't need to delve into that, but we did get a piece of listener feedback that I thought was quite interesting and perhaps amusing, which is that one of our readers wrote in to point out a piece of software that had been written. A listener named Joel Mackler wrote in, and he, saying this apropos of our conversation about trying to avoid reading certain posts on the Volokh Conspiracy, he says, “A bit ago, I wrote a browser extension to remove all Josh Blackman posts. You can read the website without real estate being taken up.”
And he provides a link. Says “It's only 26 lines of JavaScript. If anybody wants to email us, I can pass that link along,” an interesting software project. Okay, so maybe that's enough feedback. Maybe we should delve into substance.
Will Baude: Yes.
Dan Epps: As we occasionally like to do.
Will Baude: So, since our last episode was recorded, we got a decision from the Supreme Court in Abrego Garcia.
Dan Epps: Yeah. As we expected, came down, not too long after we recorded. Luckily, did not come down while we were recording, but it did make our episode a little bit behind the times, but not really substantively, because I think this came out the way you were thinking it would.
Will Baude: Yes. The court emphasized the government's continuing obligation to facilitate Mr. Abrego Garcia's return while asking the court to clarify its requirement that the government effectuate his return, which is, I think, the dividing line we predicted.
Dan Epps: Yeah. Should we quote the language? Maybe. So, order properly requires the government to facilitate Abrego Garcia's release from custody in El Salvador and ensure that his case is handled as it would have been had he not been improperly sent to El Salvador. The intended scope of the term effectuate in the district court's order is, however, unclear and may exceed the district court's authority. The district court should clarify its directive with due regard for the deference owed the executive branch in the conduct of foreign affairs. For its part, the government should be prepared to share what it can concerning the steps it has taken and the prospect of further steps.
Will Baude: Yeah. And just to add a little bit to the confusion, the beginning of the paragraph is, the application is granted in part and denied in part, subject to the direction of this order. Now, to what extent, due to the administrative stay issued by the Chief Justice, the deadline imposed by the district is now passed. To that extent, the government's emergency application is effectively granted in part, and the deadline in the challenge order is no longer effective. The rest of the district court's order remains in effect but requires clarification on remand. One report I saw of this was the court unanimously reversed the district court and said they need to clarify. Yes, I think it's the opposite. I think they affirmed the district courts order other than the deadline.
Dan Epps: Yeah, but then could you just clarify this thing for us?
Will Baude: Right. It remains in effect but requires clarification. Now, since then, there has been a huge amount of ongoing slow walking drama, which I don't think we'll try to cover here. But the district court has tried to get some of these answers from the government and gotten basically stonewalled, Although, it has required the government to file daily status updates which different people in the government keep filing and have a weird, almost poetic passive aggressiveness to them.
Dan Epps: And then the President of El Salvador was here and did not indicate any interest in sending Mr. Abrego Garcia home.
Will Baude: Yeah. So, there was this press conference in the oral argument that was televised in which everybody in the room seemed to agree they did not want to obey the Supreme Court order. I immediately started messaging some friends saying, “Okay, now are we in a constitutional crisis?” Although I will say if you listen to the press conference, which I did, it has several weird features, one of which is when Trump is asked, “Are you going to comply?” He says, “I'm going to let Pam take that one.” [Dan laughs]
And then she answers about why her claim the district court was reversed and I think she says, “We have an obligation to facilitate his return, meaning if El Salvador chooses to let him out of prison, we will let him back into the country and maybe even give him a plane, but it's up to them whether to let him out.” And then he's like, “Well, Stephen, what do you think?” And then Stephen Miller gives his theory that actually he was properly removed because the President's decision to designate MS-13 as a foreign terrorist organization somehow automatically eliminated the stay of removal.
Dan Epps: That's just wrong, right? I mean, in the light of the Supreme Court's decision.
Will Baude: Yes, many of these things are wrong. And then they ask the President of El Salvador what he says and he also doesn't quite answer the question because he says, “I mean, you're asking me if I could smuggle a terrorist into the United States? Of course I can't smuggle a terrorist into the United States. That's preposterous. I don't have that power.” And then my favorite, maybe I have the order wrong. My favorite is then finally Marco Rubio pops up and says, “I don't understand the confusion here.” So, somebody summarized in the space of eight minutes, we had the Attorney General of the United States saying, “It's up to El Salvador.” We had El Salvador saying, “Oh, I couldn't possibly-- It's not up to me.” And then we had Marco Rubio saying, “I don't understand why anybody's confused.” So, it's a real unconstitutional shell game.
Dan Epps: Yeah. I was talking to a friend last night, and he sort of said, “Well, don't all immigration judges just exercise power delegated by the Attorney General?” And I have not vetted this legal theory at all, but just what if tomorrow the attorney general just said, “I am overruling the prior immigration orders suggesting that he can't be removed El Salvador. So, now it's moot.
Will Baude: So, there are a lot of statutes involved.
Dan Epps: Yeah.
Will Baude: And there is some judicial review involved. So, I don't think-- [crosstalk]
Dan Epps: If the AG tried to do that, there would be judicial review of the decision to do that, and maybe that would violate the statute.
Will Baude: Right.
Dan Epps: This is more of an impressionistic, big principles argument.
Will Baude: At a big principles level, it is true that the AG can take control of the immigration system, and that does sometimes happen. In fact, each new often in the change of administration, the AG reverses some prior immigration appeals, ruling about various issues. [unintelligible [00:18:51] did this. And then there are things like that go back and forth, but it's not the case that would just be the end of the matter, because some courts would be involved. And the fact that we currently have a pending court order saying the government's first obligation is to treat him as-
Dan Epps: Yeah.
Will Baude: Yeah.
Dan Epps: I mean, it wouldn't take that off the books.
Will Baude: Right. No. But I do think if--[crosstalk]
Dan Epps: Might muddy the waters.
Will Baude: If the government's goal were to find a lawful way to ensure that they could still keep Mr. Abrego Garcia out of the country, I think they would either be trying to do something like that, trying to lawfully alter his status to a point where they could do what they illegally did to him, or find another country to send him there and argue that, well, they shouldn't have sent El Salvador, it's fine to send him somewhere else. I would be totally shocked if it turns out that they are secretly working on that, but don't want to tell anybody because they don't want the courts to stop them until after--[crosstalk]
Dan Epps: Yeah.
Will Baude: I don't know. But the whole thing is pretty disturbing.
Dan Epps: Not great okay, so lots of other stuff. I mean, not lots of other stuff, but other stuff we might be able to talk about. But we're trying to make this a short episode. I got to teach my final class today, which is maybe good news for the listeners. Might free me up a little bit to be more flexible on the recording sessions. We're supposed to get opinions on Thursday, so maybe opinions will be out by the time you listen to this. We will see. And if they're interesting, maybe we will grind out another episode sooner rather than later. No promises, but we are talking about one opinion, which is the Food and Drug Administration v. Wages and White Lion Investments, which is an administrative law case about e-cigarettes. Some questions about this case.
First question is, Will, why did you insist on me reading this one? You've been wanting to do this one for the last three episodes, and I kept not reading it, and we kept running out of time anyways, and you were like, “Now's the time. We're going to do it.”
Will Baude: Well, okay, so a couple of reasons. So, this is another one of these administrative law cases where the Fifth Circuit ruled against an administrative action of the Biden administration, and the Biden administration took it to the Supreme Court, and one on lopsided grounds. So, I think it's a nice pair to Vanderstock, which we just talked about. So, if you thought that Vanderstock was just about guns, now we have a similar set up, but involving the regulation of e-cigarettes. And it also has a very interesting discussion about harmless error. And I remembered that you are possibly the nation's leading expert on harmless error in the criminal context.
Dan Epps: Yes.
Will Baude: And so, I was hoping you could extend your insights to harmless error in the administrative law context.
Dan Epps: Okay, well, I will try to do that. We'll see how we feel about it once we get all the way through the administrative law principles discussion. And we'll see if we get in as much trouble with this one as we do with Vanderstock. You mentioned the case being lopsided. This is a unanimous opinion. And that actually reminded me of something, a quick thing I had wanted to talk to you about on Abrego Garcia. So, sorry to back up, but Abrego Garcia was also unanimous with only one separate opinion from Justice Sotomayor like doubling down and saying, “This is really, really bad and the government needs to follow the law.” Was that surprising?
Will Baude: Yeah. No, I think, while I tried to claim credit for having called where the court was going to land, they went both a little further than I expected in what they said to the government. And I expect it to be 5-4. I thought it was going to be 5-4 for the Wilkinson Maneuver. And it was unanimous for something a little stronger than the Wilkinson Maneuver.
Dan Epps: Yeah. So, I mean, does that tell us that the court in general is maybe more concerned about preserving its own prerogatives.
Will Baude: Maybe. I mean, packed into Abrego Garcia, there were several ambiguities due regard for the executive branch. They have to share what they can, which I assume is artfully worded to make everybody willing to join it.
Dan Epps: Yeah.
Will Baude: But that does suggest that they at least recognized it would be good if they spoke unanimously.
Dan Epps: Yeah.
Will Baude: And probably all recognize that sometime before this current presidential term is up, they're going to have a fight with the president about--[crosstalk].
Dan Epps: About this, probably.
Will Baude: [laughs]It could be this. I mean, I'm not sure it'll be this, frankly, because I wouldn't be surprised if Mr. Abrego Garcia is not killed or moved before the courts can get to it.
Dan Epps: Killed. That's dark.
Will Baude: Yes, these are dark times, Dan.
Dan Epps: Yeah.
Will Baude: I mean, he is in a prison full of people from the gang that he says they’re trying to kill him.
Dan Epps: Yeah. That's very bad.
Will Baude: Yeah. So, I don't know what to say about that, but let's go back to talking about.
Dan Epps: Okay. Yeah. So, last thought on that. When the rubber actually hits the road and we have to decide, did the government comply or not comply? I don't expect it to be 9-0. Okay, back to this one. Unanimous.
Will Baude: Yeah.
Dan Epps: One page concurrence.
Will Baude: Yeah. So, a bunch of producers of e-cigarettes want to sell what they call flavored e-cigarettes. Although, as Jonathan Adler points out in an excellent post on the excellent blog Volokh Conspiracy, all e-cigarettes are flavored? We call the ones flavored that don't taste like tobacco or menthol, but there's no tobacco in e-cigarettes. So, they add the tobacco flavor. They're all flavored. It's just whether they're tobacco flavored or other flavored. But anyway, the idea is other flavored ones, which I guess have weird fruit flavors, dessert flavors. I've never partaken of any of these things.
Dan Epps: You haven't vaped?
Will Baude: I've never vaped. Have you?
Dan Epps: Okay. Not e-cigarettes. It seems like not a habit that's really worth getting into.
Will Baude: I have enough vices already.
Dan Epps: Yeah.
Will Baude: So, one of the concerns is that the proliferation of these flavors might attract children who then might eventually turn from e-cigarettes to real cigarettes. And that would be very bad for the public health. So, the FDA is very skeptical of these, but issued guidance saying if people want to try to convince them to approve some, they're welcome to, and laid out a bunch of guidance about the kinds of things they would need to see and said, we don't need any particular kinds of studies. You have to show us what evidence you have that you're going to have a public health benefit that outweighs the public health harms and so on.
Dan Epps: And this was not binding. These were just guidance documents.
Will Baude: Apparently.
Dan Epps: Yeah.
Will Baude: And so, then people did, companies did submit their proposals after putting a lot of work into them, and the FDA said no to them all. And according to the industry groups and according to the Fifth Circuit, in an en banc opinion by Judge Andy Oldham, the FDA did a bait and switch right after telling them, “You don't need a particular study. This is what we're looking for, etc.” It turned out the answer was, “Actually we're not going to prevent any of these. We're really categorically banning them. All of that effort you put into it was a waste of time.” And so, the Fifth Circuit said, “This is an unexplained change in position and therefore it's illegitimate. You can't do that, can't pull the rug out from under people.” And so, the FDA went to the Supreme Court, which said, “No, actually, this is fine.”
Dan Epps: And, it's a principle of administrative law that if the agency is going to change, its position from what it has done in the past, it is supposed to give a reason, right?
Will Baude: Yes. It's supposed to both acknowledge that it has changed its position and it's supposed to give a reason. Now, the cases I was familiar with where this came up before, have a different posture, more like, do you remember Fox v FCC? Like, can you swear on TV cases?
Dan Epps: I remember the name from my administrative law class lo, some, 18 years ago. [crosstalk]
Will Baude: One of these rounds was decided the term I was clerking.
Dan Epps: So, I must remember it from something else. Maybe I remember it from preparing for a Supreme Court clerkship interview or something like that.
Will Baude: But there, the idea is like, the agency's about to punish you. They can't punish you if they've changed the rules on you. If they said, “You're allowed to say the F word one time.”
Dan Epps: Oh, yeah, this is the fleeting expletive.
Will Baude: Fleeting expletives.
Dan Epps: Yeah, yeah, okay.
Will Baude: That's fine. Yeah, so anyway, that makes sense as related to the rule of lenity or fair notice or something. When the agency's punishing you, it is a little weirder when we're talking about the agency approving you. The agency's not punishing anybody. It's just a question of are you going to get a permit to sell this new product? And it's a little weird to say because we originally said the criteria would be X, we're now stopped and forced to approve you. I don't know. That seems weirder. And at oral argument, that was one of the things the Justices are puzzled over was, does that even if the initial guidance was misleading and now the agency has wasted a ton of resources because it didn't make sufficiently clear, what its criteria were, it seems odd to say that stops the government and forces them to use the earlier criteria.
Dan Epps: Yeah.
Will Baude: But.
Dan Epps: Just to give people the framework, we are not in Chevron Loper Bright world which is situations where the agency is interpreting a statute. And there's a legal question about did the agency get the statute right. And did they deserve deference? Now they don't get Chevron deference. This is not that kind of administrative law case.
Will Baude: Right. These are cases about the application of law to fact.
Dan Epps: Yeah. So, everyone basically agrees on the legal framework, and then how do you apply that to specific facts? I mean, and that questions about that and the difference between the Chevron situation. I feel like came up at the oral argument in the Loper Bright and Relentless if you remember that. When Justice Kagan was like, “Okay, if you win, does that mean we have to review?” My memory is that she even used some FDA example. Every FDA, medical device approval.
Will Baude: Yes. And I think in her dissent, she has some examples that include at the technical level various scientific questions.
Dan Epps: Yeah.
Will Baude: But one of the things Loper Bright preserves, it is written in broad terms and it says something like the agency can do what's appropriate, then they still have a lot of running room. Not because we defer to them about what the meaning of appropriate is, but it's because the word appropriate, in fact, gives them a lot of power.
Dan Epps: Yeah.
Will Baude: And so here there's a 2009 statute that gives the FDA authority over tobacco, which they didn't used to have, the Family Smoking Prevention and Tobacco Control Act 2009, which says, among other things, that when it comes to the marketing of a new tobacco product, the agency must deny the application unless it is shown that the product “would be appropriate for the protection of the public health.”
Dan Epps: I'm just having read that statute for the first time, I was kind of like, “Is there any tobacco product that would qualify?”
Will Baude: Well, I mean, [crosstalk] question, smoking cigarettes.
[crosstalk]
Dan Epps: -products.
Will Baude: Yeah. So, if we had a healthier cigarette, I mean, imagine I had a cigarette that's just like regular cigarettes, but causes one tenth as much lung cancer.
Dan Epps: Yeah.
Will Baude: I take it the FDA would say, “Well, that's better.” Hopefully people will use that one instead of the current ones. So, it's a sort of harm reduction thing.
Dan Epps: Yeah.
Will Baude: And that's one of the main arguments, by the way, one of the public health arguments for e-cigarettes, which don't have to back--[crosstalk]
Dan Epps: They are less cancer causing carcinogenic than regular cigarettes.
Will Baude: Much I think because they don't have tobacco in them.
Dan Epps: Yeah. Although my understanding is they're still not great for you.
Will Baude: No, I mean, I don't recommend them, but conditional--[crosstalk]
Dan Epps: How can you choose whether to recommend them if you've never used them? You only know the cons, not the pluses.
Will Baude: I read about these things.
Dan Epps: [chuckles] Maybe this is something you have to experience for yourself.
Will Baude: Well, I'm not going to.
Dan Epps: Okay.
Will Baude: Don't do drugs.
Dan Epps: [laughs] It's not an illegal drug.
Will Baude: It doesn't matter.
Dan Epps: Okay.
Will Baude: They're bad for you. Anyway.
Dan Epps: You enjoy a nice cocktail. It's also bad for you. Isn't there lots of evidence about coming out about how alcohol is pretty bad?
Will Baude: That's true. And I try not to have too many nice cocktails.
Dan Epps: You drink in moderation, so still. I'm just saying your ironclad principles, maybe not so ironclad.
Will Baude: Anyway, the FDA is doing its part to stop anybody from moving away from cigarettes to less harmful things like flavored e-cigarettes or nontobacco flavored e-cigarettes. But that's apparently their prerogative. So, anyway, so as you said, the case really ends up coming down to like a lot of evidence. Do you or do you not have randomized control trials that show how much better these are for smoking cessation? What can you do to show us whether kids use them? A bunch of actual empirical things. You could imagine somebody with expertise reviewing and the FDA looks at them all and says, “We're not convinced.” And in the end the court says, “Yeah, that's fine.” The court concludes they didn't really change their mind in the initial guidance.
Well, they said things like, “You don't need to provide any specific study that wasn't promising that you could get an approval without studies. That was just saying, “We're trying to figure out the truth of this and you can show us whatever you've got, but you got to convince us.”
Dan Epps: And just to be clear, the legal framework here is to ask whether it's arbitrary and capricious. Whether the agency is just applying these things totally unpredictably in arbitrary fashion.
Will Baude: Yes. Although part of the way the courts have glossed those phrases is to say there are particular kinds of errors that are arbitrary and capricious. Like changing your mind without telling people you're changing your mind without good reason. That's in fact part of why, I mean, it's a little weird. And this is the kind of thing that can drive policy people nuts. The substantive question is, should we or should we not have flavors of e-cigarettes other than tobacco flavor? And there's an obvious public health trade off there. But then the legal question becomes, for the court is not the answer to that question. And it's not even really, did the FDA do a good job reviewing the evidence? It's, did the FDA's reviewing this evidence sufficiently match some stuff they said in an earlier document in 2020?
Dan Epps: Yeah.
Will Baude: Because that's the only way you can challenge it. But it's several layers removed from the actual question of what would be good public policy.
Dan Epps: Yes, for sure. And there's a lot of things that the court doesn't resolve here, which is interesting. So, footnote five, the court says, “Usually we apply this change in position doctrine, this idea that agencies need to explain when they change their minds. When an agency shifts from a position expressed in a more formal setting,” and then citing your Fox case, “given that neither party has pressed the argument.” So, here the question is does change of position apply when an agency abandons a position it first articulated in a nonbinding guidance document? The court says, “Given neither party has pressed this argument here, that it doesn't apply or that it does apply. We assume without deciding that the change in position doctrine applies to an agency's divergence from a position articulated in non-binding guidance documents.”
Dan Epps: So, this whole opinion could be irrelevant. Maybe it doesn't even apply.
Will Baude: Maybe it doesn't apply to this specific framework. But since the change of position doctrine does come up a lot.
Dan Epps: Yeah.
Will Baude: The court's gloss on it is still.
Dan Epps: Still quite useful.
Will Baude: Right.
Dan Epps: Yeah.
Will Baude: There are several more of these. There's another footnote that says, “There are several constitutional arguments like a nondelegation argument and a due process argument, and those either weren't adequately raised or we don't want to talk about them. Yeah, there are several of those. I think there's another footnote somewhere in there that notes that there's a question about violating notice on comment. So, several alternative grounds that the court flags but doesn't resolve. I assume a lot of this is like the price of unanimity.
Dan Epps: Yeah.
Will Baude: But some of it may also be, it's a complicated administrative law case and it's hard enough for the court to resolve the big questions and they don't want to get into the small stuff.
Dan Epps: Yeah. I mean, and so, but where we land, I think is a situation where the Fifth Circuit may have outs if it wants to go back to its original result.
Will Baude: Yes. And I suppose we also don't know, is the new administration going to do something different with these things anyway.
Dan Epps: Yeah.
Will Baude: I don't know. Is the FDA one of those agencies--?
Dan Epps: Yeah.
Will Baude: Yeah.
Dan Epps: I mean, I would think that RFK Jr also doesn't like cigarettes, but I don't know, maybe he's pro cigarettes. I don't know. Someone who's more familiar with his body of work.
Will Baude: Although these are not cigarettes.
Dan Epps: No, I know, but I'm just saying someone could be like pro-nicotine.
Will Baude: Right.Yeah, I have no idea--[crosstalk]
Dan Epps: Do you know who swore in RFK Jr into his office?
Will Baude: No.
Dan Epps: Neil Gorsuch.
Will Baude: [laughs]Uh-huh.
Dan Epps: Yeah. It's a mystery.
Will Baude: That's interesting. I wonder how that--
Dan Epps: I don't know. Apparently, Thomas does a lot of these swearings in.
Will Baude: Yeah, but there are a lot of Thomas clerks in the administration.
Dan Epps: Yeah. Gorsuch was the guy. So, anybody can explain that one to us. Did they go to the same? Are they from the same town or so? I don't know. Give us the reason.
Will Baude: Read some Gorsuch opinions and loved the Gorsuch style and wanted to--[crosstalk]
Dan Epps: Was Gorsuch antivaxxer? There was that thing where he didn't want to wear his mask. He was like the first one to take his mask off on the bench.
Will Baude: That is true. And I think Gorsuch was one of the most solicitous people to religious objections to the vaccination requirements, and certainly he had quite a few powerful dissents in a lot of the COVID emergency cases.
Dan Epps: Yeah. Okay.
Will Baude: Okay, maybe. I bet you anything, though, that Neil Gorsuch is vaxxed and relaxed, as they used to say.
Dan Epps: Seems quite plausible to me.
Will Baude: Okay. Now the other more interesting legal question here, which is the reason I want to talk about it, is there was one thing the FDA did that they conceded was wrong, which was they didn't look at everybody's marketing plans. One of the things they said they would look at, so it's important to tell us, like, what is your marketing plan to make sure that these things are not marketed to children. And on appeal to the Supreme Court, the FDA didn't exactly concede, but didn't challenge the Fifth Circuits holding they had not done a good job reviewing the marketing plans, in part because at this point, they'd said, “Look, we've decided these things are not enough evidence. They're not good for public health, so we don't really care how you market them.”
Will Baude: The FDA said to the Supreme Court is you should reverse the Fifth Circuit on harmless error grounds. You should say, “It's true we didn't look at the marketing plans, but it doesn't matter because even if we looked at them, it wouldn't have made a difference.” And that's an interesting.
Dan Epps: Yeah, that's a question of what we call harmless error. But I guess there is already an administrative law framework there. There had been some cases, the famous Chenery I and Chenery II, another couple of cases I haven't really thought about since I took administrative law that established a remand rule. The idea that courts are not supposed to uphold administrative decisions on alternative grounds not considered by the agency, that the better course when there's an error is to remand to the agency for additional investigation or explanation. Not an ironclad rule, but a strong presumption in favor of that.
Will Baude: And this is not a rule that applies to courts. For courts, you can affirm them on alternate grounds, that's fine.
Dan Epps: Yeah.
Will Baude: And so why are these different? What's theory of why?
Dan Epps: So, the court tells us, at least here, that the reasoning is when Congress vests an agency with authority to a determination of policy or judgment and the agency fails to exercise that authority, a judicial judgment cannot be made to do service for administrative judgment. So, a separation of powers that basically it is the executive branch's job to actually decide the policy/administrative law question, and courts cannot just become regulators by resolving these other issues. And that makes a certain amount of sense to me.
Will Baude: Right. On the other hand, there is also a harmless error doctrine which is actually in the statute in the APA, in 5 U.S.C. 706, which sets out the reviewability of agency action. It says, “In making the foregoing determinations, due account shall be taken of the rule of prejudicial error.”
Dan Epps: Applicable in ordinary civil litigation to the administrative law context, which of course includes the remand rule. I'm quoting the opinion, not the statute.
Will Baude: Right. So, I would say part of the problem is the court seems to have two harmless error doctrines. It has the harmless error doctrine in the statute which says, “Take into account of the rule of prejudicial error, and it has the harmless error doctrine in Chenery-- [crosstalk]
Dan Epps: Which predates the APA.
Will Baude: But also, I don't know, seems a little made up.
Dan Epps: I mean, many things are made up.
Will Baude: It's true. If you make something up before the APA, then often we say, “Well, the APA incorporated it, and so those two principles are in tension. And then I guess especially ironically, the court seems to sometimes lean one or the other. So, it has some earlier cases like Shinseki v. Sanders, where it cites as part of the APA, it says rule of prejudice error. But then it had a summary reversal quite recently in Calcutt v. FDIC, where it summarily reversed the Sixth Circuit for not applying the ordinary remand rule and seemed to say the only exception to the ordinary remand rule is where there is not the slightest uncertainty as to the outcome on remand because the agency was required to do something. And so, this note doesn't matter.
Dan Epps: Yeah. So, if the court interprets the law and just says, like, “The agency has no power to do something” or “the agency must do something,” there's no point writing the agency think about it again.
Will Baude: And the Fifth Circuit even said, “Wow, the government's making this harmless error argument. This harmless error argument is shocking and disturbing and one of the most alarming things we've ever heard of, we were going to roundly reject it.” And now the Supreme Court's saying, “Well, actually--" Well, what do they say?
Dan Epps: Well, unfortunately, this is another place where they don't give us a conclusive answer. They say, “For now, we agree with the FDA that the Fifth Circuit read Calcutt too broadly. It has long been accepted, for example, that a remand may not be necessary when an agency's decision is supported by a plethora of factual findings, only one of which is unsound.” So, here, I mean, maybe there's a lot of things in the record that justify what the agency did. And this is just one small piece of it. And so, when it is clear that the agency's error had no bearing on the procedure used or the substance of the decision reached, a remand would be pointless.
Will Baude: Yeah. What's also great is they set this up with a discussion of the early solicitation of the academic literature. So, they say, “Commentators have long puzzled over the tension between the ordinary remand rule in Chenery and the harmless error rule in the APA. See, first, Henry Friendly, Chenery Revisited Reflections on Reversal and Remand of Administrative Orders. Then two quite recent articles by two excellent law professors, Nick Bagley, Remedial Restraint and Administrative Law, Columbia Law Review, and then Chris Walker, Against Remedial Restraint and Administrative Law, Columbia Law Review Online. And the courts of appeals have apparently developed their own practices to reconcile the reimagined harmless error rules. See Bagley, 302, note 328, collecting cases. We will not attempt to provide a complete answer to this vexing problem here.
So, this seems to be the court saying, “Look, two smart administrative law professors are debating this. We're not sure who's right, but you all should read their articles and figure out.”
Dan Epps: We are sure that the super extreme view that this, you can only do this when the law is 100% clear. That's wrong.
Will Baude: That's wrong. And the FDA's view may be wrong.
Dan Epps: Yeah.
Will Baude: So go figure it out.
Dan Epps: Yeah. There is another-- [crosstalk]
Will Baude: So, as a harmless error expert--
Dan Epps: Yeah. I was just going to say there's this another one of these footnotes not resolving something. It says “The Court of Appeals appears to have issued a brief alternative holding at the very end of its opinion. And where the court cited circuit precedent echoing the rule of Massachusetts Trustees. But the Court of Appeals applied that case at a high level of generality in absence any analysis applying Massachusetts Trustees to the FDA's failure to consider respondents marketing plans specifically were unable to affirm the decision on that alternative basis.” Is that another ground on which the Fifth Circuit could come back?
Will Baude: I think so. I mean, again, I think they're saying you have to do more work than you did. But they're open to, I mean, they say, “We've made clear that the Fifth Circuit's harmless error doctrine was too strict.” The Fifth Circuit went out of its way to say, “This harmless error doctrine is terrible and dangerous and un-American and we're rejecting it. And we disagree. It's totally American.” Great.
Dan Epps: Yeah.
Will Baude: But how does it apply? You know, is this error actually harmless? That's up to the Fifth Circuit.
Dan Epps: Yeah. So, I mean, the Fifth Circuit has a number of outs here, if it wants to go back to its original position.
Will Baude: Yes. I mean, it may well be that in the end the e-cigarette companies will prevail, but the restrictive view of administrative law put forward by the en banc Fifth Circuit will not prevail.
Dan Epps: Yeah.
Will Baude: And I take it that's part of why the case is written the way it is. So, I have two questions for you, but maybe just one thing to say is the teeing up this issue I thought was very interesting. So, the SG clearly made a conscious choice to get the Supreme Court to issue a ruling on the harmless error doctrine that's a little more liberal than what they were getting from the Fifth Circuit, which reminds me of the attempts the SG Prelogar made to get the court to rule on the vacatur issues in administrative law. There's a broad trans-substantive principle of administrative law that's sophisticated and in the academic literature.
And the SG is like, “We're going to go to the court and see if we can get them to endorse this revision of the law in a pro-regulatory way. On Vacatur, it didn't work. On harmless error, it worked. They didn't get their view, but they got sort of a little movement in that direction.
Dan Epps: Yeah. And their direction is more supportive of agencies.
Will Baude: Yes.
Dan Epps: Right. To the extent that the Fifth Circuit and possibly other circuits, I'm not familiar with the disagreement among the circuits, but to the extent that the Fifth Circuit was applying this doctrine extremely narrowly now possibly it's a little broader, not a ton of guidance.
Will Baude: Right. But I think this also reflects a little bit of a very smart lawyering approach. If you think we work for a Democratic administration in a world with a conservative court where often our substantive merits arguments are going uphill a little bit. But are there some formalist arguments we can make about procedure and remedies that might appeal to the court that might be good for our clients.
Dan Epps: Yeah.
Will Baude: A very smart way to try to tee these up.
Dan Epps: Yeah, that's clever. I like that.
Will Baude: So, as the nation's leading harmless error expert, Dan, what do you think of this?
Dan Epps: Well, I was hoping that I would have more that was useful to say. I mean, to the extent that the debate here is about how do you reconcile, the civil harmless error rule with this administrative law only remand rule, I guess I'm not sure. You know, that seems like less a question about the substance of harmless error and more like this meta question about, how do you reconcile these two competing doctrines and how do you reconcile it with the statute? It did seem to me that maybe, to the extent that there was this remand rule, I don't totally understand why there's not an argument that the APA just totally eliminates that if it's just saying do harmless error like they do it in ordinary civil litigation.
Will Baude: Yeah. I mean, I guess because the statute says, “due account shall be taken.” Well, how much account is due?
[chuckles]
Dan Epps: Yeah, that's fair.
Will Baude: But yeah, it's a puzzling formulation. It also makes you wonder who is the Chenery Rule supposed to protect?
Dan Epps: Yeah.
Will Baude: So, I think one account of the Chenery Rule, it’s a pro executive rule, it'd be bad for the executive if courts were to substitute our judgment. And on that view, you'd think if the Solicitor General of the United States, endorsed by the head of the agency, is in the Supreme Court saying, “No, it's fine, we don't mind. We would make the same decision anyway. Please don't waste our time. There's not a lot of point in remanding.”
Dan Epps: Yeah.
Will Baude: But on the other hand, I've seen the argument that because the APA is fundamentally procedural, the agency could do whatever it wants as long as it jumps through the hoops correctly. And so, part of the point of the remand rule is like, “You got to jump through the hoops.” And some people have said the harmless error rule is at odds with the basic notion that there are hoops and they're important.
Dan Epps: Yeah. Although I guess the notion could be there are hoops and they're important. But the failure to dot one i or cross one t should not necessarily set things back by a year, two years.
Will Baude: And this is where I think it's not a coincidence that Nick Bagley, the author of this great article on Remedial Restraint is also the author of articles on, I think, what he calls The Procedure Fetish, suggesting that maybe jumping through the hoops isn't as important as we have made it out to be, or maybe-- the courts make the hoops really important because that's the only thing they can review. And we spend so much time on judicial review of administrative action that we lose sight of the actual choices in the world that really matter.
Dan Epps: Well, this might be a situation where, as you said, you said something I think is a little too strong. The agencies can substantively do whatever they want as long as they jump through the hoops. I mean, obviously that's not totally correct because there just may be certain actions that cannot be justified no matter how much process is given. If the court tries to ban huge range of products that are not even connected tobacco and says, “They might cause people to smoke.”
Will Baude: Sure.
Dan Epps: But it could be a signal for, there's some issues that are just so obviously going to be justifiable if the agency does them right. And if the agency, say included a bunch of stuff that supported the decision and just also included some stuff that was wrong, is it really necessary to make them go through the whole process again? I mean, it's kind of like Vanderstock, I mean, it's not the same legal issue, but it's kind of like, “Look, if the agency, wrote a rule that really gets at tons of stuff that they're allowed to regulate, should the fact that they wrote the rule a little too broadly and possibly swept in, some class of cases that maybe shouldn't have been in, should that change things a lot?
Will Baude: Yeah. I think this is going to get me in trouble again with our admin law, listeners. But it seems to me part of the problem is the assumption of the APA or at least the assumption of the way the procedural fetishists talk about the APA, is that these hoops actually matter. Like, if the agency got more notice and comment, they might change their mind. They would learn something and be like, “Oh, wow.” Or if they considered these things more carefully, they'd do something different. And it seems like that premise is mostly false. Mostly the agency figures out what it's going to do and then it just tries to run it through the process. And so, the process is just a huge tax.
It's not actually useful for deliberation, but it just causes all agency action to take a long time and sometimes fail for random reasons. And maybe as a weird anti-regulatory matter, that's good. Like we're going to give agencies a lot of power, but then we're going to say that if the agency wants to do anything, they have to jump through 21 hoops and if they get one of them wrong, they have to go back to square one. And that'll stop save us from overregulation. But it's a very inefficient way to save us from over regulation.
Dan Epps: Yeah. I think one way to think about it might be in terms of the standard for errors you'll accept. So, the standard in criminal procedure is an error is not harmless unless the government can establish proof beyond a reasonable doubt that the error made no difference. Now, in practice, I think the standard is not quite as demanding as it sounds. We might feel differently about errors in the administrative law context. Maybe we'd say we don't need proof beyond a reasonable doubt as long as there's a decent chance, they would have reached the same decision, why bother?
Will Baude: Right. You might imagine that if the ultimate standard review is arbitrary and capriciousness, then we should just ask is the error one that renders it arbitrary and capricious or I don't know exactly how to do that.
Dan Epps: Yeah. I mean, I guess one test could be like if you just deleted that part of the reasoning, would the whole thing stand or not.
Will Baude: Yeah. Although given that one of the agency's obligations is to address everything, for instance, because I take it, isn't this a common doctrine-- it's like the agency is supposed to address all substantive comments it gets. And so, part of the industry groups just try to spam the comments as many different things as possible with the hope that the agency will miss one and that they miss when they get in trouble. And now the agencies would love people to come back and say, “Fine, we missed that one, but it was dumb.” And I think we still don't know fundamentally is that a good answer.
Dan Epps: Yeah. Okay, so I guess I don't have insight into what the right rule should be here, other than agreeing with the basic intuition that in cases where there's some overwhelming degree of likelihood, I don't know exactly where I would want to set the standard of proof and comparing that to the criminal context, what that would look like, but that some room to just say, “Let's not make everybody go through all of this again.” Especially with administrative law, I mean, things can get really gummed up. I mean, the agency has to do its thing, has to get comments, and then it has to go back through the court system. That's a really lengthy process and administrations could change in the meantime.
Will Baude: And again, I take it that might be in a weird way for some people the point is, if all these gumming up doctrines mean that it's really hard for one administration to dramatically change the law in a four-year span, and so you only really get legal change if the agency sticks to something over a long period of time that might in a weird way backdoor a certain amount of stability into administrative law, but only at the cost of a ton of make work.
Dan Epps: Yeah. All right. Well, I think that's it for today.
[Divided Argument theme]
Don't forget to rate and review on the Apple Podcast app or wherever you get your podcasts. Visit the website dividedargument.com, we post transcripts of the episodes fairly soon after they are released. store.dividedargument.com for merchandise like T-shirts. blog.dividedargument.com, the newest addition to the Divided Argument universe, where we have a Substack with a larger crew of folks contributing pieces on the law and the Supreme Court and various other topics.
Send us an email pod@dividedargument.com. Leave us a voicemail 314-649-3790 and if there's a long delay between this and our next episode, it will because we've gotten Administrative Law so egregiously wrong that the Administrative Law professoriate has decided to remove us from our duties as podcasters.
[Divided Argument theme]
[Transcript provided by SpeechDocs Podcast Transcription]