Oral argument — Kisor v. Wilkie

We'll hear argument this morning in Case 18-15, Kisor v. Wilkie, the Secretary of Veterans Affairs. Mr. Hughes? Thank you, Mr. Chief Justice, and may it please the Court. The government now appears to agree with our principal contention. Deference does not apply in this case. The Court should arrive at that result by overturning the doctrine of Seminole Rock and our deference in its entirety. Agencies may issue a wide array of rules, interpretations of Even if the best reading of the statute is the SGs in this case? Well, Your Honor, we Making that assumption. Why do we need to reach that broader issue? Well, Your Honor, we think we have the best reading of the regulation, of course I know you think that, but this was a hypothetical Well, Your Honor, the Federal Circuit below rested its decision on complete reliance on our deference, so we think that that is the principal question that was presented by the Federal Circuit. So we think the first order of business is to determine whether or not the Federal Circuit was correct in deciding that our deference resolved this case. The government tells us it's really beside the point because not only, well, either the regulation is unambiguous or if there's any ambiguity, the Federal Circuit's reading, and the Veterans Administration's reading is by far the better reading. Your Honor, the Federal Circuit, though, relied on our deference because the government asked the Federal Circuit to do so. The government expressly argued to the Federal Circuit that our deference applies in this case, and the Federal Circuit took the government's invitation to rest its decision on our deference. So I think this case does squarely present that question because of the government's own argument before the Federal Circuit, which the Federal Circuit adopted. And that is, I believe, both the premise of the petition and the question which the Court granted was to resolve whether or not our deference. So could you turn to the government's argument where they seem to concede that our is wrong, but want us to retain some reduced or revised version of it? Why shouldn't we do that? So to begin with, Your Honor, we certainly think the government's argument is better than the status quo, and we understand it to be a version of deference under which deference would not apply in this case and the vast majority of cases coming from the Veterans Court. So we certainly think it's superior to what currently exists. We don't think, though, it's ultimately the right answer for a few reasons. The first is the most important practical and legal problem with our deference is it is a circumvention of the notice and comment requirements that Congress has imposed generally in the APA, as well as in particular statutory schemes, including this one. The government's rule still allows agencies to put a thumb on the scale without providing- I'm sorry, but that's not quite true. I don't think that here it was an issue of them trying to avoid notice and comment. New legal issues arise normally in adjudications, and that's what happened here. It's not like they should have anticipated that they needed to be more specific about this until the issue presented itself in a case and they reasoned an answer and they gave an answer. So the question really is not one of that in all our deference cases are we talking about the need to give notice and comment time. Your Honor, I agree with all of that, which is to say the agency can do these things. It can be precedential with respect to the agency. It can bind future agency adjudicators. The only question is for that agency activity to also subsequently have legal binding effect in court, what did Congress intend the procedures for the agency to undertake for the agency's action to have prospective force of law? And again, I think the agency's, the VA's own conduct here indicates that when it wishes to have the force of law, it acts through notice and comment rulemaking. Ginsburg. I don't really have the force of law because we've made it plain that our does not call for line deference. The Court must first of all agree that the regulation is indeed ambiguous and that the agency interpretation is a reasonable one. Your Honor, there are certainly limitations on when our deference applies, but the maintenance of our deference means that there are a range of cases in which the agency's notice and comment did not provide the public safeguards still will have binding effect on the courts. And it's that range of cases that this is one of which we think is the ultimate problem with our deference and why the Court should depart from that doctrine. Roberts. Only with some degree, and it's a matter of debate how much of a degree, but only with some degree of sanction by the Court, right? At least the Court has to determine that the agency's interpretation is a reasonable one. That's true, Your Honor, but what happens in cases like this is the regulated public is not able to participate in the underlying lawmaking process that leads to the ultimate rules. And that is not just some speed bump along the administrative process. This matters as a practical matter a great degree. And I think one example that I can offer is at footnote 4 of the government's brief, the government recognizes that just two months ago the VA made substantial changes to this regulatory scheme via notice and comment rulemaking. We went back and looked to see what happened in that scheme, and what we found was the VA issued a notice of proposed rulemaking in August of 2018. The VA said, here's the existing regulation, here are the several changes that we think we should make to it, here's the text of what those changes will look like. Regulated public, what do you think about this? They got comments from all over, including the Vietnam Veterans of America, the Paralyzed Veterans of America, the National Organization of Veterans Advocates, and others. Then in January, when the VA released its final rule, we went and counted. It made 45 material changes from what it initially proposed to do to what it ultimately did in the regulations in response to the public comments. Those 45 changes mattered quite a great deal because the regulated public was able to participate. If we searched through the briefs, there are hundreds of thousands, possibly millions of interpretive regulations. I mean, they give as an example one of them, where the Court deferred to the understanding of the FDA that a particular compound should be treated as a single new active moiety, which consists of a previously approved moiety joined by a non-ester covalent bond to a lysine group. Do you know how much I know about that? Right. Exactly. And that's all over the place, so they're not all like that. Do you know how long it took the FTC to make its first rule under rulemaking? I think the answer was seven years. OK. And I think a lot of them are made more quickly. But what you're doing is saying instead of paying attention to people who know about that, but rejecting it if it's unreasonable, the judges should decide. I mean, I want to parody it, but I mean, this sounds like the greatest judicial power grab since Marbury versus Madison, which I would say was correctly decided. Well, a few responses to that, Your Honor. To begin with, we think that our deference forces the agency to ask the wrong question, because under Martin and Pauly, what it allows the agency at that interpretive stage to do is to determine what it thinks the best policy is, rather than what the best. You also read everybody cited them here. All these studies that show what you say is a problem can be sometimes a problem, but rarely, and that the judges have a lot of power to reject unreasonable rules, inappropriately considered rules, they didn't think about it, rules that change position, rules that are not clear, all these interpretations, you don't have to take our literally, and later cases have not. And so do you – what is your real objection to taking those later cases and saying, of course judges are in control, of course they reject what is unreasonable, of course they reject what is inadequately considered, of course they reject things that are just changed without explanation, but in general, recognize that the FDA knows more about moieties than you do, Judge, and there are 800 judges, and they all think moiety means something different. Your Honor, the critical shortcoming of that is the lack of notice and comment, because I am sure the FDA knows quite a bit about active moieties, but the regulated public may have a – Do you want to take 7 years or 3 years or 2 years on each of the million interpretive rules? By the way, they will just go to adjudications where we have even less control. Your Honor, I think what the APA reflects is the balance this Court recognized in Perez, that agencies have a choice. Agencies can engage in interpretive rules, and interpretive rules have that flexibility and expediency. They are faster to implement, and they bring uniformity to agency actions and consistency to agency decisions. Mr. Hughes, do you think the FCC knows a lot more about the meaning of the word relevant than Federal district judges? No, Your Honor. I think that's a straightforward question of legal interpretation, that Federal district judges are – Do you know why 56D or 554D, which is the separation of functions provision of the agency, it has an exception for ratemaking, which is the FDA's job? In other words, somebody who decides they can't consult the X party with prosecutors in the agency. But there's an exception. There's an exception for ratemaking. And look it up, and you will discover why. You know why? Because nobody in the FCC really knew how to do ratemaking, and they had to talk to their staff. Okay? So you think the FDA and the judges know about the same amount about that? I think that the exception for ratemaking is precisely our point, which is to say when Congress has provided agencies authorities to act in a particular manner and has given agencies that delegated authorization, we agree that that's an area in which agencies can exercise their delegated authority if it's been provided by Congress. But the individual rates, I mean, and changes in the rates and changes in the conditions of the railroad cars and acting under, there are millions. We know there are millions. So how do you propose to deal with those millions? Every one of them goes through notice and comment ratemaking? Well, Your Honor, I think, well, again, in the ratemaking context, as Your Honor points out, Congress can establish different specific rules in specific circumstances. What we're discussing are the default rules that generally apply. And that default rule, as I was saying, is a balance between interpretive rules that are easier for agencies to promulgate that have real effect with inside the agency. On the other hand, notice and comment rulemaking, it does require more for the agencies, but it provides important safeguards for the regulated public. If our were overruled, would an agency's interpretation, particularly in areas requiring a great deal of scientific or technical knowledge, be entitled to no deference by a court? No, Your Honor, I think if our were overturned, Skidmore would apply. And Skidmore, as this Court has articulated, has exceptional importance, particularly in areas where an agency's Well, Skidmore deference is really no deference, because it applies only when it's persuasive, which is true of any argument. Well, in the context of a highly technical or reticulated statutory scheme where it's not the ordinary business of judging, like the meaning of relevant, but something like active moiety, and the FDA can explain that it's brought its scientific consensus to bear. Well, that sounds like State Farm, but Skidmore is really not any, you rely on that to say don't worry, but Skidmore deference, as I've seen it applied over many years, is not much. Well, I think, Your Honor, it's to say in one of those technical contexts, if a court is going to arrive at a different result from the agency, the court needs to have a pretty serious reason as to why it's doing so. It has to articulate real rational reasons on the record as to why it is rejecting the agency's admitted authority over particular scientific and technical areas. So Skidmore does show the respect that's due a coordinate branch of government. We think that's the appropriate alternative solution. Kagan. Mr. Hughes, may I ask you about stare decisis, because you're asking us to overrule two decisions, Auer and Seminole Rock, and really 10 or 12 more, over the past half century, where the court has talked about Auer deference or Seminole Rock deference. And what is the basis for that? Congress could have done this at any time. Congress knows that this goes on. Congress has repeatedly acted in this sphere and shown no interest whatsoever in reversing the rule that the court has long established. So why is it that overruling is the appropriate course here? A few answers, Your Honor, but to begin with, I don't think there's distance with the government on that point, because under the government's test, Auer deference — in the case of Auer, the court should not have applied deference. Under the government's view, there needs to be a principle of fair notice. Well, you know, there might be a problem of a lack of adversarialness here, but I'm asking you. I can also ask the government, but I'm asking you. So setting aside the government's position would require overturning a dozen cases on its own. To move to the point, I think stare decisis has substantially less effect in circumstances where, first, it was an underlying judge-made rule and not something that was statutory constitutional interpretation of its origin. And, second — I don't understand that. You know, what we look to is, could Congress have changed this? If Congress couldn't have changed that, that's a reason for us to change it. But if Congress could have changed it, which Congress could have done any time within these past however many decades, that's a reason for us to say, you know, we don't think that we should step in where Congress has not. Well, Your Honor, I think that was true in both Pearson and Wayfair and other cases the courts decided, where the court recognized that Congress could step in and change the rule, but the court has repeatedly said in those cases that when the underlying issue stems from a decision from the court. Well, there are very many of those cases, and we take it super seriously when we do. And we need a — I mean, we used to. And we need a good reason for it. So what's your good reason? So — so good reason, Your Honor, I would put twofold. First, stare decisis applies with substantially less force because this is not a doctrine under which the public can rely. In fact, it injects considerable instability into the legal system. Well, reliance is a kind of plus factor, and we can talk about reliance either way. But, I mean, usually we look to something terrible that's happening. This is unworkable. This is an anomaly in the doctrine. It no longer has any support in the surrounding legal landscape. Something like that. This is so grievously wrong that we can't stand to live with it anymore. Do you think Auer rises to that level? I do, Your Honor. And to begin with, Auer did not have any underpinning when it was first announced. It's never been reconciled with the APA. And the practical problem is — It didn't have any underpinning. Its underpinning is obvious. Its underpinning is everything that Justice Breyer talked about. Its underpinning is agency expertise. Its underpinning is an idea that judges are far less suited to make these kind of minute decisions of agency policy than agency decision-makers are. Your Honor, I think it's just impossible to reconcile Auer deference with the judgment that's reflected in the APA, that when an agency is going to put on its policymaking hat, which undoubtedly the agency can do, there is an ability for the public to be able to participate in that process and provide their views. You're right. You're right that it says in the APA — it says, you're absolutely right, that when a judge decides a case, and it has to do with the meaning of the regulation, it says the judge, the reviewing court, shall determine the meaning or applicability of the terms of an agency action. That's what you're relying on. And there's just one thing missing. One thing missing, and that is it doesn't say how you do it. And by the way, that isn't just made up out of thin air. They — it's not Auer, it's Seminole Rock. Auer repeats Seminole Rock. Decided in 1944. An important case. The APA is written two or three years later. I can't remember exactly when it was adopted. But wouldn't somebody have said something about it if, in fact, those words were meant to change what was pretty well-established law at the time? Well, Your Honor, I don't think that there's any evidence that the APA somehow silently adopted the doctrine of Seminole Rock. There's no evidence in the history of Seminole Rock. Well, the evidence would be that if the Attorney General's manual that discusses the APA goes through prior cases that they intend to change — I'm not saying perfectly, but to a considerable degree — and by the way, Seminole Rock is not there. And so we have both the language, which doesn't just say — it says shall determine, but it doesn't say how to determine. And in addition, you have the report, which I agree with you, says nothing. But I'm not sure that that cuts in your favor. But, Your Honor, if we look to the Attorney General's manual of 1947, as you allude, that actually, I think, cuts strongly in our favor, because that explains that interpretive rules do not have the prospective force of law. The Court relied on that manual at footnote 31 of its Chrysler Corp. decision, which is carried forward to Peres. No, I agree with you. You're right about that. But I don't see this as the interpretive — this isn't the enforce of law. We're trying to figure out what the — what the regulation means. And you read the AG's brief. He has a lot of conditions around the judge's authority. That is to say, the judge has a lot of authority to say this reg is no good. But he doesn't have to ignore what the agency says. After all, the agency knows about old Lysol or whatever it is, and we don't. And I want to be clear. We do not believe that the Court needs to ignore what the agency says, either. We think it warrants respectful consideration for reasons of interbranch comedy and the recognition that agencies do have technical expertise. We don't dispute any of that. The only question — You know, everybody's talking about this being the rule since Seminole Rock and our — but I go back to cases in the early 1800s, and one — I just picked one of 1850, where the Court said the foregoing construction, being the one adopted by the Departments of Public Land soon after the Act of 1832 went into operation, we should feel ourselves restrained unless the error of construction was plainly manifest from disturbing the practice prescribed by the Commission of the Grand Land Office. And I have a series of other cases throughout the 1800s where the courts were basically talking about, you take the interpretation of the agencies unless some manifest error was present. So Sturgeon and Auer are not more recent manifestations. They're based on fairly understandable principles. Number one, agencies have expertise. My colleagues have talked about that. Two, they are also part of an administration and often have a better understanding of what the needs are under that regulation. And three, in some ways, regulated parties need to have a starting point of understanding how their conduct will be viewed. And if you tell the world, the agencies are going to receive this generalized Skidmore deference that Justice Kavanaugh spoke about as no deference essentially, persuasiveness really isn't, then they don't really have a starting point to understand how to conform their conduct because they have to wait until 13th Circuit courts rule on an interpretation of a statute before really understanding what they have to do. That last point is one that troubles me, which is regulated parties should know where to start. And the best people who can tell them is the agency who's responsible to the public for having sound interpretations or reasonable interpretations. Thank you, Your Honor. I'd like to respond both to the point about the history as well as the stability point. To begin with Your Honor's initial point about the underlying history, I agree if we look prior to the APA, there are cases that suggest in the statutory context that a department or agency's interpretation of the statute deserves binding deference. But in the APA, Congress decided that there needed to be procedural protection to safeguard the interests of the public through notice and comment rulemaking to provide the public the ability to participate in that lawmaking function that happens within the agency. That was one of the critical innovations of the APA that imposed on past practice as a matter of congressional direction. And we think that's what's lacking here. Counsel, to get back a little bit to the stare decisis question, I think the issue depends at least in part about how much of a change you're making. And one of the things I have trouble getting my arms around is if you start with Auer and recognizing the limitations on Auer that have accumulated over the years, and you're changing from that to Skidmore deference, which I find hard to get my hands around, too. I think I know more what a moiety is than I know what Skidmore deference is. And I just wonder exactly how much of a change at the end of the day you're talking about. Well, once we take into account Smith, Kline, Beecham, and Gonzales and the Court's consistent narrowing of Auer, I think Your Honor is right that Auer has been narrowed to the point where it does have substantially less practical effects today than it does previously. But I will say Auer is still used in a way that injects inconsistencies and instability into the legal system that I believe also responds just so much. Well, I suppose it depends. I mean, it depends on the agency. It depends on the rule. It depends on the court. I mean, at some point you're applying Auer. You consider the range of reasonableness and the confidence that a court has that the agency has not found itself within those bounds is going to vary greatly from case to case. The courts are going to take a more careful look in some cases than they are in others. And maybe that's part of the problem. But I guess I'm not quite sure that I understand what you're saying when the rules have the force and effect of law, when they're subject to judicial review within a particular range, and it's really quite imprecise what the range is. And I said maybe that's the problem, but. Your Honor, I do think that imprecision is quite the problem. But as long as Auer and Seminole Rock remain, that suggests that there will be a range of rules, I think a narrowed set of rules, but some rules that if they make it through the gauntlet of Auer and Seminole Rock, do have the prospective force of law without going through the procedures that Congress identified that the agencies should undertake to have that force and effect of law. And I think this case is an example. I'd point the court to another case recently that highlights the instability. It's an en banc decision of the Ninth Circuit, the Marsh case that was decided in September of 2018 about the Fair Labor Standards Act and how the TIP credit works for employees that sometimes work under a TIP job and sometimes don't. Well, the en banc Ninth Circuit decided that case on the basis of Auer deference in reliance on the then binding Department of Labor interpretation. It was not six weeks or seven weeks later that the Department of Labor rescinded the interpretation that the Ninth Circuit en banc rested on. And now the lower courts are having to figure out is the Ninth Circuit's en banc decision still binding interpretation of that regulation? When the question- And what about the lower courts? Let's say your argument is accepted and Auer is overruled. There may have been a dozen or so cases, Auer cases in this court, but there are probably hundreds in the lower courts. So do all of those cases, what happens to all of those cases where there was reliance on Auer in the lower courts? So I think from this court's case, it would still have to start assess, but to Your Honor's question about the lower courts, I think the courts would have to wrestle to see if whether or not Auer was the rule of decision. But as the Marsh case I just referenced underscored, those cases lack the kind of stability that interpretations of statutes and regulations hold because they are constantly subject to revision overnight- You're saying that there would be wholesale cases before the lower courts, lower courts that had relied on Auer, and the losing party then says, court vacates that decision because you premised it on Auer, and Auer is not good law. I think parties could potentially advance arguments along those lines, Your Honor, but I don't think that it increases instability any more than exists in the status quo when those decisions are already subject to revision by the agency. Your argument is that notice and comment solves everything, right? Well, I think notice and comment is the scheme that Congress implemented into the APA. In other words, this issue would go away if the agency did notice and comment for the guidance. Well, and prospectively, what our argument would lead to is an area where there's far more prospective stability because once a court decides the meaning of a regulation, that has durability unless the agency changes- Can I get your reaction to the thought that the lower courts have made notice and comment too difficult through various requirements requiring detailed explanations, making it hard to change regulations that have gone through notice and comment. Do you have a reaction to that? Because that may be one of the reasons that has pushed them into the more guidance rather than notice and comment in the first place. Well, Your Honor, I think notice and comment is what Congress required and that's what the court should adopt. If the court is of the view that notice and comment has become more onerous than Congress intended, I think the solution would be to address that issue of notice and comment overreach, not to allow agencies to circumvent it. If I may reserve my time. Thank you, counsel. General Francisco. Mr. Chief Justice and may it please the court. Seminole Rock deference raises some problems in some applications, but it's been on the books for decades. It has significant practical benefits. Its practical problems can be addressed by reinforcing reasonable limitations on the doctrine. I'd therefore like to address two key points. First, in its core applications like Seminole Rock itself, where the agency provided public notice of its consistent interpretation, it has significant practical benefits. It promotes national uniformity, predictability, and political accountability, because if a rule is subject to multiple reasonable interpretations, the choice of which one to adopt is made by a single, more politically accountable agency, rather than in dozens and perhaps hundreds of different district courts across the country. Mr. Francisco, as I understand it, nobody left before us alive is willing to take our literally and it's just a matter of how much revision to it we've already made. Is it enough? How much further should we go? Or should we just give up on it altogether? And you're asking for us to keep on going. And as I understand it, there are six elements of your test. We have to decide whether the regulation's ambiguous, whether the interpretation's reasonable, whether it's consistent, whether it was made by someone at a high level, whether there was fair notice, and whether it was made by somebody with expertise. Is that a recipe for stability and predictability in the law? Or is that a recipe for the opposite? No, I absolutely think it is and it's a workable standard, Your Honor. I think our principal limitations are consistent with existing law, though perhaps not identical to it. The requirement of genuine ambiguity is really what we think this court's cases have always required, although there is language that we think ought to be replaced with the genuine ambiguity. Well, people fight over whether there's ambiguity and what ambiguity means. They fight over what reasonableness means. They fight over how consistent is consistent. And for the life of me, I don't know how high a level a person has to be before we're gonna defer to him or how much notice is fair or how much expertise counts. I'm with Justice Breyer on moieties, but the people I think have the most expertise on what relevant evidence is, probably John Cain, federal district judge of about 40 years, not an agency. And under the rule you propose, every agency could define relevant evidence differently. And then what is, well, if they have enough expertise, we're gonna get on that road. And I guess I'm just wondering, at what point does this whole edifice just fall upon itself? Well, Your Honor. Lawyers will enrich themselves and do well with this kind of test, but how are regulated people supposed to behave? And Your Honor, there's a lot built into that question, but what I'd like to bring the focus on in answering it is our strong interest in preserving Seminole Rock in its core applications where we actually think it has a significant amount of benefit to regulated parties. Because you're right, there is a lot of disagreement amongst judges as to what a reasonable interpretation is. As the court said earlier this term, reasonable jurists can look at the language and acting in good faith come to different interpretations. One of the virtues of Seminole Rock is that when you're facing multiple reasonable interpretations, you vest the decision-making authority, excuse me, in a single party rather than multiple courts. And that's actually of a benefit to regulated parties because they don't actually have to litigate that thing in multiple courts across the country. They can rely on the agency. On that, I'm sorry, but you keep saying how much of a benefit it is for regulated parties and their reliance interests, private reliance interests. And I must say, I cast a skeptical eye when the government is worried about private reliance interests. And every private party before us says their interests and stability would be better served by eliminating this rule altogether. And it's not just the Chamber of Commerce, it's the Farm Bureau. It's the national lawyers who engage with the immigration system every day and are faced with claims of our deference for single member decisions from the BIA that are unreasoned. And it's the veterans for us, the American Legion, the lawyers who represent veterans every day before the veterans courts who are outraged by our, and who say it doesn't serve their reliance interests and it provides highly unstable rules that they have to guess at all the time. Why should I credit the government's protestations that it is serving private reliance interests? Well, Your Honor, I think that it benefits both those private reliance interests, and there are a lot of private reliance interests that aren't represented here before the court, as well as interest in stability and political accountability. But if I could sort of use an example to illustrate the point. Suppose you've got a rule that is genuinely ambiguous. It's subject to multiple reasonable interpretations. There are a couple of ways to go. You could do notice and comment rulemaking. That takes a very long period of time. In the meantime, there are two things to do. You can litigate the case in dozens of district courts across the country and hope that you can convince all the courts to reach the same conclusion, or you can defer to the agency's reasonable choice amongst what is, by definition, reasonable alternative definitions. And we think that is the benefit to Seminole Rock. In the face of those multiple reasonable interpretations, you're vesting decision-making authority in a single, more politically accountable party. Judges disagree all the time, though, on the threshold question of whether something's ambiguous to begin with, and that creates a whole sideshow. And one of my broader questions is, why can't the government just do notice and comment? You said it takes a long time, and that may be a problem with some lower court impediments to notice and comment. I share that concern. But if notice and comment were more efficient, why not just do notice and comment? So there are two parts to that question. First, on the reasonableness inquiry, I completely take your point, that judges can come to a different conclusion as to what is ambiguous and what is not. And that's not a problem that Seminole Rock creates or a problem that Seminole Rock can solve. It's something that's just endemic to this process. But what Seminole Rock does do is there's gonna be a lot more agreement on whether something is subject to a range of reasonable readings than there is on pinpointing the precise, accurate, theoretically correct reasoning. So Seminole Rock reduces a large amount of uncertainty in that respect. As to notice and comment, look, we take the law as it is, as it's handed down to us by this court and other courts. And as it's handed down, notice and comment rulemaking is a cumbersome procedure. That doesn't mean it doesn't have benefits. It has extraordinary benefits. That being said, when you're looking at a rule that is by definition subject to multiple reasonable readings- Do you agree from your study of this issue that the impediments to efficient notice and comment rulemaking have pushed the government into doing more things in this manner? Your Honor, I'm not prepared to say I agree or disagree with that. I certainly understand Your Honor's point. But I guess the simpler point that I'm trying to make is that given that it is what it is- Do you happen to know what the average notice and comment rulemaking is, how long it takes? Your Honor, I don't know the answer to that question. I apologize. But I- I haven't seen a large decrease in notice and proposed rules. No, Your Honor, I haven't. The Federal Register. I haven't. And again, though, the- Your colleague on the other side, talked about one notice and comment that received 45 changes. So the rule is still being used. Notice and comment is still being used. Oh, it's definitely still being used, Your Honor. And it is a very important process, but it doesn't undermine the benefits of Seminole Rock. Because while you're going through that period, you're facing a rule that, by definition, is ambiguous. And you've got to figure out what to do with it. I do think that- One of two things happens. The judge gets deeply into the question before him or her, and does the work, and comes up with something that looks like the right answer. And once you've done that, everything else looks pretty unreasonable. Or the judge just starts looking at it and flipping through it and says, boy, there's a wide range here. Could be this, could be that. And you defer to the agency. Now, if I think that that's what happens as a practical matter, which rule should I adopt? Your Honor, I think you ought to adopt ours. Because we actually place an- Ours? You mean yours or our case? The position of the United States, Your Honor. And that's because we really do put a lot of emphasis on that first requirement of genuine ambiguity. We do think that courts should do- So you think that the judge ought to do is do extensive amount of work and come up what looks to him or her as the right answer? I think what the judge needs to do is an extensive amount of work at the front end to determine if there is, in fact, genuine ambiguity within the language of the rule itself, much like it's required to do under Chevron. But the problem is, the problem is that the judge, judges could come up with an interpretation that says the agency's interpretation of the regulation is wrong. And this is a really important interpretation. It has real effects on many people. And it's wrong. But nonetheless, rule for the agency, under your theory, because, and under the Chief Justice's question, because there's some ambiguity in it and therefore defer to the agency, even though the judges might unanimously think it's wrong. And doesn't that trouble you? No, Your Honor, because again, I don't think that is quite the nature of the inquiry. I think that there are lots of statutes- I think that I disagree. I think that's what happens in judicial conference rooms. Okay. And I'm not gonna obviously question you on that. Which is the point, I don't think this is the, I don't think the government's reading is the best reading, but it's sufficiently ambiguous that I'll rule for the government. That happens. Yeah, so I guess I- On big cases. So I guess I have a couple of responses to that. First of all, in our search for what the best reading is, I think that often the agency's interpretation is highly relevant to understanding what the best reading is of a complicated regulatory regime. Secondly, I think it is often the case that there is, it is not clear what, in fact, the best reading is when you are facing multiple reasonable constructions. And that, again, is the virtue of Seminole Rock. In those core cases, when you're facing multiple reasonable constructions and you have the benefit of the views of an agency that's administering a complicated regulatory scheme, you vest the choice on which one to pick in a more politically accountable agency rather than having to fight it out in different courts across the country, because different judges are going to come to different conclusions as to what the most reasonable or theoretically best understanding of a particular rule or regulation is. Should we be concerned about the effect that either overruling our, and Seminole Rock, or taking your position will have on cases in which courts have interpreted regulations based on those principles? And now, whichever course we take, those will be thrown into doubt. And if that's a real concern, is it more of a concern, is it much less of a concern if we take your proposed route than if we overrule our and Seminole Rock completely? I think it's far less of a concern under our rule because, under the United States rule, because under my friend on the other side's position, every single regulation that's currently on the books whose interpretation has been established under Seminole Rock now has to be re-litigated anew. So I think- Well, I guess I don't understand that response because it seems to me you've made the point that there are going to be a great many regulations where the outcome would be the same with or without our. You make that argument in this case. So we'd have to know how many of those there are. We'd have to know how many would be problematic even under your modified test, and we don't know that. A lot of these regulations get supplanted and statutes disappear and get modified, and those would have to be accounted for too. So at the end of the day, I didn't see anything in the briefs other than rank speculation on this point. Well, I think that one thing that doesn't require speculation is to know that their position would be more disruptive than ours because theirs would require everything to be revisited. Even under the most aggressive interpretation of our view, it wouldn't require everything to be interpreted. But if you look at this Court's cases, and I'm not representing them as a random sample, but if you look at them as a sample, I don't think our rule would be particularly disruptive at all. If you focus on our requirement of inconsistency, we've identified three of this Court's cases that arguably applied Seminole Rock in the face of inconsistent interpretations. In each three of those cases, the application of Seminole Rock appeared to be make-weight. In other words, the Court first explained why the agency's interpretation was likely the best one, but then applied Seminole Rock in order to confirm that decision. So I think our position would be significantly less disruptive than my friends on the other side. General, it's similar. Vest has come up about 50 times, and I'm a little curious about that. Jerome P. Frank thought, there are no cases like Justice Kavanaugh described, and I believe Justice Kavanaugh was closer to it, and so I think he's probably right, which is the best. I mean, we know one thing. We know that democratically speaking, agencies aren't very democratic, but there is some responsibility. And there are one group of people who are still less democratic, and they're called judges. So if, in fact, you believe that the best solution, where there's real ambiguity, and you just don't know, the best solution is in our country, a democratic solution, well, maybe the agency is the institution that's closer to it. Now, you're just supposed to say yes. Certainly, I would say yes. Certainly, I would say yes in the context of Seminole Rock itself, because that really does underscore our key point. Seminole Rock only applies when a rule is genuinely ambiguous, in that after applying all of the ordinary, after applying all of the ordinary tools of construction, it's subject to multiple reasonable readings. And in that context, we do think it promotes democratic accountability by vesting it in a more politically accountable agency. That's why I have a problem with Justice Kavanaugh's use of the word bad interpretation, because bad interpretation sounds to me like an unreasonable interpretation. It can only be bad if it's unreasonable. And that already is taken care of by the hour standard. By the requirement that it be genuinely ambiguous. A, genuinely ambiguous, and B, reasonable. Reasonable can't, I don't think, mean a bad interpretation that's not consistent with the statute or not consistent with either the text, the context, et cetera. So if it's reasonable, then there has to be a basis for the interpretation in the statute. I think I generally agree with that, Your Honor. And to that, I would add two points. I think that when you're talking about interpreting complicated regulatory regimes, the agency's understanding of it to determining what the, and I'm going to put it in quotes, what this quote-unquote best interpretation is, and as I've already said, if you poll 50 judges on a complicated regulatory regime, you're often going to come up with multiple different best interpretations. And that, again, underscores the benefit of seminal rock in those core applications. You agree, I think, with taking footnote nine of Chevron, using all the tools of statutory construction, and before you conclude that the ambiguity remains in this context. I think you've said that. I absolutely think that is part of the agenda. And when you do that, you usually eliminate or greatly reduce the number of cases where there remains an ambiguity. Do you agree with that? Not only that, but you also reduce the range of ambiguity, because a reasonable interpretation has to fall within the zone of ambiguity that remains in the rule after you apply those ordinary tools of construction. And so that's why, and I know my friend on the other side didn't really get into the separation of powers issue, but that's why we don't think that there's any substantial separation of powers question here, because the agencies are in fact subject to substantial control by both Congress and by the courts. Well, that's- Can I ask about a slightly broader version of Justice Alito's question? He asked about reliance, but thinking about all the stare decisis factors, when I started asking Mr. Hughes about them, he immediately said, well, the government has just as big a problem on those factors. So does it? Absolutely not, Your Honor, because what we're arguing for is that Seminole Rock be retained in its core applications, which frankly we think are the areas where it is the most important, both to regulated parties and to agencies. And it is always more faithful to stare decisis principles to retain a doctrine at its core, while perhaps imposing limitations on the edge that simply recognize that in the course of practical application, practical issues have been identified. And that's why we think that you ought to reinforce the requirement of genuine ambiguity. You ought to reinforce the requirement that you wouldn't apply it to inconsistent interpretations. And we don't think the agency should get Seminole Rock deference for secret private interpretations. It ought to give public notice of its interpretation as it did in Seminole Rock itself. But we think that when you have those principal limitations, you've largely addressed the practical problems of Seminole Rock. And what you're left with in our view are the practical benefits of Seminole Rock. So even if you think it was wrongly decided as an original matter, it's got significant practical benefits. Its practical problems are manageable. It's been on the books for decades. And this is something that Congress could fix if it believed that this court has misgaged legislative intent in adopting the Seminole Rock doctrine. You said give public notice. That's one of the requirements, right? Yes. I think the other side would say just add and comment. Your Honor, and I think that's exactly their position. The Seminole feature- And what's wrong with that? The Seminole feature of Seminole Rock is of course that you don't require notice and comment. And I think that your question gets back to the colloquy that we had before. While you're going through that notice and comment process, you're simply left with an enormous amount of uncertainty because you're left with a rule that everyone has already concluded is on its face, subject to multiple reasonable interpretations. And to say that the alternative is to have multiple courts across the country struggle with that- Mr. Francisco, you keep telling us about the benefits of that. But the benefits of notice and comment are, among other things, people will know prospectively what rules govern them. And not be sideswiped later by a bureaucracy. You can call it democratically accountable if you wish. Right. I don't know. People might disagree. At any rate, a bureaucracy coming up with an amicus brief or a single member opinion in a BIA decision involving an immigrant. Or in this case, a veteran seeking benefits who in the middle of a case is confronted with a new interpretation never seen before. All right. Those, that's the reality. And I'm not sure how that serves democratic processes or the separation of powers as opposed to having an independent judge The one thing you're gonna know is you're gonna have an independent judge decide what the law is in your case consistent with the statute that says an independent judge shall decide all questions of law. That seems to me a significant promise especially to the least and most vulnerable among us like the immigrant, like the veteran who may not be the most popular or able to capture an agency the way many regulated entities can today. Well, Your Honor, there are a few things built into that. Let me start by saying that I think that our public notice requirement ensures that regulated parties, that the agency can't rely on secret interpretation. So it makes sure that its interpretation is out there in the public. And so our is gone. You are asking us to overrule our itself because that's what happened in our amicus brief. No, Your Honor. I guess I would probably argue that I think an amicus brief filed in this court given the high profile nature of the court. Satisfies the public notice requirement because it puts the world on notice that this is in fact the agency's position. But to go to your larger point, I think as this court held in the Martin case, seminal rock deference reflects. Person who litigates against the government for years for his disability benefits as a veteran of the United States is on public notice when the case arrives here and you file an amicus brief. Well, Your Honor, we are much less concerned with the outcome of this particular case than we are with preserving seminal rock in its core applications. But to go to this particular case, remember the VA has a system where the VA itself is charged with assisting veterans through what can sometimes be a complex process. And here it was the VA itself that identified the potential reconsideration pathway that would have provided the veteran with the benefit of retroactive benefits and the VA itself that also explained why it didn't apply to the veteran. So I think in this particular context of this case, this was a very fair process. That being said, we aren't particularly concerned with this specific case as we are with preserving seminal rock deference in its core applications where we do think it has the most significant amount of benefits. If we were writing on a clean slate, what would you say is the basis for any version of our seminal rock? Is it based on some kind of delegation theory or what is its conceptual basis? I think the best conceptual basis is what this court gave it in the Martin case. And that's where it said that seminal rock rests on a presumption of legislative intent that Congress presumed that courts would defer to an agency's reasonable interpretation of its otherwise ambiguous rules as part of its delegated rulemaking authority. Now, I think that members of this court may debate whether that was or wasn't an accurate understanding of legislative intent, but this late in the day, I don't think that's any longer the relevant question because it's been on the books for decades. And usually those kinds of presumed legislative intent are based on other views, right? They're based on a view. Of course, Congress is presumed to want the agencies to do this because fill in the blanks. Is it expertise? Is it political accountability? Is it uniformity? Is it a combination of those things? I think that's fair. I think all of the above are fair considerations of what those types of presumptions are often based on. And it also reflects the fact, and this was the second portion of the point I was gonna be making on the separation of powers issue, is that when an agency acts pursuant to a lawful delegation from Congress, and we can fight over what that means, but as long as you've got a lawful delegation from Congress, at the end of the day, it doesn't really matter if what the agency is doing looks adjudicative, looks executive, or looks legislative, because in every one of those instances, the agency is effectuating executive power. It has to be effectuating executive power as Justice Scalia has made clear. You may not care about the outcome of this case, but we're gonna have to at some point. And if we overrule our, we could just kick it back, okay? But if we don't, let's assume we were to accept your approach. What did the district court, what did the court below, not the district court, what did the court below do wrong? How would you correct it? How would you advise us to advise judges to approach the our question? Sure. Write my opinion for me on that. So, a couple of points, Your Honor. And if I could first say, I didn't mean to say that we don't care about the outcome of this case, because we deeply care about the rights of our veterans, and we do care about the outcome of all of these types of cases. By the way, the biggest argument that your adversary has is that the agency didn't take into account the assumption that interpretation should favor veterans. So, deal with all of that. Sure. So, we do care about how the specific case comes out. But in terms of how it would apply to this case, at the end of the day, I actually don't think the federal circuit should have applied Seminole Rock deference to the VA board's decision in this case, for two reasons. First, and this is one you might well disagree with us on, we think we had the better interpretation of the regulation, and so we don't think you ever get to Seminole Rock. But if you disagreed with us on that, one of the key questions, and under Seminole Rock and under Chevron- Do you think their reading is unreasonable? We do. And secondly, as under Seminole Rock- Let's say we disagree with you on that, because it is the usual interpretation of relevant evidence found in the federal rules of evidence. So, it's not crazy. So, I'm going to my, my second point would be help address that. Assuming you've got some ambiguity and it would otherwise trigger Seminole Rock, under Seminole Rock and Chevron, you only defer if the determination reflects the considered judgment of the agency as a whole. And given the way the VA board is structured, there are something like 98 members of the VA board. They issue, I think, over 80,000 decisions a year. Their proceedings are ex parte. They're all individual member decisions. They're not made in panels. And I think, and none of them have any precedential value. Given that suite of factors, we don't think that any individual board decision by the VA board reflects the considered judgment of the agency as a whole. Wow, so you would have this court and courts across the country judge agency decisions as to how considered they are. No, Your Honor, that's- Isn't that a bit of, asking a bit of inner branch disrespect? I don't think so at all, Your Honor. It's exactly what this court said, that the rule was in the Meade case when you're undertaking Chevron deference. No, in Meade we said that Congress didn't delegate it in those cases. Here, we're way past that. We're on factor four or five of your six-part test. But I think both- And a judge has to decide how considered the agency decision is. Right, but I think in Meade, both the majority and the dissent agreed that you wouldn't get to Chevron deference unless the decision reflect the considered views of the agency as a whole. They just disagreed over whether or not the particular decision issued in that case, the customs letter, reflected that considered judgment. So I don't think that's an innovation that we're asking for. That's simply an elemental aspect of it. But to go to the other parts of your question, Your Honor, when you get down to the application of the veterans canon, the court, of course, didn't grant certiorari on the application of the veterans canon, but assuming that it applies in the context of regulations, we don't think that it would apply in petitioner's favor here, because we believe that that is a tie-breaking canon that only applies when two interpretations are equally plausible. And here, we think that our interpretation, even if you don't think it is the theoretically best one, we think that it is more plausible than petitioners, and therefore, you wouldn't get to the application of the veterans canon. But again, our principal concern on behalf of both the VA and the other agencies throughout the United States is in preserving Seminole Rock in its core applications, because that is an issue that transcends the facts of this case. So if I'm understanding your views in answer to Justice Gorsuch, you're basically saying a decision by, let's assume, a BIA court is not enough, unless a BIA, what? Unless the agency heads? Yeah, not at all. Tell me when they count and when they don't. Not at all necessarily, Your Honor, but I think what this court's decisions have been clear about across the board is that whoever issues the decision on which we're seeking deference has to be able to speak for the agency as a whole, and different agencies have different ways of doing that. We don't think that given the suite of factors at issue specifically with respect to the VA board meets that standard, because there are so many different indicia suggesting that an individual board decision doesn't reflect the considered views of the VA as a whole as to the meaning of its regulations. Thank you, General. Three minutes, Mr. Hughes. Thank you, Mr. Chief Justice. And I'd like to begin, and we thank the General for the clear recognition here that deference does not apply in this case or other cases like it. We certainly agree with that conclusion. But we still believe that the appropriate resolution of this case is to overturn Seminole Rock and our in their whole, because it's critical to restore the importance of notice and comment rulemaking that Congress thought was a critical check to bring democratic accountability to the agencies. We certainly agree that agencies have a very substantial role to play in policymaking, but Congress made the judgment that the way that that is done in a democratic way accountable to the population is through notice and comment rulemaking, such that the regulated public can provide their views. And that also accounts with the theoretical underpinnings of how this court has explained that deference can be appropriate to agencies. There are two things that are required. First, a delegation of the subject matter, but second, that the agency acts in the particular manner that Congress has delegated the agency to act within. In this context, as we've explained, the particular manner that the agency identified was through rulemaking that provides the public that ability to participate. And that's the fundamental problem. What is your answer to the delay? And what do we do in the interim, one year, two years, three years? Well, a few things about the delay, Your Honor. That's part of the balance the APA struck. If the agency wants to move faster, it can use interpretive rules that bring consistency to the agency, but don't have binding effect in law. In courts, they would have the effect of skid more. In the event that there is some sort of emergency situation, the APA contemplates that for allowing for regulations pursuant to the good cause exception, if the agency can show that there is something that is akin to an emergency that would warrant something like a preliminary injunction in court. So Congress has provided for those sorts of emergency situations when the delay in the regulatory process would actually pose some kind of practical problem. But to turn additionally to the practical problems that exist in our deference, as the Chief Justice was explaining, I think you get a non-satisfactory result regardless of how courts apply it. If courts apply it as they did in this case to say, we don't have to really do much statutory or textual construction to determine if both sides have an argument that looks plausible on the page, then we defer, that is not a particularly satisfactory answer. By contrast, if courts go far down the road of step one and do the interpretation, but then ultimately decide, as many courts have had to do, that although we think the agency has it wrong as a matter of interpretation, we still have to defer to the agency because it's close enough, that's also not a satisfactory answer. By the way, the General said, if we adopt your interpretation and rescind our deference in total, that every case that relied on our deference would be subject to new litigation. Well, Your Honor, I think as I explained with the Marsh example earlier, all of those cases are already fundamentally unstable. No, but they're still going to come to court for courts to decide if that's true or not, every losing party under prior our deference litigation is going to come to court to argue that under its reading, it has the better reading. It could be shot down, but it's going to still argue it. If I may, Your Honor. Yes. I don't think that increases any instability in the aggregate because the existing circumstance is completely unstable. However, if prospectively our does not apply, that is what ultimately leads to stability because interpretations of regulations would just be like interpretations of statutes that would have binding effect absent the agency or Congress going through the process that's constitutionally and statutorily prescribed for amending the underlying text. Thank you, Counsel. General, the case is submitted.