Oral argument — AIR v. USEPA

Excellent. Thank you, Your Honor. It may please the court. My name is Brent Newell. I am here for the Association of Irritated Residents. After decades of interpreting the Clean Air Act to require contingency measures that provide one year's worth of reasonable further progress, in the case of this ozone plant it's 11.4 tons per day, this court, the circuit, decided bar and EPA subsequent to that changed its interpretation and that is now at issue in this case. I'd like to make three points this afternoon. First, I'd like to explain EPA's finding that the contingency measures were valid in the event of a failure to attain the standard in 2031. That helps set the EPA's interpretation is not reasonable under Chevron Step 2. And second, in the alternative, if it is reasonable, if the court holds that it's reasonable, that EPA's finding is arbitrary and capricious. And finally, I'd like to address the Enhanced Enforcement Activities Program because that strengthening measure, as EPA calls it upon approval, is unenforceable and So first, the attainment year finding. In the attainment year, EPA's interpretation falls apart. In 2031, there is only one ton of surplus emissions that could under EPA's interpretation apply. The nominal air district measure provides one ton, so that's two tons of contingency measure reductions. But bar says that already implemented measures with continuing reductions cannot count. So we're left with one ton. EPA bases its finding on two things. The aggregate commitment, which is an eight ton commitment by the year 2031, and the Enhanced Enforcement Activities Program. Now EPA admits that the Enhanced Enforcement Activities Program is implemented before the attainment year. So if there is a finding of failure to attain, it means that that commitment and those reductions did not get the job done. Same thing with that one ton of surplus. It didn't get the job done. We are left with one ton from that nominal contingency measure. EPA also bases its finding that the Enhanced Enforcement Activities Program supports its approval, but EPA cannot quantify any reductions whatsoever from that program because it is so vague and so discretionary that it cannot be approved as a measure under the Clean Air Act. Basically, Congress wanted the state of California to deliver a plan A, the primary strategy to attain the standard by 2031, and a plan B, the contingency measures. EPA's interpretation after bar conflates the two. There is no plan B except a nominal one ton per day. So let me get into the Chevron Step 2 analysis. The one ton is the paint? The paint rule, yes. And it's a, by the way, it's a commitment that the district made. The district didn't even try to quantify it, and EPA assumed that it was worth one ton. And maybe just to focus the Chevron discussion, the statute says there have to be contingency measures, but it doesn't quantify how much the contingency measures have to achieve. So maybe you can address that point. How do you get around that in saying that what the agency has done here represents an unreasonable interpretation of the statute? Your Honor is correct. The contingency measure requirement does not specify an amount, so Chevron Step 1 does not apply, and the court decides whether to apply it or not. And as we cited in our brief, there are several cases that have applied Chevron Step 2 to EPA's interpretation of the Clean Air Act. The Utility Air Regulatory Group case in the Supreme Court, we cited an NRDC case in the D.C. Circuit, and a recent South Coast Air Quality Management District case also in the D.C. Circuit. And what those cases hold is that courts do not defer under Step 2 to an agency's interpretation that is untethered from the statute and has no connection to the statutory scheme. And here, the statutory scheme is that Congress wanted a Plan B. Congress wanted measures that would kick in if the area fails to attain or meet reasonable further progress. EPA's goes back to almost three-bar standards, which is they want already implemented measures to support their approval of this nominal measure. That's untethered from the scheme because Congress wanted that Plan B. Here, EPA has interpreted it away. It's also inconsistent with the linkage to reasonable further progress. You see, reasonable further progress is defined in the statute to equal 3% or 1 year's worth of reductions. That's why EPA's old interpretation was 1 year worth of RFP. There's that connection between the two requirements. EPA's new interpretation has broken that tether, has severed itself from this idea that if the district fails to hit those milestone years toward attainment, that there would be enough reductions to provide progress. In the attainment year, there's supposed to be reductions to keep the progress going. But here, there's only one ton. Before, our EPA said that there was a requirement, and it disapproved a number of plans, including one we cite in our brief for not making that 11.4 tons per day requirement. EPA's gone from 11.4 to 1. That's untethered from the scheme, and it's inconsistent with the reasonable further progress requirement. I mean, is it your view that the contingency measures have to achieve 1 year or 3% worth of reduction, or is the agency of discretion to decide, you know, we think this is a pretty good plan, and we have this contingency measure, and maybe it doesn't quite reach that threshold, but we want to adopt it anyway. Do they have that discretion? EPA has the ability to adopt a new interpretation. The amount is not in the statute. We agree with that point. But EPA cannot so change its interpretation that it becomes unreasonable and untethered from the statutory scheme. Going from 11 to 1, that's unreasonable by any measure and becomes totally detached from the scheme. EPA could do something else, but it has to be consistent with the scheme. And this provides a segue, Judge Miller, to my second point about EPA's explanation for its changed interpretation. So if you have any further questions about the Chevron Step 2 point, I'd like to discuss the arbitrary and capricious point. Please. Thank you. So EPA, if it adopts a changed interpretation, it must provide a reasoned explanation. It must explain why it is more consistent with the Act. EPA hasn't done that. Again, comparing this to my description of the attainment you're explaining why this new one ton per day, if this air basin fails to meet the standard, is more consistent than the Plan B that existed before and before BAR. So there must be a Plan B here, and it must deliver some reductions. EPA hasn't explained why it's finding about the aggregate commitment or the Enhanced Activities Enforcement Program provides a more consistent Plan B. Those reductions don't exist if the air basin fails to attain. EPA's finding and approving the attainment year contingency measures is arbitrary and capricious because there's only that one ton. It hasn't explained why this is a Congress wants as a backup if this air basin fails to attain, especially in the context that this air basin has failed to attain all five ambient air quality standards by the deadline. There's four different standards, and there's been five failures to attain. This action and explanation that EPA has provided has not explained the basis for finding that there is a valid Plan B or why this is more consistent. Moving on to the point about the Enhanced Enforcement Activities Program, this relates back to my explanation of the attainment year finding because EPA relied on this Enhanced Enforcement Activities Program as a basis to approve the contingency measures as valid in that attainment year setting. But the Enhanced Enforcement Activities Program is unenforceable. It is a promise by the California Air Resources Board to adopt at least one vague, undefined, unquantifiable reduction strategies such as do more inspections, do some audits. Those have zero meaning. They're like the unenforceable goals from Bayview where the state hasn't actually committed itself to do anything. Yes, it's committed itself to adopt a report. It's when you get down to brass tacks, the state hasn't adopted to do anything. EPA could not quantify any reductions. EPA had to concede after its proposed rule and pursuant to airs comments that it couldn't accept that program as a contingency measure because it had no specificity and it offered no controls in the SIF. So one of the core things that makes these programs valid, whether it's a stiff strengthening measure as EPA tried to rebrand this in the final rule, or whether it's a standalone measure or a contingency measure, they have to be enforceable. And under the Arvin case, excuse me, that we cite in my brief, Committee for a Better Arvin, those commitments were enforceable because they had specific dates and specific reductions that the state committed to do. Now, air and Committee for a Better Arvin challenged those commitments to say, Hey, these aren't enforceable. But the Ninth Circuit said they are enforceable because there's that specificity of reductions and dates. And here, this Enhanced Enforcement Activity Program lacks that specificity about reductions. There's just a commitment to exercise discretion. So may I ask, if the plan and the contingency measures, ignoring the Enhanced Enforcement Activities Program, if the plan and the contingency measures were otherwise acceptable in your view, would you have any basis for objecting to the fact that they also had this Enhanced Enforcement Activities Program? I mean, is it the program per se that you're objecting to, or the fact that they're relying on it in approving the plan and the contingency measures? Actually, both. Because the EPA relied on the Enhanced Enforcement Activities Program as a basis for its approval of the attainment year contingency measure. So that's arbitrary and capricious. Okay, if they hadn't done that. Yeah, okay. Yeah, I'm sorry, Judge Miller. Independent of that, even if this court held it was not arbitrary and capricious, that Enhanced Enforcement Activities Program, because it is part of the SIF, because it's been offered as part of the state's control strategy, it has to be enforceable. So that is an independent and separate ground that we are challenging. So they overlap a little bit. If there aren't any more questions, I'd like to reserve the balance of my time for rebuttal. Thank you. Thank you. Mr. Greenberg, you need to unmute yourself. May it please the court, Alan Greenberg, on behalf of the United States Environmental Protection Agency, its administrator, and its regional administrator for Region 9. EPA acted reasonably when it conditionally approved the contingency measure submitted by the state of California in its revised SIF for the 2008 ozone NAAQS for the San Joaquin Valley. The relevant inquiry for this court in this petition for review is whether EPA considered the relevant factors and whether it adequately explained its approval decision. And in this case, it did both. It considered the relevant factors. It considered this court's bar decision. It considered its guidance that recommends the quantity of emission reductions that a contingency measure should achieve. It considered the amount of emission reductions associated with the proposed contingency measure, the architectural coatings exemption. And it also considered, importantly, the amount of surplus emission reductions that are available. Counsel, what happens with what has been approved if the EPA and the state are wrong about the projected surplus level? Let's say that they're far less than the EPA thinks they're going to be. What would happen under the rule, et cetera, as it's now been adopted? If at any of the reasonable progress milestones or at the date of attainment, which is currently at 2031, if the area is not meeting those standards, the EPA is required and the state of California has to submit a new revised plan with additional emission control measures in order to attain the ozone max limit. Also, so then if I understand it and to perhaps borrow Mr. Newell's term, although maybe he meant it in a different way. So if the state and the plan B is to go back to the drawing board, there is no plan B as to, there are no contingency plan Bs that say if the projected surplus is way less than we think, A, B, C, D, and E are going to kick in. It would be if the projected surplus is way less than we think, we got to go back to the drawing board and come up with A, B, C, D, and E. Is that right? That's correct. The architectural coatings exemption would kick in and then the state would go back and submit a revised plan, which would be the case in any event, even if the contingency measures amounted to 11.4. If the progress did not match the milestone numbers or if the area didn't attain in 2031, the state would still be required to go back and come up with new measures in order to reach attainment. Right. But the point of the statutory requirement for contingency measures is that there's supposed to be something that kicks in automatically. And to go back to Judge Bennett's first question, if what you're saying is essentially saying, well, the contingency measures don't need to be very robust because we have these surplus emission reductions. But the only way that the contingency measures get to be at issue is if the surplus emission reductions don't materialize, the plan isn't effective. And so what sense does it make to rely on the likely success of the plan in weakening the contingency measures that the statute calls for, which only are relevant if the plan hasn't worked? First, to clarify one point, the plan does not include the surplus emission reductions. The plan, which the state submitted, has measures that on their own should lead to attainment and should meet the reasonable further progress milestones. On top of all of those measures that the state has submitted on behalf of San Joaquin Valley, you have the contingency measures and the surplus measures, which EPA considered together to determine that the likelihood of a failure to attain or a failure to meet reasonable further progress was low because those surplus measures are in the range of 90 tons per day to 150 tons per day, well above the 11.4 that would be one year's worth of surplus. Does that still mean that the contingency is unquantifiable, therefore unenforceable? Because it sounds that, and I'll use a word or a term used by Mr. Newell also, that this is untethered to the reasonable further progress because how can... It almost seems like instead of an incremental reduction, you're expecting a net reduction and the net reduction includes the surplus plus whatever is achieved by the contingency, meaning you really have no way of measuring whether the contingency meets the mark in the plan. Well, the aspects that EPA considered are quantifiable. The architectural coatings rule is quantified at one ton per day and the surplus emissions are quantified for each year and they range from 90 to 150 tons per year. The contingency measure with the surplus emissions to address the purpose that Congress identified in establishing the contingency measure provision and that is contingency measures aren't anticipated to allow these areas to attain, rather they're used as a bridge. It's something that will allow the state and San Joaquin Valley time, a year in which to submit new additional emission reduction measures. So what EPA is looking at here is the contingency measures plus the surplus emissions which are in the neighborhood of 100 tons per day, enough when considered together to allow the area to continue its progress towards reducing ozone emissions while at the same time is developing a new set of programs that will allow it to have a new SIP approved to attain... But why is it appropriate? I mean, the Committee for a Better Arvin says that if you're relying on measures to reach the target, they have to be included in the plans, that they're enforceable. So why is it appropriate to include surplus measures that aren't in the plan as a basis for making the contingency measures less robust than they would otherwise need to be? The surplus measures are in the plan, your honor. They are part of the set of programs that the state and San Joaquin Valley have identified that will allow them to, if all goes well, reach attainment. The situation is that the state of California has adopted sufficiently robust emission controls for mobile sources, cars, trucks, off-road vehicles, that there's a surplus of emissions, in other words, more emission reductions than are needed to satisfy the attainment plan. So they've got more than they need. Portions of those reductions are set aside in the attainment plan to reach attainment and the surplus is what EPA considered in connection with the contingency plan relating to architectural coatings to determine that there are 70, 80, 90, 100 tons per day of likely emission reductions in addition to what the attainment plan requires and that should keep this area on track to make progress towards its attainment goal. Are there actual deadlines for these enforcement mechanisms to be triggered? Yes, for the milestones, the reasonable further progress milestones, those occur approximately every three years from today through 2031 and if a reasonable further progress milestone is not met, that triggers the contingency measure as well as the obligation to submit a further plan. Council, is it the EPA's view that because it thinks the projected surplus is so great that it didn't even need the one ton per day paint reduction? That's not EPA's view, your honor. EPA, on the particular facts of this case, looked at the one ton per day and at the significant surplus emissions and together concluded that those were adequate to allow progress to towards attainment. And this is tethered, your honor, to the statute. The purpose of the contingency measure is to, as I mentioned, to provide time for the state and the San Joaquin Valley to come up with new additional measures in the event that a milestone is not achieved or the attainment is not achieved at the end of the 2031 period. These surplus emissions combined with the contingency measure serve that purpose to allow the states to have time to bridge that gap. I would like to return to the fundamental point here, which is that there is no requirement for a specific amount of emission reductions associated with the contingency measure. It's not in the statute, there is no reg, there is no guidance that requires that a specific reduction occur. And in that context, it is therefore not, in our view, a Chevron issue. Rather, it is simply a straightforward, arbitrary, and capricious analysis. Did EPA consider the relevant factors and did it provide the explanation for making a reasonable determination? And they did that here. Notwithstanding Mr. Newell's comments, the final rule in detail explained why EPA was taking the position it was in this existence of these surplus emissions. It's also based, in part, on addressing this court's bar decision. The background of this, and turning to a point Mr. Newell made, is yes, EPA historically has indicated that by one year, one ton per year, excuse me, one year's worth of reasonable further progress is the appropriate quantification for a contingency measure. But at the same time, EPA has historically indicated that already implemented measures are appropriate for contingency purposes. With the bar decision of this court, those and EPA in addressing the situation with San Joaquin Valley, appropriately considered the contingency measure that San Joaquin Valley had provided and the surplus emissions which provide the additional basis to uphold the approval of the contingency measure. We, EPA, does not believe that EPA engaged in any new interpretation. The prior guidance indicated that in particular facts and circumstances, the one year of reasonable further progress is not necessarily mandated. And here EPA explained why in these particular facts and circumstances that one year of reasonable further progress was not required of the contingency measure provision solely. Mr. Greenberg, you have referred to it as prior guidance, but isn't it a final rule adopted by the agency? I mean, making it akin to law, not something that's as flexible as I've interpreted you to tell me it is today. Your Honor, as we explained in our brief, it is clearly, in our view, guidance. The 1992 document talks about a preliminary interpretation that is intended to be guidance, to provide guidance to the states and to other interested members of the public about how EPA preliminarily plans to interpret the new provisions from the 1990 Clean Air Act amendments. EPA has never incorporated that preliminary guidance in any rule. They have not adopted it in any other guidance that was in the form of a requirement. So it remains guidance that EPA applies in its discretion based on the facts and circumstances of a particular case. I do want to quickly turn to the SIP strengthening measure, the enforcement program that was also challenged by AIR here. And this is a fairly simple, straightforward. Just be cognizant, Mr. Greenberg, that you're into Ms. Fierro's four minutes. That's fine, but just... Oh, I apologize. I thought that was my 16. Well, did I get that wrong? I thought we were starting from 20. Yes, we started at 20. Yes, so you are in Ms. Fierro's time. I apologize, Your Honor. Just very quickly, the point of the enforcement program is that only two aspects of the enforcement program are incorporated into the SIP. Those portions incorporated into the SIP are enforceable. The state can adopt any number of other additional Clean Air Act measures it wants that EPA doesn't have to incorporate into the SIP. And if the provisions are not incorporated into the SIP, the Clean Air Act doesn't require that they be enforceable. If the court doesn't have further questions, I will yield the rest of my time. Thank you. Why don't we put four minutes on the clock for Ms. Fierro? All right, whenever you're ready, Ms. Fierro. Thank you, Your Honor. Good morning. Good morning, everyone. My name is Jessica Hafer Fierro. On behalf of Intervenor Air District, San Joaquin Valley Air Pollution Control District, we were joined in the briefing by South Coast Air Quality Management District. The adopted, enforceable, and quantifiable control measures contained in San Joaquin's ozone plan achieved compounding emissions reductions past the attainment year. And this provides the context and assurance that the aspirational guidance of one year's worth of RFP need not be achieved solely by the bar-type contingency measure. These are not just surplus emissions reductions, but are emissions reductions that are growing over time through the continued implementation of adopted measures. San Joaquin's 2016 ozone plan reduces ozone precursors of oxides of nitrogen from 340 tons per day in 2012 to 132 tons per day in the 2031 attainment year. That represents a 62% reduction. This achievement is largely driven by the California Air Resources Board's 27 mobile source measures, which are fully adopted and enforceable, but not yet fully implemented. These are cutting-edge, transformative measures that are vehicles, on-road heavy-duty trucks, off-road equipment, and much more. So let me ask you the same question I asked Mr. Pickford, just to see if your answer is the same. So what happens if the state is wrong about the projected surplus? Is it that simply you have to look at whether new measures are necessary, or is it something else? Well, these measures are fully adopted, and they're fully enforceable. So there's every reason to believe that we can rely on these emissions reductions, that these control measures have already happened. They're just being implemented over time as the fleet turnover occurs and cleaner vehicles make it to the road. Having said that, I agree with Mr. Greenberg that if there turns out to be a failure, the proper mechanisms provided by the Clean Air Act would be that you would implement your contingency measures and you would be required to revise your state implementation plan. As fleet turnover continues under CARB's mobile source control measures, the emissions reductions are driving continual improvement in San Joaquin and South Coast until attainment of the 2008 8-hour ozone standard and beyond, as each of these measures achieves escalating growth in emissions reductions beyond the attainment year. Again, these are not just surplus reductions, but are new emissions reductions that will be achieved beyond the attainment year. These measures would, in fact, provide additional air quality benefits should the region experience an attainment failure while the region revises its SIP. These measures also provide the air quality benefits our regions need under the more stringent 2015 ozone standard, which is not addressed in the plan at issue today. Could you briefly address your argument on standing? In particular, what I'm wondering is, is it your view that nobody ever has standing to challenge the contingency measures because it's speculative as to whether they'll even come into play at all, or is it something more specific to these plaintiffs? That is the gist of it, Your Honor. It is worth remembering that contingency measures are standby emissions reductions that might never be implemented. They are only implemented in the event of a future failure to reach attainment to provide the additional air quality benefits while the SIP is revised pursuant to the Clean Air Act. So it could be that these contingency measures are never needed. Contingency measures that, by definition, shouldn't be implemented yet can't possibly be causing the current air quality concerns petitioners have, and a ruling in their favor about these contingency measures can't possibly redress the air quality injuries they're experiencing right now. So do you think we were wrong in Barr? I mean, I realize we didn't address the jurisdictional question, so it doesn't bind us on that, but do you think Barr was wrong and we should have dismissed that for lack of jurisdiction? Very quickly, the Barr court did not speak to fully adopted measures achieving compound growth in emissions reductions past the attainment year, and the holding in Barr does not preclude consideration of escalating emissions reductions as context when you're approving the other measures in the SIP as with the contingency trigger. And Barr was also silent on the quantity of emissions reductions required to have a contingency trigger. Thank you. Mr. Newell, your rebuttal. Thank you, Your Honor. I'd like to begin by addressing the question that you offered and asked Judge Bennett, that what happens if there are less reductions than planned? Basically, what happens if we get to the attainment year and there aren't enough of these surplus reductions and there's a failure to attain? Well, what happens? And both Mr. Greenberg and Ms. Fierro gave essentially the same answer, which is the paint contingency measure will kick in, that's one ton, and then the Air District will go and develop a new plan to submit to EPA to attain the standard by the new deadline. Now, the purpose of the contingency measure requirement is so that between the failure to attain and the new plans adoption and submission to EPA that there are reductions that keep pushing emissions down, that help improve air quality. Their answer to your question helps reveal how unreasonable EPA's interpretation is, because at that point, these surplus reductions don't exist anymore. They failed to get the job done. There's been a failure to attain. You only have a one-ton paint rule, a small paint can rule that is supposed to help get continued reductions until that new plan is adopted. That's how unreasonable this interpretation is, because Congress wanted a plan B. Congress wanted reductions to kick in and work during that time period between failure to attain and the new plan. So that is exactly why this interpretation is so unreasonable under the statute, why it's untethered from what Congress wanted as a plan B and why it should be vacated or remanded under the Chevron Step 2 analysis. I'd like to also address a question that Judge Pearson asked, which is that she asked about, well, whether it's untethered from the scheme, whether this interpretation is untethered from the scheme when these are projections. These are, this is the hope and the plan for scenario that air quality planners do when they draft the plan. And the point of these contingency measures is if those reductions do not materialize. They're not surplus if they don't exist. When those milestone years happen or when they don't materialize, that's when the attainment year comes around. And there's been some discussion about BAR through several questions. The BAR court looked at contingency measures from already implemented measures and said that those continuing reductions don't count. So EPA agrees. And EPA has advanced and actually conceded in the rule that it could not rely on already implemented reductions. Ms. Fierro's point about this sort of continuously compounding emission reductions somehow are different and are satisfied by BAR. That's not the case because measures, in order to be consistent with the contingency measure requirement in the Clean Air Act, they're supposed to be new and they're supposed to be undertaken if there is a failure to meet one of these triggering requirements. These surplus reductions, which are projections, don't meet that standard. And Mr. Greenberg is arguing that this is an analysis the court undertakes only pursuant to the arbitrary and capricious standard of review. That is wholly unsupported by any case that the EPA has cited in its brief. The Chevron standard of review applies regardless of whether EPA has changed in its interpretation or not because the court decides whether EPA's decision is grounded in the statute. It decides whether the statute supports it if it speaks directly to the issue. The court also decides whether it's a reasonable interpretation. So Mr. Greenberg's argument that we only review under the arbitrary and capricious standard is inconsistent with utility or regulatory group, for example. And I would also urge the court to view this as a changed interpretation situation. For years, EPA has had this interpretation that SIPs were required to have one year's worth of if this requirement has been adopted as a regulation, as Judge Pearson asked. This has been the agency's longstanding interpretation. Turning briefly in my final minute to standing, I would like to point out that the Air District did not respond at all to the point that we raised in our brief that where there's a failure to implement the remedial scheme, there is always causation. The Ninth Circuit has addressed challenges to state implementation plan approvals in several cases, at least a dozen. Many of them the court has proceeded to the merits without even entertaining standing because it was so obvious that there is injury and that if the agency is not implementing the remedial scheme, then that is causation and an order and a remand from the court where this agency is going to implement the remedial scheme is addressability. We've addressed these points in our brief and Air Hutt's standing. Thank you. I thank counsel for their helpful arguments. This case will be submitted. And with that, we are adjourned for the day. Thank you, Your Honors.