# Oral argument — EPA v. EME Homer City Generation, L.P.

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/argument%3A5648

## Record

- **Collection:** Oral argument
- **Document type:** Oral argument
- **Published:** December 10, 2013

## Text

We'll hear argument first this morning in Case 12-1182, Environmental Protection Agency v. EME Homer City Generation, and a consolidated case, American Lung Association v. EME Homer City Generation. Mr. Stewart? Mr. Chief Justice, and may it please the Court, in promulgating the transport rule, EPA sought to protect the public health and to strike a fair balance between the competing interests of upwind and downwind states. EPA's analysis proceeded in three basic steps. First, EPA performed a screening analysis to determine which upwind states would be covered by the transport rule. And in order to do that, EPA first identified the downwind receptors that were in a state of nonattainment or had maintenance difficulties, and then it determined which upwind states were linked to those receptors. And in order to be linked to a downwind receptor, the upwind state had to contribute one percent or more of the relevant National Ambient Air Quality Standard, or NAAQS, to that downwind receptor. And any state that didn't contribute at least one percent to any of the relevant downwind receptors was determined not to contribute significantly to nonattainment at that area. Second, once the states that were to be covered by the transport rule had been identified, EPA set a state emissions budget for each state. And to do that, it performed computer modeling to determine, in addition to whatever emission control efforts were already going on, what additional emission reductions could be achieved by implementation of control measures available at various cost thresholds. And the thresholds ultimately selected were, for NAAQS, $500 per ton. For SO2, the Group 1 states were at a level of $2,300 per ton. The Group 2 states were $500 per ton. And the idea was, let's see what emissions savings we can achieve if additional control measures are implemented up to those cost thresholds. Those savings would not be evenly distributed among the upwind states, right? So some upwind states that are able to make those efficient changes will be carrying more than their burden of reducing the emissions that affect downwind states, right? Well, there were two bases for distinguishing among the states. The first, in terms of the $500 per ton threshold for the Group 2 states versus the $2,300 per ton threshold, the way in which states were divided into those categories is that the states that were linked to the downwind receptors that had the most severe pollution problems were treated as Group 1 states, and they were required to make greater pollution control efforts because they had some responsibility for the most serious problems. Now, I guess the point of your question would go to the fact that even among states that were operating under constant cost control thresholds, a state that had already implemented cost measures up to that limit might have to do less, in a sense, because it would have already taken the steps that were required, at least as compared to an air quality only threshold. Well, I don't mind a state doing less. I think North Carolina said that you can use those cost figures to do less, and that's not challenged here. But what the application of the cost factor means is that some states that can more efficiently make the changes will be required to do more than merely account for their proportion of the downwind harm. Isn't that true? Well, yes or no. I mean, I think it's an easy yes or no answer. I think it is the case that if you adopted an air quality only threshold, then it would be more likely to be the case that states that had already done a lot to control air pollution would have to take additional steps, even if it was done in a non-cost effective way. Have you answered my question? Does the fact that you begin with what the statute says is that each upwind state has to account for its effect on the downwind states, but once having identified that effect, you then say those upwind states that can make the reductions more efficiently have to make more reductions than their mere proportion of the harm requires. Isn't that so? I think it would be the case that at least as compared to at least some air quality only measures, the use of cost would have the effect of distributing the burden in a somewhat different way than it would have if you considered air quality factors. Is the idea, Mr. Stewart, that the states that are required to do more are the states that haven't done much already? That's correct, and that was what I was trying to get out earlier, that if states have to do less in order to meet the $500, in order to be in a position where they've implemented all the emission control measures that are available at $500 per ton, if a particular state has to do less in order to achieve that, it's probably because that state has already implemented most of those measures on its own. And what provision of the statute allows you to take that into account? I mean, as opposed to each state, whether it's inefficient or efficient, has to merely reduce its contribution to the downwind state pollution, right? That's what the statute says. Well, the statute says that each state will adopt measures that prevent sources within its borders from contributing significantly to downwind nonattainment. And the purpose of the provision is not to allocate blame for an existing state of nonattainment or for prior pollution. It's to devise a scheme that, going forward, will prevent nonattainment from occurring. And the idea is, if each state lives up to its obligation, and if the downwind states make commensurate commitments, then the problem will be solved. And in terms of the language, contribute significantly. I think there are various reasons to think that EPA reasonably construed that term to include a component of difficulty of achievement. That is, in common parlance, we might say that dunking a basketball is a more significant achievement for somebody who is 5 feet 10 than for somebody who is 6 feet 10. We might say that a $100 charitable contribution is more significant if it's made by a person who makes $10,000 a year than a $1,000 contribution by somebody who makes a million dollars a year. I was just going to say, that's just because, in the latter case, because contribution happens to be used in both an affirmative and a negative sense. The question is, for example, whether somebody who fatally stabs somebody and someone who fatally shoots them have each significantly contributed to the bad result. Or not significantly contributed. Contributed in varying degrees. I would say, if you cause death by alternative means, then both people would have contributed as significantly. But to set out a hypothetical that involves contribution to a bad result, if you had a basketball team that lost a game by one point and the coach was asked to pinpoint the plays that contributed significantly to the defeat, the coach would be much more likely to identify a missed layup or a turnover than the missed half-court shot at the buzzer. It's true that the missed half-court shot at the buzzer would, in one sense, contribute significantly in that it was a but-for cause. If the shot had been made, the outcome would have been different. But if you're talking about significant contributions to a bad result, you're more likely to focus on errors that could and should have been avoided, not simply the failure to accomplish something that's extraordinarily difficult. Part of your answer to Justice Kagan's question and Justice Scalia's question is that it depends on the time point, the time at which you measure. That is to say, if you take a look at a state which for five years has been trying to ameliorate pollution, you can measure it from the point five years ago. And if you do that, then they're not having to contribute more. Or don't you like that answer? I don't quite want to go there. I think there's a kernel of truth in there in that. But the point at which the state's good neighbor obligation is triggered is by the promulgation of a new national ambient air quality standard. And the state is required within three years of the promulgation of the NAAQS to promulgate a state plan that includes good neighbor provisions for the particular matter. How far back do you go for the relevant NAAQS? 2006? In this case, there are two NAAQS that were promulgated in 1997. 1997. One of them for annual particulate matter and one of them for ozone. And then the 2006 NAAQS was for 24-hour particulate matter, which is harder to achieve. And so when we are asking what are the states supposed to do as of the time that the new NAAQS is promulgated, the states don't exactly get credit for what they have done in the past. That is, they can't do less than they are supposed to do in the future simply because they have done a lot in prior years to prevent pollution. But the fact that sources within the state have in the past installed various pollution control devices or are using cleaner fuels, that may make it easier for them to prevent significant contributions to downwind nonattainment going forward. Can I ask a question following up on Justice Scalia about the statutory language and how you read it? I think, you know, most people, everybody thinks that it's better to regulate with attention to costs than to regulate without attention to costs. We have this, our Trucking Association decision, where we said, well, notwithstanding that everybody agrees that regulating with attention to cost is better, when Congress says the opposite, we have to go with the opposite. And there we said Congress had said the opposite because it had talked about protecting the public health with an adequate margin of safety. Now, I'm wondering, what does it take in a statute to make us say, look, Congress has demanded that the regulation here occur without any attention to costs? In other words, essentially, Congress has demanded that the regulation occur in a fundamentally silly way. Well, I mean, in the case of the NAAQS, I think it was not the case that requiring EPA to establish the NAAQS without reference to cost would cause a silly result. That is, the ambient air quality standards were supposed to be set based on public health criteria. And the Court, in the same case, in American Trucking, said that of course you can consider costs in deciding what is the most efficient and appropriate way to implement those NAAQS. And here I take your point that in order to conclude that Congress barred consideration of costs at the implementation stage, we would have to have very clear language. And significant contribution doesn't do it. And the other thing I would say, in addition to the examples I've given of in common parlance we use significance to refer to ease or difficulty of achievement, it's worth emphasizing that this is a provision of law and it's designed to help allocate the responsibility among different actors for alleviating a shared problem. And, for example, supposing... The problem is that that allocation among different actors is done state by state. And simply taking costs into account as determining who will do what simply eliminates the requirement that each state not be required to do more than its share of the pollution it's causing downstream. It's the state by state requirement that makes it very difficult to think that all Congress wanted was the most efficient reduction of pollution, no matter where that pollution came from. That's simply not what the statute envisions. I guess... And maybe that would be a better statute. Maybe it shouldn't be state by state. I mean, the first thing I would say is we can accept the premise that each state should alleviate no more than its share, and there still may be... That each state should do more than its share, and yet there still may be different ways of determining what a state's fair share is. That is, one way would be to determine which states had been the greatest polluters in the past and say that the more pollution that had previously flowed from your borders, the greater your reduction obligation in the future. But another way would be to say, in order to ensure that each of the states that have shared responsibility for the problem in the past bears its fair share, we will ask each state to undertake commensurate efforts as measured by the cost threshold. For example, if it could be shown somehow that the generation of electric power inherently required the emission of some level of SO2 and NOx, that there was simply no way to generate electricity through any technology known today without emitting that minimum amount, I think we would certainly say, well, Congress didn't intend to bury in the good neighbor provision some prohibition against particular states generating electricity. And EPA or the states could reasonably determine that the unavoidable component of the emissions, the part that couldn't be avoided even with the best possible pollution control technology, that would be regarded as legally insignificant, that the only legally significant contribution would be contribution that could have been avoided. Now, clearly EPA has gone one step farther because it hasn't just focused on emissions that couldn't be avoided at all, at least without foregoing the generation of electric power. It has said we will treat as legally significant only the extra increment of emissions that comes after you've taken what we regard to be equitable and cost effective pollution. Just one more question on cost. In your answer to Justice Kagan's question, there is at least a possible argument that you, the regulator, the government, the EPA can take cost into account unless it's expressly prohibited from doing so. You don't go that far. You even stop short of that. You say that it might be difficult to apply the cost rationale at the implementation stage. What the Court said in American Trucking is that in setting the NAAQS, EPA was forbidden to consider cost, not because the statute said in so many words cost can't be considered, but because the criteria that were set out in the statute for what the NAAQS had to achieve simply couldn't be reconciled with consideration of cost. But the Court in the same decision said, although you can't consider cost in determining what the NAAQS will be, what air quality standards have to be achieved, of course you can and should consider cost in deciding what implementation measures should be used to determine which emissions will be reduced. If Congress wanted that, why couldn't Congress simply have said, the EPA shall prescribe minimum pollution reduction measures that have to be taken by the states? That's a quite different statute from what we have before us. But what you're saying is, you know, you reduce it this much, as much as efficiency will allow, or else you're in violation of the good neighbor rule. And that's a very different statute from what Congress wrote. Maybe it's a good idea. Maybe EPA ought to control all efficiency measures for reducing pollution, but it's certainly not the statute that Congress wrote. Let me say three things in response to that. The first is that, as I mentioned before, the good neighbor provision is addressed in the first instance to the states. That is, it's the state's initial obligation to submit an implementation plan that contains good neighbor provisions. And so if the Court says cost can't be considered in defining significant contribution, the effect is not simply that EPA can't consider that factor when it steps into the state's shoes. The effect is that a state can't consider costs of achievement in attempting in good faith to implement its own good neighbor provision. The second thing I would say— I don't understand that. Please say that again. The good neighbor provision— we're dealing here with a situation where EPA was the one that promulgated federal implementation plans. But that's only because the relevant upwind states did not discharge their obligation to promulgate state implementation plans that contained good neighbor provisions. But the language contributes significantly is in the portion of the statute that deals with what a state plan is supposed to contain. It's not in a provision that, by its terms, is addressed directly to EPA. And so if the Court said, in defining contributes significantly, we can't take into account the costs of emission control measures, that would mean not simply that EPA can't consider that factor when it steps into the state's shoes. It would also mean that the state can't consider that factor. Well, you mentioned the fact that the states didn't address the good neighbor requirement. Of course, you hadn't come up with their budgets that they had to meet at the time that they had to promulgate their SIPs. Now, at a different point in your brief, you emphasize how incredibly complicated it is for states to determine how much they must reduce their emissions to take care of the fact that they significantly contributed to downwind pollution. And yet you would impose on those states the burden to issue the good neighbor program without knowing how much you expect them to meet. Well, it's the statute that imposes the obligation on the states, and it may help to draw the court's attention to the relevant provisions. On page 1A of the appendix to the government's opening brief, the relevant provision is 42 U.S.C. 7410, and 7410A1 begins by saying, each state shall, after reasonable notice in public hearings, adopt and submit to the administrator of EPA within three years or such shorter period as the administrator may prescribe after the promulgation of a national primary ambient air quality standard, the NAAQS. And then it goes on to say, a plan which provides for implementation, and so forth. And then if you look to the bottom of, or to the top of page 2A, I'm sorry, subsection 2 begins, each implementation plan submitted by a state under this chapter shall be adopted by the state after reasonable notice in public hearing. Each such plan shall, and then if you look at the bottom of the page, it says, contain adequate provisions prohibiting, consistent with the provisions of this subchapter, any source or other type of emissions activity within the state from emitting any air pollution in amounts which will contribute significantly to climate change. So if you were working for one of the upwind states, and you were facing this three-year deadline, and EPA had not told anyone how it intended to interpret the state's obligations under the good neighbor policy, what would you have told the state to do? Well, certainly EPA's basic methodology of using cost thresholds had been embodied in the NAAQS SIP call in 1998, and in CARE, which I believe was promulgated in 2007. Right, so the head of the state, EPA, comes to you and says, how much do we have to reduce our emissions to satisfy our requirements? And you would tell them what? We would tell them, in all honesty, we don't know yet, but that's not a fatal flaw in the argument. That is, it is inherent in any legal context in which one person acts and then a second person reviews that the first person has to act before the second person has made up his or her mind. Well, but that kind of glosses over the fact that, as you say elsewhere in your brief, this is your analogy, right, a spaghetti matrix or something? And so there's no possible way for the state to know how much of a burden you expect them to address, and yet you're saying, well, you've got to do it, and you've got to do it within three years, or we're going to take over the responsibility. Well, certainly what EPA was called upon to do was far more complicated than what any particular state was going to be called upon to do, because as a result of widespread noncompliance, EPA was promulgating federal implementation plans for close to 30 states and plans for different maps. The second thing I would say is... Well, can I just address the first thing first? I'm not sure that's right. I think EPA has an easier job dealing with it as a group. They say, look, here are these states. Here's what you have to do. But any individual state has no idea what its particular role is going to be in your group resolution. Well, it certainly has the data available to it that EPA had available about how much did each state contribute to the overage at various nonattainment receptors in the past. It's certainly true that the states wouldn't necessarily know exactly what policy judgment EPA would ultimately make as to what the right cost threshold was. But that's crucial. I mean, it would have no idea whether EPA would use any or would pick $500 or would pick whatever. I mean, I don't know how it could sensibly design a program without knowing that. I guess the other two points I would make are... First, the state's role is to devise something in this area as in others that it believes will carry out its own legal obligations, not necessarily to predict just how EPA would do it if the task fell to EPA. And so, for example, when the states are undertaking the more prosaic task of devising plans that will produce attainment of the NAPs within their own borders, they have to make a variety of policy judgments about the right mix of emission controls, what sources should be allowed to emit in what amounts. If a particular state just didn't do it, that task would fall to EPA. And it's very unlikely that anything the particular state would come up with would exactly match what EPA would ultimately devise. Can you give us an example of when EPA has done this in the past where a crucial element of a NAPs has not been defined by the agency and yet the agency nonetheless requires the states to put together their SIPs without knowing what their target is? And that's the problem here. What's your best example of another case in which the agency said, you put together a SIP and we're not going to tell you what the target is? Well, the examples I would point to are in the brief filed by the respondent states that are on our side of the case who identify examples of instances where states did successfully comply with their good neighbor obligations and persuaded EPA that what they had done was enough. Well, that just means it's pinned the tail on the donkey. Some states got the tail. I mean, you know, they pinned it in the right place. That doesn't prove anything. I want an example of another instance in which EPA has hidden the ball, has said, we're not going to tell you what the target is. It's up to you to come up with a SIP and we'll tell you after the fact whether that SIP happened to meet the target that we've invented. I wouldn't characterize what EPA is doing as hiding the ball. That is, it didn't kind of fail to divulge information that it had at its disposal. It released a great deal of information at the time that the proposed rule was announced in the summer of 2010. But I take your point. The two additional things I would say, though, are that, for better or worse, Congress did place this obligation of the states. It evidently thought that, at least in the mine run of cases, states were capable of carrying out this task. And at least to the extent that adopting a good neighbor provision requires consideration of circumstances in other states, in a sense, this is just the flip side of what the downwind states have to do all the time. That is, if New York officials are trying to determine when a new NAAQS comes out, how can we bring our own air quality into compliance? What controls do we have to place on our own sources in order to get air quality to the desired level? The New York officials have to take account of the degree of pollution that is likely to travel to their borders from other states. They can't analyze emissions within their own borders in a vacuum. They have to consider what the likely contributions of their neighbors... Yeah, that just means there's some facts that they don't know, of course. There are always going to be uncertainty about certain facts. But here there is uncertainty about the target, not just about the fact. We don't know what target we're expected to hit. I guess the final thing I would say on this particular sub-issue of the case is that even if you reach that conclusion, even if you determine that it was just practically infeasible for any state to adopt a compliant state implementation plan with good neighbor provisions for these NACs until EPA acted, then the proposition of the opposing states still wouldn't follow. That is, the statute in the provisions that I pointed to says it's up to the states in the first instance to devise the state implementation plans including good neighbor provisions. And then on page 10A of the same provision, of the same appendix, I'm sorry, the statute describes what happens if a state fails to satisfy that obligation. And this is at the beginning of subsection C1 on page 10A. It says, the administrator shall promulgate a federal implementation plan at any time within two years after the administrator finds that a state has failed to make a required submission or finds that the plan or plan revision submitted by the state does not satisfy the minimum criteria. Mr. Stewart, below, the government conceded that there was a theoretical possibility that some states could be over-controlled, that they would be implementing measures that would reduce their contributions to pollution below the 1%. And I would just assume that I think there's a theoretical possibility of that, but that your approach was basically fine. What would we do about that? First of all, are there measures states can take to get out of the FIF if it's inappropriate to them because of over-control? And if not, and how do they do it? I mean, what's the process if we think there's a flow? Do we vacate the rule? Do we leave it in place? What do we do? And what's our power to do it? I mean, I think in the circumstance you describe, if you reach the conclusion that there was a theoretical possibility that this could happen and that it would be a problem if it did, but that the methodology used by EPA was on the whole rational, I think the task for the Court at this stage of the case is to rule on the more big-picture objections that are properly before it and that the Court of Appeals ruled on. Now, even if we win everything that's at issue in this Court, the case is not over. There are a variety of more specific challenges to the details of the rule that the D.C. Circuit found it unnecessary to address. And so if we won on the issues that are before the Court, the case would be remanded, and there would be an opportunity for the Court below to consider those. And to the extent... Including the over-control argument, or would that have been done? Well, to the extent that any state has... And I don't know the pending-as-applied challenges at this level of detail, but to the extent that any state has a properly preserved challenge to the effect that it is actually likely to be subject to over-control, then that could be heard by the Court of Appeals. The Court of Appeals could determine both whether that is, in fact, likely to happen and whether, if it does happen, that would render the rule arbitrary and capricious as to that state. But the real problem with the Court of Appeals' methodology was that it said, the fact that EPA can't absolutely rule out the possibility that it might happen renders the rule invalid on its face. And in other portions of the opinion, the Court faulted EPA for failing to ensure that its regime would not lead to over-control. And I think that's an extraordinary standard for an administrative agency to deal with. That is, you know, it happens all the time that federal agencies are given authority to regulate, to address one problem, and the regulation necessarily has spillover effects on other conduct. And so, for instance, if a federal agency was tasked with preventing the sale in interstate commerce of contaminated food, it might require inspections, it might require the recall of food after one item in a shipment had been shown to be contaminated. These measures might have spillover effects on food that was not, in fact, contaminated. But that wouldn't be a flaw in the rule. Of course, an agency could go overboard and impose a regime that was so onerous in comparison to the health benefits that it was arbitrary and capricious. But nobody would ever say that it's the duty of the agency to ensure that there is no other means of achieving the same health benefits at lower cost to the public. The other thing that the states could do, I mentioned that one way in which a state that believed itself to be unfairly or inappropriately treated by the rule was to pursue any adequately preserved legal challenge it may have in the judicial proceedings. And as your question indicated, there is also a mechanism by which a state can ask to have the federal implementation plan replaced by a plan of its own devising. And so the consequence of the state's failure to achieve their good neighbor obligations in time and EPAs stepping into their shoes, the consequence was not that they are forever barred from devising their own plans. The consequence was simply that the federal implementation plan would remain in effect for a fairly limited period of time subject to replacement by a state plan. If we were to rule against you and affirm the decision below, how long do you think it would take to get a new rule in place? I don't have an estimate on the time, but if the court affirms on the ground that EPA may not consider costs, part of the problem, I think it would be an extraordinary undertaking for EPA to try to achieve. That is, part of the difficulty here is that nobody has identified a concrete alternative, that is, a plan that would not consider costs and yet that would disperse the burdens of compliance among the states in proportion to their prior contributions and also would address the nonattainment problem at all of the downwind receptors. I don't know if it could be proved. Could you explain that to me, Mr. Stewart? Are you saying that the straight proportionality approach that was applied in the D.C. Circuit, are you saying that that's impossible or are you saying it's complicated and dumb? At least what we understand to be the straight proportionality approach is impossible. That is, it might be possible with respect to any particular downwind receptor because you could say that if one upwind state is contributing two units and another four and another seven, the proportional solution might be to require that any necessary reduction would be in those proportions. One state would do two-thirteenths of the reduction, another would do four-thirteenths of the reduction, and another would do seven-thirteenths of the reduction. That would be theoretically possible with respect to any receptor. But with respect to another receptor, the same states might be contributing in entirely different proportions. And so there would be no way of devising a solution that would be proportionate as to both. You could average them out, couldn't you? You might be able to average them out. I don't think that's any more irrational than picking a number like 500 bucks as to who can do it more efficiently. That's sort of arbitrary. Well, I mean, the purpose of the cost threshold was not to increase or decrease the total amount of reductions that would be necessary. It would be to ensure that the reductions that had to take place were done in the most cost-effective manner possible. And part of the irony... I understand that, but my point is that is certainly a pretty arbitrary number. And I think averaging for all the receptors is certainly no more arbitrary. Like, I think the cost methodology is one that EPA had used often in the past. Indeed, even before the term contribute significantly was added to the statute in 1990, EPA had interpreted the prior reading of the statute to allow relief for a downwind state if an upwind state was contributing significantly to downwind pollution. And it had interpreted that standard as allowing consideration of costs and compliance burdens. I mean, one of the ironic things about this case is that the only ill consequence of overcontrol is cost. That is, this is not a situation in which there is some distinct public health benefit... distinct public health problem, I'm sorry, that is caused if power plants are emitting too little NOx or SO2. The only reason that people worry about overcontrol, about reducing emissions more than they need to be, is that it costs money. And if that's the problem to be avoided, it seems strange that EPA can't take account of costs in devising a solution. I don't want you to finish your argument if you had something to say on what you started out with in describing the plan. You said there are three aspects. The first aspect was you cut out anyone, any state that's contributing less than 1%. You said the second aspect was that you used a metric of $500 per ton of NOx reduced, and you applied that to the states still in. And then you said there were three, and you never got the three, and I want to be sure you do. The third part of the process is that once each state's emissions budget has been quantified, with respect to each state, the EPA essentially divides up the emissions that are allowed among the different power plants within the state's borders. And the way that it does that is it gives allowances to the various power plants that add up to the total number of tons of pollutants that are allowed to be emitted. And it's important to emphasize that the states have not joined the industry's argument here that costs... even the states on the other side of the case have not joined the industry's argument here that states... that costs can't be considered. Those states are not quarrelling with the methodology by which EPA quantified their state emission budgets. Those states are simply saying that once those had been quantified, they should have been given an opportunity to determine on their own how the allowances should be allocated without EPA doing it first. In some situations, that might have been a rational way for EPA to proceed. That is, the statute says that once EPA finds that a particular state has either failed to submit a good-neighbor SIP or EPA has disapproved the good-neighbor SIP, once that happens, the statute says that EPA at any time within two years can promulgate its own federal implementation plan. And in some circumstances, it might be rational for EPA to wait the full two years and give additional guidance to states in order to give them every opportunity to devise compliant plans. There were basically two reasons that EPA didn't do that here. The first is that it was subject to the D.C. Circuit's mandate in North Carolina, which said get something in place that works as soon as possible, and EPA felt constrained by that to act as quickly as it could. And the second point worth emphasizing is that there are state sovereign interests on both sides of the case. It's true that by devising a federal plan, in the first instance, EPA has intruded to a degree on the ability of the upwind states to decide how emissions allowances should be allocated among their own sources. But the downwind states are subject to their own obligations to comply with the NAAQS within their own borders. And to the extent that they can't get relief from the upwind states, their task is made more difficult. I'm sure I should know this after reading all these, but if we reverse the D.C. Circuit, what would happen going forward? In other words, the states have had this time to go first and do their steps. Then they were found not to comply, so the EPA does its fit. But that's not the end of the game, is it? I mean, isn't the EPA under a continuing obligation to look at, review its NAAQS, to give the states further opportunities to come back? Yes, I mean, with respect to the particular NAAQS that are at issue here, the states, it's unclear to what extent they've been working on this in the interim, but the states certainly could, even under the terms of the transport rule, propose state implementation plans to replace the fits. Now, it's to be contemplated that there will be additional NAAQS implemented, and this Court's decision would affect the way in which both the states and EPA went about the business of determining how good neighbor obligations should be carried out with respect to those future NAAQS. Thank you, counsel. Mr. Mitchell, why don't you give us 30 seconds or so? Thank you. Mr. Mitchell? Mr. Chief Justice, and may it please the Court, EPA's actions in this case have written the states out of the Clean Air Act. to implement NAAQS, and we have no intention of doing so. EPA has proposed a good neighbor fit on the states. When EPA has left the states completely in the dark about the meaning of the phrase, contribute significantly. EPA's approach requires the states to submit SIFs that can only guess at how EPA will quantify their good neighbor obligations under Section 7410A2D. Well, it's certainly hard, but it is what the statute says, and it seems to me that if EPA had taken a different view, it would have been contrary to the statute. EPA's actions are unlawful for several independent reasons. The first is EPA's actions in this case represent an arbitrary and capricious change in the way that the agency has interpreted the statute. For 15 years, starting with the NAAQS SIF call in 1998, EPA told the states not to submit good neighbor states, good neighbor SIFs, before EPA had quantified the states' obligations under A2D. And EPA repeated that stance numerous times, including in the disapproval of Nevada SIF that we cite on pages 9 and 59 of our brief, and also in the sources of the D.C. Circuit Cites on pages 51, 52, and 56 of the petition appendix. EPA has now done a 180-degree shift, and they have told the states that they are required to submit good neighbor states, SIFs, before EPA has quantified their obligations under subsection A2D. They don't know exactly how to do it. I mean, this is a tough problem. So it sounds as if what you're making is a procedural objection here, to which the government's point was, you're right, we'd all been talking about this. We wanted to see what the states would come up with. So we look. The states haven't come up with enough, in our opinion, and so now we go to the federal process and we put out our thing, and you comment on that. And then, if you feel that their thing is no good, propose your own solutions again. That's what he's saying. But it's supposed to advance the ball. So there is a procedure for the states to come in if they can come up with a better plan. That's what you've just heard. And so do it. So what's arbitrary or capricious about such a system? Because that's the approach EPA rejected in the NOC SIF clause. Well, they objected it once. Now they think it works here. I mean, all the time it happens that people change their minds about how problems are best solved, or they select this problem's better solved one way and a better another way. So if your only point is once they did it a different way, they'll say, well, what's unreasonable about changing our way? We're trying to get the job done. EPA is allowed to change their interpretation of the statute, but if they're going to do that, they have to acknowledge in the transport rule that they're abandoning the prior construction of the statute. Well, years and years the Labor Board decided things in adjudications. Rules. One day they say, no, we think we should preside rulemaking processes like other agencies. Does anything in the law prevent that? No, they're not prevented from making the change, but the arbitrary and capricious... You're never prevented from giving a counter SIF, is what they were being told. You can counter, is what the government is saying. So it's not clear to me that they've stopped you from doing your own SIF. But we can propose and submit the SIF only after the SIF has already been imposed on states. So what's the difference if you think they're wrong? You give a counter SIF and you duke it out with them in terms of where you think they are wrong as applied to you. But let me ask you something fundamental about this. Are you challenging the transports rule using costs? Or are you just challenging the process in which that was achieved? Because if I understand all the amici briefs and the theory of this, not even you would want a command and control regulation. Is that correct? We do not have a position on the question of whether EPA can consider costs. It would be crazy if they didn't, right? We represent a coalition of states. Right, and for some of them it would really be a bad idea, wouldn't it? There's simply no consensus among the states on that question. So we are remaining agnostic on that point. Can I ask, until you propose your SIF to replace the FIP, right? The FIP remains in effect. Yes. And you're bound by that until they approve your new SIF. How long does such a transaction normally take? It depends. It really does. We don't know exactly what our obligations are. Do you think that's a quick process? You'll have to develop a new SIF. That will take you some time. Yes. And then that SIF is submitted to EPA and they chew on it for as long as they want, right? Yes. And then maybe they will say your SIF is good enough and maybe they won't. Right. We're still waiting for EPA to decide on the SIF that we submitted to implement the good neighbor obligations for the 2006 Particular Matters Standard. But at least if you've adopted a SIF or proposed a SIF, you've given reasons, you have a rational plan, and the EPA then must give a reasoned response to it. Well, it's still an unlawful use of the FIP authority for several reasons. And this gets back to Justice Breyer's question, why is this unlawful? And there are several reasons. The first is that EPA has changed its interpretation of the statute. And the key language from the NOC SIP call appears on pages 57,368 and 57,000. Where EPA tells the States, we don't want you to submit good neighbor SIFs. Let's simply take a guess at what you think the good neighbor obligations are. We will quantify your obligations in a rule first. And that means the statute doesn't require you to do that, I assume. When they say that, that's an affirmation by them that the statute does not require you to do it, I assume. Is that your point? No, yeah. They said in the NOC SIP call that they are adopting an interpretation of the statute that prohibits us from doing that. That the interpretation of the statute that EPA adopted is that EPA and only EPA is the institution that is charged with the responsibility of quantifying a State's good neighbor obligations. The States had argued for a different approach back in 1998. We wanted to have the prerogative to decide what the good neighbor obligations mean. And EPA said, no. We are the sole entity with that prerogative, and you need to wait until we issue a rule that quantifies your obligations. EPA has now changed that approach without explaining or acknowledging in the transport rule that they were abandoning their earlier interpretation of the statute. They were doing that with respect to the NACs. I thought that was them saying they had to quantify the NACs. What they said in the NOC SIP call was they have to quantify the good neighbor obligations. They have to tell the States what it means to contribute significantly to another State's air pollution. And that leads to a second statutory problem with this regime in the transport rule. The States have the prerogative under the Clean Air Act to do what the Federal minimum requirements are for clean air and to go no further. EPA's approach here requires the States, when they submit or propose SIPs, and they have to take a wild guess at what their good neighbor obligations are, it effectively compels the States to over-control and over-regulate. Because if they want EPA to approve the SIP and they don't know what their good neighbor obligations will be, they have to overshoot and over-control and over-regulate or risk that EPA will deny their SIP and impose a SIP on the State. And what EPA is essentially doing is telling the States that if you want to do only what the Federal floor requires and to go no further, the price of that is that you have to accept an EPA-imposed SIP that defines those obligations, rather than giving the States the opportunity to distribute regulatory burdens in a SIP as it sees fit. Mr. Mitchell, I might just not be understanding you, but this goes back to the Chief Justice's question. The statute says, look, after these standards are originally promulgated, the State gets three years to make its best pitch, and then the administrator shall promulgate a FIP at any time within two years after that. Now, presumably, there are lots of conversations that can happen between the EPA and the States during those five years, and maybe sometimes more of those conversations happen and sometimes less of those conversations happen. But I don't see that as different constructions of the statute. It seems to me, as the statute sets up its framework, you go first, do it within three years, then the EPA goes. It has to do it within two years. The EPA just has very substantial discretion under this statute as to what kinds of conversations it wants to have when within that broad structure. Why am I not reading it right? I agree that the EPA has that discretion. The problem is that in the Knox SIP call in 1998, they asserted exclusive interpretive authority over subsection A2D, which is the good neighbor provision of the Clean Air Act. And they said that EPA is the institution that must quantify the State's good neighbor obligations. Until EPA fills in the blanks and tells the States what this contribute significantly phrase means, it's an empty requirement. EPA could have taken a different approach in the Knox SIP call. They could have told the States, you can take the first crack at defining what contribute significantly means and we'll review your submission and approve it or disapprove it. But what they said 15 years ago was that the States need to wait for EPA to quantify the obligations in a rule. Once EPA asserts that exclusive interpretive authority over the provision, the States have no obligation to guess at what EPA might do in the future when they submit the set. And that leads to a second independent problem with EPA's transport rule. Because EPA had no authority to impose federal implementation plans for the 1997 standards on the 22 States that already had EPA-approved SIPs in place for those standards. Haven't some States already challenged that? Isn't that pending below? Why should we be looking at that issue here when States have challenged that? Three of the States have challenged their... I don't know why the rest didn't. But three of them have, so why should we enter the fray anticipatorily? Isn't that an issue we should wait and see what EPA says below? The United States is suggesting that we're somehow launching an improper collateral attack because the States could have challenged the earlier SIP disapprovals or earlier findings of failure to submit. No, no, no. This is a very discreet question, that they've already approved some SIPs. Three States have already challenged the fact that they shouldn't be required to meet a new standard because they've already had an old standard approved. That seems to me a very discreet challenge, and three States have undertaken it. But those judicial proceedings have been stayed pending the outcome of this case. Well, I don't know why, but that's a different issue. Even though it has, wouldn't it be more prudent for us to wait for that administrative process to finish before we venture into this question? That's my question. I don't think the Court should wait because the issues that we're raising are discreet from what's being challenged by those three States and their separate proceedings. The arguments we're making are that, first, EPA has no authority to impose a theft on the States before quantifying the good neighbor obligations under A2D. And, second, EPA improperly invokes the corrections power. Well, that's because EPA did it that way. You don't get that from the statute. You get it from what EPA did in the first statute. Is that right? We're not relying solely on the statute, Justice Ginsburg. That's correct. Our argument is that EPA has changed its interpretation of the statute from the NOx SIP call to the transport rule without adequately acknowledging or explaining how its new interpretation is consistent with the statute. But we're also relying on the statute as well because, as I mentioned earlier, A2D requires the States to eliminate pollution that contributes significantly to another State's nonattainment. They may not know, you know. They may not know. There are six States that contribute to the seventh State's pollution. And how much each State can cut back depends. It depends on what it costs. It depends on how much they contribute. It depends upon what the other States will do. It depends upon where the wind blows. And that changes all the time. Right. So they have a tough problem. They can't tell you exactly how much you should cut back until they know what they have in mind or what others have in mind for solving the problem. So it sounds to me as if you're asking them to do the impossible. And they had a very good reason for not doing what they did before. Namely, it would be impossible here. They have a decent, or not actually impossible, but very tough and very expensive. So that's why I gather they went the way they did. I don't know anything in the law that tells them that this statute was meant to force them to proceed in a way that would either be hugely more expensive and perhaps impossible. What's your reaction? EPA has done this before. I mean, it was the CARES steps when they first quantified the State's good neighbor obligations. They gave the States an opportunity to submit SIPs before the CARE federal implementation plans would take effect. And EPA agrees that the States have no ability to guess accurately at how EPA will quantify the good neighbor obligation. Well, they have done the impossible here, haven't they? I mean, they have the transport rule. The only question is whether it should have come out sooner or later. Right? The only question is whether it should have come out before the States were obliged to submit their SIPs. It's not impossible for EPA to decide what contributes significantly means. That's their job. They've asserted that prerogative. They can choose any reasonable interpretation of that phrase. You started to give us a second statutory reason. I was really eager to see what that was. Yes, that's the Section 7410K6 issue that we mentioned in the brief. EPA had previously approved good neighbor SIPs for 22 States that implemented the 1997 standards for ozone and particulate matter. Once EPA approves a State SIP, its ability to impose a FIP on that State expires under the statute. So EPA had a problem for those 22 States. How would they be able to impose SIPs when they had previously approved SIPs? EPA says in the transport rule that they're going to invoke the corrections power of K6. And K6 says that if EPA determines that a prior decision approving a SIP was in error, then EPA — Isn't that the issue that the three States are challenging below? Just that discrete issue about whether the EPA can call this a corrective action or not. Isn't that entire issue being determined in those proceedings? It's not being determined because the proceedings have been stayed. But, yes, three States, Kansas, Georgia, and Ohio, have challenged. But that issue is what's at issue there. Would you finish describing the issue? I really didn't hear it. EPA invokes its corrections power under K6. But K6 says that a correction must be made, quote, in the same manner as, end quote, the decision being corrected. EPA's approvals of the earlier SIPs went through notice and comment. Because of that, K6 requires that the corrections likewise go through notice and comment, and the corrections here did not go through notice and comment. There's no disagreement between the Petitioners and the Respondents on that point. And the United States tries to get out of this problem by saying that they can use the good cause exception to notice and comment rulemaking that's found in the Administrative Procedure Act. That doesn't help EPA at all, because the requirement comes not from the Administrative Procedure Act. The requirement to use notice and comment comes from K6. It doesn't help EPA to rely on an exception to a statute when the statute providing that exception is not the statute that imposes the requirement. Well, why can't this is, of course, a statute on which EPA gets substantial Chevron deference. Why couldn't we read that language to essentially mean subjects to the same procedural requirements as the original? Because that caveat does not appear in K6. Well, it's not a caveat. It's just a different understanding of what that language means. I mean, you say it has to be in the exact same way they previously acted, and I guess I'm saying it could mean subjects to the exact same procedural requirements. Was that not clear? It might not have been clear. I think your argument or the suggestion is that EPA could rely on the normal rules set forth in the Administrative Procedure Act. Whatever procedural requirements constrained EPA when it approved the SIP, those were the same procedural requirements that constrained EPA when it disapproved the SIP. But we're just asking, are they, you know, both have to be subject to the same procedural requirements. EPA can act differently as long as they're acting within that same set of rules. We don't think that's a tenable construction of K6. I mean, K6 authorizes EPA to make corrections, but it says specifically that corrections must be made in the same manner as the decision being corrected. If the decision being corrected went through notice and comment, the corrections have to go through notice and comment as well. If the decision being corrected went through formal adjudication, then the correction must also go through formal adjudication. And EPA doesn't try to make that argument in their brief about what K6 means. And they're just trying to say that the good cause exception in the APA to notice and comment rulemaking should carry over here. And there's — Well, I think that they are trying to make that argument. They're saying in the initial version, we could have done it by notice and comment rulemaking, or we could have done it if we had good cause. So, too, when we reversed that initial determination. That's what the statute means, and K6 doesn't constrain the agency much at all. I think it would follow that if you did it for good cause to apply the rule, you can do it for good cause to abolish it, not that you can do it by rulemaking when you adopt it and then use good cause when you abolish it. It seemed to me to square with the text. This text says in the same manner as. In the same manner. In the same manner. So it's looking back to the original decision and how it was made. And, you know, it's the second reason we've provided for why the D.C. Circuit decision should be affirmed. Now, if the Court were to reach the K6 issue, there's also the question of whether the FIPS can be severed, because the K6 argument doesn't — I see my time has expired. You can finish your sentence. The K6 argument doesn't knock out all the FIPS standing alone. It would require analysis of the severability question. Thank you. Thank you, Mr. Mitchell. Mr. Keisler? Mr. Chief Justice, and may it please the Court, the private party respondents are focused on the statutory limitations to the EPA's authority under the Good Neighbor provision. And I'd like to begin by addressing the issue that my friend from the government focused on a lot, which is the use of cost, and to explain not only what we think the statute requires in this regard, but why, Justice Kagan, it would neither be silly nor dumb or, Justice Sotomayor, crazy to read the statute the way we suggest. And we begin with the text of the statute, which authorizes the prohibition only of amounts that contribute significantly to nonattainment or interfere with maintenance in downwind locations. The focus of that language, we think, is quite clearly on the effects of a state's emissions on other states, and not on the cost of reducing them. What EPA has done here is assert that it has the power to increase a state's reduction obligations beyond what a focus on the effects of its emissions would for that state to bear a higher burden. And what that means is that states here, which are making only a very slight contribution to air quality problems in downwind states, are nonetheless required to make very substantial reductions, in many cases far more than states that are making far greater contributions to poor air quality in the same downwind locations. There is no relationship at all under the EPA's methodology between the amount a state contributes and the amount it has to reduce, because the entire driver is cost. Mr. Stewart said that cost was one component. It's not one component. It is the entire driver. Why is it wrong? That is, I've focused on your argument here in the briefs, which is very clear and very good, and the example that comes to my mind is we have an overgrazing problem in state A. It's caused because cows come in from state B and sheep come in from state C. The cowmen and the sheepmen are in different states. They're not friends. Now, it turns out that EPA, which is in charge of preventing the overgrazing, discovers that if the sheepmen build a fence, that will cure the problem, even though they only contribute half or maybe less. Well, if we vary it, you know, we divide it equally, you each have to cause half the problem, because that seems fair. It's going to end up that the people in state A with the cows, they're going to starve to death. So our choice is between taking two people, two states, each of whom cause half the problem, and getting an overall plan where you solve the problem at minimal cost, or just dividing it 50-50, which seems fair in mathematics, but leads to starvation, cost, and death, etc. Do you see what I'm driving at? And that's what they've done here, the second method. Rather, they're not treating each state alike. You're right. And the reason that they're not treating each state alike is they know, one, all the states are partly responsible and more than 1%, and with this plan, we get the job done at much lower cost. Now, where in the statute does it say they can't do that? I'll try to respond to that fully, Justice Breyer. Certainly it is the case, and we would acknowledge, that there are always going to be legitimate policy arguments in favor of the least cost, most efficient solution to any problem. But we would also say that there are countervailing policy arguments at issue here, and we do believe the statute sides with those countervailing policy arguments. And the countervailing policy arguments here are focused on the fact that, in your Honor's hypothetical, where there's one state that it would cost more to reduce and another state it would cost less, the only scenario in which you get a different result under the EPA's approach and our approach is where the state that would cost more to reduce is, in fact, contributing a lot more to the Downwind State's air problem than the other state. And for several reasons, we think when the statute looks at that issue and asks the question of whether EPA should have the authority to force the state which is, in fact, contributing less to nonetheless reduce more simply because it's costly, we think there are at least three reasons why the statute embodies the policy choice that says no, the state that contributes more reduces more, the state that contributes less reduces less. The first is the one that I mentioned at the outset, which is the statutory text, which we think quite clearly is focused on the effects. Significantly contribute to nonattainment or interfere with maintenance in downwind locations. But the second is the whole structure of the claimant. Amounts significantly attribute, right? It's amounts. Amounts, correct, that significantly contribute. Not conduct significantly. No, that's right, and I don't think the word significantly can bear the weight that Mr. Stewart places on it. It modifies the words contribute to nonattainment, so it's about the degree of causal contribution, and it doesn't modify at all the phrase interference with maintenance, and they've used the same cost methodology to implement that as well. But the second, beyond the text, the whole structure of the Clean Air Act is focused on treating states as separate entities which are responsible for the emissions that happen within their borders and the effect that those emissions have on other states. That's why this is in a SIPP rather than some general EPA regulation, and that's why the language of the statute is what it is, amounts of emissions within a state. And in this regard, I think it's telling that in the reply brief what the government said was that it believes it has the kind of authority here to consider cost that would be considered by a chancellor of equity in a nuisance case. Chancellor of equity in a nuisance case had private party defendants before him or her, and so, of course, they were allocating burdens on the basis of equity and efficiency and all the kinds of things a common law chancellor can take into account. The EPA has before it separate states with separate responsibilities who have a long historic role and responsibility of enforcing emissions control procedures within their border. And Congress could rightly or reasonably at least have concluded that it didn't want EPA to have the same authority to shift costs and efficiency and equity around among different states to require it. Now you remember, please, that my cow sheep example was meant to pick up precisely the disproportion that you're talking about. And keep that in mind because I found it a helpful example. Now my point is, did you find in Congress, and I'm interested in legislative history, did you find anything in the legislative history that suggests that where the EPA faces this kind of regional problem, and it's a regional, not just a statewide problem, that people in Congress thought they had an answer or a glimmer of an answer as opposed to taking this language, which is pretty open, and saying we're going to leave it to, we don't know, we don't have a clue, the EPA is there to figure this thing out, and we're giving them broad authority here. Is there anything that cuts on your side that you see as opposed to the other side of reading this language? I think there's one thing I can cite, Your Honor, and that is that the statutory history in this case is that the predecessor version to what we currently have before us simply said that states were required to prohibit the amounts which prevent attainment or maintenance. No word significantly, just prevent attainment or maintenance. I think certainly looking at that language, there's nothing in there that would suggest that costs can be taken into account. What Congress said in the committee report in 1990 when it added the words significantly contribute to nonattainment and interfere with maintenance was that it was doing that precisely because it recognized that this was a provision that addressed causation of bad air quality effects, because what it was doing was not introducing some new element of cost, but relaxing the causation standard, saying it shouldn't be something like but-for causation where the question is just does it prevent attainment or maintenance. It's enough if it contributes significantly to nonattainment. So, Mr. Keiser, I mean, you have a statute here that clearly does not have any language about no costs allowed, that also does not have what the American Trucking Association statute had, which was like public health only sufficient margin of safety, right? So none of that. What you have is exactly what you said. You have a statute that focuses on causal contribution, right? So this is a hard problem, right? Because, I mean, let me just sort of give you a numerical example, which I'm sure is as simplistic as the other numerical examples floating around this case. But, you know, let's say that the standard is 100, and there is a state that has 120. And there are two states, X and Y, that have each contributed 20, right? So we only need 20 of those. We have 40. And the question is how do you get from those 40 to those 20? The D.C. Circuit would just say, well, we take 10 from each. But if the question is only about causal contribution, and that's all that the statute talks about, there have to be other ways we can make that determination of what contribution each should be legally responsible for, right? And what the EPA said here was we're going to distinguish between states that have put a lot of technology and a lot of money into this already, and on the other hand, states that have lots of cheap and dirty emissions. And why isn't that a perfectly rational thing to do under this very statute? Well, first of all, I think in the example that Your Honor gave where you had the two states and should they be each reduced to 10, the reason in favor of doing it that way from a statutory perspective is that that then gives a consistent application to the same causal language in the statute, which means that the same causal effect from one upwind state on a downwind state isn't significant if it comes from Indiana to Delaware, but insignificant if it comes from Tennessee. And that when the statute says that states must prohibit the amounts that significantly contribute, then the more they contribute, the more they reduce. So we see that as fitting much more securely within the statutory language than the kind of shifting that Your Honor mentioned. Certainly, one could imagine, since the policy rationales that are behind Your Honor's question are certainly legitimate and more than plausible, certainly one could imagine a statute that Congress could have written which would have said, treat it as a national playing field, ignore the fact that there are state boundaries, think about what the most efficient way to force reductions to achieve attainment in downward locations, locate those reductions in the least cost areas, and impose those on the states. Surely, if that was what Congress had intended, it wouldn't have written a statute that would have said that states must prohibit the amounts that significantly contribute. The statute would have said, treat it as a national playing field, ignore the fact that there are state boundaries, think about what the most efficient way And this would have been a statute that Congress would have written, which would have said, treat it as a national playing field, have to take into account in some way the cost of reducing the amount. Your Honor, I'm here on behalf of industry and labor, so certainly we believe that there have to be mechanisms to deal with the kinds of problems that Your Honor just identified. But we don't think they come out of defining the amounts that significantly contribute to nonattainment. We think that those kinds of considerations come into play elsewhere in the process. In the American trucking case, it's been referred to. The court said that when states are implementing the requirements of the EPA, for example, by deciding to allocate among different sources how the reduction will be distributed, they can take costs into account. And there are other mechanisms in the SIP process, when that definition of what amount significantly contributes is then translated into an emission reduction obligation. We do think there are occasions, and we've noted them in Note 17 of our brief, where the state, in then formulating its SIP, can say, okay, this is the amount we have to reduce, but we're going to do it in this way because costs have to be taken into account. But that is a very different matter from saying that EPA, in defining what amounts significantly contribute, can do the same thing. And the reason it's different, the reason it's not just, oh, you know, we're locating in some different box what EPA wants to do in its box, is that the box that we're locating it in makes clear that it functions as the kind of break Your Honor described. If something is unfeasible or economic, there are ways to soften it out at the edges. Well, they found a way to do that with the cost tradeoff, with the cap-and-trade system. Because the industry itself can make that choice with the state, presumably. They're not stopping a SIP that stops the state from participating. Well, the trading presents unique issues under this statute. We support trading anywhere it's appropriate, but this is a statute which is focused on providing relief to downwind states. And if, to take my earlier example, if Indiana is contributing emissions into Delaware that credits air quality, it does no good for Delaware if Indiana purchases allowances from Tennessee, which isn't contributing to Delaware. You want me to write. Look what I would have to write if I made it very specific. Two units floats over the air from the cow state. Two units floats over the air from the sheep state, or three. It happens that if we treat them alike, we are going to tell the cow state, your unit is the same as the sheep state's unit, both make the same significant contribution, and we have to say that, even if, for you to remit, your unit causes death and destruction, destroys your economy. See? And I have to write those words to accept your argument. Don't I? Because I will. I'd like to resist the role Your Honor has been assigning me of death and destruction and starvation. I'm trying to use the most extreme words I can so that then you will either have to draw a distinction or something. That's why I'm doing it. But in some ways, the distinction is the one I was drawing in response to Justice Sotomayor's question, which is that when you get down to the level of implementing these things, you can take into account whether death, destruction, starvation will be taken into account. When the State is doing that as part of the SIP process, but that doesn't bear on how the amount of significant contribution is defined, because when EPA takes costs into account, it's not simply preventing the death and destruction and starvation. It's working the other way, too. It's saying that even though a causation standard only would require you to reduce this much, we, the EPA, can shift to you an additional burden because we think another State has already done that. They say that's not a theoretical possibility under the numbers they've worked out. So why isn't this taken care of in the process that permits individual States to challenge this as applied? Let me make a distinction in that regard, Your Honor, which is what the government has said is a theoretical possibility is simply whether a State would be driven below the 1 percent threshold. But what I'm saying really goes back to Justice Scalia's very first question, which is even apart from the 1 percent threshold, every time they are allocating on the basis of cost and displacing what you would allocate on the basis of what each State actually contributes, then you are shifting burdens around, even apart from the 1 percent. But you're saying that significant must mean only measurable amounts. It can't mean, I'll pick your word, culpability, feasibility, responsibility, feasibility. One State finds it quite feasible from a cost standpoint to reduce admissions by a factor of 10. The other State, the Justice Breyer's example, finds it can't do it except with a factor of 100. Can't you say that the contribution in one case is more significant than the other based on feasibility? Do you think? I don't think so, Your Honor. I don't think that is a proper definition of significant when it's modifying contributions to nonattainment. Mr. Stewart— It isn't contributions to nonattainment. It's the word amounts. The statute prohibits activity within the State from emitting any air pollutant in amounts which will contribute significantly. We agree, Your Honor. We would emphasize— Amounts are amounts. I mean, it doesn't— But the word significantly does import a judgmental component. I think that's what the government is going to say. I just want to— It's not a limit limit. I don't think significantly means that any factor that might be deemed relevant in a broad policy sense can be imported in. I think we can have a statute here which talks about amounts that contribute significantly to nonattainment or interfere with maintenance. All right. So there's an ambiguity here. Because, I mean, use the word amounts. I think that it does help you add the word amounts to significantly. And I think that Justice Scalia's point might be—he knows it better than I—an amount's an amount. An amount's an amount. That's what you want to say. That's my point exactly. Yes. Yes. And then the response is, well, not always, because you say an amount—you're talking about a specific amount coming out of a State and is the one, the cow one, as significant as the sheep one. All right? And that's—I think we're—I think you've hit the nail as to what the issue is. That's right. And I guess opposition is that significant may have a range of meanings, but it's not a limitless range of meanings. I think one member of the Court once said the fact that yellow is ambiguous doesn't mean it can mean purple. And here, you know, we don't think the range of meanings in the context of this provision accommodate the government's definition. But, Mr. Kessler, the nature of this problem is that there's an allocation issue. It's not just everybody gets down to a certain threshold level. It's—there's a level, and we have to allocate. And the question is, what are we going to allocate on the basis of? And the word amounts doesn't tell you what you're going to allocate on the basis of. So there are lots of different choices for what we can allocate on the basis of. We can just divide, you know, and do it all proportionally. We can take into account per capita. We can take into account the State's population if we wanted to. Or we can take into account, as the EPA did here, costs, on the understanding that costs reflect how much of an investment a State has already made in pollution technology. So the statute, neither the word amount nor anything else says anything about those different methods of allocation, does it? I disagree with that, Your Honor. I think—and, you know, I don't focus exclusively on the word amount or the word significantly. It's the entire phrase. Amounts that contribute significantly to nonattainment or interfere with maintenance, which I do think 10 out of 10 people who weren't in this courtroom and hadn't read the Clean Air Act, you sat down and asked them, what does it mean? They're talking about what the effect of the emissions in one State had on the other. I don't think this is any more ambiguous in referring to air quality effects as a standard than the next statute at issue in American trucking was in talking about safety and health as a standard. It does supply a content to what the EPA has to do, and that content isn't cost. It's this air quality effect standard. What is your answer? Do you have an answer to Mr. Stewart's basketball hypothetical? I mean, I thought that was pretty good. If you ask the coach what significantly contributed to the loss, he's going to talk about the missed layup rather than the missed desperation throw, even though as far as amount, each was going to count for two points. Well, it's very hard to— Assuming the one was within the— It's very hard for me to translate the amount concept into a performance on the basketball court, but Mr. Stewart's other example was a contribution to a charity, and I certainly would accept the notion that if Bill Gates and I each contribute $100 to a charity, I've made the more significant contribution, but that's because we're using contribution in that context to mean something else. We're using it to mean donate or give. The problem with the basketball thing is to make it parallel to what's at issue here, the question you should ask the coach, which of the—you lost 101 to 100. Which of the 101 points contributed most to your loss? The answer is that some one point— It's the one that was the layup. I mean, he would not answer the one that was the layup. He'd say, what do you mean, all of the 101? But if there were different teams playing in the league and you had an overall result, you could actually determine which team had contributed what to the overall result, and when we're dealing with states, we are dealing with groups that the statute conceptualizes as separate teams, which are entitled to be treated separately. But I would like to make just one other point, because I see my white light is on. We have raised a completely separate argument. It's the first argument in our brief, which is independent of how the court decides EPA may define the amount that contributes significantly, whether cost or air quality effects or anything else. And that is that however it's defined, EPA cannot regulate beyond the point necessary to achieve attainment or maintenance in downwind locations. And here, although in the prior two good neighbor rulemakings, it specifically said it examined the issue and avoided overkill, here it didn't say that because it didn't do that. Apart from the cost versus air quality issue, we had commenters that submitted evidence that showed that EPA could achieve attainment and maintenance at virtually all the same downwind locations at lower levels of regulation. And EPA's response to that on PEDAP 354A was they weren't going to look at lower levels of regulation because at lower levels of regulation, some sources in some states might cease operating existing controls. And that's all they said. But if sources in some states could cease operating existing controls, and as the commenter said, you would still achieve attainment and maintenance in all the downwind locations that they are linked to, then EPA has no authority under the good neighbor provision to require those sources to continue operating their existing controls. There may be authority under other provisions, but not this one. And the EPA in this particular proceeding said nothing else, gave no other reason for refusing to act on the evidence that commenters submitted that lower levels of regulation at most upwind states would still achieve attainment and maintenance at downwind locations, and they had no authority to regulate beyond the point necessary to achieve attainment and maintenance. If the Court has no further questions, I thank the Court. Thank you, counsel. Mr. Stewart, you have four minutes remaining. Thank you, Mr. Chief Justice. As Mr. Keisler indicated, in our reply brief, we cited the restatement as it bears on the common law of nuisance. And as this Court indicated in American Electric Power, if the Clean Air Act had not been enacted, the remedy that downwind states would have in a situation like this one would be a federal common law nuisance suit against upwind states or polluters in upwind states. And I think there are three lessons to draw from that fact. The first is that as the briefs and argument in this case indicate, judicial resolution of such a suit would have been a Herculean task. And the prospect of doing that through judicial processes should reinforce the wisdom of Congress's choice to replace that mechanism with the Clean Air Act. And it counsels in favor of deference to the expert agency that has been placed in the position that a common law court would previously have been placed in. The second is that as the reply brief citation to the law of nuisance indicates, the common law court in that scenario would have been able to consider the cost necessary to achieve reduction in pollution upwind in deciding whether a particular remedy would be appropriate or how much of a reduction an upwind polluter should have to make. And there's no reason, absent extraordinarily clear statutory language, to deny EPA the same authority. The third thing is, as the analogy to the common lawsuit indicates, there are sovereign state interests on both sides of this case. This is not a matter of EPA versus the states. It's a matter of EPA trying to act as an honest broker between the upwind and the downwind states. The next thing I would say about the Clean Air Act is that the statute as a whole is replete with references to economic activity and harnessing the profit motive. That is, both the states and EPA are specifically authorized to provide for the trading of allowances, the whole purpose of which is to achieve emission reductions in the most cost-effective manner possible. And I think it's worth noting in this regard that although we talk about the transport rule as regulating the emissions of states, what we're really regulating is emissions of power plants within the states. And the good neighbor provision itself talks about preventing significant contribution from emissions sources or emissions activity within the states. And one of the things that the EPA said in the proposed rulemaking was that in some circumstances, the cumulative downwind impact of a particular upwind state might be great, not because any particular power plant was poorly regulated or emitting at a high level, but because there were so many power plants in the same state. And one consequence of forbidding the EPA to consider costs is that a particular power plant in an upwind state might be required to install more expensive pollution control measures and make greater reductions simply because it happened to be located in a state with a lot of other power plants. And the last thing I would say is the statute, as I've said before, has a prospective focus. It's intended to be implemented by state officials. And if you ask how would a state official assure herself or feel confident that her own state implementation plan was satisfying good neighbor obligations when she wasn't really sure what other states might be doing. And one way is if a state official said, if everybody else did what I'm doing, I can feel confident that the problem would be solved. And that's really the approach that EPA used. It examined certain cost thresholds, and it said at particular cost thresholds, we feel confident that if everyone, upwind and downwind state alike, makes pollution control efforts at these levels, the problem will be solved or at least almost solved because there would still be slight residual nonattainment. And it seems perfectly rational to say that the significant contribution is the amount over and above what would occur if everyone adhered to an approach which, if applied across the board, would solve the problem. Thank you. Thank you, counsel. Counsel, the case is submitted.

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A5648. Public record. Not legal advice.