Oral argument — State of New York v. EPA

Case No. 17-1273. State of New York et al. Petitioners v. Environmental Protection Agency et al. Mr. Frankel for the petitioners, Ms. Shea for the respondents. Ms. Shea for the petitioners, Ms. Shea for the respondents. Mr. Frankel, we'll hear from you. Good morning, your honors. I may have pleased the court, David Frankel, on behalf of the state petitioners. I'd like to reserve two minutes for rebuttal as well. EPA may not arbitrarily deny to the petitioning states a remedy that Congress tailored to this set of circumstances. EPA's decision not to expand the ozone transport region, despite the undisputed satisfaction of the statutory trigger here, should be vacated because EPA's justifications are unsupported by the record, refuted by context and history, and generally fail to display reasoned decision-making. EPA asserted that other statutory tools were sufficient to effectively address remaining problems of interstate ozone transport, but EPA's history of partial remedies and extensive delays in using these tools refutes the assertion that they alone can redress the harms necessitating this petition. EPA's transport rule under the Good Neighbor provision, on which it heavily relied, provided only a partial remedy, resolving the obligations of the region. And that rule, addressed to only a single type of emitting source, took nine years to put in place after misdetainment deadlines from downwind states. Similarly, while EPA suggests that Section 126 petitions are an effective way to address the regional ozone problem, EPA has delayed and denied all recent petitions. EPA fails to reasonably explain why these other measures, which have provided only partial remedies on delayed timeframes, are themselves sufficient. And while EPA points to progress in reducing ozone levels as a justification for its denial, EPA failed to provide a reasoned explanation for why those trends were likely to continue. EPA relied without analysis on its unquantified expectations about the effects of various rules, but at the same time, EPA failed to explain how those rules were likely to continue to have those same expectations, given the agency's contemporaneous regulatory rollbacks. And to give one concrete example here, simultaneously with its decision to deny this petition, EPA proposed a glider rule that is projected to increase NOx emissions by 300,000 tons. That's in contrast to the 60,000 ton reduction of the entire transport rule at issue, swamping it with five times the amount of increase, as EPA's central rule on which it relies is likely to lead to increased reductions. Can I ask you a question about the transport rule? Or I guess that's the same thing as the cross-state air pollution rule? The cross-state update, yes. The regulatory hook that EPA used for that was the good neighbor provision in the Clean Air Act, correct? Yes. And that's the same good neighbor provision that requires state implementation plans to account for cross-state air pollution, right? Yes. So when the rule was enacted, had EPA acted in a timely fashion on the state implementation plans that had been submitted by the states that were subject to the rule? No, EPA had not. EPA had extensively delayed in denying the SIPs and in promulgating compliant SIPs. That would satisfy the good neighbor provision. And ultimately what we have is the CASPER update rule, which itself is only a partial remedy that, again, only solves one of those 22 states' obligations under the good neighbor provision. Though we have attainment deadlines being missed in the meantime by downwind states. We've had the 2015 deadline passed. We've had the 2018 deadline passed with the New York metropolitan area still unable to meet its obligations. And we have upcoming attainment deadlines that absent reductions in ozone transport from upwind states, additional noncompliance is likely. What do we have in the record with respect to the missed deadlines by EPA of either acting upon SIPs or promulgating FIPs for noncompliant SIPs? Well, we cite all these. I mean, these are all judicially noticeable. These are all actions that occurred within the Federal Register. And we explain how the EPA has delayed in fulfilling its obligations under these and has not ultimately promulgated compliant FIPs for any of the good neighbor provision obligations of upwind states with very limited exceptions. But on your view, if EPA were more vigorously enforcing the good neighbor regime, would that solve your problem? As a conceptual matter. Yeah. If EPA had promulgated a complete remedy under the good neighbor provision, we wouldn't be here right now. If EPA, instead of doing a partial remedy that addressed a single attainment deadline for a single kind of source, and rather than fixing the obligations of one upwind state had addressed all of the obligations of the states named in the petition, then we wouldn't be here. But that's not the case right now. And what we have shown is that the statutory trigger for this provision, that there's significant contributions from upwind states that are leading to violations in downwind states, continue to be met. And if EPA had been more vigorous. So that's not enough, right? The statute's discretion to EPA and that it uses the word may. So there's no question that EPA does have some discretion here. But EPA acknowledges that may isn't some grant of unfettered authority. I mean, it briefs this case as if it sort of can do whatever it wants because of the may. But the case law is pretty clear that may, even though it grants discretion, is still subject in agency's action to judicial review. Well, it smacks of prosecutorial discretion. The prosecutor can't bring a case unless he's got the evidence to indicate that a crime has been committed, but nevertheless has discretion not to bring the case. And we're dealing here with enforcement devices. The 126 petition, the good neighbor provision, and even the transport region are all enforcement devices to try to get to a particular attainment for whatever the NAC is. So why isn't this comparable, at least analogous, to the vast amount of literature in cases on prosecutorial discretion? Well, there are prosecutorial discretion decisions that are ultimately unreviewable, and that's the Cheney line of cases. But I would direct this Court to the Supreme Court's decision that actually came out just yesterday in the Weyerhaeuser case. And the Court confirmed that when a statute uses the word may to grant an agency the discretion to take an action or not, that word does not, and I quote, segregate his discretionary decision from judicial review. And the decision will still be overturned if the agency fails to consider the relevant factors. And here we have several items that constrain the agency's authority. But to say that the action is reviewable is one thing, and they haven't argued to the contrary, but still Judge Randolph's point has a lot of force, at least to the extent of suggesting that any review would have to be very highly discretionary. Well, there is some level of deference here. There's no question. But there are really two bounds that do provide standards against which to address the decision. And one is the APA, which generally requires reasoned decision-making. So when the agency provides its stated justifications, which they've done here, we've shown that those stated justifications ---- Matt, does the APA apply? Yes, the APA does apply. Are you sure about that? Well, the APA applies to petitions for rulemaking where someone uses the APA as a generalized decision. There's a provision in the Clean Air Act that specifically says that the APA does not apply unless otherwise noted. And I'm just wondering, I haven't looked this up. So are you familiar with that provision? I'm not immediately familiar with that. I'll tell you, you know, the Chevron case itself was a case that was decided under a provision in which the APA did not apply. Well, I would say that there's consensus between the parties here, that the agency needs to rationally articulate the facts and policies underlying its decision, that it needs to display reasoned decision-making, and that it needs to act consistently with the statute. And our point is that if you look ---- Right. But on your view, this is an APA petition for a rulemaking. Well, in our view, this standard for review should actually be tighter here. Because rather than some generalized interested person seeking some random relief, the statute here gives the states a specific procedural right. But this is a petition, on your view, this is a petition for a rulemaking under the APA. Well, it's a petition for rulemaking under the specific provision of the Clean Air Act. And that is yet another context in which our review is especially deferential. I don't contest that there is deference that's appropriate. But the agency still has an obligation to engage in reasoned decision-making consistent with the statute. If you go through point by point what the agency has provided here for its reasoning, for example, the existence of these other statutory tools that by themselves have proven not to be sufficiently effective at resolving the problem. That if you look at EPA's explanation for why it expects trends to continue, that conspicuously omits its simultaneous actions that are nearly certain to undermine those very trends on which it relies. That the agency cited, well, of course we want to address this problem, but we have resource constraints. But unlike other situations in which resource constraints really can be a legitimate reason, like these wildlife guardians or defenders of wildlife cases where agencies articulated with specificity how resource concerns would prevent them from engaging this part of a problem, and it was better to focus on the whole solution. Here we have a solution that actually requires very few agency resources, if any, to put in place. Its provisions of the OTR region tools are largely self-executing or state-driven. So when EPA says, oh, we don't have the resources to do that, that's a point that's actually refuted by the record and the operation of those tools. EPA also relied on this idea that the state needed to submit more information to prove its case, that there were some sort of analytical gaps. And not only were the states not fairly unnoticed of that, but that the states did demonstrate that there were significant ongoing violations downwind because of upwind ozone transport. And having made that showing, if EPA believed that some additional sort of cost-benefit analysis was necessary to determine what the best way to approach it, whether the ozone transport tools or some other tools, which would be cheaper, it's incumbent on the agency to do that. I mean, the states aren't even in a position to be able to have that sort of information about upwind sources and upwind emissions that would enable that sort of analysis. But ultimately, it's these ozone transport region tools that are ozone-specific, that are the most tailored tool to the ozone problem itself, that they are the fastest and least resource-intensive way of addressing the problem. And EPA, we're not saying that EPA lacked discretion, but we're saying that EPA had to grapple with the technical merits of the petition in a way that it didn't, and that it had to provide reasoned, rational decision-making for why some choice of other tools over why a decision-making tool would work. That the reliance on other continuing trends was reasonable, and that in this instance, it failed to meet that burden. In addition, there's the Clean Air Act, which also sets bounds for EPA's discretion. That the overarching mandate under which EPA manages the Clean Air Act is a responsibility to ensure that states can meet compliance and meet compliance by specified attainment deadlines. And EPA failed to acknowledge at all how its use of these tools or the other complementary tools would enable it collectively to be able to meet attainment by the deadlines set forth in the Clean Air Act. And so, in addition to the problems with EPA's reasoning, you have EPA's failure to grapple with the statute that gives it authority to administer these separate programs. At the point of denial, it's undisputed that both New York and Philadelphia, these are regions involving six states and seven states, and about 25 million people were measuring violations. More recent validated air quality data shows that both New York and Philadelphia are measuring even higher nonattainment levels than anticipated. And the modeling on which EPA relied here, the CASPER update modeling, showed projected continued nonattainment for the largest metropolitan area in the country indefinitely. And one last point before I reserve the remainder of time for rebuttal is that one other major factor that EPA failed to consider here was the import of the 2015 maps. Having opened its inquiry up to a variety of nonstatutory factors, one critical factor is that EPA has already determined that 75 parts per billion is an insufficient level to protect public health. And that even more and greater regulatory solutions are necessary. And this solution is the one most tailored to the ozone problem. I don't believe the Transport Commission has directly weighed in. But I will say that the Transport Commission is also one other entity that gives this a unique flexibility, that unlike the other solutions, it does allow the states to work collaboratively to promote further approaches to reduce the ozone problem. And with that, I reserve the remainder of my time for rebuttal. Ms. Shea? Good morning. May it please the Court, Sonya Shea on behalf of EPA. With me at Council's table are Daniel Schramm and Stephanie Hogan. EPA's denial of the state's petition to expand the ozone transport region should be upheld because the statute provides the agency with discretion, and EPA reasonably explained its basis for the denial of the petition here. The standard of review for the agency's decision not to initiate rulemaking is at the high end of deference. All that is required is that EPA explain the facts and policy concerns it relied on, and that those facts have some basis in the record. EPA met that standard here. Do you believe that the State Farm Standard applies to the reasoned decision-making here? The State Farm Standard was, I believe, in the context of a rulemaking, whereas here we have a denial of a petition to initiate rulemaking. And so while the EPA standard, if that is what Your Honor is getting at, does apply for the standard of review, the standard of review is even at a higher level of deference when the agency is denying to initiate a rulemaking. Well, I guess what I'm trying to get at is the State Farm Standard that requires there to be essentially a correct assessment of the facts by the agency, and that the agency consider all of the important aspects of the problem. Does that part of the State Farm Standard apply to our review of what the EPA did here? Well, I think what's more applicable is the standard of review that was articulated in Wild Earth Guardians and in the Defenders of Wildlife cases, where both of those cases spoke directly to the same issue that we have here, where the agency is denying a petition to initiate rulemaking. And how do you think that that standard differs from the State Farm Standard? I would have to go back and look directly at the State Farm Standard again, but my understanding is that here EPA did provide the facts in the record as needed to support its recent decision, which it explained, and it also expressed its policy concerns for why it found that expansion of the region at this time was not appropriate. So I'm not sure if the standards are... Let me tell you what my concern is with the EPA's position. Just lay my cards on the table here. Throughout the final decision, EPA talks about the fact that we don't need to expand this OTR, because we have good neighbor provisions that are being implemented. And they use words like efficiently, effectively, and they talk about the state implementation plans, and then if the state implementation plans aren't sufficient, then there can be federal implementation plans. And then the kicker is towards the end, and this appears at page 51-248 of the Federal Register in the far right column. I'll let you get that in front of you. 51248. The far right column. This is their commenter saying, but there's lots of problems with inaction, delay, and failure of the EPA to actually implement SIPs and the good neighbor provisions therein. And the EPA's response is, well, we appreciate the time and resources needed for the agency and states to take action to address interstate transport obligations. And then they say they disagree that expansion of the OTR would be the faster or more efficient way to deal with this problem, because addressing the good neighbor obligation is required of all states following NOx promulgation. And then they say thus, the CAA Section 110A2B1I process on its face provides a faster timeframe for implementation of interstate transport requirements for a new NOx than application of OTR requirements. All of that is predicated on the fact that, on the assumption that SIPs would be approved or disapproved within 12 months, which is what the statute requires, and that if a SIP is disapproved, that a FIP would be implemented within two years, as the statute requires. The problem is that the EPA never meets those deadlines. And it's replete in the record, and it was replete in the comments, that those deadlines don't get met. States have to, parties have to sue to get the EPA to act on SIPs, sue to get the EPA to promulgate FIPs. But the EPA doesn't mention any of that. I don't see one word of that anywhere in this rule. Can you show me anywhere where EPA discusses those delays? Anywhere? Well, Your Honor, I'm not sure I have an exact record site, but I do think that EPA does discuss the delay that was encountered with the EME Homer City litigation. But that doesn't have anything to do with the EPA's obligation to approve or disapprove a SIP and to promulgate a FIP. And the EPA is relying upon that process and those deadlines that are part of that process to deny the petition here. As if, you know, that process is fine, it's efficient, it's effective. When it's anything but efficient and effective. I mean, this seems to me like the EPA, you know, we're going past the graveyard here. And not accounting for, as State Farm would say, we're not considering an important aspect of the problem. I mean, we wouldn't be here if EPA acted in a timely fashion with respect to SIPs, right? Well, Your Honor, yes, there has been some delay in EPA's action on SIPs and issuing FIPs. But the CASPER update rule that issued the FIPs there was not actually late. It was less than two years after the SIP disapprovals in that case. And EPA has been working with the Good Neighbor provision for over 20 years. Were the SIPs disapprovals timely? Not sure. Your colleague is shaking her head no. Let the record reflect. I'll use my district court training to make that finding. I'll allow you to confer with her in case you want to dispute that finding. No, I think that's probably accurate. Can you tell me, as a practical matter, what the difference is between adding the states to this transport region and simply going ahead with 126 and the Good Neighbor provision? Yes, Your Honor. So with 126, it's a more tailored process, as is the Good Neighbor provision, to allow EPA to focus on the pollutants that will most effectively reduce ozone pollution in the areas where those reductions are needed. So EPA has consistently found, through its Good Neighbor rulemaking provisions, that oxides of nitrogen are the most effective for reducing ozone transport on a regional scale. And the Good Neighbor provision allows for EPA to use that targeted method of addressing those emissions and sources of those emissions that can most cost-effectively reduce their emissions. So, for example, there's a smelter in Ohio at a steel mill or whatever, and so under the Good Neighbor provision, EPA can concentrate on the ozone precursors that are emitted from that particular entity or whatever company or steel mill or whatever. Is that what you're saying? I believe that, yes, EPA can focus on sources. I think the process of how EPA does this is a bit complex for each individual source, and the states have some leeway in determining the budgets for each state. But technically, yes, EPA can focus on the sources that... So in that case, the state, let's say Ohio, would be subject to all of Section 7511C's mandatory controls, which are primarily targeted at volatile organic compounds. And as EPA has shown through its rulemakings, over four successful rulemakings that it's had, these volatile organic compounds are more effective at reducing ozone levels locally and in urban areas, whereas oxides of nitrogen are more effective on the regional scale. And so all of the state of Ohio would have to apply those controls for... The VOs that basically control the volatile organic compounds as opposed to nitrogen oxide? That's correct, yes. So the volatile organic compound controls that are in Section 7511C would apply statewide to all of the major sources of those volatile organic compounds in the state, as well as the implementation of the Enhanced Vehicle Inspection Maintenance Programs throughout the entire state area where those programs are applicable. But Ohio already has to have an inspection and maintenance program, doesn't it? They do have an inspection and maintenance program, but not in all of the counties that the expansion would require for them to include. And so I believe in their comments they mentioned that it would cost... Give me just a moment. But it would cost several million more dollars to expand their Enhanced Vehicle Inspection and Maintenance Program, according to Ohio, than what they're already implementing. Just on the inspection and maintenance, how does EPA enforce that requirement? I believe that... I'm not exactly sure how EPA enforces it, but I know that it would be a provision that would be implemented through the SIPs that the states would need to submit for the ozone transport region compliance. Well, the two ways were, one is to cut off federal highway funds, and the second was for the federal government to take over and promulgate its own implementation plan. Okay. Yeah, so in essence it would be a large expansion of controls, primarily for volatile organic compounds, which may not be necessary to actually reduce the emissions in the region or may not have a very large... Is that a scientific statement you're making, that the transport downwind is basically of nitrogen oxide as opposed to volatile organic compounds? Well, that's something that EPA has made that conclusion scientifically through each of its four good neighbor rulemakings. And there are studies that EPA has relied on that have backed that up as well, whereas volatile organic compounds can achieve reductions, but are more so at the local level than the oxides of nitrogen, which really are more on the regional scale and are the most effective on the regional scale for controlling ozone pollution. Let me ask you a question about how Section 126 petitions work. Sure. What is the interaction between a 126 petition and a state implementation plan? In other words, does the state implementation plan have to be submitted and then approved or disapproved by the EPA before a 126 petition can be brought? I don't believe the timing matters, no. So a 126 petition could be brought at any time, as far as I'm aware. So a neighboring state can say there's a problem with a source in an upwind state and file their 126 provision notwithstanding whatever the good neighbor provisions in the state implementation plan say. Right, yeah. And then my understanding is that if the EPA granted the Section 126 petition, then the state would then amend its SIF to ensure that those controls were being implemented on the source. So if a state implementation plan has been approved for, let's say, Ohio, and it requires a certain set of actions, doesn't require other actions, and a state within the OTR, let's just say Pennsylvania, I think Pennsylvania's within it. It is, yes. Says, we want Ohio to do X. Is it a defense for Ohio to say, well, our state implementation plan has been approved by the EPA and it did not require us to do X? Well, I think how the merits of the 126 petition would be evaluated would be that EPA would assess whether or not the evidence that Pennsylvania submitted showed that the threshold in Section 126 would be met by that source. And so if EPA made that finding, then certainly Ohio would have to do more. All right. Any other questions? So I think in sum, it's really important to look at the entire context of the air transport trend in this case. And EPA has had a successful history through its Good Neighbor rulemakings, through reducing ozone levels, and through the improvements in air quality through other rules. Its expectation that that trend was likely to continue is based on actual modeling that EPA conducted out to the year 2025 that shows that we are projecting to see improvements in air quality, even notwithstanding the rules that petitioners raised. EPA has authority to set up other regional transport systems, whatever. Has EPA ever exercised that authority? To create other transport regions? Right. No. No. So do you disagree that EPA relied upon timely implementation of the state implementation plan provisions of the Clean Air Act as a reason for denying this petition? I do disagree with that. I think it was more in the context of saying that the ozone transport region would not necessarily be as quick to implement as the Good Neighbor provision, and that EPA was looking at that in the context of the issues that petitioners had raised. It was not an independent basis to say we will for sure be more timely. The Good Neighbor provisions that they were talking about were the Good Neighbor provisions that are in state implementation plans or federal implementation plans, as if the state implementation plan was deemed inadequate. That's right. And it's important to note, too, that Section 7511C's controls would also be implemented through state implementation plans and through that same 110 process of EPA approving the plan or disapproving the plan and possibly eventually issuing FIPs. So back to my original question. EPA relies on that process and says it's effective and it's efficient and that it's working, but EPA doesn't mention at all that there are blown deadlines, and not just by months, but by years and years and years, of review of the state implementation plans and of implementation of federal implementation plans in lots of those upwind states. EPA just doesn't mention it at all and writes this report as if all of that is working fine. Well, I think EPA does acknowledge the delays, but ultimately the process is working in the sense that it is creating effective reductions in emissions and it is lowering the levels of air pollution both within and outside of the region. All right. Well, if there's nothing else, I'll let you conclude. Okay. So, in sum, EPA reasoned that continued use of the Clean Air Act's other provisions would better address any remaining ozone transport problems in the region. EPA found that the levels of ozone pollution had been trending downward and could be anticipated to continue to do so and that any remaining problems in the region were few and relatively isolated. And in light of the dramatic improvement in air quality and other available mechanisms of addressing the few remaining problems, and imposing Section 7511C's controls throughout the upwind states would not be warranted under the circumstances. Thus, the agency reasonably decided not to grant the petition and we respectfully request that this Court uphold EPA's decision. Thank you. Did Mr. Frankel have any time left? You have two minutes. Thank you, Your Honor. I'd just like to make a few brief points really. Judge Wilkins, you asked about evidence of delay in the record. I would just draw your attention to pages 340 and 341 of the Joint Appendix where we do reference those delays as well as we extensively talk about, there are other places in the record, we extensively talk about them with citations on pages 14 to 17 and 39 to 44 of our opening brief. On the point about timeliness, one of the real advantages of the ozone transport region expansion is that within nine months, newly added states are added and have to include in their SIPs all of these provisions. And this is unlike the Good Neighbor provision where it takes nine years for the EPA to develop modeling and count emissions budgets. And unlike that, it would give immediate relief to the states and that's a factor that EPA simply failed to give enough consideration to. The idea of 126 petitions, are they more tailored? In theory, they're more tailored. But in practice, as we've seen, neither 126 petitions or the Good Neighbor provision actions are sufficient to redress the ongoing harms that petitioning states have suffered under the 2008 NAAQS for 10 years plus. And as we look forward to the 2015 standards, ongoing non-attainment looking forward. So another point, Council spoke about VOCs and NOx and it is clear that NOx of the two are the larger contributor to interstate ozone transport, but it's also undisputed that both NOx and VOCs play a role and that reducing VOC emissions would also reduce interstate ozone transport and that the transport provisions contain tools addressing both NOx and VOCs. There's really nothing on the record to suggest that VOCs would not... What's the largest source of nitrogen oxides? I don't know the largest source of nitrogen oxides. I will say that one thing that the implementation of the transport provision tools would do was it would impose on all major stationary sources in upwind states requirements for reducing NOx emissions. And these are sources that... Are they for mobile sources as well? I don't believe for mobile sources, for stationary sources. And these stationary sources are significant contributors. And these are also sources where the downwind states have already implemented a host of regulations. For 30 years, the downwind states have lived with these controls. They're not unduly onerous or burdensome. And also because we keep missing attainment deadlines, we keep having to impose even stricter, even more costly controls on downwind sources while upwind states named in the petition haven't in many instances had to enact some of the most basic controls. I would conclude by just saying again that it's already been 10 years after the 2008 ozone acts. The states believe that they're entitled to this remedy and that EPA's conclusions and EPA's justifications don't withstand scrutiny. I would ask the court to vacate the EPA's decision. Thank you. Thank you. We'll take another round of questions.