Oral argument — State of West Virginia v. EPA

Case number 24-1009. State of West Virginia et al. Petitioner v. Environmental Protection Agency and Michael S. Regan, Administrator, United States Environmental Protection Agency. Mr. Mansinghani for the petitioners, Mr. Hoshijima for the respondents, Mr. Griego for the interviewees. Good morning, Mr. Mansinghani. You may begin when you're ready. Thank you, Your Honor. Mr. Mansinghani, on behalf of West Virginia, Oklahoma, and a total of 26 states as petitioners, I'd like to reserve three minutes for rebuttal. Under Section 111D, states set the standard of performance for existing sources that reflect EPA's best system of emissions reduction. EPA promulgates regulations governing the procedure for how states submit plans, but that procedure shall permit states to take into account remaining useful life and other source-specific factors in setting a standard of performance for a particular source. The statute thus clearly grants states discretion in setting standards of performance. The question here is whether, when a state reasonably exercises that discretion, EPA can nonetheless reject those reasonable judgments as EPA purports to do in the challenged rule. The answer is no. Nothing in the statute grants EPA that power. EPA interprets the language shall permit not as a command to honor state discretion, but as a power to restrict it. But nowhere in law do legislative commands that the executive shall permit something without more mean that the executive is free to restrict it in any way it wants. And so, too, with EPA's creation of a power to error correct, approved 111D plans are called for plan revisions, which Congress gave in Section 110K of the Clean Air Act for 110 plans, but did not give in Section 111D for 111 plans. The rule is also arbitrary and capricious because it set a deadline for the submission of state plans that was not justified by the evidence before it. EPA rejected a nine-month deadline because the data showed that most states took longer than nine months to submit their plans, but the exact same is true of the 18-month deadline that EPA chose. EPA wholly failed to explain why 18 months is adequate based on the record evidence before it, and for all these reasons, we ask the court to vacate the rule. I'm happy to answer questions on either of those topics. When you say it wholly failed to consider why if nine months to 18 months wasn't also too short, I mean, it didn't consider only the historical data about how long states took to comply. So it did explain. It explained by looking at statutes that it thought were roughly similar in terms of the activity they required of a state and described how long, you know, and identified that EPA had given 18 months for analogous inquiries. It looked at the need for prompt control of emissions. I mean, there's a whole series of things that it looked at. It looked at information provided by states on the sort of different procedural components. It looked in shifting from 15 months to 18 months at the additional work of, I think it's public engagement, states engaging with the public. And it also, it looked at a comment that said, yeah, there is historical data, but some of the states have taken longer than necessary. And it also looked, for example, at this is only a default. And it means that in two ways. One, when a particular emissions guideline is published, EPA can, and indeed it has, extended more time for states to submit their plans. And even when it turns out for a particular state that is too short, there's no sanction for going longer. So when you say they didn't at all explain why they didn't just adopt the lowest common denominator of the amount of time that states historically have taken, what's wrong with the understanding that they did explain? And they looked at, this is a timeline, right? It's longer, shorter. And so all the different factors go to the same question. And so they did explain. So there's a lot there, and I'm going to do my best to try to tackle all of it. You know it already. But feel free to interrupt me if I'm going too long. Ultimately, I think with respect to the record evidence, and there's two big pieces of record evidence here, which is the data of how long it took states to submit 111 plans, and what states told EPA of how long state procedures take to actually adopt these plans. EPA did not explain at all how their 18-month deadline was consistent with that data. Instead, as you pointed out, Judge Pillard, move straight to, we're just going to compare to other statutes, and I can get to those other statutes later. But at most, EPA said, offered at most ipsa dixit by saying, well, we considered these things. It didn't actually explain how the data that it collected, or the state plan, or the state comments on state procedures was consistent with an 18-month deadline. Now, in terms of, well, EPA can always extend the deadline, of course, the fact that they've done so a couple of times already might only indicate that the deadline, the default deadline is too short. And of course, in American Lung Association, they could have shortened the deadline for a specific admissions guideline, and yet the court did not hesitate in striking down the 36-month deadline as inadequately explained. And then the last thing I'll say on that is that the extension is with respect to a particular admissions guideline, but once that admissions guideline is there, EPA very clearly said at Joint Appendix 47 that it won't offer particular states any extensions to the extent that they need them. I guess I'm not sure what more EPA is supposed to say. So, I mean, the statute, I mean, it might be that the best reading of the statutory scheme is that it should be a 36-month deadline because that's what's in Section 110. But this court has said that's not a sufficient basis, and so EPA has to explain why, you know, how it balances different interests. But, I mean, what if they had picked 24 months or, you know, 38 months? Or, you know, once you're talking about a deadline where you're looking at, you know, wanting states to comply in a timely manner but also give them enough time, I don't know. This is the sort of thing, I mean, how much is an agency really supposed to say? Well, certainly the failure to explain is the core holding of American Lung Association. And so if they're making the mirror image error here, it should equally deserve vacature. And as far as what they should do, so, for example, they said, look, we saw that states said that their procedures take somewhere between 15 to 36 months. We can't accommodate every state, so the end, right? That doesn't explain how, what the state's comments were on state procedures is consistent with their 18-month deadline. What they could have done is what they did with their federal deadline. They said, look, here's every step of the process of doing a federal implementation plan. Here's how long that we think it'll take, and we're setting our federal deadline as long as that, at that level. They could have done the same with respect to what they think state procedures could be. They don't have to accommodate the least common denominator. But what they can't do is just ignore as a whole how long states take to submit the plan. Same with respect to the data. I think them saying that states took longer than they needed is a post hoc rationalization. I don't see it anywhere in the rule. But they at least could have explained why they think states took longer than they needed, rather than collecting the data and then doing nothing with it with respect to setting the 18-month deadline. Well, you say, how is it consistent with the data? I don't think that's the obligation. They don't have to make their timeline consistent. And you said that this is a mirror image of what happened and what we found inadequate in the American lung. But that's not right, because the American lung literally did not mention the need for speed, the need to, you know, the cost of a longer deadline. And it seems like, in a way, your argument is doing that, too. It doesn't need to be consistent with the data, because as Judge Rath's question mentioned, the task before EPA is necessarily dealing with conflicting exigencies. It needs to control, you know, have a process to control these harmful emissions and do it in a way that is orderly. And it mentioned and collected this data. It's a pretty limited data set. And it does mention it, and it does consider it, and it does actually take action in terms of lengthening the default deadline based on it. It's a little bit hard to say what more, without micromanaging the agency, what more we could demand. Well, so the reason it rejected the nine-month deadline is because most states didn't meet the nine-month deadline. But the exact same is true of the 18-month deadline. So that's inconsistent. About a quarter. But with respect to American Lung Association, I don't think the only aspect of that decision is this failure to consider the environmental protection needs. The court, basically, I read it as offering three reasons. That it had the burden of justification for timeframes. So it rejected mirroring 110 without a reason to court. And that it failed to document any problems with existing deadlines. So EPA then responded to that by collecting a bunch of data, rejected a nine-month timeline, but never explained how they were using that data to come up with the 18-month timeline, along, of course, with the state procedures problem. And so the question isn't necessarily that they had to pick the exact average. It's that they had to come to a recent conclusion based on the record evidence before it. So they had the record evidence before. And I think in this case and in any other case, if that record evidence is not somehow dealt with in a reasoned explanation, that's arbitrary and capricious. What is your best argument that that failure of reasoning that you've alleged is not just harmless error? Well, I mean, it's... I mean, unless you sort of have to tie it to the median amount of time or the average amount of time or something like that. So what I'll... One is with respect to the how long state procedures take, I think that it's certainly not harmless error because if you took into account state procedures, you'd have to get something higher than 18 months. And even if you're not getting to the average of 45 months, I think what the data does show is that the 110 deadline that Congress set as a reference point, I acknowledge not a strict, strict reference point, but the data does show that that 110 deadline is actually somewhat consistent with the data that EPA ended up collecting, unlike was the case in American Lung Association. I mean, they do... EPA does discuss that. Much more, many more different sources, emissions. Well, I think that 110 plans can be complex in that way, though not always, right? The methane rule has a ton of sources at the 111 rule. And 111 rules and 111 plans can be more complex in the sense that they take into consideration remaining useful life and other factors to tailor things, unlike 110 plans where they're just trying to hit that one target of the national ambient air quality standards. So, you know, with respect to the statutory comparators, we've, of course, made our arguments as to why they should have at least looked at 110. But the other ones that they chose are sort of picking and choosing. I don't see why they chose those two. So 129, they admit, is just not comparable because of this lack of source-specific tailoring. And 189, well, that's about non-attainment plans for the NACs. And they, for some reason, chose to only compare it to the shortest of those deadlines. This course, Wild Earth Guardians 2016 case, talks about particulate matter as being the strictest of the deadlines. But most of the 110 deadlines and the NACs deadlines are 36 months. So it just seems like an awful lot of cherry picking, cherry picking there. I do want to talk about remaining useful life before my time expires, although I know that's not necessarily the strictest. So West Virginia argues that EPA review here should be only for reasonableness. But EPA, the statutory standard is that EPA can determine whether the SIBs are satisfactory. And I'm wondering, I mean, satisfactory, why doesn't that encompass more than reasonableness? Like, a plan might be unsatisfactory if it doesn't comply with the underlying statute or, you know, violates a different regulation or standard or something like that. So, but your position seems to be that the review is exclusively for reasonableness. Not quite, Your Honor. I do believe it is for review of both reasonableness and compliance with the statute. And EPA agrees it's for compliance with the statute at Joint Appendix 68, as well as page 64 of their response brief. That's similar to the review that is, that happens in Section 110. But the way I would frame it is this way with respect to, let's say it does comply with the statute. So there's no legal inference. Right. The question, the way I'd frame it is nothing in the statute allows EPA to fault state plans as unsatisfactory if a state reasonably takes into consideration factors that Congress expressly permitted them to take into account. So if it's, if states shall be permitted to take into account remaining useful life and other factors and states reasonably do so, I don't think EPA has the statutory authority to declare that unsatisfactory. And with respect to their new restrictions on remaining useful life, understand that these are restrictions on top of the existing reasonableness requirements already in the statute, already in the regulations. So the regulations currently say if, you know, states can deviate if there's unreasonable costs. And now these regulations on top of that say it has to be also based on information fundamentally different from what EPA considered. And we think that layering on top of the reasonableness standards is what's facially unlawful. But I'd take them to be making sort of the inverse argument, which is, you know, in that the rule of shouldn't be a second, you know, double counting the same thing that the commission, that the EPA already took into account. They, for example, look at all the sources. They said something that they think is achievable for the category. And if every source that sort of was in that group that was below average and therefore was part of the data that made EPA pick the limitation that it did as realistic, and all the ones that are below average say, well, we have other conditions, and you're sort of exempting them from a standard that's been set, already taking those conditions into account, that that seems like double counting to me. I don't think it's double counting. I think EPA acknowledges that if there is something as an extreme outlier, even if it was accounted for inside their overall industry-wide data, that it would be something that might justify a variance here. And I think the question is, who gets to make that sort of line drawing, right? If a state is relying on something that's insubstantial or de minimis, I think the EPA can say, no, that's not a reasonable deviation. But in the greenhouse gas rule, EPA says, well, it might be something that's only within the 95th percentile of cost. If a state said, no, actually, this is in the 80th percentile of cost, and we think that that's an unreasonable cost, and so we're going to do something slightly different, I don't think that that's something EPA has the authority to reject. But the other thing I'd say is that the way EPA frames fundamentally different, it says that if the EPA did not consider any information pertaining to a certain circumstance in making its determination, facility-specific information relevant to that circumstance may be fundamentally different. And that's at Joint Appendix 74. But if, for example, EPA, as it should, consider costs in setting the best system of emissions reduction, what that implies is that states can't consider costs in terms of remaining useful life and other factors and setting state-specific, source-specific standards of performance. And I think the third thing they say, and this is at their brief on page 32, is if a source could meet those guidelines, it must comply. And that's just like an impossibility standard, and I think is very far from what the statute allows EPA to do. With remaining useful life, I think the important thing, of course, is that states set standards of performance. EPA does procedural rules, and those procedural rules shall permit state discretion. I mean, states have standards of performance, but they have to reflect the, what is it, the emission standards that are the degree of emission limitation that EPA has calculated in light of the best system of emission reduction. So that is not just procedural. The EPA determines a level of emission limitation that is adequately demonstrated. I wholly agree with that. The standards of performance have to reflect what EPA's determination is in the terms of best system of emissions reduction. But in terms of when you can vary for that, that is what's given to states to exercise reasonable discretion in doing so. And I think that this is bolstered by the precedent all over the Clean Air Act of cooperative federalism, where the EPA set some general standards, states submit plans to reflect that standard, but tailored to local conditions, and EPA reviews on a limited basis and can promulgate a federal plan if disapproved. This is true in the 110 context, and in case after case, like train and Union Electric, the court has emphasized that states have wide discretion and considerable latitude in making legislative choices, and EPA… …particularly to require, how to allocate that. But a choice to say, great, you've demonstrated an emission limitation, an amount of pollution reduction that must be achieved. We're going to take that as just, you know, a rough recommendation. We're going to do something completely different. I know no case that takes that approach. And of course, we're not saying we can… …as long as we reasonably…we, the state, reasonably explain, well, we, you know, all…our sources are all a little older than average, and, you know, this would be expensive. I mean, there's…I take your position to be saying that EPA can't look at those things and second-guess them because it's the prerogative of the state to take them into account. I don't think they can second-guess reasonable state judgments. And I think a state judgment that overrides EPA's industry-wide best system of emissions reduction determination… With regard to that state? With regard to that state is…so, if, for example, you know, a state said, well, all of our facilities are pretty average, and, you know, we think that EPA considered costs incorrectly in their best system of emissions reduction, I don't think that would be permissible. But if a state says, for this particular source, you know, EPA's representative data is not actually representative of this facility, and there are good reasons to, you know, choose a different system of emissions reduction or extend the compliance timeline, that should be within the state's reasonable discretion. And I think this is reinforced by the fact that in other provisions of the Clean Air Act, Congress was specific when it said that EPA's guidelines are mandatory and binding on the states. We pointed out the regional haze example, as explained by the Wyoming case in the Tenth Circuit, and the prevention of significant deterioration program, where EPA's guidelines at times are explicitly mandatory. But I think when you're in this situation, which is much more similar to the BACT, Best Available Control Technology situation, and Alaska Department of Environmental Conservation, or the state plan context in 110 of Union Electric and Trane, state discretion is well established. So, what comfort can you give us if we're inclined to agree with you that, this is a quote from our Weyerhaeuser versus Kossel, that the pinhole safety valve envisioned by Congress doesn't become a yawning loophole? I would be hesitant to describe state discretion as pinhole. And I think Weyerhaeuser involved a different situation, which the statute did not explicitly allow variances, right? This is explicitly allowed in the statute. Nobody disagrees with that. The statute provides a yawning loophole, and that's the way it is, and it's always been. Noah and I describe it a lot, yawning loophole, and I think there's a big gap between pinhole and yawning loophole. But I think states— That's for some comfort. And I'm happy to give that comfort. First of all, right, these regulations have been—the old regulations were in place for a long time, and it wasn't necessarily yawning loopholes created. But second, just like the EPA is subject to arbitrary and capricious review by this court, states are subject to a certain level of arbitrary and capricious review by EPA. And we don't think that agencies, federal agencies, have yawning loopholes merely because they're subject to a reasonableness review by the court. I think the same is true of states. And, you know, as I mentioned at the beginning, the EPA does get to determine the best system of emissions reductions, and state variances can't question those industry-wide determinations. To be fair, I mean— The court, we can. There's only a yawning loophole if EPA can set presumptive standards here. Nobody challenges whether EPA can set presumptive standards, and I'm wondering why that is. You know, in the American Lung Association case, we defended EPA's discretion not to set presumptive standards. You know, I think to the extent that those presumptive standards are not binding on states, you know, I'm not sure that there's any violation of the statute of their guidelines. And, of course, to a certain extent, right, they set the—they determine the emissions reduction available through the best system of emissions reduction. The authority to set a presumptive standard. I don't think it's necessarily—I don't think it's an authority to set a binding presumptive standard, but a presumptive standard where EPA says, like, look, we really looked at things, and if you happen to meet the presumptive standard, we're going to approve your plan, but not necessarily holds you to it beyond, you know, again, the reasonableness review and the statutory compliance review we talk about. And I'm not sure we have a problem with that. You know, at the end of the day, while it's not a yawning loophole, I also don't think that states—that EPA can take this discretionary provision that states have been granted by Congress and make it so tiny that it's practically nothing. So, how do we know whether EPA's—what's EPA's latitude in evaluating whether a state's—state plan's standard of performance reflects the degree of emission limitation and best system emission reduction that the EPA administrator has determined and demonstrated? What's the—if reflect doesn't mean match presumptively in a binding way, unless you can show, you know, whether it's a—whether you can show something that's fundamentally different that allows you to depart from that, then what is reflecting? What work is that doing? I mean, I think reflect works hand-in-hand with the remaining useful life and other factors provision in the sense that if the states have no reason to impose something different on a particular facility beyond just disagreeing with EPA's, let's just say, affirmed best system of emissions reduction determination, I think that that could be rejected as not reflecting EPA's determination. But if there are source-specific reasons to reasonably deviate from that, they don't have to match precisely one-to-one EPA's standard determination. Do you have any response to EPA saying—you say there should be reasonableness review. Of course, there's reasonableness review, but that's a far cry from fundamental differences. And I think they—their fallback argument is actually in the context of this process where we have considered a lot of this stuff, it's unreasonable to vary from it unless you've shown what they call a fundamental difference, something that is sort of new, wasn't taken into account. But that would be the categorical description of the kinds of things that would be reasonable for a state to rely on in departing from a full reflection of the degree of emission limitation that EPA has demonstrated. I guess my response to that is that if they're not saying anything more than just reasonableness review, then you could excise that portion of the Code of Federal Regulations and their standard would be the same. And that's what we're asking you to do, to vacate the rule and take that out of the Code of Federal Regulations. I think they're giving you context. They're saying, we're not going to see it as reasonable. So if you just said reasonableness review, then it would be different from saying, let us give you some explanation of what we think is reasonable given the work that we've done, our role, and your role. And they say, what would be reasonable would be if you point to something that's a fundamental difference from what we already considered. And it just gives you more information. And it does narrow where reasonableness applies. But then once you're there, they're kind of saying, it amounts to reason. I don't think that's how EPA itself characterizes this regulation. And we're going off of how they've explained it in this rule and in other rules like the greenhouse gas rule and the methane rule. I gave one example as the 95th percentile example where that, I guess, to them is fundamental. Or if you take them at their word as saying, well, it has to be information that's fundamentally different. But if they considered cost, if they considered reliability, then that implies states can't consider those things in determining remaining useful life and other source. So I'm taking them at their word. And then maybe the sort of word that causes me the most heartburn is fundamental. That has to be fundamentally different, not just that there's a reasonable deviation available, but fundamentally different, something completely outside the scope of what EPA was considering. Thank you. Thank you, Your Honors. Good morning, Mr. Hoshijima. Good morning, and may it please the Court. Suki Hoshijima, a respondent to EPA. With me at Council's table is Nora Greenglass from the EPA Office of General Counsel. The Supreme Court recognized just recently in the West Virginia case that EPA has the primary regulatory role under Section 7411D. Now, states do also have a critical role to play in the process. This rule appropriately balances the roles and responsibilities of states and EPA. Now, unless the Court... I was going to ask you about this line that EPA relies very heavily on this passing reference to the fact that EPA retains the primary regulatory role. I'm wondering, I mean, you know, the Supreme Court has also in many contexts recognized the system of cooperative federalism under the Clean Air Act. And I wonder if EPA's reading of the Supreme Court's language here as the primary regulatory role, it doesn't just mean first here in this process of setting SIPs. I mean, EPA seems to read primary as meaning foremost or paramount or something like that. But maybe in context, it just means first. You know, first, the agency decides the amount of pollution, you know, using the best system of emissions reductions, and then states then submit plans. So EPA is first and then the states are second. It doesn't mean EPA has the most significant role in the system under Section 111. We don't think that primary regulatory role refers solely to order here. One reason that's so is because EPA's role doesn't stop when it issues emission guidelines at the front end. EPA does make these nationwide regulatory determinations in which it determines how much pollution reduction is generally reasonable to require of the industry. But that's not the end of EPA's role. States then implement those determinations. But EPA, under subsection D2 of the statute, has the authority to determine if a state plan is satisfactory. Petitioners concede even that that satisfactoriness review is a substantive review, at least for reasonableness. So EPA has the responsibility to make sure that state plans aren't undermining the emission guidelines and the initial regulatory determination that Congress assigned to EPA. Now, what EPA generally does in emission guidelines is it's looking at average and representative units in the source category to determine what is reasonable to require generally of sources. And in the second step, when states are turning that determination into standards that apply to specific sources, there's a translation step at which states get flexibility and are able to exercise discretion. States also are able to decide when, instead of just translating EPA's determinations, there's reason to deviate from EPA's emission guidelines. And this rule does give states a real and meaningful opportunity to deviate from EPA's determinations based on source-specific factors. EPA acknowledges that it's making broader nationwide regulatory determinations, considering the industry as a whole. Can you explain a little bit? It helps me to have a more concrete sense of what's going on. When EPA is promulgating emissions guidelines for categories of existing sources, it sets them on a nationwide basis? Generally, yes. With the further caveat that EPA, when it finds some characteristics that are common to a subgroup of sources within a category, may subcategorize within emission guidelines. But once it's decided, here's the set, then it doesn't have a separate guideline for every state. It does it for that set. That's right. The emission guidelines are nationwide rules that EPA promulgates by looking broadly at what system of emission reduction is available and best for sources within that category. And is the level typically set based on average facility in the category? I know in some other EPA, they do the top, get rid of the best performing, and then go for some subgroup below that. But here, average, median? It's not what sources are already achieving on average. The relevant language is in subsection A1, where the standard is best system of emission reduction, and EPA looks at what systems are adequately demonstrated and what is achievable. That may be a system that the industry is not already using on a widespread basis or that the average unit is not already using, but it may be a system that is nonetheless adequately demonstrated that results in a standard that is achievable and reasonably achievable. Yes. Cost is a factor that the statute requires EPA to determine, to consider as part of that process. So it could be that there's some relatively inexpensive, easy piece of equipment that all sources could use. None of them are using it, but it's been demonstrated that it could, and it's cheap. And so you could set a much more demanding, I mean, would that there were opportunities like this, but in theory, EPA could set a much more demanding standard than what's being achieved because it could say this is demonstrated and achievable. Yes, because absent a regulatory impetus, an industry isn't going to necessarily going to adopt the best system voluntarily, even one that is adequately demonstrated and even one that results in achievable emission reductions. That's the basic idea of section 7411 is for EPA to determine what is reasonably, what is reasonable to require of the industry generally. So you, what does it mean to EPA for a state plan to be satisfactory? I thought the definition in the brief, at least as I was reading it, was a little circular. It says that standards and state plans must be consistent with the statute. But I'm not sure that that helps us answer the question of whether the fundamental differences rule is consistent with the statute or not. As one of the ways of implementing the EPA's satisfactoriness review. EPA, yes, has understood satisfactory to mean consistent with the statute. In context, what that means, and we have to zoom out a little bit to explain what consistent with the statute means here. This is a statute where Congress has assigned EPA to make these broader regulatory determinations. States have the opportunity then to deviate when there's something about a specific source. That would make it unreasonable to apply those determinations to that source. So, what this rule is saying, which is consistent with EPA's understanding of the statute for decades, is that it's not consistent with the statute for a state to deviate from EPA's emission guidelines and the determinations in them. It's not consistent with the statute unless there's some difference pertaining to that specific source that makes it reasonable or makes it unreasonable to apply EPA's determinations. I'm skeptical that this fundamentally different standard is continuous with EPA's longstanding interpretation. I mean, can you say a little bit more about that? I mean, one, those words seem to me, those phrases are very different, right? Significantly more reasonable versus a state has to demonstrate that there are fundamental differences. So, the terms seem very different to me, maybe fundamentally different, you could say, but also the fundamental difference standard. So, let me go back to the original rule, significantly more reasonable, that phrase only applied to other factors specific to the facility. And now in the revision to the rule, fundamental differences applies to all of the factors, including cost, physical impossibility, and the rest. So, not only is the term broader, but its application seems to be substantially broader. So, how is that continuous with what EPA has done in the past? Let me start by discussing the relationship between this rule and the prior regulations and then get to the application of fundamental differences to each prong. EPA explained that all it's doing here is clarifying what significantly more reasonable means. Now, I would focus the court's attention on the new version of the regulations, and this is in subsection E1 of the amended regulation, where even in this version, the touchstone is reasonableness. It says that a state can apply a less stringent standard if a facility cannot reasonably achieve a degree of emission limitation in the guidelines because of some of these factors. Reasonableness is also built into subsection E2 of the regulations, and that's where the fundamental difference language is. It says there has to be a fundamental difference such that achieving EPA's determination in the emission guideline is unreasonable for that facility. So, we do think that reasonableness is actually still the core tenet of these regulations. But basically, E2 suggests to me, at least reading this together, that a state shows what it has acted reasonably by state must demonstrate that there are fundamental differences between the information that EPA considered and facilities and information that the state considered. So, the revisions to the rule defines reasonableness as putting a burden on the state to show that there are fundamental differences. What this regulation is doing is, I think, giving a little more clarity about when... You said that, but why is reasonableness now not defined as fundamental differences? Well, the question is, when is it significantly more reasonable? That's the language under the prior regulations. When is it significantly more reasonable for a state to impose a less stringent standard? Well, it's significantly more reasonable when there's a fundamental difference between what is specific to that source and what EPA already considered in its nationwide determination, such that there's a reason to treat that source differently from everybody else in the category. And then what about the application? So, now a state has to demonstrate fundamental differences, not just with respect to other circumstances, but with respect to costs of control and physical impossibility. I mean, that is a substantial expansion, arguably. And you're referring, Your Honor, to referring that language to all three prongs. Yes. Right? So, we don't think that's a substantive difference from what was there before. It's true that in the prior version of the regulations, the significantly more reasonable language was associated with other factors. We think other is the key there, where the first two prongs, unreasonable cost and physical impossibility, are instances of factors that would make it significantly more reasonable for a state to apply a less stringent standard. Which is to say that the significantly more reasonable standard was always underlying all three prongs. It's just that textually it was associated only with this other factors residual clause. EPA thought in the process of revising this regulation that it would be clear to state more expressly what was already implicit there in that listing of the three prongs. And to say that fundamental differences is the basic idea that underlines each of those circumstances. Historically, can you tell us a sort of general frequency, just give us a feel, how often states invoke the remaining useful life of a facility to justify a less stringent standard of performance for existing sources? In the preamble to this rule, EPA identified four historical instances in which states had sought to grant a source-specific variance. Now, that's not a large number, but we would posit that shows that states granting a variance is supposed to be the exception rather than the rule. States aren't supposed to come in and take EPA's determinations as just advisory, but then decide from the ground up what's reasonable to apply to sources generally. So in the other situations they basically adopt EPA's guidelines? Are there other things that they do that change? In the circumstances when a state doesn't grant a variance? Or isn't taking remaining useful life of the facility into account? In those situations, a state has imposed standards that reflect EPA's emission guidelines. That doesn't necessarily mean mirror precisely, states are able to impose a standard that achieves an equivalent level of stringency. So if a state already has a program that regulates sources in a way that is equivalent to EPA's determinations, then a state can make an equivalent she's showing. I don't know if you can, I know this is asking a lot of the lawyer rather than the agency staff. But can you give an example of state already has stuff, EPA says here's the best system for emission reduction, here's the resulting emission guidelines. State says what we can meet the overall amount, so it would be the same amount as if we applied your guidelines, but we're doing what differently? I'm just trying to get a sense of even in the real world, what parameters are we even talking about? I'm not sure I have a specific historical example for you, but there is a little bit more extended discussion of this in the methane emission guideline. For EPA in that preamble to that rule, explained a little bit more about how it would conduct an equivalency determination. EPA set out physically a multi-step equivalency determination. Another way in which a state standard might reflect, not perfectly, but reflect an emission guideline in their standards without deviating from it would be, for example, an averaging or trading program. Where a state can allow a group of sources to, in the aggregate, achieve the amount of emission reduction that EPA's emission guidelines would require. Of course, the extent to which the state is able to do that will depend on the nature of each emission guideline and characteristics that may be unique to each industry. EPA in an emission guideline might provide further explanation of what kind of flexibility is available to states to implement emission guidelines without departing from them. I'll ask you to maybe walk through EPA's arguments about why it claims it has statutory authority to recall and revise state plans. Because 111 D.1 gives EPA the authority to establish procedures under which states shall submit plans. It's about regulations, procedures, about submission. Once a plan has been submitted and approved, what authority does EPA have under 111 to take post-approval action, like recalling a SIP or revising it? In promulgating those two procedural mechanisms, EPA relied on both subsection D.1 and subsection D.2. Starting with subsection D.1, Congress authorized EPA to prescribe regulations that govern how the existing source regulation process works. Let me answer my question. It says submission, submit plans. How is a post-approval mechanism within that language? Because the language about establishing procedures for submission is Congress specifying something that should be in those regulations, but there's not text saying that's the outside limit of the regulations. But other provisions have those specific authorities to revise or recall plans, and that authority was not given in 111. Right. Congress expressly conferred that authority to EPA in Section 7410. In Section 7411, Congress essentially used concise shorthand to say, EPA, given your experience administering these programs, we're going to give you the discretion to figure out which of those mechanisms from Section 7410 makes sense to port over or to bring over as an analog in the context of Section 7110. What in the statute says that that part of Section 110 can be imported into 111? Well, first there's the language, just prescribe regulations to govern this existing source process, and also under subsection D. It's not the existing source process. It's, you know, for submission of plans, to submit plans. But it can't be that EPA's ability to promulgate regulations stops at the submission of the states, because that would mean that the existing regulations about how EPA considers, reviews a state plan, how EPA will promulgate a federal plan. Because that's how they're reviewing the plan. They're reviewing the submission. And then after the approval of a SIP, there's no submission to be regulated. But subsection D2 authorizes EPA to ensure that state plans are satisfactory. We're talking about a state plan. That's part of the approval process of a state plan. And so after EPA has determined a state plan is satisfactory, what regulatory authority does it have after that to later, you know, maybe years down the line, revise or recall a state plan? State plans are implemented over many years after the approval step. And we're talking about a situation where a state plan has become substantially inadequate. In that situation, it's no longer satisfactory. It's no longer achieving the pollution reduction objectives of the statute, because at that point, the state plan is no longer consistent with the statute. All EPA is saying here is that its authority to prescribe regulations for this process and to ensure the satisfactoriness of state plans allows EPA to establish what are just procedural mechanisms. So it's just the, it can just decide whether something is satisfactory onto infinity into the future for all time? Well, these are not new substantive authorities. These are procedural mechanisms for then EPA to determine whether a state plan is satisfactory. So it just has like ongoing oversight authority to determine whether SIPs are satisfactory. It does. That's what EPA is essentially claiming. Yes, because otherwise you would have state plans that are substantially inadequate, and there would be no, there would be recourse that EPA would be able to take. For example, EPA. So what has EPA done in the past? Because this is a new authority that EPA is claiming. I mean, what did it do before this rule? When it thought a SIP was no longer satisfactory? What EPA might do in the alternative will depend on the exact nature of the problem. Some options that EPA described in the preamble is that it may be able to use its enforcement authority under section 7413 to enforce a state plan that a state is not adequately enforcing. EPA may also promulgate a new emission guideline that would require resubmission, but that's an incredibly cumbersome step that leads to the same place eventually. What this procedural mechanism does is to allow EPA to get to that place without having to essentially nationwide repromulgate an emission guideline and require all of the states to resubmit state plans. So one of the reasons EPA says it needs this authority is if a state fails to implement the plan. Is that right? Yes. I thought that's specifically covered by 111 D2B, where there's explicit authority of the administrator to enforce provisions of a plan if a state fails to do it. Is that different? Subsection D2B authorizes EPA to take enforcement action, such as under section 7413, to enforce a state plan. But you're saying if the state isn't enforcing it, that you could recall the plan and... And direct the state to submit a plan that includes adequate state enforcement mechanisms, or else EPA would have to promulgate a federal plan that includes implementation of the emission guidelines and enforcement thereof. Because EPA coming in to federally enforce a state plan that a state is no longer adequately enforcing is a stopgap mechanism, but it doesn't actually solve the issue. And it's a different kind of state-federal friction. So hopefully the states are going to be in cooperative federalism doing their work. So this is a way of prompting the states to do the work, rather than marching in there to do it. The feds themselves. And that's why EPA identified that as one of two types of circumstances in which it might use the state plan call authority. The other type of circumstance that EPA identified in this rule is when there's something that changes after an emission guideline and after approval of a state plan. So can you remind me what an example would be of the change, just as a technical or... It can be something technical. It can be a legal development. EPA explained in the preamble that there may be a variety of circumstances that make a state plan substantially inadequate. I do think it's important to point out that this isn't giving EPA a wide-ranging new substantive authority. It's a new procedural mechanism where states and regulated parties would also have recourse in terms of commenting on EPA's proposed state plan call, seeking judicial review of a state plan call, or an error correction mechanism. We explained in our brief that those are actions that EPA has taken in the Section 7410 context using those rulemaking procedures. In 7410, Congress explicitly gave EPA those authorities. Yes. But they didn't in 111. No, Congress did not expressly include those mechanisms in the Section 7411D process. Instead, it left EPA the room to decide whether those made sense to port over to this program. In the rule... I'm sorry. Do you want to follow up? No, I just... I mean, that's just not how we read statutes. I mean, if Congress gives a specific authority in one section, you know, for one type of program and then withholds it or doesn't explicitly provide it in the other section, the natural inference is that it doesn't exist in 111. Otherwise, you know, I mean, that's normally how we read statutes. I'm not sure that you've explained how we would overcome that presumption. Because Congress expressly linked Section 7411D to 7410. Not those parts of it. Congress was writing Section 7411D in largely a bit of shorthand that refers back to 7410, authorizes EPA to establish procedures that are similar to Section 7410 when EPA decides it makes sense to do so. Again, any of these invocations of this authority are going to be judicially reviewable. At the end of the day, if EPA... That doesn't fix the problem that EPA may not have had the authority to do those, exercise those procedures in the first place. Can we submit that Section 7411D gives EPA that authority, or Congress entrusted to EPA the decision, which of those elements made sense in this context? You said it doesn't give them any substantive authority, it's just procedural. In the rule 6427A1, Romanet 1, the recall is appropriate when the applicable plan is substantially inadequate to meet the requirements of the applicable emission guideline. Isn't substantially inadequate a substantive standard? It's a standard for directing states to submit a new plan, but then EPA's review of that resubmitted plan is going to be under the same authority that EPA has to initially review a state plan. I guess to the extent that substantially inadequate does anything, it actually protects the state because it says, can't be some nitpicky, technical, minor, inadequate. It has to be substantially. It has to be worth going back and reopening things to meet that standard. So, it's not like imposing a new standard on the state. It's actually saying, EPA, you didn't do this earlier. You can't be doing this with every little change in the world. But if it really becomes significant that it's legally or technically no longer meeting the requirements of the applicable emission guideline, then yeah. That's right. In a state plan call action, then the substantial inadequacy finding in it would be judicially reviewable. Now, if I may briefly address the timeline issue. As Judge Pillard recognized, in establishing the 18-month default state plan submission deadline, EPA was balancing competing considerations and paying heed to this court's American Lung Association case, where the court recognized that the central objective of the statute is to deal with the dangerous air pollution. Now, petitioners mostly focus on the historical data. And to make a couple of quick points about that data, we do think that data actually supports EPA's conclusion here. The data we're talking about is in a spreadsheet that's in the record. And that spreadsheet shows historically how long states have taken to submit state plans. EPA didn't look at that data and say, oh, there was one outlier state that managed to submit in 18 months. There's actually well into the double digits instances, historical data points of when states were able to submit in less than 18 months. Petitioners are faulting EPA for not taking the average, but the average just isn't the right metric. Because of this need for speed that the court identified in American Lung Association, EPA decided that it makes more sense to look at the minimum time reasonably necessary. And so if you're looking at the minimum time reasonably necessary, that's not the historical average. If the court has no further questions. We ask that the petitions be denied. Thank you. I think we have one more. Interveners, five minutes. May it please the court. My name is Matthew Greco. I'm here from New York on behalf of the intervener respondents. The petition for review should be denied because EPA's rule is consistent with both cooperative federalism and under the best reading of Section 111 D with the assigned roles of states and EPA. I want to start by explaining why the maximalist view of state discretion that runs throughout petitioner's brief is plainly incorrect and then move to why EPA's specific regulation is valid. The text of the statute alone refutes petitioner's argument. And their first error comes in treating 111 D1 as though it is primarily structured as a delegation of authority to states. The relevant statutory text here that they rely on regarding source-specific factors occurs in the midst of a sentence directing EPA to write regulations. And above and beyond that, by using the term standard of performance, it incorporates the definition of that term in Section 111 A1 to mean a standard of performance that reflects the degree of emissions limitation achievable using the best system of emission reduction that the administrator determines. This has been adequately demonstrated. So it is fundamentally a rulemaking delegation to EPA with a proviso included in the midst of it that EPA exercise that delegation in a way that provides some mechanism for remaining useful life and other factors to be taken into consideration. And when Section 111 D1 is properly understood in all of that context, it's clear why petitioner's reading of the statute has to be wrong. The statute not only permits but requires EPA, because it is a statute telling EPA to write regulations, to craft a regulation setting forth how remaining useful life and other factors may be considered. The petitioners want this court to read a sentence that expressly delegates rulemaking authority to EPA in such a way that EPA has no authority to set contours for how remaining useful life is taken into consideration. And for example, at page 15 of petitioner's brief, they argued that the statute places no limits on statute state authority other than the reasonableness that applies to all rulemaking. But that essentially reads the rulemaking rule for EPA out of the statute. And there were several other features of the text that further confirm EPA's reading is correct. The best reading of the phrase is shall permit and take into consideration is that the thing to be considered must receive a meaningful evaluation as a part of the process, not that it will be dispositive in every case. And in addition to that, we have the word satisfactory in the statutes making clear that EPA's role is to review whether the state plan is consistent with EPA's guidelines. And RULOF is meant to be an adjustment to that. And in addition to the statutory text, the Supreme Court's holding in West Virginia versus EPA could not have been more clear in stating that the agency, not the states, decides the amount of pollution reduction that must ultimately be achieved. That is in addition to the language that received some attention earlier about EPA having a primary regulatory role. I do want to say that on the question of the meaning of what the Supreme Court meant by primary regulatory role, I don't think it can mean simply that EPA starts the process at first and then it is handed over entirely to the states. It's a better way to understand it is that the states operate in the middle of the process, that EPA adopts its guidelines, states propose plans, which may in some cases involve a RULOF determination, and then it proceeds to EPA satisfactory review. Back up for a moment. You went rather quickly through shall permit, taking consideration, meaning it gets a meaningful consideration, and satisfactory. Focusing on shall permit, your sister states have said that shall permit means must permit. And your response to that is? The response to that is that it means that it shall permit them to take it into consideration. And the error lies in believing that take into consideration means that in every case, a source specific factor is going to be outcome determinative. Take into consideration means that the process provides a meaningful way to ensure that source specific factors are not overlooked during the course of the regulatory process. And in their view, that they can affect the bottom line. I mean, take into consideration is window dressing, if it can affect the bottom line. And in some cases it can. And it needs to be a meaningful review process. But the petitioner's brief suggests that once a state has made a determination that remaining useful life or other factors ought to apply, the EPA has to in all circumstances treat that as outcome determinative. And that's not what the statute says. And the statute, to the contrary, says the EPA is to set a set of procedures for determining how states shall be permitted to take this into consideration. So if a state has the opportunity to present its arguments for a variance based on remaining useful life or other factors, EPA is still within its delegated authority under the statute to set the regulation that it has set here, explaining what showing needs to be made in order for that to be met. What is your understanding when you say a set of procedures? You know, that seems like arguably it may be what Virginia would say, hiding something that's anything but procedural. You're saying not just when you take it into consideration, you know, that you have a moment to do that. But limitations on how you take it into consideration. And, you know, I think EPA is kind of saying, no, take it in consideration. And we're really subject to reasonableness review. But I guess you could say it's procedural in the sense that it only trains on things that haven't already been counted for. I don't know. Help me out with why your characterization of EPA's role is setting procedures for how that's taken into consideration. Why isn't that just three card Monte putting EPA's substantive restriction into procedural guard? Well, a couple of things. I mean, I first of all would point to the fact that the regulations have historically, ever since the original implementing regulations in the 1970s, included substantive requirements for what needs to be shown in order to make a remaining useful life and other factors determination. And in fact, the 2019 regulation retained those original implementing regulations with only minor wording changes. And many of the states who are petitioners here today defended that 2019 regulation by arguing that it simply repeats the statutory language. And the contribution that EPA has made here pursuant to its statutory delegation is that it provides additional clarity. As my friend from EPA was explaining, it provides needed context for what a state needs to show in order to make a rule of determination. And as states, we support a clearer rule because clarity in what showing is required for a variance supports transparent application of EPA's standings. And your understanding, if we were to invalidate this, is that it would roll back to the 2019 or the 2019 device terms doesn't apply 2019 standards that you just alluded to. I think it would roll back to the original regulations. 74. Right. Yeah, but even if the consequence were that it would restore the 2019 regulation, which required factors for a less stringent standard to be significantly more reasonable, a rule that not only went unchallenged by the petitioners, but which many of the petitioners intervened to defend, the delta between that rule and this one is comparatively small. I also want to point to this court's decision in the Weyerhaeuser case, which demonstrates the longstanding familiarity of this court with the term fundamental difference. And it refutes the idea that because EPA requires a showing of fundamental difference, that the fact that any information has been considered as to a particular topic rules out a variance under a remaining useful life and other factors. And the court clearly understood in that case that if a factor was considered but is present to a fundamentally different degree, that can support a variance. And EPA, in adopting the current regulation, did not choose the words fundamental difference coincidentally. And I also want to emphasize that we should want EPA to set a standard that provides a lot of clarity for when a remaining useful life and other factors determination is appropriate. Congress wanted EPA to be meticulous in its original approach in developing the guidelines. It is reasonable to require a showing that something fundamentally different was not considered in order to obtain that sort of variance. And Weyerhaeuser also makes clear that when Congress puts a safety valve into a statute, EPA is not required to interpret that safety valve as a yawning loophole. Petitioner's interpretation would prioritize the safety valve over the emission guidelines itself, which is not consistent with the text of the statute and is certainly not consistent with an understanding of how the regime works that was viewed by the Supreme Court as uncontroversial in issuing this decision in 2022. Just to briefly talk about the timeline issue. The starting point on the timeline issue has to be that EPA has to engage in some sort of balancing in order to arrive at a timeline. As Judge Pillard said during one of the earlier arguments, there are conflicting exigencies. Once you accept that there has to be some degree of balancing, the question is whether EPA reasonably cited to a basis for the deadline that it shows. As my friend from representing EPA said earlier, there are a significant number of states that have been able to meet the 18-month deadline in the past. The spreadsheet that is linked at page 104 of the Joint Appendix reflects that. That shows that it's attainable. It's reasonable for EPA to set a deadline that is consistent. It can choose something that is at the shorter end of what states have been able to attain in the past because the statute contemplates that EPA is intended to move the process along to some degree. Thank you. Thank you. Now, did Mr. Manchinhani reserve time for a vote? Three? Okay. Thank you, Your Honors. A couple of quick points. When I heard my counsel on the other side reference what is when variances are allowed, he says, well, if the information is different, that makes it more reasonable. And he omitted the word fundamentally because I don't think you can utter the word fundamentally and say this is just the same thing that's been going on all along. Now, you know, they point to this primary regulatory rule language in West Virginia. But I think, Judge Rao, you have it right that in a certain sense that EPA acts first, but also that they set the default in terms of making the generic technical determinations about feasibility and reliability and cost. And that is one of the three things we think bounds find state discretion in that they get to make that initial determination. The rule of remaining useful life and other factors determinations by states have to be consistent with the statute and reasonable. So I hardly think that this is unbounded. Can I just interrupt you on the error correction and recall piece? Your argument, I gather, is that they're not listed in section 111 or section 110C. They're not authorized. But there are the three other authorities, partial approval, conditional approval, and parallel processing that also aren't listed. But those aren't challenged. So I think all of those relate to the state plan submission process within their 111D1 authority. And, you know, so, yeah, I guess that would be my answer to that. I guess, yeah. So you're just saying that even though in some sense certainly the error correction can be said to relate to the state plan submission, EPA says we made a mistake in, you know, accepting your submission. We should have told you this or that or we just added something up wrong or, you know, so. It is partially temporal, yes, in the sense that this is relating to the state plan submission and approval process that comes in in D2. But it's not a post-approval thing like error correction or calling for plan revisions. The post-approval authority, as you pointed out, is explicit in terms of failure to implement plans and aligning it with sections 113 and 114 of the Act. And I think for all of the reasons that were given in the questions by Judge Rao, I think that issue is pretty clear. So the last thing I want to touch on where my time expires is on the timing issues. I think the real problem here is that they did a lot of work in collecting data and telling states, give us comments about how long state procedures will take. And then they did not at all, at least in the face of the rulemaking, consider that in terms of setting 18 months. The only justification for setting 18 months was this cherry-picked statutory comparators when it comes to the data. They didn't say anything about how that, what they're saying now about, well, enough states submitted their plans in 18 months. It's not in the rule. Did they move from 15 to 19 based on that data? They moved from 15 to 18 based on that data, but they never provided any explanation for why 18 was enough as opposed to something higher. Or they moved, I would say, from 9 to 18. I think the 15 to 18 was based on the meaningful engagement requirement. They doubled it based on that data, and you're saying they didn't explain why they didn't just accept it wholesale? I think they didn't explain why they chose 18 from that data. What they're saying now is enough states met 18, which is nowhere in the rule. In a certain sense, going on with the failure to consider the state comments as to how long state procedures take, that's an arbitrary and capricious decision for failure to consider the record evidence. It's also a failure to respond to comments because all they had was ifs, or dixits that, well, we reviewed the steps states need to carry out. That's a Joint Appendix 45. We won't find any other explanation of how they accounted for that factor. Just on the issue of arbitrary and capricious, the whole argument about the EPA's ability to, I guess, cabin the state's use of reasonably considered source-specific factors. What I see in your brief that's developed is an excess of statutory authority argument. I don't really see developed an arbitrary and capricious argument. Did I miss that? I think that's correct, Your Honor. It's primarily that this is outside of their statutory authority. I don't believe, with respect to this issue, we made an arbitrary and capricious argument. Thank you. Thank you, Your Honors. Thank you. The case is submitted. Thank you.