# Oral argument — United States Sugar Corporation v. EPA > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A91253 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** March 21, 2024 ## Text Case number 22-1271 et al. United States Sugar Corporation Petitioner versus Environmental Protection Agency. May it please the court, I'm Tim Bishops representing U.S. Sugar Corporation. At council table with me is Shannon Broom who represents the other industrial petitioners in the case. The first issue here is whether EPA may impose new source emission standards that it finalized in 2022 on boilers that were designed and built years earlier going back to 2010. And section 7412A4 of the Clean Air Act defines a new source as one that began construction quote, after EPA first proposes regulations establishing an emission standard applicable to such source. Now EPA says that that definition is ambiguous. It could mean the first time that EPA proposes any standard for the source category, the first time it proposes a standard under a particular rulemaking record, the first time it proposes a particular standard. And it says that ambiguity leaves it with discretion to pick and choose among those meanings in a particular rulemaking. We don't think EPA has that discretion and we think that because other provisions of the statute, the structure of the statute, and EPA's own prior explanations of the purpose of the new source will show that new source emission standards are not intended to apply to already constructed sources. And let me begin by saying, by about why the why the statute should be read to mean only a particular regulation. New source is governed only by a particular regulation. So before you get to EPA's positions, can we just start with text? Yes. And one feature of the scheme which contemplates regulations being promulgated and repromulgated and getting stricter over time. And we have the verb establish running throughout the statute, which has some connotation of doing something on the front end of the statute. And then we have the verb revise, used to denote the amendments as the regulations grow increasingly stringent. So in the context of that sort of scheme, can't you read establish to mean the first regulation and then everything after it is a revision? Well, I don't think the structure of the statute, I've got a but I don't think the structure of the statute lends itself to that to that to that reading. Once a once a standard is established, the revision process is governed by E6 and F2. These are regular risk reviews that occur at a minimum every every eight years. When those are conducted, there's no question for either existing or for new sources, there's no question that those rules are perspective only. That's all EPA has ever done. Those are perspective only. Under their perspective, but they're still, they're revisions and therefore they they might not be establishing. Well, you know, what EPA is doing is to use the fact of the remand, and not even the remand in U.S. sugar, but a remand that they asked for in U.S. sugar that went back to NACWA that was three years ago. They're using the remand here to really contort the process, which is that you have you have a standard and it's revised and that standard applies to new sources after the proposal, but that all the perspective only. It's really extraordinary to use the remand in this case that is nothing to do with U.S. sugar, but NACWA is something that they threw in at the end of that litigation to say, well, three years ago this court required a remand, required it in a different rule setting, required that we that we approach these issues this way, we haven't done that yet, so we'd like to ensure that a remand on the NACWA issue, and now 10 years later they're saying, well, we're going to apply these new source standards back to based on that remand. That's an extraordinary use of the remand process in a way that we think contradicts the entire structure of the statute, which is about revisions being prospective, and that's in section 111.2. In section 111, the parallel provision that deals with perspective only, and the RTR revisions that come under B6 and F3 are perspective only. So really EPA is just using the remand here to push back 10 years after all of these folks have invested huge amounts of money complying with the 2013 standards to say, no, you've got a retrofit. And what EPA said in the Portland Mint case, we think makes very good sense, and I think the strength of EPA's reasoning is more important these days than Chevron, which may be about to disappear. So if you look at the strength of EPA's reasoning... Can I give you a hypothetical? Imagine there had been no history, no previous proposals, no previous rules, and in 2020 EPA had proposed a standard. Notice and comment happens. In 2021, you build a boiler, and then in 2022, the final promulgated rule is slightly stricter than what was proposed. Is this where the three-year kind of leeway would kick in? Yeah, Judge Walker, and we do think that that provision that deals with that, which it's 7412I2, is important here. It's telling, because it's a special rule. If you build after a proposal, then the standard becomes more stringent in the you have the three years to comply. We think that if Congress had envisaged that that would not just be a one-off situation from the proposal, and then the final getting more stringent. But if that, as EPA says, was going to be continuous, they were going to keep looking back to the very first proposal, but it would have said so. It would have given boilers time to comply. It didn't. It only gave them time to comply when there is a proposal, a final rule that's more stringent. And everyone agrees that that provision does not apply in this case. So we think that that's one of the textual and structural issues here that is very telling. So is your position then that even if the we're still bound by Chevron, this is an unreasonable interpretation, and so we owe no deference to it, and you should vacate the rule based on that? Yes, it's an unreasonable interpretation. We think that when you look at it, when you look at the language of the definition that EPA has said is ambiguous, when you look at it in the context, the text, and the structure of the statute, and provisions like I-2, it's clear that Congress was only talking about the particular rule and not, as EPA says, being able to look back to prior proposals. And here is what the EPA said in the 2006 Port on Cement Rule. It said, quote, the whole premise of new source standards being more strict than for existing sources is that newly constructed sources can immediately install the best pollution controls without the time or expense of retrofitting. And it went on, if we were to require new sources to retroactively install controls because we have changed the rule requirements, they would have to bear retrofit costs. And you're sure here we'll face more than $20 million in retrofit costs. And EPA said, we do not believe that those retrofit costs were intended by the Clean Air Act. And it said that, you know, we cited a litany of rules in our brief where EPA is likewise... If we agree with you about Issue 1, do we need to rule on Issue 2? That's a good question, Judge Walker. I think, you know, if you rule for us on the 2013 standards and so the HCL standard that they're relying on would be gone. So strictly, you don't. But on the other hand, you know, EPA has... I don't think there's any reason at the moment to think that they would change their mind. We've commented on how they came up with the HCL standard. They have... Assuming everything you just said is correct and the only boilers that you have, Boiler No. 9 and the biovast boilers... Yes, the 10 boilers in total. Right. And they were all begun to be built before 2020, right? Yes. So I'm not sure... There's a question, not a statement, but would you have standing to challenge the prospective application of the rule proposed in 2020 and promulgated in 2022? On the HCL, I don't think, ab initio, we would have standing. You know, given where we are in the litigation, I don't think it would be improper for this court to offer some guidance to... An advisory opinion? To EPA. You know, we think that's clear. What EPA should have done on the HCL is to support, although it gives a great deal of deference to EPA on statistical issues... Do you want us to give the guidance if the guidance says we think EPA was fine? No, but that's, you know, EPA has to explain it to us, and it hasn't done that. It clearly has not done that. Related question. If you were to prevail on the new source definition, but nothing else, then wouldn't the proper remedy be a partial vacature of just that part of the rule? As to the... I believe that the part of the rules that apply going back to 2010, which affects just these 10 boilers... That's just the definition. That's just what you contend to be the erroneous definition of a new source. Yes, yes, because those are the only affected boilers here. I'd like to res... Vacate the definition is what you're suggesting? I think vacate the definition. I mean, what you... Why not just prohibit enforcement as applied to your clients? Well, you could prohibit enforcement. I think that would be an alternative to reach the same result. I mean, everyone that built after 2020 knows that they have to comply with the 2022 rule. I mean, that's unproblematic. So the only affected parties here are the 10 boilers that built between 2010 and 2020, the petitioners here. I can reserve the rest of my... What if the remedy was this? We vacate any parts of the rule that apply the 2022 standard to boilers that had begun to be built before 2020? Yes. Now, that would be broader than just your clients. That would be universal vacater, which I think is our practice. So I'm curious to hear if DOJ will push back against that as they have been in some recent cases. Thank you. Good morning. Perry Rosen for the United States. Let's look at the key statute here, this key statutory provision, Section 7412A.4, which is the definition of a new source. Applying it to the facts of this case, EPA first proposed regulations under this section establishing an emission standard for boilers in 2010. That's a reasonable interpretation of that provision. There may be other interpretations of that provision, but that's reasonable. And it's consistent with the way Congress wrote the statute. First of all, this definition differs from the definition of new source in other provisions, such as Section 7411, the new source performance standards. There, a new source is one described or defined as one created after the publication of regulations. That's very different than the definition here. Secondly, the Congress changed the definition. This is a very purposeful definition. Before 1990, the statute applied to the administrative proposes regulation under this section establishing an emission standard. In 1990, they put in the word first. It had to be a very purposeful change when EPA first proposes an emission standard. That's how you define a new source. So, how does it work if EPA revises these standards in 2028? There's a boiler that's already been constructed, a new source, because this standard is applicable to that source? The 2028 standard is applicable to that source? It would depend on the context of the revisions. If the revisions to the standards that EPA has made are subject to a whole technology review, for instance, or vacature, where EPA collects all new data and creates all new emissions standards. What does that have to do with the language in the statute? I mean, this qualification that you're putting on it, I don't see how that's connected to the statutory text at all. I think the EPA's interpretation of first proposes is when it is based in large part on the collection of data and the creation of a whole set of standards. In that case, as EPA has done in the past, it's interpreted first proposes there as a new date. So, new sources would be defined differently. What if in this case, everything were the same? The whole industry was the same, but I'm going to change one thing. EPA had used a new data set when it came up with the 2020 proposal. Then, would you still be arguing what you're arguing here, that boiler number nine is a new boiler? I'm not going to avoid that question. It would depend on all of the circumstances, but as a general matter, no. If it was a whole new data set, at least consistent with EPA's practice in the past, it has been redefined the date for a new source. This is Judge Wilkins' question again, but I don't see how, whether there's a new data set or an old data set, that turns an existing boiler into a new boiler or a new boiler into an existing boiler. The statute doesn't talk about data sets at all with regard to what's new and what's old. The statute doesn't, but again, in 2013, EPA explained how the first proposed can apply to different situations, and those situations included your hypothetical here. You're trying to get some mileage out of the remand without vacature aspect of this case. If you say that you want to apply this definition to 2028 rulemaking or the 2035 rulemaking or the 2042 rulemaking, it just gets ridiculous. So, what's different about remand without vacature? It's still a different standard formulated through notice and comment rulemaking, prospectively applied. You've made a choice not to reopen the data set, but what does that have to do with anything? We agree it can get or certainly appear ridiculous as the years go on. But in our view here, we're acting pursuant to a very limited remand here. 17% of the rulemaking is 50 pages long. It changes dozens of standards. It accounts for MACWA as well as the instructions here. It does all sorts of things besides just adjusting the numbers? Yes, it does. But if we, again, it was limited to just the errors identified by this court. It's based on, again, the same data set, which petitioners in their interim brief in the next case argue it should be based on. What if instead of happening in 2020, the proposal, for whatever reason, maybe losing court a few more times, and maybe there's some delays, and none of these things take time, maybe whatever. Imagine we're a court and we're considering a proposal that's made in 2110, 100 years after the 2010 proposal. Would you still be saying that this boiler number nine is sort of a forever young, still new, even though 100 years have passed? We'd be hard-pressed to say that in that situation. We'd certainly agree with that. Okay. So where in the statute is that line drawn? And my next question is going to be, where is the line that you're drawing? Is it 50 years? Is it 40 years? Is it 30 years? There's nothing explicitly in the statute. We'd agree with that. Our view, as EPA stated in 2013, is that the statute is ambiguous, and in different situations it would be applied differently. The way EPA views this, the way it has viewed it over the years, including in the 2013 Portland Cement Rule, which is that if it's a limited situation for remand, reconsideration, limited to just a specific issue or a couple of specific issues. So is there a scenario where the boiler number nine would be a new boiler relative to a 2110 proposed standard? Again, to try and predict the future, if EPA is possible. Yes, it's possible. But in our view, that would typically occur. I think that's quite a hard argument to agree to. Fair enough. How do you harmonize your interpretation of new source with 112I and the compliance deadlines and the way that that works? Our view is that that provision is limited to, it's a compliance or reporting type of provision. And that is different than defining a new source. It includes the term new source in there and reports, for instance, that if you're building a new source, you have to comply with existing regulations. Well, that makes sense. To get approval to move forward, you have to comply. But that is not a license that says you will not be subject to change requirements. EPA and petitioners assert impermissibly retroactive. It doesn't punish the past behavior. It just says in the future, you'll be subject to it. Then what work does the definition of existing source do? I'm sorry? Then what work does the definition of existing source do? It continues to apply to existing source. The existing source standard is a different standard. It applies to the top 12% standard. But what I'm saying is that how does anything ever become an existing source? I would say in two ways. Congress originally set down its marker and said once you, that it defined new sources and existing sources, and it basically said that once EPA sets out the standards for a particular subcategory, then anything constructed after that is a new source, as EPA said in 2010, in 2013, and again in 2022. It could become a new source if EPA makes the start all over again. It has to collect all new data and create new emission standards, just as it did in this case, because EPA originally generated the requirements in 2004, and that was vacated by this court in 2007. A vacature basically wipes things off the slate and you start all over again. I see if I'm understanding one part of your argument. You tell me if I've got it wrong or not. It seems like you're saying a new source built after a proposal gets, in that case, it gets extra time to comply under I-2. But a new source built before a proposal does not get extra time to comply, at least not by statute. I would not say that. It would depend on what the proposal is. The proposal is a proposed rule to respond to a limited directed remand of this court, that would be the case. Right. Why are you giving oil for number nine the extra three years? I know you say the statute doesn't require it. You're doing it out of the goodness of your heart. But your argument on the they rolled the dice. They took a risk. They knew that they might be building a boiler that would not comply with the standard that was coming down the road. And so to the extent they're saying that calling them a new boiler is unfair, I mean, they made their bed. Too bad, so sad. If all that's true, then why are you giving them the extra three years? I think that's consistent with the Weld County case, where the court explained that if something is imposed but gives the entity no time to comply with the requirements, then it becomes unfair. So EPA gave the same time period it generally gives under the statute in other situations of three years. I'm glad that you're giving them the extra three years, but I think the logic of your argument is they really shouldn't have built the boiler in the first place because they were on notice there might be a stricter standard. That is correct. That they were on notice in 2013 that EPA took the view, took this view. They were on notice that things would change in the 2016 decision, and they proceeded at their own risk. Now you're sort of rewarding them for doing what you say they shouldn't have done, giving them the extra three years that you don't by law. I'm not sure it's a reward, but EPA was trying to be reasonable, at least on the compliance. If I could just, go ahead. They were on notice about the 2010 standard, proposed standard that became the major boiler rule in 2013. They weren't on notice that seven years later, the standard would become 100 times more stringent. They were not on notice about what this court would rule in 2016 and how it would direct EPA to change this. What future standards you would promulgate, whether directed through a remand without vacature or on the next eight-year iteration? That's correct, but standards get changed all the time under the various environmental statutes, and as explained in the retroactivity case. They do. I don't think there's a primary retroactivity problem, for reasons you lay out in your brief, but there is this scheme which on an ongoing basis just makes it a distinction between new and existing. We believe it goes back to Congress's intent. This statute has a long history. EPA spent years trying to regulate this area, and Congress threw up its hands and said, we have to get serious and strict about this, and it set out standards. It defined a new source, and under that definition of new source, we think it meets that. Can I ask you a question? Incinerators are governed by 7429. Correct. And there's a provision there that says that those standards are to be reviewed every five years. Right. How does it work there when those standards are revised five years later with respect to compliance for incinerators that had already been constructed or were in the process of being constructed previous? I can't say exactly to that, but I believe that 7429 does not include the same first proposed language as 7412. I think that's right. It doesn't say first proposed. Right. And I think that's the difference here. I guess if we take first proposed literally, I just don't know how, you know, once there is a standard that is applicable to, you know, a class of boilers, how it ever becomes not new, because when you revise in the future, whatever this new standard is, is going to be applicable, and if first proposed just relates back to the first standard that was applicable to that boiler, then it just seems that every, your boiler is always going to be a new boiler for every subsequent standard that is applicable to that boiler, regardless of whether it's a new notice and comment or new data or whatever. I mean, it just seems like if we just look at the plain text, it supports you, but it creates, I think, an absurd result. Well, as a general matter, until the absurd result part, we agree with you, because that's the, Congress took that purposeful step to put in the first proposed language to basically say anything built after the date of the first proposal for a given subject, you're on notice that you'd be considered a new facility, a new regulated entity. Well, now I'm not sure I understand, because let's say you win this case, and then in 10 years or 20 years, you do a proposed new standard with a new data set, notice and comment, and you finalize it, you promulgate that new standard. Is boiler number nine a new boiler? Under the history of how EPA has acted in other situations and other rulemakings, probably not, no. And is that history consistent with your argument in this case? We believe it is, yes. Okay, so under your argument in this case, why is boiler number nine not a new boiler with respect to a new standard 20 years from now? Because in the scenario you gave me, it was a new collection of data process. We think that's a key distinction here. I thought your discussion with Judge Wilkins suggested that the first proposed, I mean, the first time that a standard was proposed was 2010. I actually think it was actually 2003. So if first proposed, here's my question, if first proposed means what you say means, then why wouldn't boiler number nine be a new boiler relative to a 2030 new proposal, new data set, new notice and comment, final promulgation? As EPA explained, the term first proposed is ambiguous and it's ambiguous depending on the situation in which it's applied and consistent with how it's acted on other rulemakings. In your scenario, it would be treated, the first proposed would not date back to 2010. Okay, and I think there's a line in your brief that says, although you think you went under Chevron, you don't need Chevron. Right, so if the Supreme Court overrules Chevron this term, then is the argument you just made no longer an argument that could work? Well, it would obviously depend on how the Supreme Court overrules Chevron. Your position would be either that we would say it's unambiguous in your favor, or that it's the best interpretation, even if we give you no definite evidence. Probably the latter, because you just said it was ambiguous. Yes, and EPA has said that. Your opposing counsel kind of walked through the EPA statement in 2006, and I know you don't agree entirely with how that statement applies today, but I thought I'd give you a chance. It has three sentences, and after each sentence, it lets you say whether that was correct when they said it, or whether it was correct when they said it, but doesn't apply here for some reason. You can give the reason, or whether it was incorrect when they said it. The first thing EPA said in 2006, the whole premise of new source standards being potentially more strict than for existing sources is that these sources are being newly constructed, and hence can immediately install the best pollution control without incurring the time or expense of retrofitting. So, was that a correct statement? That was a correct statement. And then, put another way, new sources know from the beginning of the construction effort what controls will be required, and do not have to incur the higher costs and the time-consuming disruptions normally associated with control retrofits. That's correct. If we were to require new sources that commenced construction prior to, in bracket, 2005, to retroactively install controls because we have changed the rule requirements, then these particular sources would have to bear retrofit costs that we do not believe were intended by the Clean Air Act. Also correct? It's also correct. So, doesn't that describe the exact situation that confronts the source owners? I'll give you a chance. What's the difference? Again, it goes back to how a new source is defined. And if, in the context of just correcting the errors that EPA made, according to this court, in existing regulations, based on the original set of data, which, again, petitioners say is what should be applied, I mean, petitioners wanted both ways. They want all of the standards to be revised based on the 2013 data set, but they don't want to be defined as new sources. Just to take one phrase from what you said is still correct. New sources know from the beginning of the construction effort what controls would be required. Do you think that Boiler No. 9 owners knew what controls would be required when they built it? They knew the existing controls. They didn't know the 2022 revisions. And you said that the 2022 revisions control, right? Yes. So, they did not know what controls would be required. They did not know at that time. That seems very inconsistent with EPA's 2006 statement, which you endorsed a few minutes ago, saying new sources know from the beginning of the construction effort what controls will be required. That is consistent, but again, it goes back to what you consider a new source. I mean, all those statements do apply to a new source as you define it coming later in time. Okay. All right. Thank you. Thank you. Good morning, Mr. Pugh. Good morning. Contrary to industry's claim, the special definition of new source that Congress enacted for Section 112 cannot be read to mean something different than it says. And contrary to EPA's claim, it is not ambiguous. Section 112.84 expressly defines a new source as one for which construction commences after EPA first proposes an emission standard that is applicable to the source. EPA can propose a standard and emission standard for the first time only once. In this court... So EPA's answer to, well, what about a standard on a boiler that's proposed 100 years from now, whether boiler number nine would be new relative to it? EPA said possibly. But I think you would say definitely. Well, I would say yes. It's, I don't, I mean, the hypothetical, I realize it's hypothetical. The boiler wouldn't exist 100 years from now. I mean, the useful life of the boiler is a lot shorter than that. But yeah, there's a law. Yes, that's the right answer. As to your question about the absurd results possibility, I don't think there's anything absurd about this. Congress wanted very strict standards. And the only difference between new source standards and existing source standards are that new source standards are more stringent. No one, not even the industry petitioners, is claiming that they are impossible to meet. In fact, as EPA pointed out in the record, more than 70% of the new sources are meeting the standard in question. Petitioners say it'll cost them $20 million just to fix boiler number nine. Their cost complaints are irrelevant under the statute. This is under the statutory floor provision, which doesn't consider cost. I know they don't like it, but it doesn't change what a new, what's Congress defined a new source at. But you said it's feasible. And I guess. It certainly is feasible. $20 million, a lot of things feasible. It wouldn't be feasible for me, but it is certainly feasible for you, Sugar. What's wrong with, there's a statutory definition here. The 2010 rule established a standard for hydrogen chloride. I don't know what it is. Let's just call it X, okay? But in 2020, EPA proposes a different standard for hydrogen chloride, which is 100X, right? So you go to the statute and ask, when did EPA first propose regulations establishing the 100X standard? That's like a perfectly natural way of describing what's going on here. The revised standard, it's a revision, but it still establishes a new standard first proposed whenever that NPRM goes out. Your Honor, I think if the statute referred to the standard, that interpretation might be correct, but it refers to an emission standard. And as this court and the Supreme Court have both held, Congress's use of articles is- 100X is an emission standard. It was first proposed in 2020. But an emission standard was first proposed in 2010. If it were that particular standard, the 100X emission standard, it would be different, but the statute speaks about an emission standard. In Otonio, the case, this court just decided, Otonio versus the United States, this court just decided last year in 2023. The court held, and this is not controversial, that when Congress uses the indefinite article N, it's referring to any one. And that's in contrast to Congress's, I'm sorry, yeah, to the legislature's use of the definite article B. Industry would like to interpret the, this language is referring to the particular standard, but that's not what the statute says. In fact, I would say that Congress's amendment of this provision in the 1990 amendments to refer to first proposes an emission standard, it would have no effect whatsoever, given the reading that industry wants. And as the Supreme Court held in the Intel case in 2020, when a legislature amends statute, courts presume that it intends the change to have a real and substantial effect. Under industry's interpretation of this statute, adding the word first before proposes would accomplish nothing. So, let's suppose you're right. So, the way that the existing source standard is, or statute is written, you look at the, I guess, the best performing 12% of existing sources. And so, over time, you know, these boilers are going to become more efficient, et cetera, at least that's the hope, so that, you know, the best performing 12% today is going to be different than the best performing 12% 10 years from now, right? So, that if EPA gathers new data 10 years from now, and then promulgates a new standard there, then existing sources are going to have to improve and maybe have to do some retrofitting, et cetera, to meet this new existing source standard. That's part of the other thing that's going on here is that you want new data to be used wherever it's available, right? So, but if your definition of new source is correct, then when would we ever need an existing source standard, really? Or when would having that standard have any sort of effect? Well, the statute defines the existing source as one that's not new, and it defines one that's new as one that's built after first proposal. So, an existing source is one that's built for the first proposal. So, what I'm saying is that once there's a proposed regulation that is applicable to, you know, a class of boilers, then it would seem that any under your definition and your construction of this definition, any boiler that is constructed at any time in the future that fits within that class is considered to be a new source, right? Yes. So, once you have, once EPA has kind of established or first proposed something that ends up getting promulgated and approved and withstands all the court challenges, et cetera, that is applicable to a class of boilers, every new revised regulation that applies to that same class of boilers gets treated to, you know, the boilers that have already been constructed as if they are new sources, right? Yes. Rather than existing sources. Yes. And I just don't understand how, so basically the existing source regulation and definition only does any work for the boilers that were constructed before the very first regulation is proposed for that class of boilers. Yes, that's our interpretation. There is a universe of boilers that are built before that first proposal. They're existing. They'll always be existing. Other sources that are built after that proposal are new and they'll always be new. This is not a, I mean, How does that comport with just ordinary understanding of the word new? Ah, I'm glad you asked that. Because I'd love to become new. Your Honor, I don't think it accords with the ordinary understanding, but as the Supreme Court held in Tansen, where there is an explicit statutory definition, courts follow that definition, even if it varies from the terms ordinary meaning. We're not claiming this is the ordinary meaning. It is the defined meaning. This court is at a very high bar for overriding the plain text of a statute. The party that wants to override it. It's also said you can look to the plain meaning of the defined term, if the definition is ambiguous. But the definition is not ambiguous. I'm glad you asked that, Your Honor, because we disagree. Then we're back to the question I asked you at the beginning, how you understand an emission standard. Any old emission standard that's ever applied to the source. That's right. I get that. But that leads me to think, you know, any old emission standard includes the 2020 one. Oh, no, Your Honor. That question about 2020, no less than 20, uh, 23. Sorry, 2010. Your Honor, what I'm trying to say is that N has to be read together with first. The first time EPA proposes N. This just means the first notice proposing 100X as opposed to X. Actually, I respectfully disagree. That's the first time that EPA proposes an emission standard is the first time EPA proposes any emission standard. What word does first modify? Pardon me? What word does first modify? Proposals. And what word does N modify? Emission standard. So it doesn't, it seems like you need N and first to modify the same thing in order to make the argument. No, the first proposal of an emission standard is the first time an emission, any emission standard is proposed. Maybe if it said, uh, you know, commenced after the EPA, I'm going to scratch first, proposes regulations under this section, establishing a first emission standard. Well, Your Honor, I don't think EPA, I don't think Congress would have had to write this differently to make it clear. It is clear by referring to the first time EPA proposes an emission standard. If we, if we agree with you and we're going to write a sentence in the opinion, I'm going to leave a blank and ask for your help to fill in. The, the, the boiler and putting aside that boilers don't last a hundred years, a hundred year old boiler is still new, just as a hundred year old blank is new. A hundred year old, I mean, we're talking about it being new only because it was built after the first proposal, right? So just as a hundred year old cement plant built after first proposal is new. So EPA proposed regulations for boilers back in 2003. Right? Yes. Were those regulations, would they have been applicable to the boilers at issue here? Yes, they were. So, so any of those boilers are also new sources under your definition? Well, we don't take a position on that because it's, it's not a statutory matter. It's a matter of construing the case law on the effect of vacature. This, this court vacated the 2007 standards and the court has held that the effect of vacature is to restore the status quo before the vacated rule was promulgated. Vacature doesn't undo the fact of proposal. Well, that's, that, that issue hasn't been decided. If that's, if that's the conclusion the court reaches, then the first proposal wouldn't back in 2003. I mean, I think that's the clear implication of your position. Again, we're not taking a position on it because our position is that EPA's definition should be upheld. We just disagree with EPA's statutes and ambiguous. And the last point I'd like to make on that is it is understandable that EPA would like to have the discretion to interpret this provision differently from one rulemaking to the next, so that it can achieve the policy outcome that it wants to, to accomplish. But that doesn't make the statute ambiguous. There is, there is no evidence apart from the fact that EPA would prefer to be able to interpret it a different way at different times. And there's no evidence in the text or the structure of the statute that actually shows this statute is ambiguous. Thank you. Thank you. Yeah. Very briefly, Your Honors, and it's worth thinking about the, the effect of what we've heard on the statutory scheme here. You know, as you've recognized, new source defines by elimination what an existing source is. And there's this whole structure about existing sources and how you regulate them and how you from the government and the environmental groups. There won't be any existing sources down the road because everything will be a new source. And it's worth thinking, too, about the incentives here, which I think are inconsistent with the Clean Air Act goals. We didn't have to build a new state-of-the-art boiler number nine when we did. It replaced three boilers that were older and dirtier that last 40 years. Typically, a boiler will last 40 years. And, you know, if we had thought that these 100 times more stringent rules were coming, it would have been totally uneconomic to build this boiler when we did. And in general, I mean, if the government's position were upheld here, why would anyone build a new boiler before they had to? You know, you'll always be at risk of being labeled a new source subject to much more stringent standards and more expensive and expensive retrofits. But you're still subject to retrofits as an existing source if the best-performing 12% Yes, absolutely. Boilers gets more efficient. And in fact, what has happened here, Judge Wilkin, is that since we built the boiler, the existing source standards for these sorts of boilers have gotten more stringent. We haven't had to do a significant retrofit, but we have had to meet those new standards. And that's how the statute is supposed to work. You know, you anticipate that under the RTR process, there will be, for existing boilers, new standards each year, but they're based on the 12% and not on the best-performing. So they tend not to be so stringent. So we think the whole scheme of the statute, which has both of these new sources and existing sources, ties in with our interpretation, not with the government's. If there are no questions, Your Honor, I'll leave it there. All right. Thank you. Now we'll turn to the other case. We'll hear from Mr. Pugh, your petition. Morning again. The Clean Air Act expressly requires EPA to base floors on the best-performing sources, quote, for which the administrator has emissions information. The best-performing 12% of the existing sources for which EPA has emissions information. And for smaller categories and subcategories with fewer than 30 sources, it's the best-performing five, for which the administrator has or could reasonably obtain emissions information. EPA does not dispute that its rule did not base floors on the best-performing sources for which it has emissions information. EPA chose to exclude the last decade of emissions information that it has from its floor calculations and doesn't disagree that by doing so, it also excluded the best-performing sources for which it has emissions information. Do you think that emissions information implies usable emissions information? Yes, certainly it applies usable information. But in this case, there's no dispute. EPA is not rejecting those data because they're not usable or there's anything wrong with them. I think they're saying it's unusable emissions information that with some work by EPA, like verification, could become usable emissions information. I don't think EPA is saying that there's any serious hurdles to using the emissions information it has. It's compliance information that was generated. I think it's a, I don't have the page of the brief in front of me, but I think their brief goes into some depth about what they would need to do to verify and that it actually is quite a difficult process, might be a stretch, but it's not an easy, quick process to just take the raw data that they have received electronically and turn that into a usable dataset. I mean, this is maybe not a great analogy, but there's like raw intel, you know, it comes from like curveball and a rock, and there's like usable verifiable intel. And I think this implies usable emissions information. And EPA says that takes some work. It might take some work, Your Honor, but it doesn't change the fact that this is information they have. It's information under their own compliance system. This is not some information coming out of left field that somebody just sent them that they have to convert. This is their own emissions information that they have. And I don't know that it's unverified. They say it's unverified. Well, they may make the claim that there is some workload, but that doesn't change. This is emissions information they have. Perhaps they have to process it, but EPA has had since 2016 to get this rulemaking done. This is a 2016 remand. The notion that EPA can't verify its own information is simply not a plausible reason to refuse to use it. Is that what your argument kind of hangs its hat on, that there was this electronic reporting, and so EPA has this information, and so they needed to use it? Would you be making the same argument if there was no electronic reporting and EPA had to go out and collect a full data set more recent than 2013? No, we would say that then EPA doesn't have the information. This is information that it undisputedly has. EPA has never argued that it does not have this information. So I take your argument, but it seems like to the extent that EPA is, for compliance purposes, is continually getting new data, EPA could be in a position where, okay, it starts, commences the rulemaking, preparing for the rulemaking process, let's say in 2019, and they gather all the data that they have and that they can verify, and they do all the calculations, et cetera, on that data. You know, then, you know, staff proposes a rule, it's got to make its way up the chain internally before it gets to the Federal Register as a proposed rule, and that might take a year or two, and then the rule is proposed, and then your clients in their comments say, well, this rule is based on data that you collected as of 2019. There's more data that was collected in 2020. So we want you to use that data. I mean, under your construction, the statute, wouldn't EPA then have to respond to your comment by saying, okay, let's gather that data, and, you know, essentially, this is a never-ending, you know, dog-chasing-its-tail situation? Your Honor, I don't think it's never-ending. I mean, there's an end when the final rule comes out. And really, this statute was promulgated with the understanding that EPA would go through the normal notice-and-comment rulemaking. I don't think it would be unreasonable for EPA to cut the data off the terms of proposal, but in fact, EPA often does consider data that is new between proposal and final. So I don't think there's anything wrong with that. But if they do, you know, I guess that's the point, is that they do, and maybe, you know, to the extent that they can, and they want to exercise their discretion to do so, certainly wouldn't be arbitrary and capricious for them to, you know, continue gathering data. But why is it arbitrary and capricious for them to say, look, you know, this is where we're going to put the end point. And even if we have more data, we're just not going to use it because we think that the data we have is sufficient. And, you know, the new data isn't really going to move the needle that much. And so we need to, you know, stop somewhere. And they have, we're supposed to defer to their technical expertise in matters like this, right? Well, not on matters that are determined by Congress and the statute. And the statute says that EPA has to base floors on the best performing data sources for which it has data. EPA doesn't get any discretion on that. I understand that, you know, that interpreting a statute isn't within their technical expertise. But I guess, I mean, you seem to be conceding that at some point, even if EPA literally has, like in my hypo, if they gather data in 2019, and then they propose a rulemaking in 2020, and then during that comment, someone says, well, we want you to use the most current data that you have since 2019 and revise this rule. I thought you were agreeing that EPA could say no to that and still be in compliance with the statute. It's possible that the agencies could say no to information gathered after a proposal. But I don't think that the statute requires that. I think the proper endpoint and one that is certainly clear and doesn't lead to sort of never-ending consequences would be the final rule. Why can they say no to data gathered after their proposal? I think that the concept of a rulemaking generally concedes of all of the data coming in at the time of the proposal. That's the point of Section 307D. Well, what if it takes you a year from gathering to kind of have a proposal ready? So you stop gathering in 2019. You do calculations. The policymakers, officials have to sign off. OMB has to review it. All of these things have to happen before it gets formally proposed. So there's a year between gathering and proposed. And somebody says, well, you know, a lot of good data came in during that time period between 2019 and 2020 when you chose it. We want you to use, doesn't the statute under your reading require EPA to go back and use that data before they officially propose regulation? This would be before proposal? Yes. I just don't think there's anything in the record to say that EPA can't use the data it has in its possession for a proposal. Again, especially given the scope of time that EPA's rulemaking has happened over. That could be never-ending, but this judge will just explain it. It's not like EPA can just gather, decide, and then propose the next day. EPA gathers, decides, and then there's an interagency process. There's all kinds of other things. Maybe that takes a year. And then if they take your literalist interpretation of the statute for which the administrator has present-tense emissions information, then the day before they're about to propose that rule, now they have new emissions information. It seems like if you're right about being literally this, then they would have to press pause, start all over again, incorporate the new data, and then begin the maybe one-year interagency process again, which, of course, would be a never-ending thing. I see your point, but I don't think that really gets to the situation here. Where EPA is excluding data- The argument here is a statutory interpretation argument. So it has to be a statutory interpretation that is sound, not just here, but always. I think it is a sound statutory interpretation. It's consistent with the text. And the possibility that EPA- You can see that if we applied it in the situation Judge Wilkins described, it would be a never-ending cycle that ultimately results in no rule being proposed. If EPA had to consider new data on the day before the rule was due, and then considering that new data postponed the rule, I don't think that would happen. I mean, there are statutory deadlines that require the rule to come out at a certain date. So I don't think that- EPA blows by those deadlines all the time. Well, that's a problem with EPA's conduct, not a problem with the deadline. Your argument here would make that problem worse. I think, Your Honor, that EPA can cut off data, you know, based on a consideration of, you know, it's just not feasible to use the data because of time considerations. But that's not what happens here. And that means that has, present tense has, does not literally mean has. Like, it could mean that has means has in the, you know, in the present, not has in the future or not, you know, hasn't, hasn't got it, you know, available yet. But I don't think it means that EPA can exclude data for the last decade, which is what it's claiming, and use that to make the standards less stringent, which is what it's, it candidly admits it's doing in this rulemaking. This is, I'm sorry. No, I said this is about as a clear an example of EPA elevating its policy desires over the EPA says that it excluded these data, not because of time problems, not because of quality problems, but because it wanted the standards not to be more stringent. If EPA can do this, if EPA can cherry pick the data that way, simply choose to ignore emissions data and draw the top performing sources, not from the sources first it has data, but from some other, from any, essentially from, you know, any category or vintage of data it likes, 2010 data, you know, 2003 data, any data it likes. If the statute doesn't tell it to base floors on the best sources for which it has data, it can make standards either impossibly stringent or meaninglessly weak. That seems like an arbitrary and capricious challenge as opposed to statutory construction. Well, your honor, I'm just pointing out that EPA statutory construction leads to that result, and that's a tool of statutory construction. Suppose we agree with you that you measure this duty at the time that notice the NPRM goes out. But suppose also we think that has emissions information, right, that phrase means something like has the information in a usable form, right, to avoid this endless loop problem. What can you tell us about whether the data in question was usable? I didn't find anything in either your brief or their brief on this question because I think they're resting not on the theory that the data was too inchoate. They're resting on the Weld County theory that it's a remand without vacatures, so they want consistency over timeliness. Thanks, your honor. I'd like to get, I guess, both pieces of that argument. I mean, the best evidence that EPA can, in fact, use the data it has is that it did use it in the beyond the floor standards. I'm sorry. In the beyond the floor analysis. In a number of these, for a number of these standards, EPA's floor analysis made the standards less stringent than they were in 2013, and EPA then conducted beyond the floor analysis to determine if it's, you know, if the standards could be made more stringent, and it used those very data for that purpose. So certainly. In the 2020 rule. In the 2020 rule. Right. There's no question that these data were usable. EPA does not have to promulgate beyond the floor standards, right? Right. The statute, Congress gives them the discretion to determine whether to promulgate beyond the floor standards. Well, the discretion doesn't go quite that far. It tells EPA to promulgate beyond the floor standards if they're achievable, but certainly they have a lot more discretion over beyond the floor standards than they do floor standards. I have a couple more. One is, if we rule in your favor and remand on this issue, what data set should EPA use to craft the next rule? It should use the data set that it, well, it sort of depends on when EPA gets around to this rulemaking, Your Honor. We hope that EPA will respond promptly to the court's remand, but there are remands pending from 2004 right now. And so I won't say that EPA should use these data because I don't know when EPA is going to get around to the remand, but if EPA promptly promulgates the remand, which I think was the intent, then I think that this data set would be fine. I mean, this data set ends in 2020, right? Why under your... Okay. Assuming that EPA promulgated in 2024, EPA should use the additional data that it has when it comes out of proposal. Your Honor, there was one more question, piece of that question, though I don't think I got to. It wasn't just about the usability of the data. Yeah, it's just their rationale seems, unless I missed something, their rationale in at the time was we're going to use old data, the same reason we used old data in Weld County, limited partial remand. Most of the standards are going to be under 2013. There's an interest in consistency and there's an interest in moving extra quickly because our remand order was rattling the saber of mandamus already. That's their rationale. Right. So the first one, like to me, I'll just put my cards on the table. If we're just doing arbitrary and capricious review, that's good enough under Weld County. Right. And you have a statutory argument which might or might not overcome that. I would say that, right, I would say that the difference between this case and Weld County is that that was dealing with statute that doesn't have a floor provision saying EPA has to base the floors on the emissions, on the best sources for which it has data. And as to the limited nature of the remand, the court didn't remand all the standards. That's true, but it found them flatly unlawful. It initially vacated the standards and it didn't expect EPA to do anything different on this remand than it would have done if it had vacated those standards. In fact, EPA stated in its rehearing petition that it was going to promulgate revised standards. As this court held in the Medical Waste Institute case back in 2010 on the remand of EPA's to promulgate revised standards, but where they do, they essentially have the same effect as a vacature. They require EPA to start processing and regulate it as though it's on a blank slate. So to the extent that, I mean, the rule of prejudicial error always applies even in these contents. I mean, to the extent that EPA considered the newer data and promulgated beyond the floor standards, where's the prejudice? You mean where are we harmed? Yeah. Well, there's two ways. Some of the standards EPA didn't do beyond the floor standards and even where EPA did do beyond the floor standards, that doesn't mean that the standards are as stringent as properly calculated floor standards would be. It just meant that EPA chose to bring the standards back up to the level they were in 2013. Do we know that? Yes, that's exactly what EPA was doing. When it promulgated beyond the floor standards, it said, the question it asked was, well, do we know that our 2013 standards were achievable? We're going to look at the emissions data and show that they're achievable. And it did. And that was its rationale for leveling up the standards to the 2013 level. But we'll make sure we're understanding each other. You're saying that the beyond the floor standards are less stringent than the mapped floor standards would have been if they had used the newer data? Well, we don't know exactly what the floors would be if EPA had calculated floors using the data it has. That's my point. We don't know whether there were prejudices, so to speak. But what we do know is that EPA excluded best performing sources from its floor calculations, and that it did so precisely to make sure that its standards were not more stringent. That's what EPA said in the record at page seven of the joint appendix. We don't want these standards to be more stringent, and that's why we're excluding these data and these sources. So I don't think there's any question that it led to the standards being less stringent. That's EPA's point. Go ahead. EPA's brief says, EPA has broad discretion in assessing the nature and timing of the data to be considered in establishing a mission standard. Do you agree with that? Not under this provision. So maybe as a general matter they do, but the statutory text here takes that discretion away. Right. I can imagine another part of the statute where EPA might have discretion. Similar to what you and Judge Skat just wrote. Yes. Okay. And then I think this is my last question. As you may know, in the Ohio v. EPA case at the Supreme Court right now, one of the key arguments against EPA is that when you have a system of interrelated rules, they can't function properly when something happens to make the application of those rules not uniform. I wonder if you think that sheds any light on this problem. Maybe this would cut against you. So maybe my question is why does this not shed light on our problem? But if EPA loses that case and the Supreme Court says, there's just something kind of not right about subjecting 83% to a strict data... I'm going to now wedge into our case, but segue in our case. Subjecting 83% to a generous data set and 17% to a lenient data set for the very kind of arbitrary reason that, in our case, some of them came later than others. Well, Your Honor, I have said, and I'm not familiar with the details of the Ohio case, but I would say that it's not construing this particular statute. And the policy argument you're talking about is one EPA articulated in the record. It said that that's one reason it didn't want the standards to be more stringent, but it's just a policy argument. It doesn't override the statutory requirement to base floors. And it's only a loose analogy, but the 10 second version of what's going on there is they did nationwide standards for different states. And some of those courts have vacated the rules for the calls for some of those states. So some of the states are subject to the more rigorous, some of the states have gotten relief, at least temporary, and the ones that are left say, hey, it's not right that it was supposed to be uniform. Now it's not uniform. Why should we be left with the extra strict stuff? The other point I would say, I don't think as a statutory matter it matters, but EPA agrees that if this is really a concern, if consistency across the board for every source in the category is truly an important concern for the agency, it can level up and make them all similarly stringent based on the same data. Once again, we're here on a limited remand where EPA was directed to revise limited number of standards without examining all the new data. There's nothing in the court's opinion that even implied that EPA should go back and use all new data. As Judge Costas, you pointed out, the Weld County case really applies on all fours. It was under a different statute. But maybe yes, maybe no, right? We have some statutory text here that wasn't present there. Best performing for which EPA has commissions information. Suppose I asked you, you're a football junkie. You follow all the statistics, right? And I ask you, what is the passer rating for the best performing quarterback in the NFL? And you answer by giving me Tom Brady's rating from eight years ago. Would that be a fair answer to the question I put to you? Not to that question. It probably would not be fair. But that's not the question that the U.S. Sugar Decision posed to EPA. It said, go back and figure out the correct sources that are in each category. And just as to the ones that you find, make any necessary corrections on that. It was not saying, look to the future and give me a whole new set of emissions. It said, you made some errors, correct the errors. And when you go back and correct the errors, you do that. I mean, you do a new rulemaking, right? We didn't vacate the old standard. It continued to have the force and effect of law. And you have to do a new rulemaking. There's a statutory standard for that. And it says, look at the best performing standards for which you have data. And it leaves to EPA the discretion to determine what data to look at. That's been stated by this court over and over again. That is in the two provisions that achievable from new sources. That's achieved in practice by the best controlled stimulus source as determined by the administrator. This is not merely reading numbers on a graph or a chart. EPA goes through a comprehensive process to do that. The provision uses the word deemed, not just what is laid out on a page in a report. And it's deemed in the judgment of the administrator. The court in U.S. Sugar said the Clean Air Act directs EPA to establish emission caps that result in a maximum degree reduction in emissions that the EPA determines is achievable. And the statute says nothing about when EPA is supposed to look at that. What temporal application they're supposed to look at in the data. But what about the fact that under, you know, this definition for new and existing sources in subparagraph B, it talks about the best performing five sources for which the administrator has or could reasonably obtain emissions information. So, there, doesn't the Congress require the administrator to seek to obtain information if it's reasonable to obtain it? So, it's going beyond, like, what the administrator has in their possession. But they have to seek to obtain it. And how does that make sense? If has, does it mean that what you already kind of have in your possession? So, the process, the compliance reports are basically simple reports of emissions levels reached by the reporting entities. The process for establishing emission standards that everybody is subject to is a whole different process. It requires a collection of data from all of the to the source. It's based on an average of the emission standards that's reached. After that, the upper protection limit is applied to determine variability. It's based on a whole number of different factors that takes typically years to assess. And those are the efforts that the simple reports of how a particular entity is doing should not form the basis for a set of standards. And it doesn't in practice. As far as the language in the statute about it, for which the administrator has emissions information, which is, which petitioners hang their hats on. As this court said in Mississippi Commission Environmental Quality, interpreting similar language, quote, the act calls for EPA to make designations on the basis of available information. It's similar to here. They're repeatedly found similar language should be ambiguous when assessing whether to defer to an agency's construction. EPA therefore may interpret the statutory language as it sees fit, so long as its interpretation is reasonable. We think that the interpretation here as to what information is possessed by the administrator. Is quite reasonable. So we need to, I take your point, but we need to interpret the statute so that, you know, this subsection 3A makes sense. And also so that subsection 3B would also make sense. Correct. What does, what does, has or could reasonably obtain emissions information in 3B? What does that mean? We think that the reasonably obtain means conducting a whole, you know, review process of all of the, every one of the boilers subject to regulation. Of going through all of the stack tests that are required. Applying the small data requirements that if you don't have enough stack tests, that certain formulas have to be. That EPA has to review and consider normal patterns versus log normal patterns and the distribution of the data and the timing of the data. That you believe that's the difference here and that's the reasonable steps EPA makes, which are not reflected in simple compliance reports. So your position is that Congress gives EPA the choice to go with A or B, here 3A or 3B, and EPA chose 3A and so they have some discretion there since it doesn't include the language or reasonably obtain emissions information to do what it did here. Yes, and as the court has said many times, EPA does have a lot of discretion in the data collection, the data consideration, the data analysis portion creating these standards. Your colleague said that the disputed data here were actually usable because EPA did in fact use them for other purposes. Do you want to respond to that? Yeah, that's a different process. That's the beyond the floor process and EPA did not create new standards based on beyond the floor process. All it did was look at the compliance reports, see that people were complying with the original 2013 standards and said well part of the beyond the floor analysis looks at the cost, has to look at the cost unlike the max floor. And it used that information to determine that these entities would not incur costs to comply with a stricter standard, their beyond the floor standard, because they had already been doing it and they had already had put in the equipment to do that. But it was very, it was limited. That suggests that this data is not, the problem here is not that that data is just a jumble and you haven't had time to convert the numbers that are coming in through the compliance monitors or whatever into something that you can use to set standards, one of the two things going forward. Well again, through these technology reviews that are to occur every eight years, EPA does an analysis of whether the technology has advanced to a level that really would require a whole new establishment of a set of new emissions. And when it does that, it starts a extensive review process, which again collects the data from every single source, scrutinizes that data, applies the... Is that review process different than for beyond the floor purposes? Yes. Okay, so your primary rationale here was that you, in what you characterize as a limited remand without vacature, you don't have to, the statute does not require you to reopen the rulemaking record. That's correct. Consider new data. Okay, I get that. But are you pressing an alternative theory that the disputed data here, you didn't have it because it was just too much of a jumble? You hadn't had time to digest it? We didn't have it in a form that could support a whole new process of resetting. Okay, and where do I get that? Because I didn't see that in your explanation. Five through seven of the rulemaking, it's all about the Art Weld County point of, you want, you just want a consistent data set across all standards. That's true. That was, you know, some of the bases here. But there were some explanations, as I think Judge Walker referred to earlier, about the process that you have to go through to create a whole new set of standards. But to clarify, I was talking about stuff you explained in the brief. Okay. I don't know if Judge Katz is asking this or not, but I'll ask it. Where can we find that in the record? I don't have the references for you on that. I can't get those for you. I'll ask a slight variation, which is, have the alternatives that I've articulated, there be a Chenery problem with considering that as a basis for affirming? Could we affirm on the basis that the data, the late-breaking data, is just too messy to use? That doesn't seem to be the rationale that you've rested on in the administrative order. Probably you could. I'm not sure that would be the correct word. Probably incomplete. Okay. I'm sorry. I didn't understand your answer to Judge Katz's question. You're saying that would be a Chenery problem? That it could be. We could affirm on that basis. Actually, we could deny the petition on the basis of this argument you're making, that the data wasn't usable, even though Chenery says we can't consider an agency reason that wasn't part of the petition when they made the decision. We believe that it's basically in the EPA's explanation. I'll be explicit. Okay. Then, just to follow up. This is, I think, a friendly question. Where in the EPA's explanation can I go to see that you preserve this argument? Again, I have to go back to the rulemaking and get you the exact pages for doing that. I'm assuming I can find it. You're implying that I can't. I have it at five to seven, if there are other things of the JA, if there are other things we should look at. I'd be curious to know what they are. The petitioners say in their briefs, EPA does not dispute that it excluded best performing boilers for which it has emissions information from its floor calculation. Do you agree with that statement? No, we do not agree with it. Why is it inaccurate? Because, again, you cannot just go on compliance reports. EPA would have to go back and go through its whole analysis to establish that. Okay. Hopefully, we'll see if that is what EPA said when it made its decision. Can I ask the presiding judge for permission to ask one question about the remedy in the earlier issue? Sure. I talked with the petitioners on the earlier issue, the industrial petitioners, about whether if we agree with them on issue number one, the new versus existing rule, you know what I'm talking about? Yes. Should it be a universal vacuum or should it be that would apply to any boilers built, begun to be built before 2020? Or should it be a more narrow remedy that only provides relief to boiler number nine and the nine other? It should be limited to those boilers, those identified boilers. Did you argue that in your red brief? I don't believe we did, but that was the remedy that petitioners were seeking. They were not seeking, as far as I know, a global remedy here. In fact, I think they talked about the fact that this was limited. Our usual practice, I think, for decades has been unless we remain without vacuum. That aside, when we vacate, it's not a party-specific vacuum. It's a universal vacuum. I think that's been our practice for decades. You want us to argue that? Yes, we think it's limited to the claims made in this case, which they limited to standards for certain pollutants, limited pollutants here for a limited class of the 34 different subclasses that are in the rule. Can you just explain to me as a layperson trying to understand this, how is that good enough to be used for rule-making standard for beyond the floor, but it's not good enough for the floor standards? EPA can look at different factors when it goes beyond the floor. With the focus on costs, use the limited data that it had to determine that this would not result in incurring costs at all, perhaps, but any kind of reasonable costs, monumental costs, in just requiring these entities to comply with the standards that were in place before the change. But the beyond the floor standards are universal, right? They're to all boilers that fit within that classification, right? That's correct. Just like floor standards apply to all boilers that fit within the classification. So I guess I'm just not understanding why new data is good enough to create a kind of cross-the-board standard that goes beyond the floor. So it would seem like you need even better data to do that because you're exercising your agent, your client, the agency is exercising its discretion to kind of set a more stringent criteria. So it would seem like you need the best data to justify doing that, but you're saying somehow the data was good enough to do that, but it wasn't good enough to set a floor? That just logically, that doesn't track. Well, it goes beyond the floor of the 2020 standards, but it doesn't go beyond the floor of the original 2013 standards, which the most entities were complying with. So that's the difference here. Some of the standards changed in 2012, but EPA made a determination that we can use the beyond the floor standard to make sure that at least people continue to comply with the 2013 standards. It did not use data collected through the subsequent years to alter those standards and say, you know, we're raising the standards on that basis. It went in the beyond the floor. It used it to ensure that whether the standard has changed, if the standard, for instance, has lowered, you could still be required to comply with the 2013 standard. Okay. I'm not completely sure that I understand, but I'll go back and review that. Okay. Any other questions? Thank you. I think you were out of time, but we'll give you two minutes. Thank you. Very quickly, I'd like to come back to what EPA actually did say in the record. EPA has never disputed that it has emissions data and that it excluded best sources for which it has emissions data. It didn't dispute it in the record when we made the written comments. It didn't dispute this point throughout our opening brief. What EPA's rationale was is that page six of the joint appendix, page 70 of its brief, EPA's rationale is a statutory interpretation that the statute does not mean what it says, that the only function of the requirement to base sources on the best performing or base floors on the best sources for which it has emissions information is to relieve EPA from having to collect data for all the sources in a category and set standards based on all the sources in the category. The problem with that interpretation is that it's simply not what the statute says. The statute expressly directs EPA to base floors on the best sources for which it has emissions information. Thank you. Thank you. We'll take the matter under advisory. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A91253. Public record. Not legal advice.