# Oral argument — United States Sugar Corp v. EPA - Part A: Industry

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/argument%3A14485

## Record

- **Collection:** Oral argument
- **Document type:** Oral argument
- **Published:** December 3, 2015

## Text

Case No. 11-1108 et al., United States Sugar Corporation Petitioner v. Environmental Protection Agency. Case No. 11-1125 et al., American Forest and Paper Association et al., Petitioners v. Environmental Protection Agency. And 11-1141 et al., American Chemistry Council Petitioner v. Environmental Protection Agency. Part A, Industry. Before you begin, counsel, let me just get a couple of housekeeping things straight. If all, counsel would introduce himself or herself as you speak, because normally the deputy clerk does that, but we've got so many, we thought we'd save time. And speaking of time, we are going to hear from a lot of people this morning, so we're going to be strict about the clock. And when you see the red light go on, finish your sentence, but then please sit down, unless you're answering questions. All right? Good morning. Good morning. May it please the court, my name is Bill Wareham. I'm counsel for a group of industry trade associations, including the American Forest and Paper Association, the United States Chamber of Commerce, and the National Association of Manufacturers. I'm arguing this morning on behalf of industry petitioners on the issue of EPA's failure to set standards for emissions during periods of malfunction. The bottom line is that the final rules are flawed, because EPA did not set a standard that accommodates emissions during periods of malfunction. Instead, the final rule requires that the standard developed for periods of normal operation applies during malfunctions. This is a problem because malfunctions unavoidably occur. In short, things break. And sometimes malfunctions cause emissions at levels greater than the standard for normal operations. EPA admitted as much in the record, this is a JA 197, where they said that malfunctions are expected to occur, and at times those malfunctions should be expected to result in emissions that exceed the standard for normal operations. This means that EPA made an express decision to set standards for periods of malfunction that it knows cannot be met. But as I understand it, EPA has tried to solve that problem by providing for case-by-case review of malfunctions. So that's one way to solve the problem. Why does industry think that doesn't work? For three reasons, Your Honor. First of all, we believe the agency's approach, which is case-by-case, and to address malfunctions case-by-case through enforcement discretion. And that's very important. And we think there are three problems with that. First of all, we think it doesn't meet the law. We believe Section 112, Section 112D2, and Section 112D3 says that EPA must set standards that it has determined are achievable. And on the record, EPA said it knows these standards are not achievable for periods of malfunction. So waving its hand and saying we'll deal with it during enforcement actions and we'll exercise discretion simply doesn't meet the law. So how would we deal with this otherwise? In other words, I'm assuming that what industry wants is an emission standard that would be higher so that even if you had these spikes, it would somehow fit within that limitation. But how, since malfunctions are ordinarily not something you could predict, how would EPA do that? Correct, Your Honor. What we're asking for is a standard that's either applicable to malfunctions or a standard that's designed to accommodate malfunctions. And that didn't happen here. The record is very clear. The standard that EPA applies to malfunctions is a standard that was developed using data from boilers operating under normal conditions. I guess my problem is how would they develop this data set that would allow them to figure out what that margin should be? Right. In reading the briefs, it sounds like we're dealing with a great unknown. We've come to call it the chaos argument. But this is not chaos. We're dealing with things that we know. We're dealing with industrial boilers. We're dealing with industrial incinerators. We know what they are. We know how they operate. We know what kind of emissions controls are used. We know how they work. Elsewhere in the statute, we don't call it the rate precision, do we? It's a reasonable estimate? Is that what you're getting at? Do you think a numeric standard is possible? Or should they resort to a work practice rule? Both might be appropriate. But we don't know because EPA completely failed on the record in this rule to gather data on malfunctions, to do the kind of assessment I just described. Again, we're not dealing with a great unknown. We're dealing with a type of emission source that's very discreet. There are a lot of them out there, and they do a lot of different things. But at the end of the day, people know what a boiler is. People know what an incinerator is. And EPA could do an inquiry into how they operate, how they fail, and then use that information, using the standard setting methods available under the statute, to set a standard. It might be a numeric standard. If EPA develops data that allows that to happen, it could be a work practice standard. And we already have that in our briefs. And we think that probably is more appropriate because, Judge Brown, of what you said, which is there is a certain element of unpredictability as to what a malfunction is. But that doesn't prevent EPA from setting a standard. And it certainly doesn't excuse EPA from setting a standard under a law that requires EPA to do that. So, Your Honors, we've touched on a lot of this. But we think what EPA has done here is flawed for three basic reasons. The first is we think it is entirely inconsistent with the statute. As I said earlier, Section 112.d.2, which is 7412.d.2 of the U.S.C., says that EPA must set the standards at a level the administrator, quote, determines is achievable. So the statute flat out says that EPA must set standards at levels that it has determined can be met. And, again, on the record, in this case, J.A. 197, EPA says it recognizes that equipment that's properly designed and maintained can sometimes fail and that such failure can sometimes cause exceedances of the relevant emission standard. The law says EPA must determine standards are achievable, and EPA determined the opposite on the record. It determined they're not achievable for periods of malfunction. We think that violates the law. There is supporting case law of this circuit on this point, Sierra Club 353 F3rd 976, where this court has said that EPA's obligation under this part of the act is to set standards at a level that the best performing peers have been shown to be achieved. So that's a reflection of the obligation of the statute. All right. The second reason we think what EPA has done is wrong is it flies in the face of clear circuit precedent. And there's a lot of cases going back to the formative days of the Modern Clean Air Act, when EPA first started setting these kind of technology-based standards, and at that time it was under Section 111 of the Clean Air Act, a little bit different program than we're dealing with here. We're dealing with Section 7412. But in any event, you know, the question as to whether EPA should consider malfunctions and accommodate malfunctions in standards setting was litigated as early as 1973, just a couple years after the advent of the Modern Clean Air Act. And this court three times over said it would be arbitrary and capricious for EPA to set standards and not accommodate periods of malfunction. This court said that in what we call Portland Cement 1. In our briefs, 486F2-375. There was a case just three months later called Essex Chemical, where the very same conclusion was drawn. And several years later in what we call National Line 1 in our briefs, 627F2-416, where the court reflected on those two prior decisions and reiterated the holding that it would be arbitrary and capricious for EPA not to accommodate malfunctions when setting technology-based standards. What's wrong with their case-by-case basis? Judge Henderson, as I said, first of all, we believe that that... I mean, they have discretion. I realize you're talking about our precedent and so forth, but that doesn't violate our precedent. A case-by-case approach? Does it? We believe that it does. Okay, what case? In... One moment, Your Honor. In Portland Cement 1, again, Portland Cement 1 is 486F2-375. And this is in footnote 91. The court says, this view of enforcement, so they were talking about using enforcement discretion to deal with malfunctions. This court said, this view of enforcement discretion would defer the question of available technology to the enforcement stage, an approach not contemplated under Section 111. Companies must be on notice of what will constitute a violation. And a similar statement was made in what we call National Line 1, and again, that's 627F2-416. This is in footnote 46, where the court said, the flexibility appropriate to enforcement will not render achievable a standard which cannot be achieved on a regular basis. So, Judge Henderson, this court has considered this precise question, the precise question of, is it okay, is close enough good enough? Is it okay for EPA to say, we know there's a little slippage in these rules, but trust us, we'll deal with it during the enforcement action and we'll exercise appropriate discretion. This court has said, that's not appropriate, because EPA's obligation in the first instance is to set standards, that everybody knows what the standards are and can take actions necessary to comply with those standards and not wait until an enforcement action for somebody to make a decision as to whether what they've been doing is lawful. Well, part of my difficulty here is I'm not sure that I understand exactly how this works. Is the effect of their case-by-case determination actually an enforcement action? Is that the only context in which that would come up? I sort of read this as saying, when there are exceedances, the facility would be able to explain, right? But does that only occur in the context of an enforcement action? Yes, Your Honor. This has been a complicated case, as you well know. And there was a point in time at which EPA said, we're going to create an affirmative defense. And we're going to create criteria, and if you meet these criteria, then that will be a defense you can raise in the context of an enforcement action. And this court overturned the use of an affirmative defense in a separate case that actually occurred during the pendency briefing in this case. And so what we're left with is the agency asserting that we'll deal with it on a case-by-case basis, but purely in an enforcement context, and unguided even by an enforcement discretion rule or policy that would at least establish criteria. Just so I understand, now, one part of that affirmative defense would have been against citizen suits. But it sounds as if you're saying that the affirmative defense would also have been available against EPA or in an enforcement action brought by EPA. That's correct, Your Honor. And you raise a very good point. And another reason why, you know, solving this problem through enforcement discretion is a problem, is that EPA is not the only entity that can enforce here. You know, the Clean Air Act makes it clear states have the authority to enforce, citizens have the authority to enforce. So even if EPA has the best of intentions and follows through on the best of its intentions to use discretion not to take action under appropriate circumstances, where I think unavoidable malfunctions actually are the source of the violation, that doesn't prevent others who don't have the government's perspective from trying to enforce. And even if they're not successful, we have gone through the time and the effort to have to defend against those cases. We face the risk of losing. And, you know, we have to do all of that in a world where there's no standard to govern, you know, how that decision should be. The basic principle is that you can't make an unlawful rule lawful by promises to enforce it lawfully, right? That's correct, Judge Griffith. That's a fundamental principle, isn't it? And I believe that... Is that the gist of your argument? As it relates to this enforcement discretion point, that's exactly right. And I believe that's exactly what the footnote 91 in Portland Cement and Wine and the footnote 46 in National Law and One say. They say, you know, what EPA should be doing is setting standards. If it knows there's a period of time where the standards that would otherwise apply is not appropriate, it can't just wave its hand and say, we'll worry about that later. What they need to do is set a standard. And that's what we ask the court to do here. All right. Thank you. Thank you very much, Your Honors. Mr. McWilliams. May it please the court. My name is Douglas McWilliams, and I represent industry petitioners in our challenge to EPA's non-pollutant approach to setting MACT floor standards. When setting a MACT floor standard under Section 112.d.3 and the comparable CISWE language under 129.a.2, EPA is required to meet clear limits. The floor for new sources in a category or subcategory shall not be less stringent than the emission control that is achieved in practice by the best-controlled similar source. Congress directs EPA to set the floor based on reality in this sentence. What is achieved in practice by the best-performed similar source. EPA failed to follow this directive for the new heavy oil subcategory, as well as for the small remote incinerator category under CISWE. Sir, what in the statute tells us what best means? The debate here is whether the statute requires EPA to look at best overall, the treatment of it, or best for each individual. That's it. You're arguing that the statute requires that the EPA set a standard on best overall at a category, perhaps. But where in the statute does it say that? It just says best. It says the best achieved in practice, though. And I think our focus is that what is achieved in practice grounds this standard, what the EPA determines to be best in reality, what is achieved in practice by the best-performing source. I shouldn't agree with the EPA, but isn't their point, it is best achieved in practice pollutant by pollutant. Their argument would be that you're putting a gloss on that and saying best achieved in practice is best achieved overall by any single effort. But I don't see where the statute requires that. And obviously, if we're in the realm of ambiguity, we have to defer to the reasonable reading of the EPA. So where in the statute does it require it? What are the words in the statute that tether it to your reading that it's got to be best overall as opposed to the best pollutant by pollutant? Well, the words in the statute, achieved in practice, I think that is part of the way there. But as you point out, there's a need to look at the context of the statute as a whole. And the context of the statute as a whole dictates that Congress intended for all sources to be brought up to the level achieved by the best-performing source. Now, under EPA's approach, the pollutant by pollutant approach, the best performer for pollutant acts could be required to shut down if it could not meet the MAC standard for pollutant Y. Because the statute requires that you set a minimum floor below which you can't go. So if the best performer for pollutant X cannot meet pollutant Y's MAC standard, it has no choice but to shut down. There's no discretion to apply cost or other factors. When you look at our approach, and that is to set a standard achieved in practice by the best-performing source in the source category, you set a MAC floor standard that can be achieved by definition. It's been established by at least one source that they have determined it's cost-effective because they do it in practice across all pollutants. So the effect of the congressional structure here, the statutory structure, is to protect those best performers so that a best performer for one pollutant isn't required to shut down as a result of not being able to meet the other. Now, EPA's inquiry does not need to end there, and I think EPA and environmental petitioners have a parade of floors. Are you making an argument that the EPA's approach here runs into problems if they are mutually incapacitated to control technologies? In part, yes. Are there such instances? Do you have instances where that's happened? We don't know that based on the record today. What we have are standards, the new heavy oil subcategory as an example, where you have no source that has achieved all of the standards in practice. So no source has achieved the level that EPA set for the subcategory. We think that runs afoul of achieved in practice by the best-performing similar source. But EPA has the ability, if you set the max standard at the place where reality... It's been achieved in practice somewhere, hasn't it, by someone. Not for all pollutants. Not for all, but for that half in particular. For that half, but what that does is it creates the dilemma that if the best performer for pollutants cannot meet pollutant Y... I can see your point if you're saying it's impossible to do that, that there are mutually incompatible control technologies, but absent of showing that, I don't know why that isn't allowed. EPA, under the vague terms of the statute, what does that mean? Your Honor, the pollutant-by-pollutant approach sets up a scenario where that is a feasible outcome. And why would Congress have created that as the statutory structure when the purpose that Congress intended was to bring everybody up to a level achieved in practice by the best-performing source? If the best performer for pollutant X has to shut down for pollutant Y, it's our contention that that's inconsistent with the statutory structure that requires that you set max floors based on what's achieved in practice. But EPA doesn't need to stay there. EPA has the authority under 112D2 to go beyond that max floor. So if EPA believes that a pollutant-by-pollutant approach is appropriate under the statute, all they have to do is apply cost, energy, and other factors under 112D2 to justify that decision that a beyond-the-floor approach is appropriate in this instance. That protects the best performer for pollutant X from having to shut down because they couldn't meet pollutant Y's max standard. See, it's the very pollutant-by-pollutant approach that creates a dilemma that Congress could not have intended in the structure of this statute. With this context in mind... You need to wrap it up. Sure. I would just say that the analysis is similar for existing sources and is presented in our briefs. EPA exceeds its statutory authority by setting standards that have not been achieved in practice. So we ask that the court remand the subcategories for existing soaker coal, for new heavy oil, and for the small remote incinerators under CICLI. Thank you. All right. Thank you. Mr. Friedland. May it please the court, my name is David Friedland, and on behalf of the industry petitioners, I'm addressing the energy assessment requirement that is a term in both the major source and the area source spoiler map rules. The bottom line is that the energy assessment requirement is unlawful because it's beyond the scope of EPA's authority under Section 112. What I'd like to do is describe that authority, describe precisely what the energy assessment requires, and then show you how the energy assessment violates Section 112. So Section 112C requires EPA to list all categories and subcategories of major and area sources and then regulate them under 112D. So 112C says EPA first has to list the categories and subcategories and then regulate them under 112D. 112D in turn says for the categories and subcategories that were listed under 112C, EPA then has to regulate those categories and subcategories. So what did EPA do here? EPA listed industrial, commercial, institutional boilers as the category. And a boiler, as we all know, is nothing more than a device that provides steam, electricity, hot water to a facility. So EPA then went about setting numeric standards for the boiler. So far, so good. But then what EPA did is it promulgated the energy assessment, which is a study that goes far beyond the category or subcategory, which is the boiler. The energy assessment applies to, for example, facility heating, ventilating, and air conditioning. It applies to the building envelope. It applies to lighting. It requires a review of architectural and engineering plans. It requires a review of facility operation and maintenance procedures and logs, all of these things that are far, far beyond the boiler. And therefore, the energy assessment, it goes far beyond what the statute requires, which is a regulation of the category set forth in 112C, and therefore it's unlawful. There are a couple of other reasons why the energy assessment is unlawful. First, it is a work practice standard. Under the Clean Air Act, there are either emission limits or work practice standards. Emission limits, as we all know, means numbers, pounds per hour, tons per year, pounds per million BTU, et cetera. The energy assessment is not a number. The energy assessment is a study, which is a work practice standard. EPA argues to the contrary in footnotes in both its major source and area source briefs, but it is a work practice standard. It is not a number. It is not an emission limit. And therefore, EPA had to have gone through the 112H requirements for promulgating a work practice, which is EPA has to determine that it is not feasible to prescribe an emission limit if it's going to prescribe a work practice standard. EPA made no attempt to do that. It is a work practice standard. It's not a lawful work practice standard, and therefore it's unlawful. Finally, the energy assessment is unlawful because it doesn't require any emission reductions. The energy assessment is a study that does not require emission reductions, and the statute requires emission reductions. 112D2, maximum achievable emission reductions. 112D5, if we're talking about area sources, also requires emission reductions. The study does not require emission reductions. Therefore, it's unlawful under the statute. That also makes it unlawful as a beyond-the-floor standard. EPA claims that the energy assessment is a beyond-the-floor standard, but it's not a lawful beyond-the-floor standard because beyond-the-floor standards for EPA to go beyond the floor, they have to take into account cost. No question about that. Everybody agrees with that. They have to take into account cost. And if there's no emission reduction in a denominator, when we look at cost, you look at, for example, dollars per ton is how EPA usually does it. If there's no tons, no emission reduction, there's no way to take into account cost when EPA promulgates a beyond-the-floor standard. And EPA looked at the cost of the energy assessment itself, surely. EPA looked at the cost of doing an energy assessment, but that's not the cost that's in play here. We have to look at the cost of actually doing an emission reduction. And because the energy assessment does not actually require any emission reductions, we can't do a dollar per ton, we can't do a cost analysis, and therefore we can't have a legitimate beyond-the-floor standard here. So if I understand this correctly, there is a numeric standard here, an emission standard that they do for boilers, but this requirement is going beyond that. That's correct. There are numeric standards for various pollutants under both the major and the area, so standards for carbon monoxide, standards for mercury, standards for various pollutants. But the energy assessment reports to be a beyond-the-floor standard and is not tied at all to any of the emission, the actual numeric emission limits. It's not tied to any of those, and it can't be tied to any of those, because it does not require emission reduction. Okay. So I'm trying to understand, again, exactly how this works. Because on the one hand, they have set the numeric limits required by 112D. But you make what seems to be two arguments. One is they can't do this because what they're asking facilities to assess goes beyond just the boiler, right? It goes into these other aspects of the facility. And the other is it can't be a beyond-the-floor standard because you can't do a cost-benefit analysis, right? Correct. Okay. I should sit down. Correct. So I see why you make the beyond-the-floor argument. I'm not quite sure about the argument that boiler doesn't encompass anything else. I mean, the boiler isn't just sitting there in a vacuum, right? It's got to be within a facility. It's got to operate. And so can you really make the case that the language of the statute only applies to the boiler and nothing else? I think you can, Your Honor, because in this respect, the Clean Air Act, I hesitate to say it, but the Clean Air Act is actually pretty elegant in the way 112C and 112D operate together. In other words, 112C says EPA, write a list of all categories and subcategories of major area sources of air pollution. And then 112D says for that list that you just described, regulate those categories and those subcategories. And so EPA then defined boiler, quite clearly defined boiler, as basically the device, okay, the device. But EPA then, the energy assessment that I described, to the extent it requires looking at architectural plans, heating and air conditioning, this could certainly be, if we have a box at the facility and we have a boiler here, the architectural plans could be at sources way far away from the actual boiler. And the reason, that's unlawful because the statute doesn't allow it. But from a practical standpoint, EPA writes MAC standards for various types of equipment. We have boiler MAC'd. We have refinery MAC'd. We have chemical MAC'd. We have all types of individual types of sources. And if EPA is allowed to kind of take the boiler MAC'd and run roughshod over the whole facility, and there are usually always other MAC standards applicable at the facility. So my client, and this is the American Chemistry Council, so most chemistry plants, most chemical plants, have numerous MAC standards already applicable. They have the polymers and resin standard. They have the refinery standard. They have the leak standard. They have all sorts of vinyl chloride standards. They have all sorts of MAC standards applicable to particular pieces of equipment. What the boiler MAC'd does is it says, we're not only going to regulate the boiler, but we're going to regulate everything else, notwithstanding that we already have MAC standards for everything else all the way upstream and downstream, because it requires us to look at the architectural plans, the facility operation and maintenance laws, and the like. And when I say the Clean Air Act is elegant, the Clean Air Act in this regard is elegant because it required EPA to write those MAC standards for those individual sources. The Clean Air Act doesn't say regulate area sources, as EPA says in their brief. Regulate major sources. Regulate area sources. The Clean Air Act says regulate categories and subcategories, and that's what EPA has done. That's why when the cases come before you, when we have the copper smelter MAC'd, or when we have the cement MAC'd, or when we have all the various MAC cases that have come before you, it is not the area source MAC'd or the major MAC'd. It's the particular category or subcategory, and that's how EPA has always done it in the past, and that's not how they've done it here. That's not how they've done it here with the energy assessment requirement, and that's why I believe the energy assessment requirement is unlawful and should be vacated. Thank you. Thank you. Jason Morgan. Good morning. May it please the Court. My name is Jason Morgan. I represent the Alaska Miners Association, the Alaska Oil and Gas Association, and ConocoPhillips Alaska, and I'm here today to represent argument in the consolidated rule challenges for commercial and industrial solid waste incinerators. I'll represent an argument on the issue of small remote incinerators and variability, and also touching on the record-keeping issue addressed in our brief. I'd like to start with the variability issue that impacts small remote incinerators. The statutory rule impacts 106 known incinerators, and 28 of those incinerators are in remote locations in Alaska. These small incinerators essentially burn camp garbage, kitchen waste, sewage sludge, packaging materials, things associated with oil and gas operations in remote locations where there's no other practical or reasonable means for disposal of waste. The batches of these things burn waste in batches. In reading your brief, I didn't understand what you wanted the EPA to do. What should they have done differently? So EPA decided in this case that it was going to use lowest emissions to represent best performers. And I think with this argument, what we're saying is they don't really have the data necessary to determine which ones are actually the lowest emitting and which ones are actually the best performing units. So what we want EPA to do is go back and work with industry to figure out how do you determine, based on the batch nature of these units and the variability, how do you develop a testing protocol that can accurately determine what are the best performing units? If we look at what they did in this case, they sent out- How do you imagine that would work? Would they create new subcategories? It wouldn't have to be new subcategories. I don't know if this is clear in the briefing, but the XGO energy unit, the one that was burning pallets and oily rag, that's not all that it burns. It also burns sewage floods. It also burns food waste. It burns the same range of materials as incinerators. It wouldn't necessarily need to be subcategorizing. It's more about making sure that you're testing that incinerator when it's burning those other source categories. If you only test- So you're arguing they just didn't take a representative sample? I think, in essence, they didn't have representative samples. What we could do is work with the agency to figure out a testing protocol. There are some issues associated with the fact that these units do operate in batches. They burn what's on hand at the time that it's available, because the whole goal of these incinerators is to avoid accumulation of things like food waste that attracts polar bears and things that present a hazard to the operators, et cetera. There's a batch nature of these incinerators, and EPA didn't take that into account when it issued this information request. It said, give me three test points, three data points from end-of-stack tests, and do it under normal operating conditions. Have you presented data? Is there data in the record to show that, in fact, the data they're relying on is not representative? We did not. We submitted data showing that XTO Energy was burning a limited subset. We presented that data to EPA, and EPA didn't respond to that data. So I think, you know, if you think about this, if you're talking about – so they sent out an information request, a 114 information request, that said give us a representative sample, and XTO Energy did. It was representative then to be at one point in time burning just pallets and oily wrecks. That's how the unit works. And EPA got data back that was essentially, you know, for sulfur dioxide was essentially zero. The results were zero, zero, and 0.39 parts per million. And that then was used to set the floor for all the incinerators burning all kinds of waste material. And I think you can see, you know, the absurdity of what results from that. You look at J245, which is the table that shows a summary of all the subcategories and what their emission limits are for SO2. And at the bottom of that table there's a line that shows each one. The limit of 1.2 parts per million for small remote incinerators is between 10 times and 600 times more stringent than all of the other subcategories. These are small units that are operating in Alaska in remote conditions, and they do not employ end-of-stack control technologies. It's a required standard that's 10 times or 600 times more stringent than those that are. It just simply doesn't make any sense, and EPA didn't have data. I think you can also see in some sense how this feeds into the problem with the pollutant-by-pollutant approach, because when EPA sets the floor for the small remote incinerators based on XTO Energy's burning of pallets and says those who are burning kitchen garbage must meet that standard, I think you can see that it creates this sort of absurd scenario. You've got the only new incinerator that can come online is one that burns pallets, and that's not the point of the subcategory at all. EPA's response is that we gave them data too late, and I think that's not accurate. EPA did not use the XTO Energy data point, which is the one that I think really reveals the problem. They didn't put that in their proposed rule. They didn't have it in their data set available. It was only available when the final rule came out. They put XTO Energy in the final rule, and so we timely submitted comments showing that XTO Energy is not a bad data point, and we timely submitted a request for reconsideration on that and submitted comments subsequent to that, so I think we're timely. And as for the issue of variability and how the output prediction limit can account for variability, I think it's really hard to understand how you can do a statistical manipulation on XTO Energy burning wood pallets and figure out what that unit is going to be burning when it burns camp garbage or sewage sludge. It doesn't make any sense as a method for capturing the variability of these batch units. And, again, I think it feeds into the entire pollutant-by-pollutant argument when they do it on a pollutant-by-pollutant basis. I think maybe I'll briefly address the record-keeping issue. So the SISWI rule defines a SISWI unit as one that – to include one that burns waste other than fuels that have been discarded and does not keep certain records. Showing that that material is, in fact, not discarded. Basically, the rule says if you don't keep records showing your materials are not solid waste, then your materials are solid waste. And the result of that is that you become an incinerator. This is sort of an odd rule if you think about it. It's put into the rule that applies to incinerators. Incinerators are burning garbage. They don't need to be keeping records showing that they're not burning garbage. And this really applies to other units that aren't incinerators. And the result of the rule is really quite draconian in the sense that if you don't keep records showing that you're not burning garbage, you're not in a record-keeping violation where you're subject to EPA giving you a fine for not keeping records. You're turned into a SISWI, and you're stuck that way for six months. And you're subject to all kinds of violations as a result of being pulled into a category that you don't belong in. This is another one of those things where just as a practical matter, it's hard to understand exactly how this is working. I think you've answered the question that I have, which is my question was really, are there units which are not SISWI which would get caught up in this? And I think you've said yes. Yes. But that seems strange because I was assuming that these were all incinerators. But you're saying that's not the case. No, you can have a boiler that is using secondary materials that are defined under the RCRA rules as not solid waste. And so they're not an incinerator. And that unit, even if EPA has sent them a letter saying, yes, we understand your process. We agree it's not solid waste. They show up at the facility and say, well, what is your record showing it's not solid waste? And they say, well, we didn't keep records. You told us it wasn't solid waste. We all agree it's not solid waste. And they said, sorry. Now, you're a SISWI and you're stuck that way for six months. Okay. Thank you. May it please the Court, Douglas McWilliams again to address the CO work practice issue. EPA disregarded its statutory obligation to convene reconsideration for coal-fired boilers on the issue of whether to set a work practice for certain organic hemp, as EPA did for similar coal-fired utility boilers. My clients at American Municipal Power are municipal electric utilities who are subject to the Boilermact because they use generators of 25 megawatts or less, but in all other respects, they're similar to utilities. So we were very interested when EPA, just weeks after finalizing the Boilermact, promulgated or proposed utility mats with a work practice set for the organic hemp instead of the numeric standard that they had just recently set in the industrial boiler rule. Isn't this argument foreclosed by Sierra Club? In Sierra Club, didn't we say, if you're going to find error, you've got to find it in this rulemaking. You can't go to another rulemaking and argue by analogy. In a petition for reconsideration context, though, when new information arises within the context of 307D, we have an opportunity to bring that to the agency's attention, and each EPA has its statutory obligation to consider that if we meet the criteria. The criteria in 307D are that EPA shall convene a proceeding for reconsideration if we can demonstrate that the grounds for such objection arose after the period for public comment, which we can show here, because the grounds for the objection arose when the utility mats rule was proposed, and if such objection is of central relevance to the rule. Those are the only criteria we have to show to trigger a mandatory duty for the petition for reconsideration. At that point, EPA must provide us with an opportunity for additional notice and comment to complete its rulemaking process. How is the mats utility central to the relevance of this? What EPA did in the mats rule was relevant in two respects. One, they have a pilot plant, a pilot coal-fired boiler that they use to generate additional information. That boiler was identical. In fact, if that boiler were operating commercially, it would be subject to boiler mat. So it generated data that was directly relevant to determining if organic hazardous air pollutants could be emitted in excess of the detection limit or were otherwise reliably quantifiable. The second is that EPA applied a different methodology when they evaluated the data for utility mats than they did for boiler mat. For boiler mats and utility mats, the first step was the same. Determine what the below the detection limit level was. And in that regard, the two databases were identical. But EPA applied an additional layer of methodology to determine that there was unreliable data up to 89% in the utility mats rule. But that part of the methodology was never applied to industrial boilers. If it were applied, we would be in the same situation. So our objection arose when utility mats determined that there was a pilot plant that generated data that's relevant to our rulemaking and in applying a methodology that was inconsistent with the methodology they used for us. Similar sources treated differently created an arbitrary objection to the industrial boiler rule. So we timely filed a petition for reconsideration, which EPA summarily denied. We're now before this court to ask that you reverse that determination and direct EPA to grant the petition for reconsideration so that they may consider this data in the true light that it deserves. I'll reserve the rest of my time for rebuttal. Thank you. All right. Mr. Rosen for the EPA. Thank you, Your Honor. Perry Rosen from the Department of Justice for EPA this morning for all of the industry issues. I'd like to start with the pollutant by pollutant issue. Congress gave EPA one express mandate to establish emission standards for both new and existing sources that shall require the maximum degree of reduction in emissions of hazardous pollutants. And as this court said in cement kiln, what that means is that EPA is required to set the most stringent standards. And for florists, it's supposed to do that based on what's actually been achieved in practice by the best controlled sources. There is no dispute in this case that EPA has, in fact, set standards for each of the covered hazardous pollutants based on an actual source that has actually achieved that standard. And for the existing sources, it's based on actual sources that have achieved the standards for the best 12%. There's only one real issue on the pollutant by pollutant, and it's whether achieving practice refers to each pollutant or if it refers to a single unitary source and their emissions based on the entire suite of pollutants. And to put it more bluntly, the only issue is the statute unambiguously states that the unitary source approach has to be used rather than the pollutant by pollutant, which will result in the least common denominator, which will result in not the maximum reduction of the covered hazardous pollutants. Petitioners make a few arguments. What is the deal if you have mutually incompatible control technologies? First of all, we're not aware of any case that that situation has arisen. We say in our preamble that there is no case that that's been identified for these particular rules, for any of these rules. We've done this for about 150 subcategories of different types of burning facilities, and we're unaware of any conflicts. But I think Senator Durenberger answered that in the legislative history when he said that in such a case, and then EPA has some discretion to figure out what one source, how to rectify a technology which may cause an increase in emission in the others. But that's a pretty limited situation. In fact, Senator Durenberger's statement actually proves the fact that this needs to be determined by a pollutant by pollutant approach. Because if a conflict could not possibly arise, if you apply the unitary source approach, a conflict couldn't possibly arise, because you're only picking one source, and you're doing it based on a subjective view of what the best is on all pollutants. It's only when you apply each pollutant when a conflict can arises, and as Senator Durenberger said, there's a process for dealing with that. Is counsel correct that you could have a circumstance where a facility which perhaps meets the strictest standard on one pollutant would fail the emission standard on a different pollutant? Yes, and that's the problem. The purpose of this statute is to require sources, all sources, to meet the standards of the best performing operators. So we certainly expect that there will be sources that don't meet those standards, or the top 12%. The whole purpose of the statute is to require those other sources who cannot yet achieve it to achieve those standards that are actually being achieved. It does seem a little bit ironic that you may have a situation where there is no single source that can do it all, right? That can do it. There are very few situations where that's true. In almost all cases, there are existing sources that can do it all. But if the extra step has to be taken for a source that meets 90% of the covered pollutants but doesn't meet the last one, then it does have to take some action to meet that standard. That's what really fulfills Congress's requirement to make sure that the standards are based on the maximum levels actually achieved by existing sources. And there's a source out there that has achieved that level on that last 10%, on that last pollutant. And in that case, what we're saying and what Congress said is require those other sources to put in a baghouse, to put in some technology, which will then meet that standard. But if the max standard doesn't allow you to look at cost versus benefit, correct? That is right. Okay. So what you would be saying is if you – let's say you had a facility that could do everything but couldn't meet that standard as to one pollutant, and it was cost prohibitive to make that last step, would they be required then to close down even though they're 90% compliant? We don't think there's any evidence that it would be cost prohibitive if there's another facility that is meeting the standard for that particular pollutant. And this court has faced this issue before at least three times. At least three times in Cement Kiln, in Sierra Club, and in NACWP. The industry has come in and said you can't set the standard that low because we have sources that can't achieve those standards. And in each case, this court said that is irrelevant. That cannot be considered in considering the source, so much so that in NACWP case, they said that argument has been roundly rejected by this court. Achievability is not part of the analysis in establishing whether the floors are properly set. And that's all we're talking about except for the energy assessment, which I'll get to later. That's all we're talking about here is floors. And the way Congress said it is that EPA has to set at least as a minimum requirement what's been achieved. And in fact, if you look at how would this be applied, there's 189 hazardous air pollutants under Section 112. Did Congress really intend for EPA to go look at the emissions levels on 189 potential hazardous air pollutants and pick the one source that does the best on that? It's contrary to the legislative history, and it's contrary to the specific legislative history of the 1990 amendments. Because remember, prior to 1990, Congress said to EPA, you figure out what the hazardous pollutants are, and you figure out how to rank them and determine what's most important. And that didn't work. And Congress came in and took that discretion away from EPA. And they said, we're going to define the pollutants, and you are not allowed to rank those. If they meet whatever standard they meet, you set the standard at the technological level that those particular sources, that active sources, have actively achieved. In fact, in Sierra Club and again in White Stallion, the court very clearly said that that 1990 amendment said that EPA cannot pick and choose those pollutants. And that's exactly what would happen here. Take, for example, one of the hazardous pollutants is mercury. Another that represents a whole bunch of hazardous pollutants is carbon monoxide. So let's say we have a source that does great on carbon monoxide and is the lowest emitter, lowest emission. And EPA says, okay, that makes sense. I guess we'll choose that as the source, the one unitary source, because it covers a whole bunch of hazardous pollutants. But it doesn't do anything for mercury. It doesn't bother to pay any money to put any controls on mercury. So there is no emission limit on mercury. Mercury is as high as it can be. And EPA is put in the position of making the very valued judgments that Congress said it shouldn't make. It could literally result in standards set where the best level of breach on one pollutant is actually the worst level of breach on another pollutant. Congress expressly left this in the hands also of the administrator. It directed the administrator to set the maximum degree of reduction in emissions achieved in practice by the best controlled sources as determined by the administrator. As you know, EPA gets a lot of deference here. This is a complicated statute. This has been its longstanding interpretation since the 1990 amendments came out. And in this case, Congress expressly left the discretion to the administrator. Let me move on to malfunctions. Petition's argument is fundamentally flawed because EPA does consider malfunction periods and start up and shut down. I'll treat that a little differently. But mostly malfunctions here at the beginning. Just not in the one size fits all manner that petitioners would like. The first issue is it has to be factored into the numeric limits. Now, how does that work? Not only do we think that's impractical, petitioners said in their brief it's impractical. We don't know the frequency of the malfunctions. We don't know the duration. You deal with variability all the time. Yes. Variability. Why is that any different? Because there's no data. There's no data in this record, certainly, and there's no data otherwise that were available that can factor in. A pilot light goes out and maybe that emits some extra emissions for a half an hour. There's a boiler explosion. And maybe that emits all kinds of hazardous pollutants for months at a time. Why not a work practice rule? If the numeric data can't be obtained, why not a work practice rule? The work practice rule is, again, at the discretion of the administrator. So let's see what the administrator looks at. First of all, on a work practice rule, that's completely inapplicable to Section 7429. So the SISVI units can't use that. But let's look at the major and the area sources. The EPA is, again, they're looking for a single work practice standard that applies to malfunctions. So it's just as difficult to apply that to malfunctions as it is to almost as difficult to apply numeric standards. Again, how do we create a work practice standard that applies equally to the half hour pilot light or the boiler explosion or the bag house ripping or different level of emissions? I think Mr. Worms said that EPA admits that sometimes during malfunctions emissions increase. Well, sometimes they do and sometimes they don't. And how much they increase, we don't know that. There's no data in the record. No data has been presented by petitioners in the record. There's really no way to analyze that. So there's no reasonable way to factor that in. This court has said that we don't have to be exact in sending the standards, as Judge Griffith said, but it has to be based on data and it has to be based on a reasonable analysis of the data. And that data is completely absent here. If you could just take a moment to explain how this works. I think counsel said this question where you may exceed the emission limit because of some malfunction only comes up if there is an enforcement action. Is that the way the case-by-case determination would work? No. It can come up in enforcement action, but the case-by-case analysis is when an operator says, okay, here's our data for a certain period. And during this period we had a malfunction. And here's the circumstances of the malfunction. So EPA will look at that and determine whether on a case-by-case basis should we take any action? Should we just require them to take some certain procedures based on that equipment? Should we look at that particular piece of equipment, ask them to get a new piece of equipment? EPA is not going to go out and prosecute every case. I mean, when you think about it, what better approach? Think of malfunctions. Think of, you know, the biggest one you can think of. And I don't know how we define that. And under their theory, the standard, if malfunctions are included, would have to be based on the worst possible emissions. Because that's what we're talking about. We're talking about some malfunction, some horrible, horrific malfunction that's used as pollutants. And their sources, in their words, would not be able to achieve the standard without factoring in those horrible, enormous emissions. So the case-by-case approach, which is used in other contexts, as the Weyerhaeuser case is used in the Clean Water Act context, it says, you know, at some point with these types of events, case-by-case approach is the best. And that's all EPA is talking about. EPA tried other approaches. They tried exemptions. They tried, you know, other things. And this court struck them down. We recognize malfunctions can be different. We're trying to deal with them in the best way that we can. And case-by-case analysis seems really the most reasonable, and it's the one we can also apply to the SISV units. Because we can't apply a work practice standard to the SISV units. This is precluded by statute. But we can apply a case-by-case approach here. In the end, the question is, and in the case-by-case approach, EPA can require, as opposed to the work practice standard that was proposed by petitioners, it can require certain other things, like what's called a root cause analysis, where you just don't look at what happened in the emissions. You also look behind that. You say, well, why did that happen? What can be done about it? And that's what we think is the proper approach. But what EPA recognizes is a different situation. Help me to understand, why can't you get data on malfunctions? Why isn't that possible? You've got lots of authority from Congress to get data. Why can't you get that data? Presumably, EPA could probably promulgate a regulation requiring specific data on each malfunction, defining the malfunction. But that's not provided for at this point in the regulations. But there's no provision calling for that at this moment. That's EPA's fault, isn't it, if they haven't sought for that? Yeah, but again, we don't know how that would occur, how the emissions levels can be measured for each particular instance. And in the end, where does that get us? Do we define hundreds or thousands of different malfunction events, which leads to hundreds or thousands of different work practice standards? It becomes unmanageable to try and get this one-size-fits-all solution. The industry is concerned about leaving it to case-by-case enforcement. That's a tough way to go. How do you plan ahead of time when the enforcement authority is going to say, you know, no real standards here? Well, the best way to plan ahead of time, obviously, is to make sure you operate your facility in a way that doesn't result in malfunctions. Or if it does, it happens very seldom and with little effect. That's all you can give them, huh? Don't make a mistake. Well, I recognize we have made your life somewhat more difficult. Perhaps just having an exemption would have been a lot easier, but we've said that you couldn't do that. So I guess the question is, is there any way to collect data that would provide a pattern or an intelligible commonality among incidents that would allow you to establish some sort of baseline? As I stand here today, I can't say that's impossible. I can say that EPA has looked at that situation in very limited circumstances, such as a particular piece of equipment on a particular type of boiler or operator, and if they can get enough information about that particular piece of equipment, perhaps a work practice standard or something can be worked out like that, where there's enough of a pattern of malfunctions and what the levels of the emissions are that occur during those malfunctions. Maybe that's a possibility. In the end, though, again, this is the discretion of EPA, and really your issue here is by taking this approach on a case-by-case basis, just as in the warehouser situation, is that at least a reasonable approach of dealing with malfunctions after the other approaches have been struck down by this court? Let me turn to the energy assessment. This is the single beyond-the-floor standard that's before the court today in the industry arguments. Now, counsel says that cannot be a beyond-the-floor standard because you can't do any kind of cost assessment. What's your response to that? No, I think he said that it can't be a work practice standard because of that. We are not claiming it's a work practice standard. So whatever counsel said on that, you can believe him, because we're not claiming it's a work practice standard. This is a beyond-the-floor standard, and in one limited situation, it is a DAC standard for the area sources, which is close to the same thing. Now, again, this beyond-the-floor standard, again, Congress requires EPA to take these actions, so long as it looks at costs and energy effects and other issues. He says the administrator shall do this. This is just not willy-nilly at EPA's option. And that statute says look at the processes, the methods, the systems, the measures, the techniques to impose that type of standard. It doesn't limit it to numeric standards. It says you can look at the whole process. You should look at the whole process. You should look at the whole system. Now, this assessment— What's the limit on that? I think industry's argument is that you've gone too far here. You've gone beyond the scope. You're looking at things well beyond the process. So what is the limit? Well, all we did was, and this is in response to comments made on the proposed rule, we made sure we limited it to just those pieces of the source that are related directly to the boiler that we are looking at. So, for instance, we have a boiler at a university that heats the schoolroom portion of the university. And so part of an energy assessment might require the source to say, can we do something in those schoolrooms to cut the heat down during the night when nobody's there, and that will use less fuel and therefore emit less hazardous air pollutants. Those are the types of things that we're talking about. As long as it's directly related to the amount of emissions coming out of the boiler that we all agree we're talking about, we didn't say that you have to look at the whole facility and all of its operations just because we're regulating the boiler there. It has to be directly related to that boiler, and it has to reduce the volume of the emissions. So this assessment is very limited. It's not facility-wide. And it is just that. It's an assessment. It doesn't require implementation of any procedure. There are no numeric emission standards that are in this requirement or that are going to come out of this requirement. This is designed to let the marketplace decide, the regulated parties decide whether they want to actually implement anything. This is designed after similar studies that have been done under the auspices of the Energy Department, where entities have determined that they've cut their fuel costs 10% to 15% by going through this assessment. And with that, they cut their hazardous air pollutants. So the idea is if the source says, hey, you know what, I looked at this, and I can't cut my energy costs 15% and with the hazardous pollutants, but I'm really more concerned about cutting my energy costs. I'm going to go ahead and implement that. That's all EPA is asking you to do. Take a look at it. It's the ultimate common sense approach that won't even be implemented unless the facility itself says it will save us money. And Petitioner's argument is sort of upside down. They basically say what's okay is if EPA gets some studies, looks at analysis, and says, okay, here it is. We're going to require everybody to put in the following procedures and operations because we've determined overall at the different facilities we've looked at that it's going to reduce some emissions. We don't know if it's going to reduce your costs, but it's going to reduce emissions, so we're requiring that. We're going to mandate that. That, they say, would be okay. What's not okay is if we leave it up to the facility to say, A, let me look at my own facility and see what makes sense, and then, B, let me make my own monetary determination as to whether I want to Again, it's sort of the ultimate marketplace provision. It's the type of beyond-the-floor standard that Congress contemplated. As to looking beyond the boiler itself, Council makes an argument that because it's a subcategory, we can't go look at those classrooms. But that's not accurate. Clearly, the statute allows EPA to set categories and subcategories, and as I think Judge Williams explained, the whole purpose of that is to ensure that in setting numeric emissions standards, you're not requiring whole different types of sources to meet the standards met by other sources. You're not requiring the small remote incinerators to meet the same standards that the large boilers meet. So it tries to, in a fair way, say when we gather data from a group, we're going to try and make it as close to the same group as possible. But just by setting those categories doesn't limit what's regulated. The statute regulates the sources. The statute doesn't regulate the categories. The statute regulates the sources. 7412D2 says it regulates new or existing sources. The source is defined as the entire facility. The source is defined as stationary sources, and it refers to 42 U.S.C. 7411 to define stationary source, and that's any building structure, facility, or installation. And so that is what allows EPA to regulate the source. But again, in this case, EPA has targeted it very specifically to only those other portions of the sources that can have some effect on the emissions. Clearly, in our view, Congress's intent by setting up categories wasn't based on a statute that requires the maximum reduction of emissions. They didn't set up the categories to hem EPA in. They set up the categories to make those emissions fair, but they didn't set up those categories to say you're limited to those categories. As far as a few of the other arguments raised the cost, EPA did look at the cost. EPA did analyze the cost. They said the cost of this emission is $2,500 to $50,000. $2,500 to $50,000, that's it. And they also found they don't have to make a cost-effective determination. I think the Sierra Club said that. They just have to look at NRDC versus EPA, the 2014 case. They said they can look at cost-benefit analysis, or they can just do the general review of the cost. And they did that, and they looked at those energy department studies and said in all of those cases, in the end, when things were implemented, that it was cost-effective because, again, the operator is the one that makes that decision. Let me turn to the record-keeping requirement. Quickly, it wasn't raised this morning, but on startup and shutdown, which was raised in the Siskiyou case, we believe that argument is waived because it wasn't discussed at all. And in the end, the data in the record shows that emission levels will actually be lower during startup and shutdown than the standards that are set, so there is no reason to establish standards during that time period. And, again, for our CISWE units, we can't create a work practice standard for them. On the record-keeping requirement, we think this court has already decided that in the companion case, in the coordinated case that the court decided in June, that was coordinated with these four cases, the NHSN case, the secondary materials case, which helped define which category is which, which you receive, I think, briefs on, pursuant to the order. But we think the opinion in that case decides this issue. 7429 regulates solid waste, and that's defined as any garbage, refuse, or other discarded material. So if you incinerate discarded material, you are subject to 7429, not 7412. Secondary materials are defined in the regulations as any material that's not the primary product of manufacturing. So under 40 CFR 241.3a, unless specifically excluded, non-hazardous secondary materials that are combusted are solid waste. That's what it says. This court in the ECHO services case, in the coordinated case, said by default, the rule classifies secondary material as solid waste unless otherwise excluded. Now EPA has set up a procedure where you can take secondary materials, and if you show that you comply with the legitimacy requirements, and it's managed as a valuable commodity, and it has meaningful heating characteristics, and then its contaminant levels are at least as low as what the replacement fuel would be, you can qualify as secondary materials that would not be considered solid waste. But that's not an entitlement. That's a showing that has to be made. Again, ECHO services court said a party must satisfy the legitimacy requirements to qualify for this exclusion from being regulated as solid waste. Now, how do you qualify? You keep records showing that you meet those legitimacy requirements. And absent the records, EPA assumes that the discarded material is secondary material, and that's the problem. That's the problem that they raise, because we assume that if you don't have the records, that's it. But that's all they have to do. That's all they have to do is keep records. We're not talking about an evidentiary hearing. We're not talking about presentations made to EPA to show that their secondary material is not solid waste. That's all they have to do. This is not a presumption, as I understand it. It's basically a definition. If you don't keep records, you are a SISLI. Is that correct? That's the way EPA enforces many parts of the Clean Air Act. And they have broad authority, I think it's 7413, which lays out that broad authority. And EPA has been admonished for not using that authority to enforce the actual requirements of the Clean Air Act. But, again, these materials are secondary materials that come under 7429, absent some showing from the operator. Here's again, one more time, from Echo Services and what the court said. Quote, in deciding which materials to regulate under RCRA, Congress wanted to err on the side of caution, citing the American Chemistry Council. Thus, EPA may place the burden upon the regulated entity to show that it is material that should not be regulated. EPA is well within its statutory authority to assume that transferred material is solid waste until an interested party demonstrates that the material has not been discarded and is indistinguishable in all relevant aspects from the fuel product. That's exactly what this regulation does. Again, it is not much of a burden, but it makes that presumption and it says you have to prove otherwise. And part of that presumption is if you don't have the records, then we're going to assume, as this court has said, we can and must say that that subject is 7429. Let me turn to the small remote incinerators. EPA understands that this is a different type of category of incinerator. So what did it do? Here, it created a special category for what is, at least right now, 28 incinerators to make sure that the data collected was just from this small group and it wouldn't be competing against other types of units. Like larger incinerators or like less remote incinerators or like kilns or something like that. And then it refined that category in response to industry's comments. And then it asked for the data from these units because that's how the standards are created. Give us the data of your operations and we'll create a standard based on that. And they asked for data specifically on waste streams. And they asked in the information collection request in 2008. And they asked again in the proposed rule in 2010. And in issuing the final rule, EPA explained we got no data. We received no data, meaningful data. Part of the argument is that the data that you did get wasn't representative of the types of burning that's going on there. How do you respond to that? Well, whose fault is that? That's the fault of the industry that didn't give us the representative data. All EPA can do is say please give us representative data, do your stat test, and give us the data that represents your operations, your complete normal operations. And that's the data that we received. And that's the only data upon which EPA can base its standards. They can't guess that there's other data out there. What do you make of their point, at least in their briefing, that relying on waste segregation for these SRIs isn't really practical? I mean, these are out in the remote Alaskan wilderness. They're in the landfills. Well, first of all, we're required to rely on waste segregation data. That's in the NACWA case and that's in the cement company case. It can't be based just on technology or other things, and waste segregation is a key part of the whole process. In fact, the statute says... How are they supposed to effectuate waste segregation in the landfills in the Alaskan wilderness? What are they supposed to do? Well, there's no evidence in the record that other parties can't do it. This is anecdotal evidence, anecdotal discussion, that we have problems with polar bears and things like that. We don't know what the effect of that is. We don't know if they can't meet standards based on that. We don't know if there's not enough waste that can be segregated, other than food products or food waste or things like that. In fact, I think they really only focus on one standard, SO2, and only for new sources based on this XTO facility. They don't appear to have any complaint with any of the existing sources. They don't appear to have any complaint with all of the other pollutants. I believe it's just SO2 for this one facility. Based on that, they're trying to overturn the entire set of standards and regulations for small, remote incinerators. This is sort of basic administrative law. You ask for the information, and you ask again, and you ask again, and you can only base it on the information and data that you get, not on anecdotal stories that, in this one instance, this doesn't seem to make sense. Now, they can come back to us. They can petition for a new rulemaking, but they've got to give us the data to make that determination. And that's still absent from the record. There's a story about these guys maybe burned a different type of waste, and on the day that they did the test, they didn't burn the same type of waste, and that affects us. It doesn't seem to affect us on existing sources. They're not asking for existing source limits to be overturned, just for right now non-existing sources and just for that one standard. When you say you ask for data, I thought, I heard counsel say, that the way the stack test is conducted is sort of a spot test. You do it at a particular time or something like that. How does it actually work? I mean, can they define how they do the test or how long it lasts? It's supposed to represent their normal recurring operations, and there's a wide variety. There are lots of incinerators that burn the same type of waste that they're talking about, food waste and things like that. These are not kilns. These are not boilers. These are incinerators that incinerate what we all seem to call general refuse, and those other incinerators seem to meet the standards, and there are other ways to meet it. It's not just waste segregation. There are control equipment that can be put on these types of facilities. There are afterburners that can be used, and maybe that's the case for a facility that can't meet the same standard that XTO met. But, again, as of today, that's for a non-existent source because they make no claim that the standards for all existing sources are inadequate. Let me turn to what I think is the final issue, if I'm keeping count right, is the carbon monoxide surrogate, and this is not a surrogate issue that the environmental people raised. This is just whether a work practice standard is required in lieu of the carbon monoxide emission standards that were set, and, again, this is a work practice standard that they're claiming. A work practice standard requires certain things. First of all, it requires emission standards to be infeasible, unmeasurable, and that's a determination to be made by the administrator. It's also only applied in the judgment of the administrator. It's also only applied with the bias that numeric standards are the ones that are to be applied. And, again, it's under the general parameter, the general directive, that EPAs require to set the most stringent standards that are available. The real argument is that we did something different in the utility rule than we did here, and Judge Griffith, we agree by Sierra Club and other cases say that's foreclosed. Each standard stands on its own, and the fact that a different standard was set in a case that has different types of facilities and different types of energy generators is completely irrelevant here. But let's look at that. In the utility case, EPA found that between 57% and 89% of the test results were useless because they were either all undetectable or at least some of the results were below undetectable levels. So it met the feasibility requirements. Is this argument in case I'm not reading Sierra Club right? Is that what you're saying? I'm sorry? Is the argument you're making now in case I'm not reading Sierra Club right? Yes, exactly. So EPA made their technical determination that that was just too many non-detect levels. We could not set numeric standards, again, based on the cases of this court. We don't have to be exact, but it has to be based on reasonable data, and it could not do that. In this case, the opposite was true. There were some non-detect levels, but at least two-thirds were present, and EPA made its determination that that was enough upon which to base the numeric standards. Now, in the reply brief, petitioners come back and say, well, that's because we applied apples to oranges, and we need to apply apples to apples. Let's look at the hazardous pollutants that were common to both rules. But their explanation is both irrelevant and wrong. It's irrelevant because in their chart on page 20, they go through their analysis. And what they say is that, in fact, if you look at the non-detect levels in the utility rule, they were about the same as the boiler rule. They weren't that high. Well, if that's an issue, it's wrong. The data they're looking at is wrong. But even if it were right, that would be a problem with the utility rule. That wouldn't be a problem with this rule because they don't say that the non-detect levels in this rule were any higher than we said they were. And EPA made the determination that two-thirds worth of the detect levels is good enough data upon which to base the standards. And petitioners don't contest that in their reply brief. It's also, as I said, wrong because, in fact, there are three data standards that EPA used. One is a below detect level. The second one is a detection level limited. And in the utility rule, EPA said that we don't consider either of those categories good data. In this case, there's no such similar thing. The non-detect level, the level of test that EPA determined it couldn't use is still about one-third. And when you actually compare what they say are apples to apples, the four common hazardous pollutants, looking at the below detect level and the detection level limited standard that they used in the utility rule to say we can't use the data, it came out to about 73%. So, as I said, not only is it irrelevant, it's wrong. That level of non-detect levels in the utility rule is still too high to use anything. That is not the case here. Work practice standard is the exception. It's not the rule. There's the strong bias that this only be done if EPA, in its discretion, makes the determination that is infeasible and then further, in its discretion, makes the determination that this is an appropriate case for work practice standard. And given all the deference on a very technical issue like that, it can't be said that EPA's determination was unreasonable. And based on all those reasons, we think the petitioner's petition should be denied. Thank you. Thank you. Thank you. Mr. Pugh for the environmental intervenors. Good morning. Jim Pugh for environmental intervenors. I'd like to make just a few points on three of the issues covered in the industry arguments. One of those is the argument that standards should be set at malfunctions or to accommodate malfunctions. The other, or the next, is the argument that EPA should set standards based on some imaginary unitary source, for example, and the third is on the carbon dioxide of the sewerage system. Very briefly on the malfunctions argument, EPA, industry's entire argument, it's in their brief and even just now, rests on this remarkable claim that EPA has to demonstrate that its floor standards are achievable. That's what counsel for industry petitioner said just now. It's in their brief at 35 and 42, and it is simply untrue. This court has held over and over again that floors have to reflect what the best sources actually achieve, not what industry thinks or even EPA thinks is achievable, and there's no obligation to demonstrate that the standards are achievable during malfunctions or any other time. That's simply the holding of this court. The other point I would like to make, and by the way, that statutory argument is the whole basis for their claiming that EPA's alleged refusal to accommodate malfunctions is unlawful. It's that they didn't demonstrate that the standards were achievable during malfunctions. There is no obligation to make that demonstration. Their backup argument is that standards have to reflect what sources achieve, always have to reflect what sources achieve under the worst perceivable circumstances, which they read to be malfunctions. They neglect to mention that this court held just in MACRA that EPA is not wedded to that interpretation. That's at page 144 of the MACRA decision. In fact, as the MACRA court pointed out, that's actually dicta that refers only to new source standards. There's no holding that EPA has to set standards at a level that will accommodate the worst perceivable circumstances, even if those were included malfunctions, which they don't. And those are the only points I'd like to make on the malfunction argument. On the pollutant by pollutant argument, I have even less to say, really. I agree largely with what EPA says. The only difference that we have with EPA is that we contend that the statute actually requires EPA to set the course the way it did on a pollutant by pollutant basis. I don't plan to go into our arguments on that point. They're all in our briefs. The only point I would like to make this morning is that the court doesn't need to decide whether the statute requires EPA's pollutant by pollutant approach or just permits it, and we would ask the court not to decide that issue because it's not properly teed up here. And finally, moving on to the carbon monoxide as a surrogate issue, I realize that industry officiators didn't emphasize their claims about the inadequacy of carbon monoxide as a surrogate, but in their brief, their argument is that Matt's rule shows that carbon monoxide isn't a valid surrogate for organic hazardous air pollutants. And they claim, in that case, EPA, or in that rulemaking, EPA determined that there is simply no meaningful relationship between carbon monoxide and organic hazardous air pollutants. They claim that the same problem is here. Now, we don't, we're not interested in trying to rely on the record for another rulemaking, but to the extent that EPA contends that carbon monoxide is not a valid surrogate, we agree with them. Where we disagree with them is where that, what the significance of EPA's inability to link carbon monoxide and hazardous air pollutants takes the agency. The agency can only set, they contend that means EPA has to set work practice standards. The statute says that EPA may only set work practice standards if it is infeasible to set numerical emission standards for a hazardous air pollutant. Carbon monoxide is not a hazardous air pollutant. So even if it were infeasible for EPA to set numerical emission standards for carbon monoxide, EPA's statutory obligation to set numeric emission standards for the underlying hazardous air pollutants would still be there. I have no further remarks. All right. Mr. Gormley? Good morning, and may it please the Court. I'd just like to make a quick, a couple of quick points on the record-keeping issue in Case 11-1125, the SESWI case. We agree with EPA that the agency was within its authority to require SESWI to keep records. But I'd like to specifically address the record-keeping language in the SESWI definition. First, I'd just like to point out that this is not a sweeping requirement that all units keep records or be considered SESWI. It's narrowly limited to units that are burning nontraditional fuels that are discarded, but that nevertheless wish to take advantage of a special dispensation that was provided for in the waste definition, whereby they can still not be considered waste burners. So it's only in those narrow circumstances that EPA has allocated the burden of proof to the operator to have records to show that they're meeting the terms of that special allowance. Second, industry suggests that EPA was required to put that record-keeping language in the waste definition rather than in the SESWI definition. And we agree with EPA that that argument is in serious tension with this Court's decision in the related case, Ecoservices. In that case, environmental petitioners challenged the waste definition's lack of procedural safeguards, and EPA relied on the Clean Air Act rule's procedural safeguards. So evidently, the Court, in rejecting the challenge there, determined that it's proper for EPA to rely on the provisions of the Clean Air Act rules, such as the SESWI rule, in order to effectuate the waste definition. Thank you. All right. Let's see, Mr. Wareham, you reserved five minutes, right? Yes, Your Honor. And we intend to share the five minutes, as among my colleagues, with permission of the Court. All right. I was going to let, I think everybody else used his time except Mr. McWilliams. I was going to let the other three lawyers have two minutes each. But that would be great, Your Honor. Okay. All right. Okay. But I will be quick. So, again, Bill Wareham for industry petitioners on the malfunction issue. Just a couple of points in response to the government's arguments and Mr. Hughes' arguments. First of all, I predicted it and you heard it. The chaos theory came through in the government's argument. And there are a couple of things to say about that beyond what I said before. First, EPA, in part as a result of the cases we talked about before, Portland Cement and Wine, the Essex chemical case in the national line, has had provision for malfunction in its standards for 40 years and continues to have provision for malfunction in its standards. And chaos has not ensued. So, EPA has found a way to deal with this. And the programs have worked effectively. And I think that's a very important point in history. It teaches us a big lesson here. The second thing about we're dealing with something that's unknowable. And I addressed this before. But what we're dealing with are boilers. And we know what they are. We know how they operate. And we know how they break. And if, for instance, EPA did a survey of all this in this room and said, tell me about the maintenance history of your car in the last five minutes, you'd know what it is. For me, I had a broken fan belt a couple years ago and a couple flat tires and a couple other things. I can make a list. And then EPA can look at the list and say, okay, well, this looks like the world of malfunctions. And now I can do something about it. So, we're not dealing with the great unknown. We're dealing with the finite world of sources that everybody understands and break down in finite ways if EPA would do the inquiry it's supposed to do under the law to set standards that are achievable for these periods. A brief point on Section 129 work practices, which I didn't address before. The government did address quickly. As we say in our brief, but it's important to emphasize, the statute makes it abundantly clear time and time again that standards established under Section 7429 are also standards established under Section 7411, the new source performance standard provision. And Section 7411 expressly allows for work practices. So, we believe the connection that Congress intentionally made between 7429 and 7411 provides a clear pathway and clear authority for work practices under Section 129 standards. The last thing I'd like to say, I didn't mention remedy earlier. This is an important issue for us, but we do not seek for all of the emission standards to be vacated. What we've asked for in our brief, and I'd like to emphasize, is that we would like the standards vacated only as they apply to periods of malfunction. Because the problem is the standards were developed for periods of normal operation, and by and large we're not contesting that. So, we're contesting the standards as they apply to periods of malfunction. Thank you very much. Mr. McWilliams, if you would address both the pollutant-by-pollutant and the carbon monoxide in two minutes. I'd be glad to, Your Honor. With regards to the pollutant-by-pollutant issue and Judge Griffith's question as to whether there are sources that face this sort of technological barriers to simultaneous achievability, I refer you to page 30 and footnote 10 of the industry petitioner's brief, identifying a top-performing source for particulate matter and HCL that is the worst performer for CO and mercury, identifying a potential technological problem for meeting both, and certainly an issue where cost could be a substantial concern. EPA seems to be particularly concerned about adding cost to this analysis. If we establish simultaneous achievability, a source that actually meets the standard, but EPA wants to do more, in fact, EPA wants to do more because its pollutant-by-pollutant analysis indicates that it may be feasible to do more, achievable to do more. They can go there, but they need to consider cost first, and considering cost under the 112D2 analysis is not the burden that EPA suggests it might be. Now, if they do consider simultaneous achievability in the context of resetting the floor, it does not require that they consider health or risk-based standards. All it does is require that they establish a ranking system and attribute to each best-performing source the appropriate rank and then determine the best-performing sources of all the best-performing sources based on that rank. It doesn't require that they do anything to step into the risk analysis. With regards to CO work practices, EPA misses the fundamental point here, and that is that we're asking for a petition for reconsideration to be granted because there's new information timely available that's of central relevance to the determination before this Court as to whether it's appropriate to set a numeric standard or a work practice for the coal-fired sources for those organic haps. Thank you, Your Honor. All right. Mr. Friedland, I wish you would address something that Mr. Rosen said. When you were arguing before, I thought it sounded like you were arguing for more regulation rather than less, and he seems to have picked it up. We're not arguing for more regulation. The energy assessment should be vacated. If EPA wants to come back and do a standard and it relates that standard to the category or source category, to boilers, and if they're doing a work practice standard, they do 112H for work practice, and if they're doing beyond the floor, they do all the things they're required to do for beyond the floor, et cetera, all those sorts of things, then industry would certainly look at that and comment on that. But the current standard, the standard we're looking at, is a study that has to be vacated for all the reasons that I've described earlier. I'd like to just make three brief points. First, counsel says that the energy assessment is not a work practice standard. So it really doesn't matter what counsel tells us today. It matters what it is, and the energy assessment is not a number. It is not a numeric emission limit. It is a work practice standard, and even EPA believes that because if we look to the Federal Register notice for the area source, EA, it's contained the area source energy assessment is contained in Table 2. Table 2 is called work practice standards, emission reduction measures and management practices. Table 1 is called emission limitations. So the energy assessment is clearly a work practice standard, and you don't have to just take my word for it. The environmental petitioner is in their brief at page 15, quote, to be sure non-numeric MAC standards like energy assessments must typically be promulgated as work practice standards. Now they go on to have an argument about that, but they say non-numeric MAC standards like energy assessments must typically be promulgated as work practice standards. So it is a work practice standard. Secondly, counsel says that the agency has broad authority to use measures, processes, methods, systems, et cetera. This is 112D2. That's correct. They do have broad authority to use measures, processes, methods, or systems, but what counsel doesn't quote is the beginning of that sentence in 112D2 where it says that determines is achievable for new or existing sources in the category or subcategory to which such emission standard applies, which is back to the argument that I, the fundamental argument we made that you have to create a category and then you have to regulate that category. 112C create the category, 112D regulate that particular source category. And finally, counsel argues that the industry overestimates the breadth of the energy assessment requirement. Counsel says that it was tailored as narrowly as could be possible. Again, if we go back to the language of the rule, the actual language of the rule, as I mentioned in the opening presentation, it requires a review of, for example, architectural and engineering plans, facility operation and maintenance procedures, et cetera. It goes far beyond the actual boiler to require the energy assessment of all or a manufacturing facility, all manufacturing operations at that facility, not just the boiler. So it is really not narrowly tailored. Thank you. Thank you, Your Honors. Just a couple of brief points. Counsel for EPA says lots of incinerators that burn the same kind of waste are able to meet their standards, so what's the problem with these units? Again, I would direct you back to GA 245 where we look at the table for SO2. They can meet their standards for SO2 because their standards are between 10 times and 600 times more lenient than the standards that are imposed on smart remote incinerators. That's part and parcel of the problem. We have more stringent standards for units that actually have no end of stack controls. And as to that point about, you know, counsel's point that there's things that we can do end Afterburners are not an end of stack control technology. They are part and parcel of some incinerators mechanism for combustion. Some combust all in the primary chambers. Some combust in the afterburners. It's not a control technology. The control technology that's available for SO2 is a wet scrubber. And EPA concedes in the record that wet scrubbers don't work in Alaska. You can't have a wet scrubber at, you know, minus 40 below. It's just not going to happen. So is there a reason why there was not more data available? I mean, counsel says if we had more data, we would have looked at it. Well, I think you have to take it in the context of the time. These sources had never been regulated before 2010. They never had done a stack test on these units before. So EPA found a request. The operators did the best that they could under the circumstances. And the initial standards that came out in the proposed rule looked somewhat reasonable. It was only when the final rule came out and we saw that the XGO energy unit was way off that we realized that there was a serious problem with the data. And so we did what we could. And I know EPA says whose fault is that? Well, EPA is the one that has a burden of showing that their estimate of best performance is reasonable. And they have to do that based on the record. And they haven't done that based on the record that they have in front of them. And last point is that they say there's no argument that we can't meet these standards. I mean, that argument is in our brief that we can't meet the standards on pages 24 and 25 of our brief. So I direct you back to that. Thank you. All right. We'll take your break.

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A14485. Public record. Not legal advice.