Oral argument — Committee for A Better Arvin v. United States Environmental Protection Agency

Good morning. Good morning. May it please the Court, my name is Colin O'Brien and I am here on behalf of petitioners. I would like to reserve five minutes for rebuttal. Continuing with the topic of poor air quality in California's San Joaquin Valley, this case challenges EPA's abandonment of its 30-year interpretation of the Clean Air Act's contingency measure requirements for PM 2.5. Rolando. Counsel, one of the things that EPA says, they assert certainly in their brief and they indicate that it has a long-standing practice of allowing states to adopt contingency measures that provide for less than progress standards if the state provides a reasoned justification. What's your response to this and has it been, do they have this long-standing practice with regards to 172C9? Yeah, Your Honor, I would invite you to look at the particular citation to the record that they give for that assertion. Going way back to the early 90s, EPA said unambiguously that contingency measures should achieve one year's worth of reasonable further progress. Their citation in the briefing is to their 2016 SIP promulgation rule where they for the first time introduced the idea that maybe a reasoned justification would be allowed for less. So they went from like 1992 to 2016 saying there was no qualification and then as we highlight in our briefs, since 2016, EPA has in fact on two separate occasions disapproved of proposed contingency measures on the basis that they did not meet that one year's worth of reasonable further progress standards. So this action before us is the first time that EPA has ever allowed a state off the hook the way that they're proposing to do it. Just for clarification, it seems like we have a couple things going on here. One is the move from reasonable further progress to progress. The other is infeasibility. So on infeasibility, is it your position that that's a new development too? I think EPA concedes that reasonable further progress to progress, that is new. They acknowledge that. But as to infeasibility, I understand them to be saying we've always allowed for people to come forward and explain why something's not feasible. Do you dispute that? Yes, Your Honor, I dispute that. And I would point you to the citations in their brief, both their brief and the brief of the California intervenors, where what they are citing is the 2016 PM 2.5 implementation rule. 7502C9 was adopted in the 1990 Amendments to the Act. In the immediate aftermath, EPA said unequivocally what we want is one year's worth of reasonable further progress. It's not until 2016 that they even explore the idea of an out for feasibility. But as I mentioned in my previous answer, in the interim from 2016 until now, they on two occasions actually disapproved proposed contingency measures because they did not meet the one year's worth of reasonable further progress requirement. And so this action is the absolute first time that EPA has approved a plan based on the feasibility out. Now, counsel, Congress has set out in different sections feasibility language. So I'll borrow the statement from EPA's counsel, Ms. Martinez, in the last case that statutory silence. There's statutory silence allowing feasibility with regards to contingency measures. So do you agree that had Congress intended to put some feasibility language in this particular section and the contingency section, it would have done so? Your Honor, I think I missed the last part of your question, but I think I can still answer. Which is, across the Clean Air Act, Congress was very intentional about how and when it wanted feasibility, including costs, to be considered. In certain provisions, it made references to feasibility that are explicit. And in other provisions, it makes no mention. So for example, under section 7513a.d., which is the 5% plan provision that was part of the earlier discussion, that section makes no mention of feasibility. Just at the end of the day, a state has to come up with 5% reductions every year, period full stop. And so I would point, Your Honor, I think to the American Trucking Association's case, which I think is instructive. In that case, EPA wanted to import costs into portions of the act where cost considerations were not mentioned. And Justice Scalia said that Congress does not hide elephants in mouse holes, like it's a famous line from Clean Air Act jurisprudence, and that without an explicit textual commitment to considering costs, they shouldn't be part of a section where they don't appear. And we take the position that the same is true of 7502c9. So we have people on the other side saying, how can we be forced to adopt something that's infeasible? So how do you respond to that? Yeah, Your Honor, I have several responses. First, I don't think that we should confuse claims of infeasibility with impossibility. Nobody here, or at least on the other side, they haven't established that authorities have literally run out of things to do in the Valley. What they're really talking about, and I think the Colorado brief, Amicus brief says the quiet part out loud, what they're really talking about is the absence of low-hanging fruit. Things that are easy are gone. So it might be a little bit harder, it might be a little bit more expensive, but that doesn't make it infeasible. Also, and this is maybe one of the... What do you contend? I mean, this is a somewhat different point, right? Because there's, I had understood your position to be just sort of as a legal matter, you can't even do infeasibility. But maybe there's a backup plan, which would be to say, even if you can do it, you can't just claim anything is infeasible, which of course I would agree with. But the question here I would have is, what measures do you think should have been adopted that were feasible that the agency has decided are not feasible? Yeah. Well, I have a few examples in record citations that might be useful to your honor. First, maybe one of the most important record citations in this case is from the excerpts of record at page 12, column two. And that's where EPA says, in response to our complaint that they shouldn't have allowed one set of measures to apply to three standards, especially when it was known that one or more is on the verge of being tripped, EPA says, don't worry. If a triggering event happens, California can and will find more contingency measures to adopt. Basically, EPA has said in the record that they don't think it is truly impossible or infeasible for California to do more. But in terms of examples, I'll give you a few. One is that there is an ability that the local air district and state have to regulate indirect sources of pollution. Think of like a warehouse that attracts tens of thousands of trucks, for example. The state actually possesses authority to adopt regulations that affect those big sources that attract a constellation of vehicles, but the valley did not consider strengthening its existing indirect source rule. So the relevant citation for that is the supplemental excerpts of record at pages 150 to 152, where EPA explored the idea for a potential federal implementation plan and said, we as EPA don't have authority to do that, but that is something that the state should consider. The state did not consider it here. Another example is that, as you know, one of the contingency measures before you is about residential wood burning. And the valley and California have strengthened that measure so that curtailment sort of burn bans apply more universally and at a lower level of pollution. But one thing they did not consider is that wood burning devices can be certified as lower polluting. When they're marketed and sold, some are basically uncontrolled and others are certified as producing less pollution, but the valley has consistently refused to consider requiring homeowners to install those lower, cleaner burning wood devices. Another example, and the relevant citation for that is the supplemental excerpts of record at 79 to 86. I'll also note that new technology has emerged to effectively regulate these enormous charbroil grills that some restaurants operate. The valley has an incentive program to try and encourage this technology. There are other jurisdictions that require the installation of that technology, and there are, in fact, I think five restaurants in the valley that have already installed these controls, but the state and EPA declined to adopt that as contingency measures, and that's in the supplemental excerpts of record at 133 to 138. One thing I want to highlight is that one of the key reasons why the state refused otherwise viable measures is it said they couldn't be adopted within a really short one to two-year time frame. Now, that's problematic for a few reasons. First, they kind of control the timing. Also, we have to take their word for how quickly it can be implemented, but let's think about that for a second. There are measures that they acknowledge are available and could be effective, but because they're concerned that they can't be implemented within a one to two-year time frame, they're electing to do essentially nothing. So that is the type of unbounded, unreasonable feasibility excuse that we think should be disallowed. So this is only for nitrogen oxide that we're talking about? All of this is only for that? Right. The focus is really the nitrogen oxides because they did meet the contingency measure emissions reduction standards for direct particulates. So what is the, just before you move on, what is the relevance to this case of the reasonable further progress versus progress debate? Because it seems that the particulate matters met both of those and the nitrogen oxide failed both of those, and we're talking about feasibility as to the latter. So is the former issue just not an interesting issue, but not actually important to resolve here? No, Your Honor. We think that there are two independent reasons why EPA has violated the statute and then a whole host of reasons why they've acted arbitrarily and capricious. But we think the best reading of the statute is that one year's worth of reasonable further progress should be required. So EPA setting up one year's worth of progress as the standard violates that statutory requirement. I hear you on the argument, but what does it matter as to these two, as to the particulate matter and to the nitrogen oxide if one is meeting both and the other is failing both? Well, at the end of the day, the Valley and California have chosen a strategy for addressing fine particulate matter. They want a certain relatively modest amount of air pollution reductions to come from reducing PM2.5 directly. And then really the major part of their strategy is to reduce NOx because NOx can be converted in the atmosphere to PM2.5. So the biggest, most important part of their strategy is reducing NOx emissions. And that's the part where they've allowed the state to skate by with a tiny fraction of the emissions that they would have allowed otherwise. And so due to the feasibility allowance? Well, it's a double-barreled problem. First, they set the target low, right? Unlawfully low. And then second, they undermined even pursuing that target with the feasibility analysis. And we think both of those, both dimensions of their analysis and approval are unlawful. I know that we've jumped right into feasibility. So I want to underscore in the briefs, we believe that the best reading of the statute is that one year's worth of reasonable further progress is the statutory requirement. And just a few quick thoughts on that so you understand the argument. The idea is that reasonable further progress is the ultimate underlying drumbeat of progress that Congress expected states to make through the SIPP planning process. And consequently, since Congress expected that minimum increment of progress every year, it doesn't make any sense at precisely the point when a state is falling behind to allow it off the hook from that level of progress. But being mindful of where I am time-wise, I do want to lift up that there are, apart from the pure statutory questions, and I'll try to get these out quickly, sort of what we conceive of as five different reasons why EPA has acted arbitrarily and capriciously. First, most of the arguments in their brief, or at least many of them, are new. They're post hoc. They weren't offered in the record, and so they don't offer a basis for the Court to uphold the decision at issue here. For example, Your Honor highlighted that maybe we were going to have a debate about one year's worth of reasonable progress. They don't actually really defend that in their brief. Instead, they've pivoted to saying that essentially any quantity of emissions reductions that they can characterize as rational is consistent with the Act. So they've taken a more extreme and unmoored position in their briefing that does not borne out in the rulemaking itself. A second issue, and this is tied to the statute, is that ultimately they're not committing themselves to any statutory minimum for emissions reductions, which renders Section 7502c9 essentially meaningless. As I noted, their bottom line position is that literally any positive contribution meets the requirements of the Act. Consistent with that, we've seen the de minimis reductions in nitrogen oxides. And this is inconsistent with State Farm and other precedent that says an agency's approach can't be inconsistent with a statutory mandate or frustrate the congressional purpose. Let me ask you, and I know we'll make sure you have time for rebuttal. If one were to agree with you on this point, on the reasonable further progress being the better reading of the statute, but one were to also conclude that EPA could give infeasibility allowances and that its reasons for doing so here were sufficient, does the distinction between progress and reasonable further progress matter here? Your Honor, I think, yes, it does, at least to the extent that the Court will be providing direction to further decisions. So what EPA's process here was it set a target and then allowed that target to be eroded by feasibility. We think it's bad enough that there's a new kind of non-statutory carve out for feasibility. But if you lower the target, you're setting a lower bar to begin with. And so it's important both, and we also want to think about a rule that makes sense where states can't or don't claim feasibility concerns, right? At that point, you want it to be clear, did Congress intend for them to shoot for one year's worth of reasonable further progress as the starting point, or can they off the bat shoot for a lower target? But there are two additional reasons why we think the decision is arbitrary and capricious. Your Honor, it sounds like you, Judge Brest, might be thinking about this differently, but we don't see anything in Section 7502c9 that allows EPA to create a new feasibility exemption out of whole cloth. And I would refer your honors to the American Trucking Association's case and this idea that Congress was intentional about feasibility and about costs in particular. Also I want to note, feasibility is not monolithic, right? It could include technology concerns, it could include cost, it could include this new excuse that they've made that if it takes too long to implement, they're just going to do nothing. So ultimately, that brings me to this sort of related point about arbitrary and capriciousness, which is they haven't applied any standards. And here I want to point your honor to the excerpts of record at 286 through 297. Essentially, EPA identifies factors that it said can be considered in making a reason justification, but they're not actually objective criteria. And here's kind of the big tell, and this is another really important part of the record. If you look at their new guidance in the excerpts of record at page 288, they tell states that there's essentially a sliding scale of how much documentation EPA expects to address feasibility. They say if you're trying to excuse most measures and adopt very little, well then you really need to document that a lot. But if you're just trying to get away with a few feasibility exceptions, you don't need as much paperwork. That highlights that this is not an objective assessment of what's truly technologically possible or what isn't or what might be cost prohibitive or not. It's an exercise in papering over things that states don't want to do because they're a little bit hard. The last point I want to make is that EPA did problematically adopt one set of contingency measures for three standards, knowing full well that at least one of those standards was almost immediately going to be found in nonattainment and therefore triggered, leaving the state without sufficient contingency measures for the remaining standards. It was arbitrary and capricious for EPA to acknowledge but disregard that reality and not ensure that the contingency measure requirement would be meaningfully effectuated as Congress imagined across the valley. Okay. Thank you. Let's hear from your opposing counsel. We'll put three minutes on the clock for rebuttal. May it please the Court, Sarah Isfar for the United States. With me at counsel's table is Jefferson Whaling from EPA's Office of Regional Counsel. The Court should deny the petition for review and uphold EPA's approval of the contingency measures. This case begins and ends with the statute. Under Section 7502C of the Clean Air Act, state implementation plans for areas that are in nonattainment must contain extra components. Among those are contingency measures. The text in 7502C9 describes contingency measures as specific measures to be undertaken if either of two triggers occurs. Counsel, hold on, hold on. In 172C9, there is nothing in that section that says anything about feasibility. So where do you get the feasibility, the EPA's authority to read into that section feasibility? So all that's required for contingency measures is that they are specific measures to be undertaken without further action by the state. So the best interpretation of the statute is that Congress delegated EPA authority to fill in the gaps and set contingency requirements. Because EPA is charged with administering the act, EPA has to determine whether the state has submitted approvable contingency measures. So EPA... Hold on, Counsel. The Supreme Court has told us that Congress generally acts intentionally when it uses particular language in one section of the statute but omits it in another. And here, there is no infeasibility allowances in this particular section of the statute. So the absence of infeasibility language in the section, don't you think that that's a deliberate decision by Congress? I would disagree, Your Honor. I think because the requirement for contingency measures is so vague, there does not need to be this explicit feasibility... Sorry, not so vague, so broad, that there does not need to be a specific inclusion of feasibility. The only place when we're talking about control measures in the statute where feasibility specifically mentioned is in the requirement for most stringent measures, that 7513E of the statute. And that's when we're past moderate nonattainment, we're past serious nonattainment, and the area is going to miss its deadline and it needs to get an extension. To get an extension, it has to adopt... Sorry, go ahead. I think that elsewhere also in the statute, Section 172, for example, talks about the availability of feasibility of pollution control measures. In Section 112, they're also talking about things that may or may not be feasible with regards to enforcement of emission standards. In Section 202 of the statute, it talks about feasibility allowances. So Congress is pretty clear when they want to add in feasibility language that they do so in separate sections of the same Act. So I don't understand why your contention is that only in one particular section, there's feasibility language. So contingency measures are control measures like any other. One difference is that they go into effect if triggered. But once they are triggered, they become part of the plan. And if you look at starting with the most basic requirements for a SIP, starting with Section 7410A2, the state has to adopt control measures that are necessary and inappropriate. That's an inquiry that permits consideration of feasibility. But even when they're in nonattainment, and a moderate nonattainment area has to adopt reasonably available control measures, there's no mention of feasibility, but EPA has adopted or issued regulations that define reasonably available control measures to allow consideration of things like cost. And then when we get to serious nonattainment areas, the area has to adopt best available control measures. And there, the court, or not the court, the area, the district, can consider cost. And so it is not reasonable to conclude that the one area in which control measure, the feasibility, are contingency measures, especially when considered in the context of what the purpose of contingency measures are. What's the historical practice of EPA on this issue, on infeasibility? So my friend on the other side is correct that the first mention of a reasonable justification was in 2016 when EPA promulgated the SIP implementation rule. EPA has never provided guidance until it provided this recent new draft guidance on what that kind of demonstration would actually look like. But I would say that EPA's past practice is not as important, even though we think this is consistent with EPA's past practice. EPA's past practice is not as consistent with the text, the statutory structure, and the purpose of contingency measures, which are short-term measures that are actually supposed to get into effect. It's a statutory requirement that's focused on practicalities. Because if there is a triggering event, like the failure to attain or the failure to make reasonable further progress, the statute requires that there are further steps that unfold. The state either has to make a revision of its SIP, it's got to submit a new revision, and that could contain additional control measures that may not qualify as contingency measures, because, you know, they might take longer for the emissions reductions to occur. Or it's got to make an amendment to its reasonable further progress. So contingency measures are like this parallel process that's going forward to ensure uninterrupted progress. And EPA was very clear in the rulemaking as to what the purpose of contingency measures I guess I kind of see this as sort of contingency measures are sort of like the plan B option. Yes. And they still, on the first, we have this plan B. We have these contingency requirements that Congress has set out. But what the agency has done now is not just said that we have this contingency, but we're going to lower that standard. We're going to lower the contingency standards that we have met, and we're going to just require that these areas only produce progress, not reasonable further progress. So I'm not so sure that looking at it from the point of view that you're suggesting makes sense from a statutory construction point of view. Your Honor, I would disagree that EPA has lowered its standards when it comes to feasibility. And I do want to separate the two different aspects of EPA's guidance. There is undeniably a new position that EPA has taken that emissions, that EPA now recommends that contingency measures achieve emissions reductions equivalent to one year's worth of progress. And that's calculated differently than one year's worth of reasonable further progress. And separately, EPA has said that if a state cannot identify contingency measures that meet that recommendation, they've got to make a reasoned justification, which can take the form of an infeasibility determination. And in the guidance, EPA outlined what a state has to do to document what an infeasibility determination actually looks like. So they've got to identify the emission sources. They've got to identify the existing control measures, the ones that are on the books and the ones that are on the way. They've got to look at technological feasibility. They've got to look at costs, you know, the cost per ton. And then they also look at whether the emissions reductions can be achieved in the short term, which is the purpose of contingency measures. And I can give you a concrete example of some of the measures that were considered that EPA agreed were infeasible. One of them is on charbroilers. So we've got these commercial charbroilers. And there is this technology that you can attach a control device, which would lower the emissions. However, that wasn't a feasible contingency measure because there were concerns about whether there could be certifications for fire safety because these control devices have safety concerns with them. There was also controls about whether the technology was widely available and could easily be implemented. Because recall that contingency measures, by statute, have to go into effect without further action by the state or EPA. So they've got to be adopted at the beginning of the SIP. And if they're triggered, there can only be minimal administrative action that takes place for the contingency measures to actually go into effect. So it's not feasible for a contingency measure to require, you know, the installation of cost prohibitive technology or expensive technology. There are supply concerns. And so all of those practical considerations, EPA allowed. And that's consistent with the specific statutory text, but also the structure and the purpose of the statute. And real briefly on the statutory interpretation, Judge Bress, I agree with you that the issue of whether one year's worth of progress or one year's worth of reasonable progress, that's not relevant here. And it's not going to resolve the dispute because the crux of the dispute is really going to turn on feasibility. However, I do want to clarify that they, that if EPA lacks the authority to recommend one year's worth of progress, it also necessarily lacks the authority to recommend one year's worth of reasonable further progress because the statute does not specify a specific quantity. And it is within EPA's discretion so long as it is within the bounds of reasoned decision making. The best interpretation of the statute is that, yes, there's a statutory connection to reasonable further progress and to attainment. And EPA's recommendation has to fall within that. Well, but so what is the limit on that? Because your friend on the other side says, well, we should just kind of have a continuous level of progress. And EPA says, well, it can just be some progress. So how is that going to be evaluated? So EPA did make a concrete recommendation that, that contingency measures should achieve one year's worth of progress. That's a specific concrete amount. And it is different from what EPA has previously recommended. And it is unquestionably a change in position that was undertaken because EPA had a deliberate reconsideration of its longstanding policy after intervening case law, after 30 years of implementing contingency measures, because the initial guidance was adopted right after the 1990 amendments, when contingency measures were first included into the statute. And now, after 30 years, new iterations, new plans, EPA wanted to address concerns. It wanted to address the Barr case. It wanted to address the 2021 Ayer case. It wanted to address feedback it was receiving from states who were informing EPA that they could not comply with their guidance because they were not able to identify contingency measures. So EPA took a look at the statute and came up with a new recommendation. But it was a concrete amount. It doesn't eviscerate the contingency measure requirement. It requires material amounts of reductions. But here, what we're talking about, and just for, and just for perspective. I mean, this issue, this maybe would have been an easier argument for you under Chevron, right? But now we have to interpret it for ourselves. And so the, you kind of look at the statute and say, is there a reason to think that what Congress intended was that EPA could essentially have full discretion to set the standard, at least as long as it was above de minimis? I'm not sure that the statute's necessarily best read for that. And I, and, and, Your Honor, we did not take that position in our brief. We said that there had, so EPA's discretion is within the bounds of reasoned decision making. And so EPA's interpretation has to have some sort of statutory connection, a rational relationship to reasonable further progress and attainment. And the, and the recommendation that EPA made, one year's worth of progress, complies with that. It is a material amount of recommendation. It's a material amount of emissions reductions. If implemented, it complies. But as we said before, it's really not relevant because what's relevant here is feasibility. I do want to make, unless the Court has further questions about the statutory interpretation arguments, I want to make a couple of record-based observations. Contingency measures are exactly what they sound like. They are a Band-Aid to staunch the bleeding while the state regroups and comes up with a new plan or a new revision. And the contingency measures that were adopted here, there was one for wood-burning heaters and one for rural open areas. They, if triggered and adopted, will meet this purpose. They will achieve a material reduction of direct PM2.5 emissions. And for the, the infeasibility determination with respect to the nitrogen oxides, the Valley and CARB followed EPA's guidance. They provided an infeasibility determination. And what's unique here is that EPA also conducted its own infeasibility determination. That's in the supplemental excerpt of the record. It's a lengthy document where EPA goes through all of the different sources. It explains in detail why certain contingency measures are not practical, practical for the purpose of contingency measures. And based on this record, EPA reasonably approved California's recent justification. The last point I want to make about approving contingency measures for separate standards. This is consistent with the statute. There are two triggering events, but EPA has never required that there are different contingency measures for different triggering events and, or for different standards. And there are good practical reasons for that because the same contingency measures can achieve progress for all of the different standards. And here we're talking about, in the earlier case, one of the contingency measures were for the 1997 NAAQS. And since then, EPA has proposed attainment. And so those contingency measures may never be implemented, especially if the court upholds the previous extension. And so at the time in which EPA made its decision to adopt or approve the contingency measures for all three different standards, it did so with open eyes, recognizing that there was a shrinking pool of contingency measures. So unless the court has any further questions, I would urge the court to deny the petition for review and uphold EPA's approval of the contingency measures. Thank you. All right, thank you. Good morning, and may it please the court, Corey Moffitt for Respondent Intervenors. I want to start with just a brief context for the purpose of contingency measures, the Clean Air Act, and then I would like to jump directly to Judge Mendoza's questions about feasibility before answering any other questions that this court might have. So to understand what the Clean Air Act requires of contingency measures, it's important to first clarify the role they play. And as all parties recognize, including the petitioners in their opening brief at page 10, the purpose of contingency measures is to serve as a stopgap should the primary attainment measures not prove as effective as originally planned, to provide interim emission reductions while the primary attainment measures are revised to perform their explicit statutory role of leading to reasonable further progress and then ultimately attainment. The purpose of contingency measures is not to serve as any sort of safeguard to ensure that the overall SIP achieves reasonable further progress. To interpret them as such would obviate the need for Section 7502C2, which explicitly requires that the attainment measures, again separate measures, achieve reasonable further progress, and 7502D, which provides an explicit method to correct a planned deficiency such as the failure to make reasonable further progress. So jumping to Judge Mendoza, your question about feasibility, ultimately that question boils down to whether Congress intended EPA to require States to adopt contingency measures that even EPA acknowledges wouldn't be feasible to implement. Petitioners didn't argue in their opening brief, they didn't argue except through a side comment to the amici briefs, and they didn't argue more importantly in their comment that EPA missed any potential contingency measures. Their argument today is the first I've ever heard of that. And there's actually some really good reasons why the examples that petitioners raised are not feasible. In addition to the charboil example that EPA raised, petitioners also raised the potential for indirect source rules. Why wouldn't California adopt an indirect source rule? There's actually a very good reason. Section 7410A5A explicitly prohibits EPA from requiring indirect source rules as part of a SIP. If that issue had been raised earlier, it could have been taken off the table well in advance of this argument. So as Judge Mendoza, as you mentioned, Section 7502C9 does not mention feasibility. However, neither does Section 7513A, which petitioners concede in their reply brief at 22, allows for the consideration of feasibility in the setting of reasonable available control technologies and best available control technologies. So to that point, whether or not the word feasibility is included is not dispositive. Rather, the role of this Court is to look to Congress's intent. And there's two reasons to understand Congress to have intended EPA not to require infeasible measures. The first, as EPA mentioned, is the take effect language in the actual statute. And I'll leave that there for time. The second one, and I think perhaps more relevant, is the context in which contingency measures are found. As described earlier, contingency measures are not a final safeguard. They are an incremental step in a series of escalating requirements if an area is not able to achieve attainment. Once triggered, they become permanent attainment measures applicable to the next milestone. However, if that next milestone is not met, and as Judge Mendoza mentioned, Section 7509D2 requires that the SIP be further revised to include, quote, all measures that can be feasibly implemented in the area in light of technological achievability, costs, and any non-air and other air quality health and environmental impacts. It simply doesn't make any sense and, therefore, would violate obligation to read the text with reference to the entire statute for contingency measures to requiring feasible measures, but for the subsequent kind of escalated requirements to allow feasibility. In that context, I think it's fair to intuit that this was not Congress's intent. And finally, and perhaps just simply as a reference to our brief, petitioners' interpretation would actually have a potential counterintuitive effect of worsening public health and welfare in the Valley because it would result in a situation in which the state would be left in a zero-sum situation in which it would have to pull or withhold attainment measures in order to bolster the contingency plan. It would have to weaken Plan A in terms of strengthening Plan B. As the entire purpose of the Clean Air Act is to attain the NAAQS as expeditiously as practical, I don't believe that's a logical way to interpret that. I see my time is out. I'm happy to answer any questions or to step back. Okay. Thank you, Mr. Moffitt. We'll hear rebuttal from Mr. O'Brien. So I want to start real quickly with statutory interpretation and then pivot to arbitrary and capricious. So as the Supreme Court recently emphasized in its Loper-Bright decision, a lack of detail or even ambiguity in a statute doesn't mean that it should be regarded as an enormous grant of unbounded delegation to an agency. That's essentially what EPA is claiming here, right? In their brief, they say, well, as long as it is material, which we assume is some positive movement towards reducing emissions, they consider it to be consistent with the statute. But Your Honor should be mindful that as EPA's counsel acknowledged, this is a new interpretation. And so I encourage Your Honors to apply the wisdom of Loper-Bright, including their cross-reference to Skidmore, which says an agency's interpretation, which is contemporaneously issued with the statute and applied consistently, is the one, or at least those are markers of a reliable and likely the best interpretation. Also, in this sort of context of what did Congress mean, we heard counsel for EPA say that essentially contingency measures are a Band-Aid. They're an interim measure. And Congress just wanted something to be done to improve air quality. But there are a few problems with that. This idea that anything is enough is not only unbounded, but it can't, that requirement can't be read into meaninglessness, which is essentially what they've done here by adopting de minimis measures. And that's clear on the face of the record, but also by comparison to other decisions of this Court, like the Committee for a Better Arvin case. But I want to talk some about the logic that we heard from counsel for the state. Contingency measures, everybody agrees, are intended to be a bridge between a primary plan and the next better, stronger plan because a crisis has been revealed where air quality is not being met. So at this moment where planning has failed, does it make sense to let the state off the hook and do barely nothing? Or does it make sense for the requirement to be meaningful and robust to keep the state on a strong, continuous trajectory toward attainment? At the end of the day, this is not just about like numbers and abstract figures, but this is about public health. Real quickly on the feasibility of particular measures, I gave examples in my argument, but our brief and our point throughout is that categorically they've approached feasibility in an unlawful way, both because it's not allowed by the statute, but ultimately because they haven't set up any objective criteria for evaluating what is truly cost prohibitive, what is truly technologically impossible, or what might otherwise be infeasible. And so departing from the statute in this absolutely unbounded, standardist way that they have proceeded cannot possibly but what Congress intended for those circumstances when the public is suffering and when real affirmative action is needed to address air pollution. Great. Okay. Thank you very much. We thank all counsel for the briefing and argument. This case is submitted. That concludes our calendar for the morning. We'll stand in recess until tomorrow.