Oral argument — American Public Gas Associatio v. DOE

Page number 20-1068 et al. American Public Gas Association petitioner versus United States Department of Energy. Mr. Day for the petitioner Spire Inc. and Spire Missouri Inc. Ms. Wiener for the petitioner Air Conditioning, Heating and Refrigeration Institute. Mr. Starcher for the respondent. Ms. Wu for the intervener Natural Resources Defense Council. Morning, counsel. Mr. Day, please proceed when you're ready. Thank you. May it please the court, Bart and Day for petitioners. I'd like to address two related reasons why the court should set the rule aside. My co-counsel will then address reasons why vacate or is warranted in this case. It's undisputed in this case that DOE could only adopt the standards at issue if it made a determination based on clear and convincing evidence that the standards were economically justified. DOE admits that its justification for the standards was inadequate. And it was. Not for the reason of inadequate explanation, as DOE now suggests, but because of a basic failure to engage in recent decision making. DOE claimed that the standards would provide economic benefits for consumers. And it based its economic justification for the standards specifically on that claim. However, there were obvious reasons to doubt the basic premise that the standards in this case could provide economic benefits. Commenters argued that there was no basis to conclude that economic benefits would result. And DOE offered no information or argument to the contrary. It did provide an analysis, and its analysis showed economic benefits only because it was based on a facially unreasonable assumption that grossly overstates the potential for standards to provide economic benefits to consumers. Commenters challenged that assumption as completely absurd and unsupported by any evidence whatsoever. And DOE did not even attempt to justify it on the merits. That's a failure to engage in recent decision making. I'm sorry. Can you just be a little bit more specific? So what assumption are you talking about? Is this the imperfect or irrational consumer assumption? Is that what you're referring to? In the modeling, with respect to the economic analysis, yes. And I'll get back to that in a moment. What I'd like to do is address these issues in a little bit more detail. There are actually two related issues here, as I suggested. One is that the basic premise of efficiency regulation is that consumers are leaving economically beneficial efficiency investments on the table, and that that's the problem to be resolved here. The problem is that DOE never examined that question at all. It simply has a standardized analysis, it plugs the numbers in, and it crunches the numbers. And what actually happened here is we were all scratching our heads trying to figure out where are these benefits coming from? And the result, it was dug out of the analysis by a consultant. It turned out that they had this assumption embedded that generated artificial economic benefits. But let me address the basic... The assumption being that the purchasers were not concerned with economic gains. More than that. The assumption, actually, the way it plays out is that no matter what the economic stakes are, purchasers acting on their own are no more likely to make even the most economically beneficial efficiency investments than they are to decline them, and are no more likely... Not indifferent to this consideration. I'm sorry? That they're totally indifferent to this consideration. Yes. Okay. And these are commercial purchasers. I'm sorry? They're commercial purchasers, not homeowners. Exactly. And that's the... The facts here are very important. Commercial package boilers efficient enough to satisfy these standards are already available on the market. They already have a significant percentage of all product sales. And those investments are not necessarily beneficial for the purchaser. The economic outcomes of those investments vary considerably. In some cases, they provide significant economic benefits. In others, they impose significant net costs. And as we're just pointing out, commercial package boilers are substantial pieces of equipment. They're typically purchased as capital investments by business or institutional purchasers that routinely consider the economics of this type of investment and act in their own economic interest. So commenters made the obvious point. Under these circumstances, there's every reason to believe that purchases of commercial package boilers have a tendency to make investments in these more efficient boilers when it makes economic sense to do so. They have a tendency to decline those investments when they would impose net costs. Can I ask you a question? Go ahead. Thanks. There is a reference in the regulatory materials to a couple of arguable market failures in the sense of informational asymmetry or the lack of reliable information and these misaligned incentives. And so my question is just a conceptual one. Suppose, just hypothesize that we agree with you that there's a fundamental question here about the soundness of the assumption that's reflected in the use of a random array. But then suppose the agency were to say, well, we actually think the random array turns out to be a reasonable approximation because there's no perfect information here. And they have a sophisticated analysis that says when we look at the degree of informational asymmetry or the unavailability of information, and we take into account the misaligned incentives, we think that the random array actually gets us pretty darn close. And they expound upon that and say that, therefore, by clear and convincing evidence, we find that the findings that you think are required to be made by clear and convincing evidence are in existence. You don't disagree that conceptually that would be okay. You might disagree that factually they might not be able to make that assessment in a way that's sufficiently sound. But conceptually, that approach would work, right? Yes, that's correct. The problem here is that these issues were pointedly raised and simply ignored. The references that you're referring to are stock language that show up in every efficiency standards rule. They're not even in the economic analysis. They're in the part of the rule that goes through the motions of responding to executive orders indicating that agencies have to identify what the issue is that they're supposed to address. And if you look at that analysis in the face of all this pointed comment that there's no reason to think standards would provide economic benefits, they don't even refer in discussing those sort of generic market failures to commercial package spoilers. So it's a situation where in response to the suggestion that they simply failed to address these issues, they had no response at all. And again, it's difficult to understate the extent to which the assumption that purchasers never considered the economic impact of their analysis has on the analytical results. And in this particular case, there's a figure on page 56 of our opening brief that illustrates the basic problem, which is there's a wide range of economic impacts of individual investments. And so the justification for a standard is going to depend on what's the distribution of those outcomes that purchasers leave behind acting on their own. And without looking at that question, and DOA made no effort to look at that question, you end up with a situation where here you've got standards that are pretty close to break even. And they're justified only because of these extreme outlier economic outcomes that are the most likely outcomes to be influenced by economic considerations of the purchaser. They used the same kind of analysis in the residential furnace rulemaking, and there there was an extensive technical study done actually breaking out all the data points they used, et cetera. And the bottom line in that case was that over half the regulatory benefits that were being claimed were in cases where the more efficient product had lower installed cost. It was the low-cost option. So the whole premise that standards are necessary because people only focus on initial costs didn't even apply. I ask you, Mr. Day, do you have an alternative? My understanding was that the sort of modeling that you all did was not supposed to be a substitute for the Monte Carlo method or whatnot. So are you suggesting that there's a particular kind of analysis that the government could have and should have done to capture what you say is a problem with the analysis that they did? Yeah, and there are many things they could have done. I think in this particular case, if they simply went through and said, okay, we're not going to count as results of a standard purchases where the installed cost of the more efficient product is the low-cost option. I think that would probably be enough to show that the standards were not economically justified. Well, of course, that's not the goal. There might be ways that they can adjust it to reach the conclusion that you'd like to see. I'm just asking from the beginning, are there ways that they can account for what you say is the market forces at work that they've left out of this analysis, no matter where that leads ultimately? Yes, they can. The factor I suggested was specific to one type of outcome, but really- Top 10 percentile? I'm sorry? Was that your suggestion of ignoring the top 10 percentile? No, because I don't know. There's no way to tell from looking at the data that they provided where the cutoffs are. All they needed to do was come up with some reasonable way to account for the fact that as economic benefits get higher and higher, there's a greater chance that those investments are going to be made by consumers acting on their own. And as net cost situations go deeper and deeper, the odds that purchasers would make those decisions on their own go down. Right, and I'm just asking, do you have an idea what that reasonable way would be? Because you also have this concept in the statute of practicability and whether or not the government actually can do anything more to perfect its analysis. So, is there some clear thing that they could do to capture this concern? Oh, absolutely. I mean, their model is set up to do that, and that's why we couldn't figure out where the problem was. There should be an algorithm that simply tries to capture what the model looks like, what consumer purchases look like in the absence of standards. And that comes down to where they assign base case deficiencies. And instead of having an algorithm in there that was designed to show what the market is probably like, there's just a random assignment function. So, the fix is very easy. There's nothing wrong with the model. It's just the garbage that went in. That's the problem. Was this a similar problem with regard to shipments? Now, the shipment issue is different because the way they do their economic analysis, they look at all these individual variations on installation. So, they do an analysis of the economic impact of efficiency investments in a wide range of individual scenarios. And then they look at those individual outcomes called life cycle cost outcomes, and they generate an average of what the average life cycle cost impact is. And shipments goes to two issues, but in determining national impacts, it's simply how many investments would occur as a result of the standards. That's the issue. How many of these more efficient products are out there already? So, you multiply the average life cycle cost result by that number of efficiency investments to get the national impact of the standard. And the issue we focused on is that that average life cycle cost outcome is driven by these extreme outliers that are likely to be influenced by the decisions consumers make on their own. And yet, that drives the outcome that turns a marginal break-even kind of standard that if you look at how purchases are made, it's probably going to be a negative LCC outcome and turns it into a positive life cycle cost outcome. And that's the fundamental problem with the analysis. Can I just ask one more question? I'm wondering if you ever concede any sort of irrational behavior by consumers, because it would seem to me that you would never have economic justification if consumers always acted perfectly rationally, right? So, there is some world in which we're going to have these big institutional consumers not doing what is in the economic feedback. Absolutely. And I would concede that in the case of commercial package spoilers, there are going to be some cases where consumers don't make sensible economic decisions. But it's not a binary, and that's the false conflict that DOE set up in responding to this. They said, well, we can't assume that people are making perfect decisions all the time, and so we're going to model a world in which these issues are not considered at all, ever. And that is just ridiculous on both ends. I think the appropriate thing to do is to try and get a sense. And again, I think it's a matter in this case of simply recognizing that as the economic stakes go up, the probability that economics are going to affect the decisions that consumers make goes up. And to your broader point, the solution is not one size fits all. I think you have to look at the circumstances with respect to the product at issue. And here, we're dealing with a particular type of product where the usual sorts of factors that one would be concerned about in terms of justifying a standard just don't apply. And that's not the case for all products, not by any means. Thank you, Mr. Day. Let me make sure we don't have further questions for you at this time. Thank you very much. Do you have a comment, Ms. Weiner? May it please the Court. Stephanie Weiner for Petitioners. Given that it's not disputed that the clear and convincing evidence standard applies to this rulemaking, and given that the Department agrees that remand is required, I'm going to focus my time today on why the final rule should be vacated. Vacatur, which is this Court's normal remedy, is required under the Allied Signal Test for two reasons. First, vacatur will not be disruptive. It is what Congress intended should happen in this circumstance, and in fact would prevent the disruption caused by leaving the rule in place. Second, vacatur is necessary to remedy the Department's serious legal errors, particularly in light of the statute's anti-backsliding provision. First, vacatur is not disruptive here. There's no disruption because these standards have not yet gone into effect. There's nothing to unwind here. There's no egg to unscramble. If you vacate the rule, these products will remain regulated. They're regulated at the efficiency level set by the Society of Engineers, codified previously by DOE, and those are the basis of building codes around the country. And that's exactly what Congress intended to happen. Is it not the case that the industry is taking preparatory steps on the assumption that more restrictive requirements are going to be in place? I think there has been an overhang over this rule due to litigation. And while, you know, I represent a trade association that has multiple members, I can't speak to what any individual member has done. But the Department itself has estimated that it would cost the industry $21.2 million to bring equipment into compliance here. If these rules are left in place, and I believe some manufacturers want to know whether these rules are in fact unlawful, as the Department says that they are, and remand is required, we believe that that is disruptive enough and the industry should not have to incur those costs for rules that the government has said are unlawful and may go away on remand. Can I just ask you, because I was curious about your point concerning what Congress intended. If I read the statute that underlies this whole thing, Congress, it seems to me, clearly intends for these standards to be revisited and potentially revised, even when the society itself, whatever the private organization is, doesn't touch them. So it almost seems like maintaining the status quo is against the intent of Congress. I mean, the no-rule world. Because Congress says every six years, we really want you to review, revisit, and potentially increase. So how do you reconcile that with your conception that Congress does not want change, does not want disruption in terms of the imposition of a rule? Congress clearly wanted the society standards to be reviewed and revisited, but it specifically said what needed to happen before more stringent standards could be adopted. And what it said was the agency had essentially two choices. Either it found clear and convincing evidence to support more stringent standards, in which case they could go into effect. But if the department failed to find that those standards more stringent than those in place around the country were not supported by clear and convincing evidence, then the default rules, the Society of Engineers, were to remain in place. Sorry, there's a little feedback on my end. And that's what's happened here. The department has said it did not make the clear and convincing evidence standard required by the statute. And in that instance, I think the statute is quite clear that it's not the department's turn on remand to go back and do it correctly. But the standards should not remain in place in the meantime, because Congress was very clear about what had to happen before they could go into effect. All right, well, setting aside what the department is now saying about what the rule determined, the rule itself indicates in many places that the agency was making this determination that it had found economic justification by clear and convincing evidence. And so given that now, the question is whether or not, even though we have these potentially meritorious challenges to the manner in which they found it, to their explanation for some of the data, aren't those kinds of defects readily curable, such that, you know, when it goes back, we're not talking about ripping out the fundamental underpinning, necessarily, of the rule, and so in the interim, under allied signal and the like, everybody's on notice, this is what the department has found, and we should just keep going with the rule. Well, the department has not, in our view, found that these standards are supportive of clear and convincing evidence, and that's because they have failed to actually apply that clear and convincing evidence to the standards that they adopted. What happened here, Your Honor, is there was a notice of proposed rulemaking that said, we believe tentatively conclude that these standards are supported by substantial evidence. Commenters raised arguments in the record, including those made by my co-counsel, which the department has yet to defend in this court, to say that there was not clear and convincing evidence for the standards. And the department, in its final rule, instead of addressing that, said, we do not believe that the clear and convincing evidence standard applies to this rulemaking after all. As to the references… Wait, before you leave that, that's not all the department says. The department says that in a footnote, but it's not clear at all that the department actually applies any different standard when it's doing its analysis. And there are several places in the actual final rule, JA583, JA590, 591, 658, where when it is making findings, it says, we find by clear and convincing evidence, X. So, at a minimum, we have some sort of ambiguity about what standard they were applying, but one could argue that notwithstanding the footnote that talks about the potential for some other standard rather than clear and convincing evidence, they were, in fact, making findings by that standard. Or so saying, anyway. Or so saying. So, with respect to those references, it is our position that, particularly with the department saying that it did not actually apply the correct standard, that those references under this court's cases, referencing a standard is not the same as considering it. Stating that something was found is no substitute for finding it. And while we acknowledge that there are places where the clear and convincing words were used, we do not believe that the department actually held itself to that standard. And I would suggest, you know, standards of evidence matter, right, as this court knows well. And one place that you could see what the department actually held itself to here is if you look at the conclusion of when the department summarized the rule, which is on JA582, right, where after going through all the key findings in the record, the conclusion paragraph does not include the requisite clear and convincing evidence standard. And if you compare that, for instance, to rules that intervenors cite on page 6 of their brief that were held where the department did find clear and convincing evidence, those corresponding concluding paragraphs stated that expressly. Yes, understood. But I guess I'm confused as to what they did then. Because they don't ever say another standard. And it's hard for me to wrap my mind around the argument that just because the agency failed to meet what we now all agree as the test means that they were, in fact, applying a different test as opposed to they just didn't get to the requisite level. In other words, it seems like you're making a separate argument that, you know, we have reversible error as a matter of law because the agency was applying a different test as evidenced by the fact that they didn't meet the applicable test, but not as evidenced by the fact that they actually said we're applying some other thing. I think we are saying that they didn't actually meet the applicable test because they didn't apply the right standard, which is in the rule itself. Your Honor, not just the footnote. It says clear and convincing standards need not be based. And this rule need not be based on clear and convincing evidence. In the alternative, we believe the reference is that this is not complicated. They said they reached this conclusion by clear and convincing evidence. A later repudiation of that by a later brief is not the same as rescinding it. They've said it. It's out there. They also said an alternative. If we don't think the alternative applies, they have to meet the clear and convincing standard. We do believe they have to meet the clear and convincing standard. And we do not believe that they have done so, which is why we don't think The only fair point that you can make. Exactly. Everything else is irrelevant. Either they did or did not meet the standard that they purported to and that we all agree, seems to agree anyway, is the appropriate standard. We believe that they did not. And that's why we believe that remand with vacatures is required because when the department goes back on remand and takes into account the concerns that my co-counsel has raised and creates an updated record for things like gas prices, which were inaccurate, we now know. I guess the bottom line is it's different to say, and this is salient for applying out a signal. It's different to say they thought they met clear and convincing, but they didn't meet it. They didn't meet it. And then you could go on to say, and they never could. And therefore you ought to vacate. That's different from saying they thought they were applying clear and convincing and they didn't even apply it. That argument is a feature of what you're saying, and it's a feature of what the government has said in its briefing too. And that one is just hard because when an agency says it's applying clear and convincing, I don't know what else to infer other than that they are. Now, you might be you might well be right that they haven't met it. They thought they met it and they haven't met it. But to say that they didn't even apply it is a little difficult when they tell you in the alternative that actually clear and convincing to the extent that's the relevant standard is satisfied. We believe the department has said that they did not apply the correct standard, even in their most recent brief. Right, they said they didn't apply the correct, which, which just the correct standard is clear and convincing correct government's own argument on the agency's own argument. And the agency said in the rule that in the alternative to the extent clear and convincing is the correct standard. We don't think it is. That's what they said at that time. But to the extent clear and convincing is the correct standard clear and convincing is satisfied. But even if that is correct, it is. Why do you say, even if what's what's arguable about that? Because that just. Oh, I agree. Clearly, the department used the word in the room. Yes, I agree. You haven't telling us anything standard was met. Well, what the agency is telling you is that they did not apply it in a way that can be sustained by this court. That's not in the rulemaking. That's in a subsequent brief. The rule itself is, in our view, does not apply the correct here because that evidence is not there to support. That's fine. Just forget the stuff about later. I believe that that is the case that the evidence as my co-counsel has discussed does not support this alternative funding, whatever the department is saying now, I think, is what you're Directing me towards. And I believe that is the case. I also believe that when the department on remand takes up this role, it needs to do a new rulemaking. It needs to update the record. And when it does that, Your Honors, in our view, addresses the concerns raised by my co-counsel and updates the record, we do not believe that it can adopt the same standards On remand. And as a result, it would force manufacturers and the industry to incur costs that for a rule that is unlawful and that would otherwise go away. To the extent Can I just, I know your time is out, but I'm so curious about this anti-backsliding provision argument because it seems odd coming from your side of the table to suggest that on remand, the government would not be able to, per that provision, Change the standards. So our concern is that because the anti-backsliding provision is written, which, you know, prevents the department from relaxing standards once they are in place. Our concern is that the department or others may believe that the department cannot change these rules if they are not vacated on remand. If the department is revisiting them, they're still in effect, but they're doing so under a constraint of a judicial order. It seems to me that your concerns are not very weighty and they can certainly be addressed by the court's order itself. And surely that's not your position. I hear you say we're just worried that the government might take that position and therefore not change. Correct. Our concern is we don't know what the government's position is on that because it wasn't addressed in the briefing. We believe that absent Without it could be an open question. We have not found a case in which a standard was remanded without vacature and how the anti-backsliding provision. I assume your position is that if we were to remand without vacator, your position would be, well, don't be constrained by the anti-backsliding provision. You can do whatever you'd like because you have a judicial opinion that requires you to revisit this. Yes, we would. Our position would be that the department should follow this court's order on remand and if the court's order says that. More broadly, we believe the department should have a clean slate on remand to conduct a new collection of evidence in the rulemaking and make the determination that the department, in our view, has not met The clear and convincing evidence that has not found sufficient clear and convincing evidence to support the more stringent standards. That's what we, why we believe the rule should be vacated because otherwise You know the entire industry will have to take steps to comply with a rule that may or may not stay in place. Once on remand and we believe that that is, you know, under this court's allied signal test, a reason why vacature is required. Right. We need to enable the department to make The determination that was required by the statute in the first instance without any overhang of anti-backsliding and prevent the disruption caused by leaving the rule in place. We have tried this ground several times now. And with no further questions. Thank you, Ms. Wiener. We'll hear from respondents side now. Mr. Starcher. Yes. Good morning, Your Honors, and may it please the court. As already discussed, you know, there's agreement at this point that the clear and convincing evidence standard does in fact govern these kinds of rulemaking. So I think the most sensible place to start is with Something that was touched on briefly during Ms. Wiener's time, which is the adequacy of the alternative conclusion given in the rule. I certainly agree with the panel's, you know, indication that That it is an alternative conclusion that it was presented as such, and that the court has to address that alternative conclusion and to deal with it, but as explained I'm sorry, I don't know that I'm persuaded that it was an alternative conclusion. And so can you just spell out why that seems to be your position. Right, so I think the best way to get there is to look at where this discussion appears in context. And if you look at where this assertion about whether or not I have the JS site here. This assertion about whether or not just say 591-592, I believe. This assertion about whether or not the clear and convincing evidence standard applies during these sorts of six year look backs Is provided in response that, you know, the rule first summarizes comments received from some of the petitioners here. About whether or not the evidence presented to the agency was, you know, putting aside whether or not it could meet the sort of normal evidentiary standard of preponderance of an evidence of evidence. Whether or not the evidence presented here could meet this heightened standard and whether or not there were certain features of the evidence in front of the agency that That would render it insufficient under clear and convincing evidence. And the first response given to those comments is DOE reciting The text of the relevant six year look back provision and then stating both above the line and in an extended footnote There are features of this statute that indicate to the department that clear and convincing evidence is not required during the six year look back. Can I just stop you right in that point just to say that because this was a response to a comment or a question Does not, to me, signal that in all the prior discussion and all the discussion subsequently DOE was actually applying a different standard. It's sort of like a person pops up. They're doing something. They're talking about it. They're saying in, you know, 583 before this that they are making certain findings by clear and convincing evidence. And someone says, wait a minute. Are you sure that clear and convincing evidence is applicable in the six year look back context? Footnote, long musing about circumstances under which it may not be. But in any event, they say I see it as in any event, rather than alternatively because I don't know what other standard they would be applying. Did they say what else they were doing in the rest of the rule? So the alternative standard is the default standard in rulemaking, which is preponderance. And again, and I think that was the sort of The thrust of these comments. They were responding to was certain commenters had suggested during the comment period. That when you're under a clear and convincing standard certain kinds of evidence or certain kinds of assumptions cannot be made by the agency. In engaging in rulemaking when you're subjected to that higher evidentiary standard, but I think, you know, Anywhere in the rule. Do they have the two or three pages or paragraph going on to describe preponderance in the way that they do clear and convincing No. And so I do think, you know, I think there's agreement that the court has to look, you know, whether you call this an alternative conclusion or principle rationale alternative rationale or the only rationale. There's a great, you know, the department doesn't disagree that that the court has to look at the adequacy of this alternative explanation. That was it, whether or not we call it alternative the explanation that was given. And I think, you know, if you look at what is actually said about the clear and convincing evidence standard. In the rule in the entirety of the rule. There are these two, I think, three paragraphs that essentially describe in general terms what the department understands that standard to require and then proceeds to just state. Both in these pages that we've talked about. And also, also the pages. That the department refers the court to where later on in the rule. Again, the words clear and convincing evidence are used. It's a reference to a standard and then a conclusion that the standard is met without any sort of, you know, explanation or reasoning about why that standard is met. Another question in a similar where else in the rule. Does it ever say we find by preponderance of the evidence that Right, I would, I would hear you that they're doing two different things or that this is an alternative if there's any evidence of actual application of another standard. So, so this, this, this, the pages that we focus on are the only place in the entire rule where the implications of this higher standard are discussed. At all. And so, so I it's perhaps not surprising that there's no later point in which preponderance is described as sort of expounded upon, I would, I would point this court. I think it's helpful. To just compare what the agency did here and its discussion of clearing convincing evidence here to the agency's discussion of clearing convincing evidence. In in the last time that the agency looked at commercial package of boilers, which is in 2009 And there in that discussion, there was, you know, granted that was a finding that the clearing convincing evidence standard was not met, but in some ways, you'd expect Less discussion when they're finding the standard not met when they're finding it met. But if you look at the discussion that occurs in that rule. There is, you know, it's not just clearing convincing evidence requires X, we believe that to be met. It's under the clearing convincing evidence standard here are a number of specific pieces of data or factors. You know, in that rule, they identified a relatively small, you know, LCC, they identified things they weren't sure about. This is 74 Federal Register. I think I'm sure it's right. Then you can, there's either real world situations or you can imagine situations in which the incantation clearing convincing is repeated often. So it just brings home that clearing convincing is being applied, but the agency said that it was applying clearing convincing. It just did. I mean, if you look at J658 The paragraph, this is a concluding paragraph. This is the end of the entire analysis summary. After carefully considering the analysis results and weighing the benefits and burdens of TSL 2 and based on clear and convincing evidence. Setting the standards represents significant improvement that is technologically feasible and economically justified. It just says that. Right. So it seems to me that what you're asking us to do is to say that even though the agency said that it didn't really do it. So, I mean, I don't, I don't know what's the What's the basis by which we could do that when a jury says we find bias beyond a reasonable doubt. They just concluded. Right. Don't need to have an exegesis about why beyond a reasonable doubt. In fact, the standard and that's exactly what we applied and we did something more than preponderance. We did. They just concluded and we just assume it to be true because they said it. Right, but I don't disagree with any of that. But, but, you know, there's there's a line of cases that this court. A line of cases that explain that there is a point at which a recitation of a standard or a statement that the agency did something Is inadequate and that there's a point at which it can be too conclusory or too unclear that the agency actually did it to the point that That a court is left with the sense or the conclusion or the fear that the agency, despite stating that something was done or was met didn't actually do the work to back it up. One way to think about this, in my mind, is that suppose you had the exact same conclusions that clear and convincing was satisfied without the The ruminations on 591 to 592 about whether clear and convincing in fact is the right standard. And somebody could come along and say, well, all you have is a couple of conclusory statements that clear and convincing was applied. We've got these decisions that say you can't just recite a standard. You actually have to apply it. That's all we have. And I mean, I guess I would have a hard time knowing what else they were applying if they said they were applying clear and convincing. And then you come back to the real world here and you insert The ruminations about whether clear and convincing in fact is the right standard. It seems to me that after the agency engages those ruminations, it's all the more telling that they in fact say clear and convincing because they've already engaged with the possibility That they might not apply clear and convincing, but yet the conclusion says they did. I understood. I understand this concerns and I certainly agree that this is In some ways, a difficult application of this line of cases about when when something is too conclusory or too Too scant to provide the assurance that the work was actually done, but there is that line of cases. And I think, well, this case is different from those line of cases. Particularly when coupled with the proximity of this discussion to what now everyone agrees is, you know, an incorrect assertion of authority. I think the totality of this discussion that appears on 591 to 592 does fall within, you know, the kinds of concerns that animate these courts, this court's cases which are cited in the brief In which the concern is that there is a recitation of a standard without a meaningful application, but Can I also focus our attention on what is actually said about the possibility of another standard. You would expect that in this 591 text before we get to the footnote or around the footnote, they would say The statutory text indicates that the standard need not be clear and convincing evidence and indeed with respect to our Consideration of the evidence at issue here. We have not so applied that standard. However, if it is, you know, footnote explain why they didn't do it. If it is, then we would say that clear and convincing even, you know, applies under these circumstances. I see no affirmative representation That the agency is doing anything other than applying clear and convincing evidence. And in fact, the footnote just muses on the possibility that they really didn't need to go as far as they actually did. I don't disagree that it's what what what the footnote and the statement above the line were intended to do Is to some extent ambiguous. But again, I just returned to, you know, those were the first level responses given to a series of comments about the clear and convincing evidence standard. I don't think there's any disagreement that the court does in fact have to grapple with the statement that clear and convincing evidence was satisfied. I am happy that, you know, I see my time is out. And I do want to move on to some of the other features in this case, you know, particularly Talk about the remand. So what is so fine. Let's say we agree with you. Why does that have implications for whether we vacate whether we don't you say don't vacate. But does that mean that on remand, you would do the kind of fulsome whole start from scratch review that Ms. Weiner is talking about. Are you just going to go back and like insert the right standard cross out footnote 21 So, so I can't definitively committed to anything that the agency would or wouldn't do on remand. I think, you know, if you look at this court's decision and EMA Homer versus EPA. You know, there's a discussion at the end of that decision that is a case for remand without vacator was was utilized at the end of that decision. You know, the court notes that what happens in remand without vacators on remand the agency petitioners other parties. You know, can do a lot of different things they can stick to the exact same record and just provide more explanation, they can Engage in further comment they can update figures to represent to reflect changes that have occurred since the rulemaking took place. And so in any or all of those things could happen here, which is, I think, precisely why. And I think this, this, this is equally true for For for the sort of Methodological challenges that petitioners raise, you know, this is a rule that if the rule be sent back to the agency without vacator You know, different additional things could be said different explanation could be given about About why the methodological choices that that have been, you know, the focus of petitioners challenges here are in fact justified. I'm happy to talk about those Hurt you and not help you. Right. Why doesn't the uncertainty, make it worse from the government's perspective of you should leave the rule in place. Says we have people who are relying on these people who are spending millions of dollars. So maybe I don't know she's saying this, but It makes sense to keep the rule in place. Oh, if they could come back and redo the whole thing, then you should actually wipe it off the book. Because it's not fair to us and we've got reliance in Right. So to be clear that the sort of ambiguity. I was talking about, or the uncertainty. I was referring to is not uncertainty from the agency's perspective and the ultimate conclusion that will be reached. So this is, you know, as represented in The agency supplemental brief, you know, at this time, the agency does have full confidence in the record understands the sort of challenges that have been raised in this litigation, but does You know fully expects on remand to ultimately reach the same conclusion. What I was referring to in uncertainty is, you know, Exactly. How will they get to the same conclusion, will it be as you suggest sort of a crossing a paragraph or footnote 21 and then providing additional explanation of clearing convincing, will it be Will there be changes made to the model. Those, those are questions that I can't answer. But, but again, the agency Pre judged it and they'll figure out a way to support Certainly not. Again, that's why I can't speak definitively but but of course this is, you know, this court deploys remand without vacator frequently Certainly not infrequently in environmental cases and cases that touch on environmental impacts in particular Right. And in some ways, it's not just that they would they you're predicting that the agency would reach the same conclusion in terms of thinking that more stringent standards are justified. It's that you're predicting that the agency would reach the same conclusion that the more stringent standards are justified by clear and convincing evidence because that's what they said they did. So the fact the Right. Clear and convincing evidence actually adds meat to the bone on whether they're likely to meet arrive at the same conclusion. Right. Let me ask you this question for them to do for the agency to do that. They let's just engage in this hypothesis. Suppose we were to agree. None of us think it necessarily would. But suppose we were to agree with the challenge that petitioners make To the main underlying assumption about the random array and whether it's fair to just impose a random array, rather than one that takes into account the possibility that purchases actually make important decisions that are acting in their beneficial self interest. If we were to agree with that. And we were to disagree with you that the agency, in fact, didn't even apply a clear and convincing standard, but rather they said they did. And so we just assume that they did. Do you have a substantive defense of the rule. Yes, I mean, yes, I think that that's something that events largely tracks what what petitioner or respondent interveners laid out in their brief. But it's also reflected in the rule itself. I mean, I guess, beginning with, with, with, with the sort of challenge that took up majority petitioners time which is which is this the rent the choice to randomly distribute base case models. Across base case instances during the Monte Carlo analysis I a number of responses and feel free to cut me off if I'm too far over time here, but a number of responses to the points that petitioners make many of which are in respondent interveners brief. The first of which is this this assertion that the random distribution, the choice to engage in a random distribution is effectively affirmative assertion that consumers never engage or never consider economic benefits. One second. Sorry, but I do want to hear the summary the top line summary of the points, but I just want to make sure I understand that the government's position is that if we disagree with you that clear and convincing was only in Canada did not apply at all. Even though you don't have a substantive defense in the brief, but you actually do want us to go ahead and. Rather than remand for the agency to supply more elaborate explanation of the rule, you actually want us to go ahead and sustain. If the court were to get past the yes. Short answer is yes. Once you get past, I think the only disagreement between the department and respondent interveners at this stage is the disagreement that we've discussed earlier, which is, which is the alternative explanation sort of sort of topic on the random distribution point. At a high level, the issue here, the issue that petitioners are highlighting is an issue that is not uncommon to Department of Energy rulemakings or really to any other agency rulemaking, which is that oftentimes in rulemaking you reach a point. Where were there just is no further data to allow further refinement of a model. That is what happened here. The department did have a lot of data that helped inform its base case assignments. I think it's important to keep in mind that this was a random distribution, but a bounded random distribution in at least the following very significant way. Which is the department actually knew what types and efficiency levels of boilers were being shipped out to consumers for two of the eight categories because the industry provided that data. For the other six categories, it actually knew total numbers of boilers being shipped and average efficiency levels of those boilers, but did not know efficiency distributions. And that's something that petitioner challenged, but the point being is that the boilers that were being randomly assigned out The department has a lot of confidence that that set of boilers is an accurate reflection of what is in the market. And then the question becomes, how do you assign those boilers to individual base cases when there is no additional information as the agency, you know, explained that J622 That would allow the agency to engage in this sort of complex consumer choice model that petitioners now erred, but I think as the panel sort of hinted at At no time, including in the briefing before this court, has petitioners ever really come up with any concrete way to actually actualize The idea that you would somehow need to account for this kind of consumer behavior. And I think it's important, you know, petitioners really are relying solely on a common sense. You know, I think, I think, intuitively appealing sense of, oh, well, this is a commercial market. We know consumers must be behaving rationally and therefore this is some sort of gross overstatement of economic benefits. But I think as the department pointed out, there are a large number of reasons in which petitioners don't contest to think that in fact consumers in this market. Both commercial and residential consumers in this market are frequently, we don't know exactly how frequently, but frequently engaging in behavior that is not strictly economically beneficial. So can I just add, we have a respondent in Venus to hear from, and is there a explanation that you think they haven't made that the government has? And if so, can you just give the top line summary of that? Certainly. So, yes. So the top line response is, you know, there's lots of literature out there about the reality. I think the Nobel Prize in economics in 2017 was actually won for this. The reality that consumers in many markets do not behave strictly rationally, but there are a number of things in this market in particular for just to give one example, a huge amount of the boilers that go out every year are replacement boilers. So it's like, I think 20 to 25% of any given market is new construction where people are buying these things in the first instance and making informed, possibly economic decisions. But a huge amount of the boilers going out every year are replacement boilers, and a huge number of those are what's known as emergency replacement boilers, which is, in other words, a boiler fails in the middle of winter in Wisconsin. It's absurd to think that in those circumstances, the owner of that commercial real estate building is really engaging in this careful process to replace the boiler. There's evidence and research showing that most emergency replacement and replacements are like for like, which is to say that when a boiler fails, you don't engage in a search for a new one, really. You just ask someone to come in and install a new version of the one that you had. So that's one example. Okay, that's helpful. Let me make sure my colleagues don't have additional questions. Mr. Starcher, the secondary literature suggesting that, and empirical literature, that remands without vacature often result in essentially no further action being taken. The agency has no incentive to take the matter up again. And so these things languish for years and again, sometimes never revisited at all. And there's a host of examples involving just the EPA. And here we have the rule going into effect, if it does, without vacature in 2023. Don't you think we ought to, if we abide by your request, not to vacate and remanded instead that we ought to put a deadline on the agency to issue its revised regulation or whatever it's going to do expeditiously before all this money is spent on the matter is effectively moot? That's certainly something this court has the authority to do. It has done that in past remand without vacature cases. As long as the agency needs to. Again, I think precisely because of the sort of pre-decisional problem. I can't give an exact number, but I think if the court were to impose a deadline, a reasonable deadline, or to direct the agency to move with alacrity or whatnot, the agency would certainly abide by this court's order. What if we just said 90 days, unless you all come in the next 10 days and show us why you need more time than that? I can't, without running that by the client, I can't make any representations about whether or not the agency would be able to meet that short of a deadline. You mean they couldn't, within 10 days, tell us that they need more than 90 days? Oh, I'm sorry. Clearly they could do that. Yes, likely they would. I think we have a possible resolution of that problem, potential problem. Mr. Starcher, can you tell us where in the record, if at all, the emergency boiler point comes up? So there's, sorry. Please. So there's in one of the technical support documents, I don't know that I have this particular, I have many JAS sites, but not this particular one handy. There is discussion about the reality that in the market, new shipments account for only like 20 to I think 25, again, depending on which category you're looking at. I think the specific detail about emergency replacements, I don't believe is in the, is specifically pulled out in the technical support documents. But of course, you know, this court has lots of case law reflecting the reality that when an agency regulates any kind of market, there is some amount of difference due to just the agency's familiarity with the realities of that industry. And so, you know, I'm not sure that that, again, because the agency did explicitly say there are lots of market failures here and listed a few. I'm not sure the agency was in fact required to list every market failure it could have thought of. Was the agency required to provide some support for the listing? Yes, the agency listed a number of potential market failure issues. But they remain purely hypothetical, I would suppose, unless the agency also includes in the record some support for it. And here's a study that shows that this sort of thing happens in this environment. For some of these market failures, I'm sure that literature exists, but I think that question actually highlights an important point to keep in mind, that there was no evidence in the record either about the percentage of consumers, commercial or residential or otherwise, who were actually making the kinds of strictly economically beneficial decisions that petitioners were raising. And so this was in response to an assumption made about how consumers behave provided by petitioners. DOE's response was, we can also come up with a lot of assumptions, but you've given us no data that would allow us to actualize the assumption that you're making and to actually place it in our model. And of course, this court has a lot of cases dealing with challenges such as this, where really the challenge boils down to an assertion that the agency didn't have perfect information or should have pushed further to develop more perfect information or better information and didn't. And this court has repeatedly said that's not a challenge that this court will countenance. An agency isn't required to go down every rabbit hole if it's satisfied that the evidence it has is adequate. Let me make sure my colleagues don't have additional questions for you. Thank you, Mr. Starcher. Ms. Wu, we'll hear from you. I may appease the court. Michelle Wu from the Natural Resources Defense Council for Respondent Intervenors. This court should deny the petition for review because the department reasonably concluded that it was highly probable that the standards were economically justified. And I'd like to pick up on some of the calls we had with Mr. Starcher about why the department's method of modeling consumer choices was reasonable. In addition to the emergency replacement issue that Mr. Starcher pointed out, the department did explain that there are also other sorts of market failures that are impeding purchasers from capturing the full benefit of efficiency investments. That includes, for example, as the department explained, a situation in which a developer purchases a building and can thereafter sell it for a profit. It also includes, as we pointed out in our brief, situations in which landlords are purchasing foilers and can pass on the operating costs to tenants. So there are many reasons to think that people aren't, or that purchasers of commercial packaged foilers aren't identifying the most economically beneficial investment. But as a matter of agency responsibility under, you know, arbitrary capricious review and whatnot, to what extent is the agency required to not just theorize about these circumstances, but actually, to some extent, establish that these distortions of the types that you're talking about are in fact happening in this market? Do they have that responsibility and did they do that in this case? Well, Your Honor, again, as Mr. Starcher pointed out, there is case law that suggests agencies can rely on their background regulatory experience and their knowledge of the market that they regulate. And in this case, the background of Congress's mandate to set mandatory energy conservation standards is the idea that voluntary market forces are inadequate to advance energy efficiency in the way that Congress wanted. So that is, that's the background of this type of regulation. And then as far as things in the record, I think one thing that supports the agency's assumption of imperfect information or insufficient information is the nature of the Monte Carlo model itself. There are numerous inputs that feed into the agency's model, separate from the ones that petitioners have identified, and show that whether a boiler is safe and even well-informed, rational consumers would have a difficult time identifying which boilers are economically beneficial in the long run. As for misaligned incentives, petitioners' own evidence suggests that in 75% of cases, the person who purchases the boiler also is the person who's paying the operating costs. But that still leaves 25% of purchasers in commercial buildings, not to mention 12% or all residential buildings, which constitute 12% of commercial package boiler shipments, where there is more of a problem with misaligned incentives. So the misaligned incentives and the informational issue, which are the market failures that you're pointing to, am I right that the only place mention of that is made in the rule is at J.A. 660 in the section that deals with procedural issues and regulatory review under the executive orders? And there are two other places, Your Honor, where the agency discusses it. The agency, for example, rejected a model put forth or an alternative put forth by petitioners that would have assigned base case efficiencies based solely on economic criteria. And it says that that reflects an overly optimistic and unrealistic working market and presumes information that may not be available to all consumers at that J.A. 621. And then the second point is that at J.A. 622, the agency explained that developing a model that would have accounted for economic criteria, which petitioners have suggested the agency needed to do, would also require accounting for all the other factors that influence purchasing decisions, including green behavior, government incentives, and things like whether the purchaser is a landlord or a tenant. And that's at J.A. 622. But then they just said it would be too difficult, right? Well, they said it would require information that isn't available and data that isn't available and a way to recognize all the other factors that influence consumer choice separate from the economic criteria that petitioners have identified. Can I go back to your answer to my question about the agency's responsibility to show its work and actually amass evidence of the kinds of things that they think are happening? When we're talking about clear and convincing evidence as the standard and a statutory requirement that they evaluate factors to the maximum extent practicable, does that give them more of a responsibility to not just rely on their own background principles but actually figure out the extent to which these kinds of distortions are happening in the context of this market? Well, Your Honor, of course, we agree that the clear and convincing evidence threshold is higher than the prompt ponderance threshold. And the agency did recognize that as well, and it said it concluded that that higher standard was met. It needed to conclude that it was highly likely that the standards would provide economic benefit. But as far as how the clear and convincing evidence threshold comes in, there are two responses to that. The first is that, as you pointed out, the agency is required to conclude whether the standards are economically justified by clear and convincing evidence, but only to the maximum extent practicable. And that's because Congress recognized that this is an inherently predictive exercise that requires the agency to make predictions about future savings and consumer savings based on information that is uncertain. And the second point is that the Supreme Court has indicated that evidence to meet the clear and convincing threshold may not be mathematically precise or based on definite present or future conditions, and that approximations and reasonable estimates are necessary. And that's from the 2018 Florida versus Georgia case, that's 2527. Ms. Wolin, those predictive judgments, of course, are necessary. We have a record in which the agency predicted gas prices for five years, which have now elapsed. We know the predictions were erroneous. Are they obligated now to make new predictions for the next five years or whatever it is, or can they rely on facts that we know have been disproved? Well, as for the question of the adequacy of this rule, I don't think the agency needs to look at what actually happened with gas prices, because the agency can supply a more comprehensive explanation, you know, if it wanted to revisit it, to justify this rule based on the record at the time and the decision at the time. And that would be entirely appropriate. And when that's challenged and subject to review, we could find that there was someone could reasonably believe that was clear and convincing evidence when they knew it was, in fact, untrue? At the time, in 2016. The agency relied on the best available data that it had about gas prices. I mean, the Congress has set this thing up so that it's not just ordinary, you know, minimally necessary evidence. Well, the agency looked at the best data that was available in 2016 from the Energy Information Administration. Once we remand, the record is potentially reopened. Well, it could be, but another option, though, and what the agency has requested is an opportunity to provide a better explanation on this record. Wouldn't the failure to reopen the record, knowing that the data on which it relied are incorrect, make it hard, then, to be upheld as having clear and convincing evidence? No, Your Honor, I think because the rule would still be assessed based on the record at the time. An agency has an option on remand of just providing a better explanation based on the record at the time. But on the point of, you know, a remand, I do want to emphasize that. If we remand it, it's all open. Well, this court also has the option of structuring the remedy in a particular way. For example, this court could direct the agency to provide the more comprehensive explanation it asks for an opportunity to provide. It could, as Your Honor suggested, impose a deadline on that explanation, such as 90 days, and then it could adjudicate those with the benefit of the agency's fuller explanation. And that would also mitigate Petitioner's concerns about the compliance deadline as well, because this court would be, you know, would be able to rule on the motion or on the case expeditiously. And I also just want to make the point that it's critical that any sort of remand be without vacater in this case under the Court's Allied Signal Test. Vacater would be both disruptive and environmentally harmful. As far as the disruptive consequences, Petitioner's own declaration from one manufacturer indicates that it will take three years to adapt operations to come into compliance with these standards. Because this 2023 deadline is on the books, and nope, Your Honor, I see my time has elapsed. May I briefly continue? You can finish the thought. And because this 2023 deadline is on the books and nobody has sought a stay, presumably manufacturers across the country are already beginning to adapt their operations to comply with these standards. And it would be environmentally harmful because it would leave less efficient boilers on the market for longer, locking in for those boilers more harmful pollutant emissions, and higher utility bill costs over the course of their 25-year lifetime, and postponing the benefits to consumers. Emissions reduction, even though these standards are already long overdue, they were finalized in 2016. There are boilers on the market that would meet the standard today, correct? I mean, the new standard. Correct. Would it be possible, just in principle, for the industry to comply with the rule by simply removing some of its existing offerings from the market? I'm not sure the answer to that, Your Honor. On the market, those that would be compliant. I believe the compliance deadline applies to boilers that are manufactured after that particular date. So I don't know that the rule requires us to remove anything from the market. If they have some models that cannot meet that standard, they couldn't go on producing them after the deadline, right? Correct. Okay. We don't know. But that might mean they just discontinue those models and stay with the ones that do comply. That do comply, yes. And presumably, again, manufacturers are already making those decisions and may have already made those decisions now. Well, as I say, they may have decided to do nothing. And then they'll take them off the market if the new rule goes into effect. Yes, they could. Okay, let me make sure there's no additional questions for you, Ms. Wu. No. Thank you. Thank you, Your Honor. We'll hear from petitioners to rebuttal. We'll give you three minutes to wrap up. And I don't know which one of you is doing the rebuttal. This is Martin Day for petitioners again. I'd like to respond to a couple of the issues raised about random assignment because there are some serious red herrings here. Nobody, the petitioners have not argued that there's some need for complex modeling or anything else. DOE has addressed consumer purchasing behavior in the context of field switching analysis in other rulemakings. It's not a big deal. It's simple. And it's something DOE has done. The idea that there needs to be some elaborate model that accounts for every conceivable consideration is simply an effort to make the unnecessarily complicated the enemy of the reasonable. What DOE did in terms of its random assignment approach isn't reasonable. It's not remotely reasonable. It grossly overstates the potential for standards to provide economic benefits for consumers to an extent that I think it's improbable, very improbable, that DOE could justify these standards based on the same record. And if it has to reopen the record, which, of course, it should, obviously, they relied on a huge increase in the price of natural gas to justify these standards, and that didn't occur. So the idea that this random assignment… I just want to make sure that I'm clear on what you're saying. So your position is that basically the major failure here is we don't know where all this data came from. They haven't taken into account certain relevant assumptions, but that it's not going to require a whole new rulemaking from scratch necessarily if we were to revamp. No, I'm suggesting they have serious issues they have to address. The context is important here. DOE started out this rulemaking acknowledging that it needed clear and convincing evidence, and then it received comments pointing to all the holes in the record, of which there were many. And its response to that was to say completely unprompted because no one raised the issue. It was a completely settled issue. And DOE suddenly blurts out, oh, the standard doesn't apply. And then it announces an interpretation of the standard that focuses on its subjective level of confidence rather than evidence in the record. DOE's response to the fact that it didn't have evidence in the record was to say we reject D.C. Circuit precedent saying that we have to have our key evidence in the record so that it can be exposed to reputation during the notice and comment process. I mean, the idea that they needed to have evidence in the record to support their standards was not something DOE accepted in concept. And so they're going to have to look at issues, you know, like natural gas prices, like the use assumptions where, you know, they cobbled together something that resulted in increasing these high benefit spikes that drive the average results that they rely on. So they're going to have to do some work. That's not to say that this process has to be all that lengthy, but there are some specific issues where they do not have the record to address these issues in an appropriate way. And, again, a core issue that they didn't even consider whether there's a need for standards and whether they would provide any economic benefits at all. Yeah, that can be addressed by looking at what the factual circumstances and trying to come up with an approach that's reasonable as it has done in the context where it's looking at the circumstances under which purchasers would move from a gas product to an electric product, for example. Okay. Let me make sure my colleagues don't have additional questions for you in your rebuttal time. Okay. Thank you, counsel. Sweenor, you're opening your mouth a little bit. I misunderstood that I had a little time on rebuttal to talk about disruption, but if the court is ready to... We can give you a minute to give your rebuttal. I didn't have the direction beforehand about who was taking rebuttal, but go ahead. Why don't you take a minute? I appreciate it, Your Honor. Just real quickly, I do think what the government has said, it's not clear what happens on remand here. We do think what happens on remand is the government needs to update their records to, at a minimum, use current evidence as well as address the concerns raised by my co-counsel. That means that there is inevitable disruption in the industry. It will have to take steps to comply with this new rule on remand, and that, to your question, Judge Ginsburg, when a new standard goes into effect, manufacturers have to redesign their products, retest them, and certify them to the new standard, so it is quite an undertaking. Contrast that with the lack of disruption from vacating the standards, particularly if the department were to act quickly on a new rulemaking. While environmental concerns are at issue, there's no exception that says environmental concerns are disruptive. In those cases, petitioners, including the ones led by the government, petitioners have agreed that vacating would be disruptive. We don't agree here. We don't think it would be disruptive. In fact, it would be oxidative. Thank you. Let me make sure I understand that. So if they have a model now that they say is compliant with the, would be compliant with the proposed standard, they would have to get that officially certified should the standard come into effect. There are annual certification requirements that happen, and so those would apply to even existing products on the marketplace. Thank you. Thank you, counsel. Thank you to all counsel this morning. We'll take this case under submission.