Oral argument — American Gas Association v. DOE

Case No. 22-1030, et al. American Gas Association, et al. Petitioners v. United States Department of Energy and Jennifer M. Granholm, Secretary, U.S. Department of Energy. Mr. Mansihani, Petitioners, Mr. Hazel, for the respondents. Mr. Mansihani, go ahead. Thank you, Your Honor, and may it please the Court. The challenged DOE rules will eliminate non-condensing appliances from the market, but millions of existing homes and commercial buildings were designed to be used with non-condensing appliances. This uncontested reality makes the challenged rules unlawful for at least two reasons. First, non-condensing appliances possess a distinct performance characteristic under the plain text meaning of that term, and so they are protected by law under EPCA. Simply put, they operate differently, and that different means that they function within existing homes that have non-condensing appliances. Meanwhile, condensing appliances won't perform in the millions of buildings designed to be used with non-condensing appliances without sometimes substantial renovation to safely remove exhaust gas from the building. Second, DOE claims their rules are economically justified by using a technique called random assignment, which assumes that appliance purchasers act with no regard to their economic interest. This Court has twice before rejected DOE's attempt to rely on this technique, and the Department has once again failed to justify its usage. DOE picked a random answer to the important question of how often do market failures occur, but random decision-making is arbitrary decision-making, and arbitrary and capricious action violates the APA. I welcome this Court's questions. Mr. Mancingani, it's your position that the DOE definition in the 2021 interpretive rule is incorrect, the definition of performance characteristic? Of the December 2021 interpretive rule, yes. And so if we were to agree with petitioners on that question, do we need to provide a definition for performance characteristic? And what would that definition be like? If you agree with the Department, well— I mean, if we think that the Department's definition is too narrow, what would be an appropriate definition of performance characteristic? Or do we need to even state one? I think it would be appropriate to state one, because I think you need to define what it is before you decide whether this is a performance characteristic. And the way I would define it is that it's a product attribute that provides utility to consumers desiring to use the product. And I think if that's the definition, then there should be no contest that this meets that definition of performance characteristic, regardless of any factual disputes at the So say that again, a product attribute that provides— Utility to consumers that desire to use the product. That desire to use the product. Yes. I mean, it doesn't include those who don't want to use the product, and it doesn't include things unrelated to the performance of the product. So it has to be something having to—about using the product. And here, the way that non-condensing appliances work is that they can vent gas out of existing vertical vents, and that's about how it performs. And consumers derive utility for that, because they wouldn't have to renovate their homes without that feature. How does your definition square with the statutory language? Right. So the statutory language uses the word performance characteristic. So performance encompasses with the idea of how the product performs. And characteristic, of course, is a product attribute. So there's that part. And I think that the idea of utility is inherent in the idea of what's a performance characteristic, but it's also the word used in 6295Q1, which we consider the companion provision here, which perfects performance characteristics by creating new subclasses. Tell me this again. 6295Q1. Of course, this is all in Title 40. Q1? Q1, yes. Q as in queen. Go ahead. Finish your sentence. Oh, yeah. And there, it says that the department must create different classes for groups of products that have performance characteristics, and in doing so, the secretary must consider the utility of the product. So that's where the notion of utility comes in. But it seems like there's a tension within the statute in that, I mean, the department says, yes, there may be costs, but those are considered, appropriate under the statute, considered in the economic justification. And in fact, in the non-exhaustive list of statutory criteria bearing on economic justification, the Congress included decreased utility or performance of covered products, which, so there's utility and performance in that side, too, which seems to support the notion that not everything that affects, for example, use of insulation is necessarily a performance characteristic. So, I agree that that feature you just pointed out undermines our cost argument, because otherwise, you wouldn't consider utility at all either in a performance characteristic. But to answer your question, I think utility is considered in that, but that doesn't mean that utility is irrelevant to the idea of what is a performance characteristic. So, maybe there is some de minimis, a decrease in utility that wouldn't be considered a performance characteristic, but I think utility still has to be the focus of determining what is a performance characteristic. And of course, there's a non-exhaustive list of examples of what are performance characteristics, and within that list are things like sizes and reliability, which talk about, for example, a size as distinct from a capacity and a volume is important because it determines whether a consumer can actually install it in their existing space. I mean, maybe that size is for... This statute covers a wide range of products, and to me, size is a performance characteristic if it's like a hairdryer that you want to travel with, or a mixer that you want to be able to stow in a drawer, or, you know, so it seems like there are lots of products where size is obviously a performance characteristic for the consumer, but doesn't necessarily mean that that is always a performance characteristic. I don't know if EPCA covers hairdryers, but with respect to... I knew I was going to get myself in hot water by imagining some product that's covered that isn't, but I assume that there are many products that are... where size has a different degree of germaneness to the consumer's ongoing use of the product as distinct from an installation relationship. So I don't know of any examples, but what I do know of is examples within the department's past rulemakings and within the statute as to where some of these things have been taken into consideration. So, for example, the statute Congress created separate standard for furnaces, quote, designed solely for installation in mobile homes, Congress recognizing that installation characteristics justifies a different standard. In the department's prior rules, they had different standards for standard size and non-standard size packaged terminal air conditioners, which are installed within like a sleeve within the wall, and a standard size is more efficient, but not everybody's wall has the same hole size in it. So they created a separate less efficient standard for non-standard size because, quote, some could be forced to invest in costly building modifications to convert openings to standard size dimensions. So I just wonder how we, you know, following up from Judge Rouse's question, how we draw the line if, I mean, at some level, every, at least installable style appliances, putting aside whether, you know, my imagined category of a lot of more affordable things, but assuming installable types of appliances, every appliance that's already in use provides, in your terms, the utility of functioning in the purchaser's existing space. And so any time that the Department of Energy mandates a new type of appliance that requires some form of extra installation work, it's depriving consumers of that appliance, of the performance characteristic of being able to function without requiring whatever that extra installation work is? Right. So you have to have a product class that would be rendered unavailable and create more than minimum installation problems, right? So obviously every product can be installed, but if you're, if the department is eliminating an entire class of products that is generally available, that's the language of the statute, and rendering it unavailable, and then that creates significant installation problems, then I think that, yes. You're saying you're, the end of your argument is where the product class is rendered unavailable and you create, it creates substantial installation costs, I thought you said? Well, substantial installation issues, right? It's not just a cost issue here, right? Because they have to potentially renovate their home, and that not only creates changes to the consumer's space, but also takes time to do, which that time is also a detriment to the consumers that can be very serious. And I would say, you know, I don't know if significant substantial is the right word, but I would say it's beyond de minimis. Is the venting that's required just an installation cost? I mean, you know, in some sense the venting, you know, having to install in different kind of venting, you know, possibly losing interior or exterior space seems to be a feature that could thought to be part of the ongoing use of the new product, right? Like the venting is part of using the new product in your home with the new type of installation that was required. I agree, Judge Rowell, and I think it's two things beyond just cost. So one is the disruption that's caused during the installation and during the renovation, and the second is the ongoing harm. And I think both would qualify, and the department can't just say, ignore all installation issues, because they don't count for performance characteristics. That's nowhere in the statute. But I do think it's both. So that's sacrificing balcony or windows. That's losing shelf or storage space in a business. That's the problems created by common venting with an appliance that's still there that's non-condensing. And then it's, of course, the time disruptions caused during installation, including lost business for businesses, displacements, potentially freezing pipes during cold weather. And all of these things are what's involved in what the department admits is 40% of installations where you're replacing a non-condensing with a condensing appliance. Do you have figures on, like, if we disagreed with you that the disruption at installation, the time, the cost, I mean, sometimes just getting something out of the packaging is enough to, you know, cost you time, cut your fingers. You know, let's say we set that aside and we're just saying ongoing sacrifices. Judge Rao's question was suggesting ongoing sacrifice of space. Do you have any information on how common that is? So the best the best information we have is that one 40% number I gave, and that's what putting this all together. Yes. And that's what the department. That's your 39 percent figure. Thirty nine percent. Yeah, I rounded up to 40 percent. We can give it a thirty nine percent of the same figure. And and and and I will refer the court to a joint appendix 976 to 77, where the department talks about what's involved in difficult installation situations. This involves potentially multiple wall penetrations, concealing of flue vents, orphaned water heaters, interior wall displacement, vent or equipment relocation. And I joined Appendix 67. They also admitted to potential changes in living space. These are not just simply cost issues. Your first figure was nine seventy six to nine seventy seven. Nine seventy six to nine seventy seven. I do want to reserve remainder of my time for rebuttal, but before I do that, I do want to talk about random assignment. Their economic analysis was based on the assumption that consumers don't consider costs. This court already covered the problems with random assignment. And APGA one. And so now the question is, have they justified the use of random assignment? And yet again, they have not. There's an I didn't read it that way. I thought that they were although they're describing it somewhat differently when they say, OK, what we're trying to do here is create a baseline where we are not counting the benefits from. Of. Households or businesses that would, without the rule, purchase condensing appliance, that's the point of that exercise, right, to put those aside, and so they're looking at the the figures on who's already purchased based on shipping information. Right. And then they're trying to say, well, who's likely in the future to purchase? And they looked at a bunch of variables they could have used, and they determined that the ones that seemed material were size of dwelling and newness and that those were the entities or the owners, the situations in which the even without the rules, most likely that condensing units would be purchased anyway. And so they also included that in their modeling. And why wouldn't it be double counting? I mean, if if they're right, probably those overlap to some pretty significant extent with the people who have the biggest savings. And so if you say, well, just also look and say, say who would have the biggest savings and throw them out, it seems to me that that's very likely double counting. Do we also said and we looked at a bunch of other potential criteria and we thought no, they're not actually showing any any difference. So we're not going to use those. How do we know that's not what's going on here and that your proposal would result in double count? Oh, I don't think it'll result in double counting because what they did in those various factors to determine the share of consumers, that would be, say, choosing condensing and non condensing. So I'm just going to give broad numbers here. Let's say let's just say said let's just say they said in replacement situations, 60 percent of consumers would choose condensing and 40 percent of choosers would consumers would choose non condensing. Now, which consumers would choose which of those? They said, well, we're just going to determine that by random and entirely. I mean, I think they have those two factors. They make small adjustments for that. But what they didn't recognize is that consumers do actually consider their economic interest. That's why consumers in the north considered that. Right. They had all the behavioral economics modeling and and I mean, you don't you don't dispute that there are also situations that the rule describes where people don't act entirely rationally. So we don't dispute. I mean, nobody here would be among the people who act irrationally, but there are people. We don't dispute that market failures can exist. But the question, the relevant question here is how often do those market failures occur? And the way that the department answered that question is we are going to pick an answer. So didn't they ask you to give us the answer to that question and your clients didn't. So we said that real world data shows that economic choices by consumer that the choices by consumers generally align with their economic benefits so that that's the reason why more efficient appliances are more popular in the north. That's the reason why condensing has an increasing market share. That's the reason why it's more common in new construction. And we actually provided regression graphs to show this. You can see that in the American Gas Association comments around joint appendix 744 to 746. And so we said, yeah, economics is clearly one thing that consumers are considering. And the department said, no, we're not going to assume that they consider economics at all. And so and this is even their examples of market failures don't completely align with what they did here. So so there are examples of situations where a consumer currently has a condensing furnace and the model assumed that they would replace that with a non-condensing furnace. So they would choose a less efficient appliance that would require them to install new venting, which means higher installation costs and higher operating costs. And there are hundreds of cases like that in their model. And there's no explanation for why a consumer would ever pay two thousand dollars more in front costs just to get a more or less efficient appliance. Mr. Ransing, I assume for a minute that we agreed with the Department of Energy that this sort of random assignment in the no standards, no new standards case is OK. Wouldn't it still be a problem that they are assuming greater rationality in the standards case? I mean, so there seems to be a gap. It's sort of random assignment, you know, at point A, but like after the standards, they're assuming rational switching either to electric or to a more economically rational choice. I agree. Isn't the gap between that also a problem, irrespective of whether the randomness itself is a problem? I agree there is an inconsistency there. Now, of course, they now say that with respect to fuel switching, they didn't rely on it. And so we don't have to consider the economic benefits of fuel switching. But to the extent that they did rely on it, that is an inconsistency there because they actually developed a model there with that that judged when consumers may switch. And what they said there was said, hey, if an installation cost is going to be higher and the payback period is going to be really low, we're going to assume that the consumer would switch to electric. And they could have done the same thing with their model on the base case scenario here, but they didn't. So I agree that there is an inconsistency. Not even just with fuel switching, I mean, it seems that the department has assumed there will be some rational switching. So there's random assignment. But then if you randomly assign the, you know, kind of the wrong kind of furnace, you would then switch to something that made more rational sense. Well, so when you say switching, I now take it you mean condensing to non-condensing condensing. Yes. So with respect to that, there is no choice on the new standards case because everybody in the new standards case gets a condensing furnace. So that's not quite what happened here. So your answer to why the Canada has required condensing consumer furnaces for several years is they don't have the legal constraints that we have, or they have some kind of technology allowing them to install condensing furnaces in older buildings that we don't have. So the department didn't really do any study into what happened with Canada. They said, we know it's been done in Canada. And you must know that you must know. Well, what I do know is that Canada is really cold and. And not and so therefore, condensing appliances probably already consumed a very large portion of the market share like they do in very northern states in the United States where it's 95 percent of the market share. So the fact that Canada didn't have significant problems is maybe not too surprising because they probably already had a lot of condensing. There's going to be uptake there. Rule or no rule. Right, exactly. Whereas in places like Georgia and Texas and Florida, that's only 5 percent of the of the market share of condensing appliances. If you're in Miami, you know, you're not going to turn on your furnace very often. So it doesn't matter really too, too much about how efficient it is. In Miami, what you're worried about is air conditioning. Exactly. I shouldn't have to go ahead. No question. Oh, I was curious about that Meyer declaration, so there's a challenge to that as, oh, you know, not in the record, not in comments. Well, the data, I appreciate that the data is in the record, so that doesn't seem to be the problem. But it's almost like it's a page limit problem. You're taking data and making an argument that is appended to your submissions that expands the argument space that you have. I mean, I've never seen something like that. That just says, let us have let us hire an expert to walk you through something in a way that no expert in the record has walked us through. So, you know, I don't remember what our word count, how close we were to the word count on our on our opening brief was. But I will say that the top line conclusions, the top line conclusions in the declaration were all repeated in the brief. The only thing that's additional is just giving the court these steps. You look at footnotes two and three of respondents brief. They basically do the same thing, except they don't give the courts the steps. They just said we got this from the data. So, you know, we were there. It was there primarily for the court's convenience to understand how we got there. But the top line conclusions were both in the brief and in the declaration. I don't really think that that's that's a word limit problem. All right, and we'll give you some time for rebuttal. Thank you, Your Honor. Good morning, Your Honors. May it please the court, Stephen Hazel for the government. You could speak up a little bit. Of course, Your Honor. Congress has directed the department to regularly revisit the efficiency standards for the appliances at issue here. Consistent with that directive, the agency has determined that raising those standards will yield billions in net benefits without sacrificing product performance. That determination reflects a detailed technical and economic analysis spanning hundreds of federal register pages, and it readily withstands petitioner's challenges. I want to start by clarifying all of factual misunderstandings that came up earlier today and in petitioner's briefs as well. Actually, if you could start. I mean, throughout throughout the government's briefing, there's a reliance on deference to expertise and the agency's predictive judgments. After Loper Bright, though, we need to figure out what the best meaning of the statute is, and I see much less in the government's brief explaining how its interpretive rule in December 2021 is consistent with with the words of the statute in terms of what a performance characteristic is. So if you could speak to that. Absolutely, Your Honor. So let me take that in two parts. We're not asking for deference on any legal interpretation where deference would come in when we're talking about what's factually going on with how furnaces are vented or installed. And this is a market that the agency has looked at for decades. Respect for the agency. Sort of the zeitgeist of your brief, though, is, you know, this is all about science and expertise, and we can't really figure that out for ourselves. So. So, Your Honor, I wouldn't say the court can't figure that out for itself. But when we're talking about sort of how pipes are installed in different kinds of buildings, you know, that that seems like something where the agency is talking about the statute, though, and the agency's interpretation of the statute. Absolutely, Your Honor. I'm not sure how much distance there really is between us and the other side on the on the interpretation of the statute. This is about performance characteristics. It's not, you know, every characteristic of a product that would go too far. It's about how it actually performs. It's about how it operates. I think if it were really true, just to give an example where I think we actually agree with the other side, if we were really true that condensing units couldn't be installed in a substantial portion of buildings, I think that wouldn't be a performance characteristic. It couldn't perform in those buildings. That's the it's the venting dryers rule. But what's so the department has said that a performance characteristic is what what a person experiences just using the product. And since the condensing and non-condensing furnace both produce heat, there's no difference. That's right, Your Honor. Obviously, if the sorry to interrupt you, obviously, if the furnace couldn't be installed at all, you would experience that, right? It wouldn't it wouldn't perform or it wouldn't function in your space. Our disagreement with the other side here is just they haven't shown that what the agency found is they hadn't shown either the magnitude or even the existence of any of these kind of non-cost problems. That's a J.A. 74. We've talked about the Canada evidence. Canada is cold, but it has a building stock very similar to that in the United States. They've had these requirements for a long time. They just haven't experienced any of these kind of non-cost. So you don't seem to want to talk about the words of the statute. I'll just ask you a more specific question. So so it talks about features. So why is it not a feature, the type of venting of a furnace? Why would that not be a feature that is a performance characteristic? Because it affects the way the furnace is working in a person's home, you know, in terms of like how it's being vented, right? Because maybe they've had to lose a closet or a window or something like this. So, Your Honor, again, I'm really not trying to avoid the statutory question. I mean, I think if to take another version of this hypo, if if the furnaces really were taking a lot more space, if the size were different, we agree that would be a protective characteristic. That's not answering my question. You're not. What about the venting? So I understand these furnaces are roughly the same size, the furnace itself. Right. But but one kind of furnace requires a lot of different venting because it has to be vented out through a window or, you know, in some other way. So, Your Honor, I think if the argument was that the performance of the actual sort of operation of the product is the same, but the venting works differently and we think that different venting takes up more space, we disagree with that, actually. And I'd like to clarify that, too. But if it was true that the venting took up a lot more space, that would, of course, be relevant to the cost analysis. Doesn't the department concede that about five percent of installations would be difficult, right? And that that relates to the venting. Your Honor, that that's right. So five percent. So five percent of of consumers. So that seems to be a performance characteristic that takes something off of the market. So, Your Honor, people are using if I could just clarify what difficult means to the department. Difficult means the costs would be higher than the norm. And so they should be included in the model. What it absolutely doesn't mean, and I want to be really clear about this, is that usable space would have to be given up. What the agency actually found, and this is J.A. 67, is that typically there is no loss of usable space. Again, at J.A. Typically, there is no loss of usable space and that in any case where you're replacing non-condensing venting with condensing venting, you can actually use the same pipes. There's commercially available solutions that can do that. Typically sort of give away the game. So so then it's the department's position that even in those cases where it does take up additional space, that's not a performance characteristic. So, Your Honor, it's not clear from the record that there ever is a case where it takes up additional space, in part because of the technological solution that I just mentioned that that that it does sometimes take up additional space. Why is that not a performance characteristic? So, Your Honor, even in that case, I think it would be incumbent on the other side to show not just that there is like one building somewhere where there would be a problem, but that would be in a substantial portion of buildings. I mean, the statutory text refers to, let's say, five percent, right? The difficult installations or something. Why would that not be a performance characteristic? So, Your Honor, I think the reason we win this case is because that's factually just not true. You know, the text of the statute focuses on performance characteristics of a particular product, not sort of the venting or pipes or whatever it is that might be associated with that product. So I think the scenario you're positing would be analyzed as part of the economic analysis. If people were giving up rooms, windows and balconies, that might well swamp the economic analysis. And so the rule would just not be justified. That's just not what we have here. They haven't pointed to any actual building where a window or a balcony would have to be given up. And that's not what the agency found in its factual findings. You know, if they had wanted to argue that those findings just weren't supported by substantial evidence, they could have done that. They haven't done that in their brief here. And Congress put the burden on them. It's the burden on petitioners to put forward evidence showing that this really is a performance characteristic. They just haven't done that. And I do, I'll just a couple of other JA sites that I would encourage the court to take a look at on the facts here. JA 948 always or almost always the technological solutions to these problems. JA 973 is talking about the evidence from Canada and other jurisdictions that have implemented these rules without issues. JA 974 focuses on windows and balconies and says there's not any evidence that these can't be successfully installed in any particular place. So, again, I think the root of our dispute here is a factual one. I also just want to emphasize the implications of the other sides. Legal interpretation, which, as I understand it, is sort of any time there's diminished utility, that would constitute a performance characteristic that just can't be right under the statute or sort of as a practical matter. If that were true, by design, when efficiency improves, there's often going to be increased installation costs or increased maintenance or something like that. Congress anticipated that and said that goes as part of the economic analysis, but it doesn't necessarily preclude sort of all rules in that area. This case, I think, is a good example of the problems with their argument. Any time you increase the efficiency of these types of products as a scientific matter, they produce condensate, which has to be vented through these plastic rather than metal pipes. And so if we're really true that that difference in venting is a performance characteristic, DOE would never or almost never be able to raise the efficiency standard for this whole class of important products. And that logic would threaten to extend to other appliance categories as well. I'm happy to answer, I wonder if you could talk a little bit about Mr. Mancini's position that the department ignored the cases in which it would be most economically advantageous with even without the rule for individuals or businesses to switch to condensing units. And that instead the department insisted on assuming randomness in which purchasers would turn to condensing units without the rule. Absolutely, Your Honor. And maybe if I could start by just giving an example of how the model would would project people's behavior in the base case without standards. Say, for example, that we're trying to project a building in Virginia, whether that building is going to have a condensing furnace or a non-condensing furnace 10 years from now. The agency started by looking at past shipment data for that particular state, then added trends. What do we think is going to be happening 10 years from now? So the trend is generally towards condensing units. And so that's what the model reflects. The agency also found data showing that there's some correlation between household square footage and whether it's condensing or not. So it included that variable in the model. It found evidence that whether it's a replacement scenario or new construction is relevant. So it included that variable in the model as well. The agency looked at all of the all the data available, looked at its own data, looked at third party data. Anytime there was a significant correlation, it included the variable in the model. All of this sort of focus on random assignment is because the agency used what's called a Monte Carlo method. The that method, the sort of term of the last step is called random assignment, but it doesn't mean you're just flipping a coin. What it means is you include all the variables that you have in the model. And when when that runs out, you make your predictions. So I think I thought you just said that any time there's a significant correlation between when the data shows that a variable is significant, a variable and the likelihood of condensing. And and I think their argument is, well, a variable is how much would I save? I think it's I mean, maybe it's a circularity, maybe it's a double counting, but I'm trying to better understand as not someone who is not a statistician what your sort of common sense response to that is. So, Your Honor, I think if there was data that showed that even after you control for all of the other variables, after you control for square footage and state and these different trends, even after you control for all of that, sort of the amount that you would save is significantly correlated with with purchase decision. The agency would have included that in the model. That's just not what we have here. And to put it another way, if there's like one factor we know that consumers like actually don't include in their decisions, it's the it's the agency's analysis of how much they save. Right. The agency went through hundreds of pages to figure out everything from what's what our energy price is going to be for 20 years from now. All of these other factors to calculate what what the sort of optimal economic allocations are. I suspect that there's no consumer in history who's actually done that kind of analysis. That's why things like, you know, what's going on in your state, what a contractor is in your area providing things like that. That's that's what's determining the agency's analysis here. Mr. Hazel, though, this this Monte Carlo type approach, I mean, isn't it something that this court rejected in the AFCA case just a few years ago? I mean, how is how is the agency's analysis here different from what we rejected in that case? Your Honor, as I understood the court's opinion in that case. It was not sort of rejecting as a substantive matter that model. It was rejecting the agency's particular explanation there. I think in that case it was something like a sentence or a paragraph in that particular rule, which obviously is not the rule at issue here. We have dozens of pages sort of explaining how this analysis works. And we've added many of the parameters that I discussed a few minutes ago. So you think it's just a failure of explanation in the AFCA case? That's that's what the court said in that opinion, Your Honor. I also just want to be very clear. Monte Carlo methods, they're well accepted. Many agencies use them, not just the Department of Energy. They're used in the private sector and academia as well. So this isn't this isn't some kind of novelty. It was actually the agency trying to be very, very careful in how it did this analysis. Monte Carlo just sort of sound like it's arbitrary and capricious. I understand. Roll the dice. Yeah, I mean, it's fun. Yeah, more fun than furnaces. Your Honor, it's certainly not the title I would have chosen for the method, but it really is what statisticians use. And it's actually a method that is supposed to make the results more accurate, not less because you're doing, in this case, sort of 10,000 simulations rather than just sort of a simple model where if you take my Virginia hypothetical, right, where we'd say like, you know, you put all those variables in and there's a 75 percent chance that 10 years from now, this house will have a condensing unit. You actually went one step further to make this even more accurate. And I do want to emphasize, I think the other side has said, you know, we think at a minimum there should be like a middle ground. Right. Sort of some people behave rationally. Some don't. That's exactly what this model does. The outputs estimate that actually most people in the no new standards case are taking the picking the sort of economically justified earnest. So it is a middle ground. Many people in the estimates are making that quote sort of correct choice. Some people are not. That's not surprising given what we know about market failures. We know, for example, that people tend to undervalue long run efficiency and undervalue or overvalue initial costs. We have examples, for example, like landlords who will rationally focus on initial costs and and downplay the sort of long run operating benefits of a condensing appliance. What about Mr. Monsignor, he's questioning the department, assuming that some people will go from condensing to non-condensing. And his claim that that's part of the model and that that's irrational. Is that a kind of irrational that you have risen to believe occurs? I'm sorry, I didn't quite follow. From condensing move to non-condensing. That's I don't know if that's irrational or not, but the reason that's in the model is because it's part of the real world. The past shipment data that the department used shows that in new construction, some developers will put in a non-condensing unit, even though those have higher initial costs and higher long run costs. You can theorize about why that might be. You know, maybe contractors are just sort of used to installing non-condensing units. But because developers actually do that in the real world, it's part of the data that the agency used. It's also reflected in the model. It's not that they're switching from condensing. Well, I guess maybe if they're developing, they're gutting some building, let's say. So I thought that I thought his point was that your data assumes that there already is a condensing unit and that when when the purchaser goes to market, they buy non-condensing. And if he was positive that would not happen for the same reason that he thinks it's too complicated to go the other way. But I think you're saying something maybe a little bit different, which is brand new construction and or gutting and redoing. Yeah, I thought that that was what the question was getting at. I think they have it's from their declaration, the statistic that sort of 80 percent of new construction in this particular scenario behaves in this unusual way. And our answer to that is just it happens in the real world. It happens in the model. I think that's that's a broader answer to many of their questions. You know, we're looking at past shipment data for these particular models. So what did consumers of purchasing furnaces? What have they done in their state? So it's grounded in what people have really done. So I am also trying to understand the fuel, the department's position on fuel switching. Maybe that's enough of a question. Thank you, Your Honor. So fuel switching is often considered as part of these rules. Anytime you change efficiency standards, there's a possibility that some consumers on the margin are going to say, I'm not going to use that product anymore. I'm going to switch to something else. That's why the agency has at least considered the possibility and rules going back many administrations. In this case, fuel switching didn't turn out to play a key role in the agency's conclusion that these rules were economically justified with respect to the water heaters rule. The agency looked at fuel switching and said, we actually don't think there's going to be a significant amount of that with respect to the furnaces rule. The agency said there could be some amount of fuel switching here. It's hard to estimate exactly. And so we're going to explain that even if there was no fuel switching, the economic justification would still be there. The agency said that explicitly in the rule. That's not something we're sort of adding in our briefs now. And the agency also said that, look, even if we assume the maximum amount of fuel switching, in other words, even if we assume that, you know, everyone fuel switches when it's sort of economically justified to do so, we would still come to the same conclusion. Then I think that came up. Maybe Judge Rout was one of your questions earlier. That wasn't the agency assuming that everyone behaves rationally. It was a sensitivity test to make clear that, you know, even if that was really the fuel switching behavior, it still wouldn't affect the agency's conclusions. It wasn't a declaration by the agency that we think people behave perfectly over here and imperfectly over there. So they make an argument, they I'm sure make it better than I do, that EPCO requires because EPCO requires different energy conservation standards for products that use different types of energy, that the department cannot adopt an energy conservation standard that it renders that requirement meaningless. If you can issue a standard for gas powered appliances that for many consumers makes gas powered appliances so expensive that they have no practical option but to switch to electric. So it's sort of doing de facto what the statute appears to disallow in if you were to do it explicitly. So no matter whether you have an electric or a gas powered furnace, here's our standard. Congress says, no, no, no, no. You have to give standards for those separate. Right. Am I right about that? Product classes generally need to be different based on fuel source. That's right, Your Honor. Right. Based on fuel source. So then I think their argument is that if you issue a standard that is only formally for gas powered appliances, but it's demanding enough that for many consumers it makes gas powered appliances so expensive that their practical option is to switch, that you are doing the prohibited thing. So I don't think that's right for a couple of reasons. I mean, it may not be exactly their argument and they can correct me. Understood, Your Honor. I mean, one, of course, is just actually this isn't this is it's like five percent projected fuel switching, right? So this isn't a huge number of consumers. And that's not necessarily a situation where someone would say, like, you know, practically I have no choice. Right. People can sort of make decisions on the margins without feeling that kind of financial pressure. I think, you know, more more broadly, there were a couple instances in the statute where Congress said, you know, we we want to make sure that when you issue rules in this area, it does not cause any significant shift in fuels. Congress said that explicitly in a couple of narrow areas. It didn't say that here. And that makes pretty clear that that was not something that that Congress was worried about. Again, even assuming that these were rules that would produce a significant amount of fuel switching, which, as we discussed, is just that's not what the agency found. And you said even though the benefits were calculated, that of fuel switching, the cost savings to consumers, something that DOE did calculate. Right. It did, Your Honor. And the conclusion was that with fuel switching or without it, these are heavily economically justified on the order of billions of dollars a year in the case of net billions of dollars a year in the case of the furnaces rule and hundreds of millions of dollars a year with water heaters. So this isn't a situation where the sort of balance was closely poised and one small change could have shifted it from justified to not. These were strongly justified under the the incredibly careful analysis that's reflected in these three rules. Unless the court has any further questions, yes, they do deny the petitions. Thank you. Mr. Honey. You can fix all my mistakes. I think you have it mostly right. You know, I will briefly cover fuel switching and say that I think you have it right with respect to the sixty not sixty two ninety five Q1 argument. What I would add to that is that the economic justification provisions in both sets of statutes require them to determine the economic benefits of the new standards for the covered products. But because electric appliances are not part of the covered products, which counting the benefits of switching to them would would not be part of the correct economic analysis. And with respect to the impact of fuel switching, this is over 50 percent of their cost savings in the furnace rule. So it's primarily a rule about getting people to switch to electric and without fuel switching and without random assignment in the in the model. It goes from the 10,000 consumers in the model saving one point nine million dollars to the 10,000 consumers in the model losing two point five million dollars. So those two things together, that's not one small change. I think this is where it's those two things together flip the economic justification. What are you saying without fuel switching and without and without we disagreed with you on random assignment and they say, well, fuel switching doesn't put us into the red. Do you have a response to that? Well, they didn't rely on fuel switching as the sole basis for their economic justification. So you still have to go back if you agree with us on fuel switching. They're saying no fuel. We're putting setting aside fuel switching. Right. Well, so you're saying, oh, with without random assignment, I think it's agreed with you on random assignment, but thought maybe it's not. Allowable for them to count fuel switching, maybe for the reason that you said, because they have to look at the benefit for the covered products. And so we're not going to count that, then isn't it enough, though, if we credit their Monte Carlo modeling? Yeah. You know, with respect to that, I don't I don't think so, because I think they'd have to decide even if it's a positive number, whether that is still economically justified. I think if there was like one dollar in savings without fuel switching, that they would have to say, well, maybe forcing consumers to go through these things wouldn't be justified. I do want to spend some time on Monte Carlo here. It may be a valid statistical concept in other cases, just like correlation is a valid statistical concept. But if an agency said correlation equal causation, you would say, no, you can't do that even if correlation is a valid statistical concept. And here they have not just one variable that's changing over the same case. The 10,000 cases are all different cases. So they have a variable randomly picked in 10,000 different cases rather than the same case with 10,000 different options for that one variable. And and to the point that Judge Pillard, you were asking my friend on the other side, that the situation I was pointing to was a replacement situation on a new construction situation, replacing a condensing with a non condensing. If you want to look at this giant Excel spreadsheet, I would refer you to trial case number 24. Okay, go ahead. What'd you say? Trial case number 24 on the giant spreadsheet is a situation where it's an it's an existing home. They're replacing a condensing with a non condensing and paying two thousand dollars more in upfront installation costs to do so. That's just irrational. Can you also speak, please, to the government says you haven't met your burden here of showing that this actually affects consumers or would, you know, that both that you have the burden and that you fail to meet it as a factual matter. So with respect to the performance characteristics arguments, we've met our burden of showing with the preponderance of the evidence, and that is not a very high burden, as the department itself argues in trying to justify its own rule. But there is no dispute of all those things I listed, like multiple wall penetrations, etc., that these things will have to happen. They're not just cost issues. They say, well, you can always still install it in the same place. First of all, that would also be true of a space constrained appliance if you renovate your laundry room to install a not very compact washing machine. But second, their reliance on this commercially available products we pointed out in our comments at Joint Appendix 51 and Footnote 1 as to why those actually won't work in a lot of situations. And when we did so in the final rule, they actually backtracked and said, well, we're not actually relying on these products because we understand they are developing technology that are proof of concepts that are a potential solution. You can find those words at Joint Appendix 74, and they recognize the limitations of that new technology. You can find that at Joint Appendix 975. You just mentioned trial case number 24. When I read your brief, I thought that the going to non-condensing was about, in your brief, I thought it was about new construction. And now you're drilling down and finding what you say is an example of existing construction. Did you mention in your brief anything about existing construction moving from condensing to non-condensing? And if so, where? So in the brief, we talked about how 60 percent of replacements were the less economic option. And so that includes the less economic option of both non-condensing to condensing and condensing to non-condensing. And that's paragraph 7 of the Meyer Declaration, which we've talked about before, but it's also repeated in our brief. I don't have the exact page number. I think that concludes the morning. The case is admitted. Thank you all very much. Thank you.