Oral argument — Sierra Club v. EPA

Case number 15-1487 et al. Sierra Club and Natural Resources Defense Council petitioners versus Environmental Protection Agency et al. Mr. Pugh for the environmental petitioners, Ms. Bowers for the respondent EPA, Ms. Felicia Good morning. Just to provide the Court with some guidance here, are you going to devote this opening argument to the merits of your challenges and then, consistent with the language in our most recent order, respond to the challenges? To whatever is said by other counsel regarding abeyance. I'm just trying to order, have an understanding of how you'll present your case. I think that makes sense. I'm happy to answer any questions the Court might have on abeyance, but unless the Court has any right now, I'll start by addressing the merits. The first issue I'd like to address is EPA's treatment of the potential for cancer from the pollutants from which it set standards under Section 112.4 of the Clean Air Act. 99% of brick kilns' hazardous emissions is made up of three hazardous air pollutants. That's chlorine gas, hydrogen chloride, and hydrogen fluoride. Based on nothing more than a lack of evidence about whether these pollutants cause cancer, EPA asserts that a health threshold has been established for them under Section 112.4 of the Act. Are you blending the question of carcinogen with the question of health threshold being established? That sentence seemed to refer to both. I'm sorry, Your Honor. I meant to just address cancer right now and to address other health effects. You seem to be leading over into the other question. Maybe I misunderstood you. I'm not sure after reading your brief and thinking about it and reading it, what is it you say would be enough to establish that there's not a carcinogenic property to a chemical? You say in the brief and you say here that they haven't proved it's not. Proving a negative is not something that we normally require anybody to do under any circumstance. If you just don't have any evidence to propose to you, isn't that enough to establish they're worthy? Your Honor, it's certainly possible. If EPA had conducted substantial studies and determined in these studies there's no evidence of cancer, we could have concluded that these pollutants aren't carcinogenic. But we're not asking for some… When they looked at what was out there and found nobody had ever found carcinogenic properties, why isn't that enough? They could give them the standard review we have in the Administrative Law. Well, a couple of reasons, Your Honor. First, they didn't actually find that they don't cause cancer. They didn't classify. What they actually found is that none of these… That's what I'm saying is how do you determine that something doesn't cause cancer? If there's no evidence that it does, why isn't that… Well, EPA's guidance in the record explains how it can be done. Page 695 and 696, EPA says there can be convincing evidence in both animals and humans that a pollutant, an agent they call it but it's a pollutant, is not carcinogenic. And the agency goes on to outline how that kind of evidence can be marshaled and presented. And, in fact, EPA itself has, on the basis of that kind of evidence, which, you know, based on robust data as the agency says, has reached a conclusion for other pollutants that there is no cause for human health hazard concern about cancer. So we're happy with that kind of conclusion. EPA has made it in the past. EPA knows how to do it. It's written the guidance on how to do it. That's all we're asking for. But here, EPA has something very different. What EPA says here about these pollutants is that not one of them has been classified with respect to carcinogenicity. And that the only evidence it has is old, in fact, decades old. It's limited, as the agency says. It's unreliable. And with respect to one of these three pollutants, hydrogen fluoride, it's equivocal. It is unreliable, but it's not the term they use. Well, Your Honor, it's a term they use with respect to a lot of them in the Joint Appendix. I thought the term was of low... Of low confidence. Pardon? Of low confidence. Yeah, of low confidence. They do use that term with respect to the non-cancer threshold, but with respect to the data that they had for cancer, they describe it, among other things, as unreliable. So that's, you know, essentially that's the difference. What EPA has when it actually determines that a pollutant... When it's established that a pollutant doesn't cause cancer, it has convincing evidence, and it says that it doesn't cause cancer. Here, EPA has never said that. All it says is that we don't have evidence one way or the other. They try to spin it as saying we don't have affirmative evidence of carcinogenicity, but the truth is they don't have evidence one way or the other, and they admit that by saying these pollutants haven't been classified. What EPA is saying as a statutory matter, I think, is important. EPA is equating the absence of evidence with an established threshold. The agency's interpretation of the statute is, first of all, it agrees. So you now are leaving a pure aggressive carcinogenic and going over into the general ability to operate under K2 or whatever the numbers are there that says if you're going to stab me, kill me. Well, I don't mean to be, Your Honor. I'm still, at least, I believe I'm still talking about it. I'm trying to still talk about cancer. I don't mean to start talking about the other pollutants yet. The point I would like to make is that by equating an absence of evidence with an established health threshold, EPA drains the statutory term established health threshold of meaning and punches a hole in the most fundamental requirement in the Clean Air Act for hazardous air pollutants. The listed hazardous air pollutants will be subject to technology-based standards that reduce emissions by the maximum degree that's possible. Congress wanted Section 112b.4 to be a limited exception to that requirement, not to eviscerate it. But if EPA is allowed to switch the burden of proof so that it can decline to set technology-based standards any time it doesn't know that a hazardous air pollutant causes cancer instead of having to establish that it does not cause cancer to proceed under Section 112b.4, the agency can do this for many other pollutants as well. There are 122 out of the 187 hazardous air pollutants listed in Section 112b that are in exactly the same situation as these three. EPA doesn't know one way or another whether they cause cancer, and EPA doesn't dispute that under its interpretation of the Act, it would be free to do exactly the same thing for those pollutants as it is doing for these three. It isn't inherently the case that they would be wrong if they did this. You argue as if you've won your case once you say, well, there's a lot of others out there that aren't established to be carcinogenic either. Okay, so that may be the case, and so maybe they can do the same thing with those. That doesn't say that it's unlawful for them to do this. Well, I guess I'd like to make a few points. That's a non sequitur, I think. Well, if I can, I'd like to make a few points about the language of the statute as well as Congress's intent. The first point is to start with the language that Congress referred to in the established health threshold. That is very specific language, unlike other language that Congress used. I'm sorry to interrupt you. I'm not very sorry, but I'm going to interrupt you. You can win on the health standard, established health standard, without winning on carcinogen, can't you? Yes, Your Honor. It's not the same. Okay, thank you. Excuse me, go ahead. I seem to keep running into times when I think you're bleeding one end to the other. Excuse me, go ahead. I would like to stick with cancer to make a few points. One point on EPA's statutory interpretation is that EPA agrees that a pollutant cannot be carcinogenic for it to proceed under Section 112.84. That's at page 23 of the brief. On page 24 of their brief, they say that EPA doesn't have to conclude that a pollutant isn't carcinogenic. That defies the plain meaning of what an established health threshold means. Because it's not the statute that requires non-carcinogenicity, if that's actually a word, and it's their regulation. And so do we know that the word establish is the operative word for purposes of determining non-carcinogenic status? Or is it just could it be a different standard in the regulation? They conceded that word established, and the statute requires them to establish that it is not carcinogenic for purposes of compliance with their regulation. No, I think there is some disagreement over the meaning of what established is. So what's the standard under their regulation by which they have to determine that it is not carcinogenic? But if I can go back to the beginning of that question, I think they do agree that it's statutory that the pollutant can't be carcinogenic because there's a requirement for an established health threshold. And what EPA is saying is because there is no safe level of exposure to a carcinogen, to establish a health threshold for a pollutant, that pollutant can't be a carcinogen. Right, but there could be – if we know it's carcinogenic, right, that works. If there is debate about whether it's carcinogenic, you could probably find just about anything declared to be carcinogenic on the website these days, right? There has to be some measure of judgment on this. And so what is the standard for determining that something is not sufficiently – there's sufficient evidence that it's not carcinogenic, sufficient reliable evidence that it's not carcinogenic. We can't establish it with 100% certainty. But there's sufficient reliable evidence for us to then determine that that, unless no other health effects, as the statute requires us to look at, is sufficient. Or does it have to be established in scientific literature under the regulation? Yeah, I think it has to be established not just under the regulation but also under the statute. It needs to be established that it doesn't cause cancer. And if it didn't have to be established that it didn't cause cancer, there would have to be a health threshold. They would have to be establishing a health threshold below it which it didn't cause cancer. Even if they didn't agree that for cancer that has to be zero, they would still have to establish some level below which it doesn't cause cancer. No, but say something is only carcinogenic at absolutely high volumes, constant exposure, and infrequent exposure or very diluted exposure is not carcinogenic. They could factor that in, right? They could proceed, assuming other statutory requirements are met, to set a health threshold for that. Could they not, even though it would be carcinogenic in some circumstances? I think I agree with you, Your Honor, that the EPA doesn't have to resolve every last molecule of scientific uncertainty. It just has to do what it's already done for other rules. And, again, the EPA says in the record there can be convincing evidence based on human and animal studies that a pollutant is not carcinogenic. That's at page 695 and 696. We're not asking for something that goes beyond that. We're just saying the EPA has to make a determination. You want them to use the same standard they use elsewhere. Right. Or, I mean, perhaps there could even be another standard, and the EPA could adopt that and explain that. What could the other standard be? Well, I'm not sure what the other standard could be, but the distinction I'm trying to draw here is that here EPA hasn't made any claim that these pollutants don't cause cancer, and neither has any other scientific or regulatory body. Is there any evidence that they do? Well, I'm not aware of the evidence, and, frankly, I think the problem is that There's nothing in the record that says they do. Right. Well, with respect to hydrogen fluoride, it's actually equivocal. There is something in the record that suggests that hydrogen fluoride does. But we're not trying to make the case that they do. I think the statutory point we're making is that Congress intended these pollutants to be regulated under technology-based standards and put the burden of proof on EPA. To move over into a D-4 if the burden would be on EPA. To put the burden on EPA. That's also consistent with the origins of the statute, not just the language of the statute, but the origins of the statute. Congress amended, as this Court has noted several times, Congress amended the Clean Air Act in 1990 and rewrote the hazardous air pollutant provisions precisely because the previous risk-based discretionary approach hadn't worked. And it was especially concerned about cancer, and it wanted to make sure that the scientific uncertainty that had stopped EPA from regulating hazardous air pollutants before didn't stop it now. In fact, that's what the U.S. Sugar Court noted at the beginning of its opinion just last year. And so for EPA to say that because it's uncertain, it doesn't have to set technology-based standards, turns that statute on its head. What it's really claiming is that it's not enough for a pollutant to be hazardous to be regulated under technology-based standards. It has to be hazardous, and somebody has to prove it's carcinogenic. That's not what Congress intended. Even if there were, and I guess I'd just like to make one point on sort of step two and go on to sort of a record-based argument. For EPA to claim that the same Congress that undisputedly made clear that a pollutant can't be carcinogenic, proceeds under Section 112.3.4. EPA agrees on that point. Didn't care whether EPA knew whether a pollutant was carcinogenic. That doesn't make any sense. And EPA doesn't explain that position anywhere on the record. Even if these words didn't exist, even if EPA were just based on a record agreement that there has to be a pollutant can't be carcinogenic for EPA to proceed, EPA would still have a burden to proceed to demonstrate with substantial evidence that that's true. This Court has held that over and over again in the prior air toxics cases under Section 112.3.3, and it's just as true for Section 112.3.4. And nowhere does EPA go forward with that affirmative demonstration. EPA just says, because we don't know, we can assume that this pollutant doesn't cause cancer. If I can just make a... What about the more general health threshold? What do you understand an ample margin of safety to be? My understanding here is that there is no ample margin. I'm just asking, what does that phrase mean? Well, I think that it means that a margin of safety that addresses risks that have not already been addressed and considered. And that's what the... It's a fairly old case, but there's an Environmental Defense Fund case, I believe in 1978, that discusses... It's a Clean Water Act case, but it discusses the term ample margin of safety and contrasts it to what an adequate margin of safety is in other parts of the Clean Air Act. Don't you actually have a much better case on this than you do on the carcinogen question? Your Honor, I think the carcinogen question is really strong, but... You seem to think so, but I'm not sure that that's the niche you want to die in. The amplitude, I'm not sure that I see where EPA has done something we can defer to on that one. I'll ask them about it, but you seem to want to spend all your time on the carcinogen, which may not be as strong as... Okay, well, we'll certainly move on, Your Honor. Well, you're out of time. Go ahead. Thank you. There are two points on the non-carcinogen. First, it's precisely because EPA is relying on low-confidence studies and data and analysis that it says only that it doesn't expect these pollutants... It doesn't expect that there will be adverse effects under its threshold. And I'm saying that not expected is different than established. The terms are not synonyms. Even industry interveners agree that what a threshold is is a level below which adverse health effects don't occur. To move on to the adequate margin of safety analysis, it is undisputed that EPA's rule allows exposures right up to the threshold it determined. That can't be what Congress meant by directing EPA to consider the threshold with an ample margin of safety. And EPA essentially argues that its decision not to provide any identifiable margin of safety at all should be overlooked because it thinks the assumptions that it used to develop the threshold in the first place are conservative. That doesn't address the statutory issue, which is that Congress wanted both a threshold and a margin of safety. And in any event, and I don't want to go into the details, because I realize I'm running out of time, but the allegedly conservative assumptions aren't even conservative. I don't know if the court wants me to address the technology-based standards. Very briefly. I'm sorry. Very briefly. There are two issues on the standards for all the other pollutants, and one of those is EPA's use of small data sets. The court has already held in the NAQA case that the upper prediction limit formula that EPA applied here is flawed, that's the court's word, with respect to small data sets because it produces demonstrably irrational results, absurd results even. EPA, following NAQA, needed to explain either explain why the upper prediction limit works or come up with something else, and it didn't do either one of those things. The last point I'd like to make is that even if that could be overlooked, the idea that EPA can set three different standards. Well, just to back up on that, what was before was they said you've got to explain how you can rely on this when you've got such a low data set. And they say this time we gave you the explanation for how we were doing it, so why isn't that consistent with our cases? And then what we have to grapple with is not what old cases said but whether the explanation here was sufficient. So why don't you tell me why the explanation here wasn't sufficient. The explanation doesn't go to why the UPL works. EPA doesn't even claim that the UPL works. What the explanation is, is yes, we know it doesn't work, we know we can't rely on it, but we can figure out for each and every rule, we can come up with some approach. We have come up with some approach that allows us to tell which of these 34 standards that had small data sets are defective and which ones aren't. And there's no way in the record to tell, actually. Well, they have to be able to do something when they have small data sets, so what are they supposed to do? Oh, well, one thing they can certainly do is collect more data. As the Court held in the sugar case, if EPA needs more data, it certainly has authority to collect it. But we're not trying to tell it to you. We've got a small set. I mean, there's nowhere else to fish, right? Well, there is, Your Honor. EPA has authority under Section 114 of the Clean Air Act to ask for data. It has authority to have asked for testing and get more data. So get more data from the companies themselves. If you get more data. Another reason it has such small data sets is that it chopped this category up into so many different categories. If it hadn't done that, it would have more data for each category. And then beyond that, there's nothing in the Act that says EPA has to use this upper prediction limit. The Court said it can, but it didn't say it has to. This is the same situation very much as the cement film recycling case, where EPA wanted to use this technology-based approach and then complained that it wasn't working very well. And the Court said nothing requires EPA to use this technology-based approach in the first place. If it's not working, it needs to devise another one. All right. Thank you, Your Honor. Counsel for Respondents. Good morning. Good morning, Amanda. Kate Bowers here on behalf of the United States EPA. I think you have to bring the mic closer. I will do that. Is this better? Yes. Great. Good morning. I'm Kate Bowers here on behalf of the United States Environmental Protection Agency. With me at counsel's table are Sonia Shea, also from the Department of Justice, and Scott Jordan from the EPA Office of General Counsel. My plan is to also start by addressing the issues raised by the environmental petitioners. And then at the end of my time, I'd be happy to discuss the abeyance question. I think you want to keep your voice up. I'll do that. I apologize, Your Honor. I'd be happy to address abeyance questions after. No, we got that. Sure. Okay. All right. I'll start with the carcinogenicity issue, since I think this is a point where we would all benefit from a little bit of clarification about what the statute says and doesn't say and what EPA did. We'll start with your regulation, right? Your regulation, not even the statute, says it has to be not carcinogenic. Right. Doesn't that mean someone has to decide that it's not carcinogenic? Well, there are two, to back up for a minute, there are two considerations that go into determining whether a pollutant can be a threshold pollutant. Well, I just want to, again, not the statute but your regulation, says it has to be non-carcinogenic. And so I guess I just have a direct question to you, and to be non-carcinogenic, doesn't someone, you, someone else, have to decide that it is not carcinogenic? As a practical matter, there are a handful, I'm aware of, I think, two occasions where EPA has actually classified a pollutant as not carcinogenic or not likely to be carcinogenic. That's two out of the 187 listed houses. No, all right, so you're able to classify things as either not carcinogenic or not likely carcinogenic, and is that what the regulation, isn't that what the regulation, I'm not going to say, isn't that exactly what the regulation requires you to do? There's no overarching regulation beyond what EPA has set out in this rule and in prior rules where it considered promulgating health-based standards. But the practice that EPA has consistently applied in its analysis. I don't understand what you mean by there's no, your position is, and it's in your briefing here, is that to be listed it has to be non-carcinogenic. And the EPA is in the business of determining things are not carcinogenic or not likely to be carcinogenic. And isn't that what your regulation, your regulatory position requires you to do? Well, EPA has to be able to say that for purposes of setting a health-based emission standard, yes, that a pollutant is not carcinogenic. Okay, did you say that in this case? I believe EPA said, I can pull up the exact language. Did you say that it's not or not likely carcinogenic in this case? Well, it's a little bit different for each of the three pollutants. Did you do it for any one of them in this case? EPA, let me, I can pull up the exact language if that's the easiest thing to do. But EPA did not issue a conclusive statement that to a level of scientific proof. I'm not scientific. I'm saying nobody's, I guess you can't point me to anybody who has said even that these things are not likely to be carcinogenic. Is that right? Am I just right in understanding the record that you can't point me to someone, anyone who has said that these are not likely to be carcinogenic? What EPA said in the record was that no scientific body had classified these pollutants as known, likely, or suspected carcinogens. That just says that nobody else has done the job either. I guess I'm, one more time. At any point, did anybody say that any one of these three is not likely to be carcinogenic? EPA did not make a formal classification. What EPA pointed to. Did anybody, okay, go ahead. What EPA pointed to was negative carcinogenicity data. So what EPA looked at was for each of these three pollutants, what were the studies that were available on cancer risk? And for each of the pollutants to varying degrees, there were studies that indicated that there was not a risk of cancer. There was not conclusive proof, which, as I think this Court recognizes, is not something that is necessarily possible. So then why didn't EPA just make a find that these are not likely to be carcinogenic? Because the weight of the evidence would not have been sufficient to support that judgment. Right. So didn't you just confess then that you don't have substantial evidence that these things are not carcinogenic? EPA believed in this rule that it had sufficient evidence for purposes of setting the health-based standard that the pollutants were not carcinogenic. You have substantial evidence? You just told me they couldn't make that determination because they didn't have enough evidence. I'm not understanding what the substantial evidence of non-carcinogenicity is that your own rule requires. Well, the rule doesn't specify what kind of determination EPA has to make in terms of the weight of the evidence. So what do you tell me? What is it then? Well, so there are these weight-of-the-evidence classifications where EPA may group a pollutant after it does a full toxicity assessment as either a known, likely, suspected carcinogen, definitely not a carcinogen, not likely a carcinogen, or this middle category where there's not enough data to make a conclusive determination as to either carcinogenicity or non-carcinogenicity. And it's in that middle group that we find ourselves here. Well, is being in the middle group good enough to qualify for this method of non-tech-based analysis or decision-making? Well, what EPA does in that particular case and what it did here is do a case-by-case evaluation to pull up the studies that either EPA or other agencies have done. And when it gets through with those, is being in the middle group good enough? In this particular case, with respect to these three pollutants, EPA concluded yes. But so normally when an agency has a regulation and its plain text says it will do something and it doesn't do it, it's not a sufficient answer to say, in this case, for this day, this ticket, good one day, one way only, we're not doing that. But there's no requirement that EPA conduct a full weight-of-the-evidence assessment. There's a big world between those two options, right? I think all we're asking is if you've got... No one's really given us what your standard is. You said we don't... You didn't say we determined not likely using our test for what not likely is or that you're interpreting the language of your regulation a certain way or what your position is, as I understand it, but we don't know. We don't have enough evidence to decide this question, even though our regulation requires us to decide it. Well, the regulation requires EPA to decide, but not necessarily to affirmatively prove or to classify a pollutant into one of the different... But did you decide that this was not likely carcinogenic? Did you decide that? No, not in that exact language. What language did you... Did you make that judgment in different language? EPA stated in the rule, based on the negative carcinogenicity and on the EPA's knowledge of how the different pollutants react in the body and their likely mechanism of action, the EPA could consider them to be threshold pollutants. That's not making the finding at all. That's just hopping right over the carcinogenicity determination. But that finding in the context of the rule comes at the conclusion of a pretty extensive consideration of studies of carcinogenicity that turned up no evidence of cancer risk. Is that going to carry the day for you if there's enough in the record that the EPA didn't make the decision, never made the decision on carcinogenicity? Is it good enough to say there is evidence in the record that would support a finding of non-carcinogenicity? Is that good enough? Well, there certainly is evidence in the record that would support... Whoa, whoa, whoa, whoa. Did I ask you that? Did I ask you whether or not... I asked you whether having evidence in the record is good enough if there's not an express decision to that effect. Well, it's certainly our position, Your Honor, that EPA was not required to make the express decision. Well, doesn't that make us into a fact finder then if the threshold required by the statute is to not be a carcinogen and the EPA has not made a decision by what you tell us. The EPA has not made a decision on whether or not it's a carcinogen. Aren't you asking us to be fact finders then in order to get us to affirm? No, Your Honor. We would not ask the court to serve in that role. In that case, shouldn't we send it back and ask the EPA to make a decision? If the court... I thought you told me earlier that one of the reasons they didn't make the decision is they didn't think the evidence in this record was not sufficient enough to support that judgment, so they couldn't even just do it on this record is what you told me a few minutes ago, right? EPA believed that there was enough evidence of non-carcinogenicity to support threshold. I got that. You're just...we're at the predicate question that comes before that as to carcinogenicity. And you did tell me earlier that the reason they didn't make that specific determination is that while there were studies here, they didn't consider it to be sufficient to make that determination. I'm just asking what you told me earlier. If I misunderstood it, that's why I asked you about substantial evidence because you said it wasn't sufficient for them to make that judgment. It wasn't sufficient for EPA to, in the context of this rulemaking, make a weight-of-the-evidence classification. Okay. But that's not...that's typically a multi-year process that just...it serves a... certainly there's overlap, Your Honor, but the multi-year process of doing a weight-of-the-evidence classification and a full toxicity assessment is just not something that EPA was contemplating in the scope of setting standards here. Can I ask, unless my colleagues have more on this, if you move to the question of the ample margin, where is the EPA told us what the margin is and why that's ample? Sure. To answer briefly and then to elaborate, EPA did not quantify what an ample margin of safety would be, either in this case or generally. EPA didn't have and wasn't obligated to provide that kind of global definition of an ample margin of safety. But what EPA did say in the rule is that it was expressly setting out to set the health-based emission limits at a level that at least assures that for the sources in the controlled category or subcategory, persons exposed to emissions of the pollutant would not experience the adverse health effects on which the threshold is based. The statute, however, requires an ample margin of safety when establishing standards. That's correct. I didn't hear you tell me how EPA had complied with the language of the statute. You may have told me EPA did another good thing, but Congress decides what's the good thing they have to do. Congress decides, but at the same time, Your Honor, Congress did not define what an ample margin of safety is. Therefore, when we call that a Chevron question, Congress left open for the agency to determine. That's exactly right. In order to affirm on a Chevron analysis, we have to have the agency doing something that we can affirm. If we can't read the record and tell what EPA meant by an ample margin, what they understood the statute to mean by an ample margin, how can we denounce it? EPA did explain in the rule, however, that in the context of providing an ample margin of safety, it calculated the emission limits in such a way that people exposed to emissions would not experience the adverse health effects, even if the emissions reached the threshold level. And in both the proposed and particularly in the- So that sounds like the threshold level. But you have to have a statute that requires a cushion. Right. So it seems to me that in fact what you're saying is that there is no cushion because all you found was that right at that threshold, we don't think it's going to happen. But the EPA did provide a cushion, Your Honor, and that's fully explained in the rule. EPA made several conservative assumptions in the context of its modeling in order to calculate the emissions limitations. Those conservative assumptions were all things that could happen in the world. You think they're unlikely, but they could all happen. That is a possibility. Okay. So if it's possible that people could be exposed right at this threshold level, then we're back to no cushion. But all of the assumptions taken together would generate a scenario in which EPA was comfortable with the unlikelihood that exposure would, in fact, happen at the threshold level. That sounds like you're telling us how EPA maybe could have found an apple margin here, but it still doesn't supply. We can't use the arguments of counsel to post hoc justify what the agency didn't find. And it doesn't appear that the agency made a finding of an apple margin in safety, does it? It did, though, Your Honor. I asked you to quote me where it did, and what you quoted me didn't say that. I don't want to waste the Court's time by looking for the specific language, but what I can tell you, Your Honor, is that that discussion did come up specifically in the response to comments regarding the apple margin of safety. I do believe it is. And therefore, the agency responded in terms of the conservative assumption. That's correct, Your Honor. Yes. And the agency believed that the conservative assumptions taken together in the modeling that it set up would provide that apple margin of safety. To require the agency to actually quantify, would it be a 90 percent likelihood, or would the exposures be 10 percent below? That simply goes beyond what's actually contained in the statute. Well, we have to have something to decide that it's ample, right? Ample has to have meaning. And unless we know, just to say, well, we were conservative, conservative doesn't mean it won't happen. Conservative means it could happen. The conservative assumptions mean that EPA was not envisioning a scenario in which the exposure limits, the exposure would happen at the threshold level. That was the purpose of the assumptions. Well, wasn't one of them you didn't think that it would happen? That's the whole point of the thresholds, right? People will go up to the thresholds. That someone wouldn't be subjected to it for 24 hours a day. Wasn't that one of them? Yes, that's one of them. Okay, so is that even rational? If somebody's in a nursing home nearby, they don't get out, they don't have a lot of mobility, they're going to be there 24 hours a day. Well, that just doesn't seem like a very conservative one at all to think that people won't live in a certain area or be in a certain area for 24 hours a day. You're correct, Your Honor, that there is certainly a possibility. Not a possibility. This happens all the time every day. That's not even conservative, I don't think. I'm not aware of information on the record, Your Honor, that identifies specifically what the demographics are like at the particular receptors where emissions occur. That would be a problem. So I can't speak to that with more specificity. If the Court has no other questions on the health-based standards, I can turn quickly to the question of the upper prediction limit. Why isn't the obvious thing to do when you have small data sets just either to get more information from the companies or to repeat tests more frequently so that you have more data instead of going good enough for government work? That is an option that's available to EPA, Your Honor. That's correct. Why wouldn't that be the reasonable thing to do, then, if you're trying to justify application of this UPL to small data sets? It would be one way to address the uncertainty that's inherent in the UPL at small sample sizes. But what EPA was able to do and what it fully explained in the limited data sets memos in the record is how application of the UPL methodology to sample sizes as small as three does still allow a reasonable inference as to the average emission levels of the best-performing source or sources. Did one of the tunnel kilns involve only two data points? I believe one of the sources involved two data points, and for that one, EPA— And you agree that's too little? EPA determined that the UPL methodology was not appropriate for sample sizes that small. So in that particular case, EPA then selected the next best-performing source and calculated the amount— But then that could have you not using the best performer instead of just doing repeat studies. Well, it might not be the best performer, but the second best one was the best one for which the administrator had available emissions information. Instead of getting information from the one that is actually an issue, shouldn't it do that first? Doesn't the statute require that? Well, I don't believe the statute specifically sets out what level of data gathering is necessary in order for EPA to evaluate these scores. And that's something that the court has acknowledged as recently as in U.S. Sugar, that EPA has wide deference to determine when it has enough data in order to set the max standards. Here, because EPA was able to determine that the application of the UPL methodology to limited data sets allowed with enough certainty in most cases to reasonably infer the emissions levels of the best-performing sources. Beyond that, EPA developed a case-by-case further evaluative process to make sure that the calculated emissions levels were reasonable. When EPA then applied that process to the UPL calculations in this particular rule, the outcome there was reasonable. If the court doesn't have further questions on the substantive issues, I'd be happy now to turn to the matter of abeyance. As the court knows from the letter that we submitted last Friday, EPA is now proceeding to reconsider the aspects of the rule that were challenged by the industry petitioners. EPA has set out a timeframe for that reconsideration process. And the agency's positions on these issues may change from the positions that were taken in our brief. Here's one thing I'm worried about, and that is this rule was supposed to be out in 2000, right? And, right? The statute required a rule by standards by 2000. That's right. And under your scenario, we're now going to be 19 years past that. And I get, you know, in an ordinary course the timeframe might be adequate, but we're not in the ordinary course here. We are pushing two decades after the statutory deadline was set. Why is the timeframe you proposed reasonable? And I get that you're only talking about two years out of that almost 20, but it's the same EPA for all 20 years. Right. Why is that timeframe acceptable under these circumstances? Don't you have some duty to act with exceptional urgency? EPA certainly aims to act as expeditiously as it can. The rulemaking process is going to involve review of statutory interpretations, policy, scientific data, potentially the submission of Well, you've already presumably done some of that when you decided that their claims were valid. And in other cases, EPA has been able to move faster when it's flown past statutory limits far less aggressively than it has here. So I'm just really concerned about this timeframe. We really are winking at an egregious violation of the statutory deadline. Understood, Your Honor. The agency believed that this was a reasonable amount of time given the amount of information. Yeah, reasonable. And it might be reasonable. The question is whether it's expedition. Is that the best the agency can do? At the time the letter was issued, yes. That's the absolute level best the agency could do? That's what you're telling me? I don't know, Your Honor. Okay, you don't know then. That's why our order was very specific. We want to know. The industry has a lot of this information. The decision has already been made that some of or maybe all of the industry challenges are worth reconsidering, correct? That's right. So industry has the information you need. It's in its own interest to give that information to the agency. So this is not as though the agency is starting at square one, developing statutory interpretation analysis, developing data sets, developing scientific methods, et cetera. So that's why the two years doesn't seem like the agency is acting as expeditiously as possible. So the question is, why is all this time needed? Your Honor, I don't have more specific details for you. Well, that's why we issued the order, all right? So that you and your colleagues would be on notice. That we do want to know specifically what new studies need to be done, what data don't you have. Now, we're not trying to run the agency. But it seems that in this particular case, looking at the allegations that are made in the industry brief, why is two more years required? I apologize, Your Honor, that I'm not going to be able to walk you through with more detail today. So who can? Is somebody at counsel table in a position to provide us this information? You introduced them at the beginning. There's somebody here from EPA. I don't think that EPA is in a position today to be able to walk through what specific new information it plans to either solicit. How did you know you could do it in two years? I don't know, Your Honor, what specifically went into that calculation. How could you tell the court then that the agency had a duty to act as expeditiously as possible? Our understanding was that EPA needed that amount of time in order to assess the large... So can EPA tell us why? We would be happy to submit something additional as quickly as possible. Well, we're trying to move this case forward. And we have the letter from the administrator. And we have the industry allegations, not allegations, contentions. We have the administrator's letter saying these are serious concerns that need to be addressed. So what did you think our order meant when we said be prepared to address with specificity whether an additional period of abeyance is appropriate? Specificity. I apologize, Your Honor, that I'm not able to provide specific information about a reconsideration process that EPA is only just beginning now. What do you mean only just beginning? You told us in the letter that you have apparently already studied the concerns raised by industry and made a determination that they're valid and require revisiting the rules. So how much process went into that? What's already been done? I believe that EPA reviewed the issues that the industry petitioners raised. Reviewed the issues or reviewed evidence? The issues don't require doing anything more than reading the brief. I believe that EPA reviewed the issues and determined that it should begin the formal reconsideration process. Did it review any evidence yet? I don't know, Your Honor. So reviewed the issues. So you haven't done any legwork but determined that there's a change has to be made. I would have thought some of the homework would have been done when you decide that a change needs to be made, that they're valid enough to require changing the rules. I don't have the details about what specifically was before the administrator and others at EPA when the decision was made to initiate reconsideration here. I mean, your brief was filed April 28th of this year. So somebody's been looking at these issues in order to file that brief. Yes, the final briefs were filed in April. The proof briefs had been filed in January. Somebody did not notify the court in April that we should not proceed on the basis of the brief that was filed later in April. Your Honor, we notified the court as soon as EPA decided that it would give these issues further review. And again, when EPA decided that it would begin reconsideration proceedings. We regret that that decision was not made sooner. We understand. No, I understand entirely that a new administration needs time to figure out what it wants to do. It is now November 9th. Industry has been knocking on EPA's doors to let it know what it wants done. And here on November 9th, we still have no information. And you're representing the agency. And you're saying counsel from EPA cannot provide any information? I'm not aware of more specific information that counsel from EPA would be able to provide today. Can counsel from EPA make that representation? Your Honor, I'm Scott Jordan from the Office of the Attorney General Counsel. Thank you. I can clarify a couple of things. I'm not sure I'm ultimately going to be satisfied with everything you've asked. A couple of things to clarify. So October 1st, the administrator decided that we would be reviewing this rule on all the issues. It's October 1st. That was when that letter came out. And what happened that led to that decision on October 1st? All we were told here was they looked at the issues. I assume that what you're going to tell me is when they decided something was valid, they've actually already done some factual study and got new information or additional information to decide that. I was not personally involved in those briefings. I do know that there was a series of briefings up through the management chain that ended with the administrator and the decision that we saw in the letter. And following that, that review then began or continued. A week or so ago, we had the initial decision with respect to the environmental group issues that there would be no further review of those. I'm just trying to understand what the letter said. It doesn't just say we're looking at it and we want time to look at it. It says we've looked and we've made a determination that these claims are valid and we're going to have to. That's what I want to clarify because I think what we said, the letter said that after initial review, we decided that we needed to do further reconsideration. We have not reached the point where we've concluded that any of these issues are valid and will necessarily result in a change. So what's that initial point? You're not confessing error. I know agencies don't generally do that, but you're not doing that. We've not reached a point at which we can. But is there anything, if we decided the case on the present record, is there anything that would prevent the agency, the administration, from going ahead and reconsidering anyway? Wouldn't you still be able to do that? Right, Your Honor. But as we've argued— If we deny your motion, we haven't taken away from you the ability to do exactly what it is you want to do. We simply put something in place in the meantime. Well, Your Honor, I'll make one point and then I may hand this back to the Department of Justice attorney. I think the point that we are making is that this is under review. At this point, it would be speculative to predict anything about how the rule would be changed other than that it may be changed, and this is a point where I'll hand it back to you. What I'm saying is that if we deny your motion, it would still not be changed. You could go back and have another rulemaking. Agencies can do that. They're not locked in for the entire future of mankind. I don't see what we're taking away from you if we deny your motion. Well, I would just make the suggestion that, as we've argued in our motion, that it would be more in the interest of the court's resources and judicial economy not to be reviewing a rule that's under review. Thank you. Could I ask counsel one other question, if I may? In terms of our concern about the length of time, who within the agency is the unit or division that would be able to provide that type of information? I know the administrator and the deputy, but in terms of giving those high officials the information they need to make a representation to the court, who within the department do we know? What I do know is that the letter announced the time frame that came from the administrator. The question is, is it possible to do it in shorter time? That's certainly an issue we could take back, but again, the agency has to balance the many different things it has to do. One thing I will point out, which may have been overlooked here, is this isn't a question of putting a rule on the books two years from now. There's a rule on the books. And that question could still be out there, even if we decided occasionally it didn't have to be. You could still revise your rule. That's correct, Your Honor. Even after a court's decision, we would be obviously subject to whatever decision you make and comply with that. But that would not prevent us from further reconsidering these issues. All right, anything further? No, thank you, Your Honor. Yes. Counsel for the industry. Perhaps you can provide some information. May it please the Court? I think we should dive right into talking about the abeyance issue. So the first thing I want to point out, Judge Millett, you were concerned about this rule being in place. There's two important things. One is that the compliance date is December of 2018. Nothing that has happened so far has changed that deadline. But as a practical matter, we hear all the time from industry it needs time to rev up, to plan, etc. So December's right around the corner. Yes, we're very concerned about that. I think we were saying the same thing. Yes, we are with you. We have a vested interest in this going as quickly as possible. So why do you want a decision now on your claims? That'll give you a year to get ramped up in case they don't do anything next year. Because next year, all they're not even promising is saying they hope to have a proposed rule, revised rule, by next fall, which won't eliminate the existing obligations. It'll be due right around the same time. So I don't understand. Are you planning to get a stay? Is the agency giving you a stay as long as they're revising this? They have not. Well, I mean, you all agreed to this. So I'm trying to understand how to reconcile your statements about being so acutely concerned about legal deadlines that are pending out there, and your willingness for this just to stretch out. Well, we believe that the agency here should be making the decision in the first instance. Judge Centell, you'd made the point, well, if we issue a decision, that won't prevent the agency from revising the rulemaking. That's true. And also, I think that there's an important consideration. It wouldn't be in the interest of this court or in the parties here to have a case that's ultimately mooted. You know, a lot of these issues here are Chevron 2 issues, right? I don't think it's going to be mooted. Well, there is a case law. I'm happy to provide to your honor that if the facts, if there's no longer a case or controversy at the time. No, I understand that. But we're talking at least two years. It doesn't take two years for us to rule on your 90 days on cert petition time to run. Well, I mean, those are the facts that would lay out. I mean, that's nothing that anyone here can speak definitively to. So I'm looking at the issues that industry has raised in its brief. Yes. Are you in a position to explain to the court why two years is needed? I don't know specifically what the agency has in mind. Rulemaking's take a long time. Two years actually isn't very long when you look at, you know, OMB review, proposed final rule, notice and comment, response to comments. So what new studies does industry think need to be conducted? I think that for the second issue, I think that we would want EPA to reassess the data, include more data. I don't know that a lot of new studies really need to be done. A lot of this has to do with their legal interpretations and the way that they evaluated existing data. Is there any reason you guys couldn't give them whatever information they need from your perspective? As far as I'm aware, we're very happy to. I mean, as you already have from the prior proceeding, there's already a full agency record. And I assume during this months-long process that led to this October 1 letter, you may have provided them additional information. But I take it there's nothing else that industry thinks it needs to provide for a decision to be made. Right, right. And this court's function is to decide cases or controversies between parties seeking their assistance. And to you, I want to be sure I understand your answer to it. Her last question, as to whether we provided all relevant information? Yes. Yes, I did hear it. I'm not aware that there are important studies or additional factual findings that need to be made. And you saw our order asking for specificity. Right. So your representation on behalf of your clients is that the agency has all the data it needs. That what we're talking about is either statutory interpretation or we're talking about interpretation of data it already has. Right, exactly. And it's just a question of the agency's bandwidth, which is something we don't have any control over. The agency's bandwidth meaning what? The number of employees it will dedicate to this task? Right, in addition to its other obligations, the other deadlines it's under. We will do everything we can to assist the agency to do this as quickly as possible. And at a certain point, we can't do anything more. I'm very confused about your agreement to a deadline that comes a year after or not a deadline, a proposed final rule that comes almost a year after your compliance deadline under the existing rule. Why aren't you demanding that they act in time for you not to have to expend an enormous amount of money complying with the existing rule? I don't understand that position at all, that aspect of your position at all. Well, I think that we would prefer to find a way to work with the agency cooperatively and get this done as quickly and as well as possible. Well, this is going to be a rulemaking process. I hope it's not going to be just industry. Of course not. You should want a time frame for this transparent public process that is about a quarter of the time that they've proposed. I think, Your Honor, at this point we're happy that they are opening the reconsideration process. In the meantime, you're still planning to come in compliance with the existing rule by December 2018? At this point, that is the law. Is there anything that would prevent us legally from deciding to deny your motion and deciding this case on the briefs with respect to your petition? Well, I think that it would be disadvantageous for you to decide it without the opportunity for oral argument. What did you understand my question to be? I thought you were asking is there anything preventing us from deciding our issues on the briefs? That's what I was asking. Did you give me a yes or a no to that? I did not. Would you give me a yes or a no to that? I don't think that anything would prevent you. I mean, oral argument isn't technically required, but I think that it would be in your interest. Would you like to speak more about the advance or can I speak briefly about it? Yes. Let me just be clear, though. What did you have in mind addressing other than advance? I wanted to speak about the environmental petitioner's issues. Is there anything new that is not in your brief? Well, I wanted to clarify something. What did you want to clarify? Judge Millett, you keep speaking about EPA's regulation. It's my understanding that there's not a specific regulation setting out how to do the 112-D4 analysis. You have statements in the preamble. I don't think that's the basis on which the discussion proceeded. Okay. Anything further? I guess I would just like to leave you all that safe doesn't mean no risk and that with threshold pollutants, finding out that it's very – there's a very low probability that that threshold would ever be reached is a margin of safety because below the threshold would be no effects, and that is quite exceptional. Well, you're familiar with January. All right. Thank you. Thank you. Thank you. All right. Counsel, we'll give you a minute or two here. Thank you, Your Honor. I don't know if the court has any questions for me on the subject of advance, but if not, I really only wanted to address one point on the merits. All right. That point goes to the upper prediction limit, and I think counsel may have been understood – it may have sounded like she was saying that EPA had confidence in the upper prediction limit formula for small data sets and was using a case-by-case evaluation to confirm that. That's not what the record indicates. EPA doesn't claim that it knows this – that formula will generate rational results in any instance. What it has is 34 different standards based on a formula it knows to be flawed, and that formula is the only way it has at measuring their actual emissions. So because the formula doesn't work, it has no way of checking its analysis. It still says it did this case-by-case analysis and somehow eyeballed the results, and somehow that eyeballing makes the results reasonable. That's the same kind of assertion that EPA has made in so many prior air toxics cases where it offers assertions that its results are reasonable rather than demonstrating it. And if I can just give a quick example. So there are 29 floors that EPA didn't change, and there's nothing in the record that shows how any one of those floors actually reflects what the best source has achieved. There's just EPA's statement that it thinks its approach to eyeballing the results and checking them is reasonable. If that's – if the Court has no further questions, that's all I have. Thank you. We will take the cases under advisement.