# Oral argument — United Parcel Service, Inc. v. PRC

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/argument%3A67922

## Record

- **Collection:** Oral argument
- **Document type:** Oral argument
- **Published:** January 13, 2020

## Text

The next case on the docket originally, which is number 19-7036, Aaron Ball v. George Washington University, the court will consider it on the briefs. Case number 19-1026, United Parcel Service, Inc. Petitioner v. Moser Regulatory Commission. Ms. Sullivan for the Petitioner, Mr. Shee for the Respondent. Good morning, and may it please the Court, Kathleen Sullivan for UPS. The Postal Accountability and Enforcement Act requires that the Postal Regulatory Commission ensure that competitive product packages collectively cover an appropriate share of the institutional cost of the postal service. That's 39 U.S.C. 3633A.3. And in doing so, 39 U.S.C. 3633B requires that the Commission shall consider, shall consider, the degree to which any costs are uniquely or disproportionately associated with any competitive products. But the PRC expressly declined to follow that statutory mandate. And that's why we say this order has to be vacated and remanded as contrary to law and arbitrary and capricious. And you don't have to... They are not correct that A2 subsumes B. They are not correct as a matter of law, Your Honor, and here's why. I would refer you... No, actually, the language is sufficiently... I generally agree with the language is sufficiently equivocal as such and have it so that if there's a possibility but it doesn't pan out, I read the statute and say, well, good. Your Honor... I'm still curious as to whether they're correct. Let me try to ask you some questions. I just need to have an understanding from both sides as to what we're talking about here so I can understand what's really the issue. Am I right in assuming there are inframarginal costs that are indisputably institutional costs? In other words, no one would argue those inframarginal costs are attributable to costs of a competitive project. Is that right? It's not right, Your Honor, and I think where we should go is we should go back to this Court's prior decision in the A2 docket. No, no, wait. It's open. I want to take you to the graph there, Your Honor, to answer your question. If you go to the graph at 890F3rd, in this Court's prior decision on the A2 docket, the docket pertaining to Order No. 3506, which interpreted the A2. So to answer your question, Your Honor, if you look at the graph that's at page 1060 of your prior opinion, and for ease, Your Honor, it's reprinted in the blue brief at page 30. And I think I can answer your entire question, Judge Edwards, if we look at the graph on page 3. I'm not going to do that. I'm going to give it. Your Honor, this Court already considered the question of what costs are attributed to competitive products under A2. Now, let's remember the statutory language under A2. Let me ask you a question. I'm really trying to ask a preliminary question so that I can understand what I know you want to address on the other side of that. Are there inframarginal costs that are indisputably only institutional costs? Yes. And they are not attributed to competitive products under A2. Right. Wait. Bear with me. I mean, you know this better than I do because you've all been working with it, but I have to get myself up to speed. So that category is not indisputable. In your mind, everyone's mind, there's no doubt that there are inframarginal costs that can be institutional costs. That's not the issue here. That is the issue here because the inframarginal costs that are institutional costs are the very ones that we say the Postal Regulatory Commission was required to consider under A3. And they didn't. And, Your Honor, I think the graph is helpful if I could kindly refer you to the blue brief at 30. Inframarginal costs occupy the white space of the graph. And I think it's helpful to go back and look at the statutory language. And, Your Honor, your precise question was, why didn't the A2 analysis that this court reviewed in its prior decision subsume the costs that we say need to be considered under A3? You know, I know the court said there's some inframarginal costs. That are attributed and most aren't. And the ones that are attributed under A2 are represented by the green triangle. Some that are attributed under A2. And there are some that are not. Correct, Your Honor. So there's a question. And that's why I'm asking the question, my first question. Is there such a thing as an inframarginal cost that is properly characterized as an institutional cost, period? In your wildest moment, you would not say no. You must attribute that or say it's associated with a competitive product. Your Honor, there's two different buckets. The attributable costs under A2. No, I really wish you'd answer this question. I am answering the question. No, I may be simplistic, but I really need the answer. Your Honor, yes, there are indisputably institutional costs. Inframarginal institutional costs. That's the white triangle in the graph on page 30. That's none of those have ever been attributed to competitive products under A2. No, well, we, in the last. No, no, you're not hearing my hypothetical. You're jumping ahead of it. All I want to know, it's a very simple question. It's very simplistic, possibly. Is there such a thing as an inframarginal cost that is purely institutional? No one would say it can be attributed or associated with a competitive product. Is that right or wrong? That's wrong, Your Honor. We believe that institutional costs must be associated with competitive products under A3. And A2 says that costs that are reliably identifiably caused by competitive products are attributed to competitive products under the A2 analysis. That's the blue rectangle and the green triangle in the graph on page 30. But not all institutional costs are associated with or attributable to competitive products. That's correct, Your Honor, but A2 requires that you attribute to competitive trial products all those that have a reliably identifiable causal relationship. That's a very narrow category, Your Honor. I understand. That's not my question. The remainder of the inframarginal institutional costs, the institutional costs that are not attributed, so all of the white space in the graph on page 30, part of that must be considered to the extent that it is uniquely or disproportionately associated with competitive products. And that's what the PRC failed to do. And, Your Honor, I think it's worth looking at the statutory language. I need to keep this picture. Go ahead. I want you to get your questions up. Yeah, well, I'll wait to get the other slides up, too. Mine really is a very succinct question. I'm trying to see if I can clean out of my mind a certain category of inframarginal costs that neither you nor anyone else would suggest we would ever set because institutional costs are different from the costs that are clearly either associated with or attributed to competitive products. It's a broader sweep. Let me give you a simplistic answer. Part of the white triangle is associated with competitive costs, competitive products. Part of the white triangle may not be associated with competitive products, but we don't know what portion of the white triangle is associated with competitive products. The answer to my question is yes. Yes. That's all I wanted to know. That's just to help me to sort this through. I'm trying to understand the fight. You're saying some inframarginal costs have improperly been characterized as institutional costs when, in fact, they should have been associated under B. No, Your Honor, that was our position the last time. We lost. But the reason why we have to win this time is that we lost on the very grounds that the area of the white triangle here was none of that. The PRC has decided in the current docket that you are reviewing that nothing in the white triangle is uniquely or disproportionately associated with packages. That cannot be true. And why can't that be true, Your Honor? Because, remember, we're in the age of e-commerce. Package traffic is rising while mail is dropping, and 30% of the revenues of the Post Office now come from packages, and that's 30% of the costs that are attributed under A2. Now, Your Honor, why – But you can see there are some institutional costs or some inframarginal costs that are not uniquely or disproportionately associated with competitive products. Yes, Your Honor, because the language Congress gave was uniquely or disproportionately associated. There may be some costs that are associated but not uniquely or disproportionately with packages. There are lots of joint costs that may be associated with both packages and letters. The Post Office is in the process of buying billions of dollars' worth of larger trucks. Why do you need larger trucks consuming more fuel than letter traffic is going way down because of e-mail? Because packages take up a lot of space. So when it buys these larger trucks, you can still carry letters and packages on the trucks, Your Honor, but that cost is disproportionately associated with package delivery. Why? Because the larger trucks and the additional expense to purchase them and fuel them is attributable mostly, disproportionately, to the higher cubic volume of packages compared with letters. Your Honor, I wanted to reserve time for rebuttal, but I've already lost my time. May I keep going? So, Your Honor, I want to speak to the key to the case, Judge Edwards. Why doesn't the A2 analysis subsume the A3 analysis? And the answer is because the A2 analysis did not take into account as attributable to packages the larger trucks that were involved in delivering packages. It didn't take into account other uniquely associated costs uniquely associated with packages. When the post office goes out and it makes a contract with an e-commerce distributor, those transaction costs are all about packages, and yet none of those costs, none of those costs are attributed under A2. What is attributed under A2 and where the Postal Regulatory Commission prevailed in the last case was the extremely narrow set of costs represented by the blue rectangle and the green triangle. Those incremental costs of packages and the definition of... But some percentage of, if I understand it correctly, you've got a nice big area there. I guess you're quite a triangle, but we'll call it a triangle. It's a leftover area. It's a polyhedron with a roughly triangular shape. Got it. Okay. We'll call it a triangle. Good. Is some portion of that white area costs of the competitive products that are not reliably calculable? Precisely, Your Honor. Okay. Those would not be institutional costs. Those would just be costs in the statute's terms, its costs, the competitive product costs, but we just can't reliably compute them. And that would be different from things that are not its costs. Institutional costs are not its costs. Sorry, Your Honor. In this world, institutional here does not have a common-sense definition like its common costs. Institutional here is the residual of attributable costs. Institutional... The white space is by definition institutional costs. It isn't all institutional costs. So everything's not reliably calculable as an institutional cost? Well, no, no, no. See, I think it's helpful to go back to the language of the statute because it's the key to my argument. A2 and A3 are not coterminous. So the A2 analysis you approved last time, look, Your Honor, for reliably identified causal relationships. That's 3633A2 plus the definition in 3631B. Costs attributable under A2 must have reliably identified causal relationships. Reliably identified causal relationships have been defined exceedingly narrowly by the PRC, approved by this court in the prior panel decision, as just the incremental cost. The blue rectangle, which represents the cheapest marginal cost of the product, and the green rectangle, which represents the intermarginal costs, Your Honor, that are attributed because they're reliably identifiably caused by the competitive product. That leaves this vast white space. And last time the PRC was here, Your Honor, and I commend to your attention, if you do nothing else in this case, listen to the prior oral argument at minute 39, second 11, where the PRC very articulately stated, don't worry if we're undercounting the costs that packages cause the post office under our A2 analysis. Don't worry if the blue rectangle plus the green triangle are too small because that vast white space, all those costs go into the institutional bucket. And under the A3 bucket, which is now before you under order number 4963, the PRC will allocate, and I quote the PRC, reasonable sums that are uniquely or disproportionately associated with competitive products. And that's what the PRC refused to do. They didn't make good on that advice to this court. And I did listen to that, so I understand that point. But under B, I want to know on your statutory text. So under 3633B, when it says, look at the money language here, the degree to which any costs are uniquely or disproportionately associated with any competitive product. Correct. Does any cost there mean any institutional cost or any attributed and institutional cost? Both, Your Honor. Any cost means any cost. I think I agree with that. And the government may say, oh, well, we took into account some uniquely or disproportionately associated costs over in the A2 analysis because we did account a little bit for cuts over in what we attributed as causes that are reliably, identifiably caused by packages. But my argument is that A3 requires more. And the answer to your Honor is yes. Go ahead. Go ahead. Go ahead. Why doesn't A3 subsume? And the answer, Your Honor, to come back to your question, is if we look at the language of B, uniquely or disproportionately associated with any competitive product. And it's any cost. You have to look at the institutional cost. It's not enough to just look at the attributable cost. Any cost. There has to be costs over in the white space that are at least disproportionately associated with competitive products. Because it's residual, the things that they're not accounting for that you say they ought to be accounting for under B are going to be considered by them as institutional costs. They didn't consider them as institutional costs. That's the whole point, Your Honor. It's not considered. They're in the category of institutional costs. They are, Your Honor. Right. They're in the category of institutional costs. So, hop up to 3633A3, and let's look at what the Postal Regulatory Commission failed to do. It is supposed to ensure that all competitive products collectively cover what the commission determines to be an appropriate share of the institutional cost of the postal service. And in doing that, Your Honor, down in B, it must consider the degree to which any cost, including institutional costs that weren't attributed over under A2, are uniquely or disproportionately associated. I'm emphasizing that language because I thought any cost would include both, I think it's self-evidently institutional here, but also things that were already attributed under A2. And so, when it comes down to deciding, I'll just throw this out there for you, because yours seemed to be get whatever you missed under A2. And they say, well, that's nothing. We think we got everything under A2. And I don't know why this language, please tell me, doesn't mean, I don't care if you already counted it under A2. When you come down to determine appropriate share, you've got to sit back and look at any unique or disproportional cost, whether they're in the blue box, the green box, or the white box. And you have to factor those in to determine appropriate share. And they didn't do that. That's exactly correct, Your Honor, 100%. They didn't do that. Any cost means you can't just finish the job. I think, oh, well, we captured a little itty-bitty bit of trucks over in our blue rectangle and our green triangle. They've got to look at the proportion of truck purchases, big truck purchases, or new contracts with e-commerce distributors that were done disproportionately or uniquely because of competitive products. Whether they're in the green, blue, or white. Whether they're in the green, blue, or white. I thought you were just talking about white. We're just talking about white. Seriously. They have to do it. I didn't ask you what your matter is. My thought, again, on this statutory language is that I don't care if you already counted it under A2. When you read this, it says when you decide an appropriate share, stop and go, hey, is there any cost associated with competitive products that is unique or disproportionate? Whether you counted it under A2 or it was the white area, I don't care. However many things are unique or disproportionately associated with competitive products, you've got to factor that in and determine an appropriate share. That's exactly right, Your Honor. And that's why we win because that's what they didn't do. Okay. Then you said you were arguing about the white box, but you shouldn't. You should be arguing about they didn't get. Their argument that we already got unique or disproportionate under A2 is a so what. You have to count. It's a double counting. Right. It's actually not a double counting, Your Honor, because remember these are the white space and the blue and green space are residuals of one another. They are mutually exclusive. So our argument is A. No, no. Mutually exclusive whether it's called institutional or attributed cost. Yes. Right. What I'm suggesting is that when it says any cost in B, that means any attributed or I'll call them residual, residual cost, any attributed or residual cost that's code for institutional cost that are unique or disproportionate. Those together have to be factored in in calculating the appropriate share. That's correct, Your Honor, but we know that A3 requires the Postal Regulatory Commission to ensure that competitive products collectively cover the appropriate share of the institutional cost. Failure under A3 pertains just to the white space. Maybe the way to put it, Your Honor, is the failure under A3 pertains just to the white space where the PRC declines. But it could be a part of your appropriate share of institutional cost could be you're an 800-pound gorilla over here when it comes to truck buying. Yeah, you're capturing some of that under attributed, but we're going to stand back and go even if you're capturing a little bit of it, you're an 800-pound gorilla here and so as to things that you should carry a little more weight in some sense under institutional cost. Because the appropriate share is not so much a mechanical calculation. It's a judgment about you bring factors in, but if they stand back and they make a judgment about what's appropriate for them to do. That's correct, Your Honor. That's why I call it a double-accounting. Your Honor, we agree with you that any cost that must be factored in covers both buckets, but A3 requires you to ensure that an appropriate share of the white space bucket be factored in. I understand that, but you can factor in. If I'm just saying that these things were reversed, if this were all blue and green and it was only a small amount and it was white, you would still say fine. The fact that they have all these things that are, let's assume some percentage of this is unique or disproportionate. We're going to look at the big picture of how unique or disproportionate your costs are in deciding what your appropriate share of institutional cost is. Absolutely, Your Honor, and I think we're in agreement, and the PRC failed to do that, and that's why you must agree, ma'am. And how we know that the PRC failed to do that, Your Honor, is the multiple times in the order. And I would cite you joining me at 657 as a particularly clear example. Multiple times in the order, the PRC says there are no, at 657, second paragraph, carryover paragraph, middle, there are no costs uniquely or disproportionately associated with competitive products that are not already attributed to those products. This is because all costs that are uniquely or disproportionately associated with competitive products exhibit a reliably identified causal relationship with a specific competitive product group. That simply is false. It cannot be true because reliably identifiable causal relationships produced a very narrow test in the last docket in this court's last order that just gets you to the incremental cost at the right end of the graph, the blue rectangle and the green triangle. Everything else is institutional, and the Postal Regulatory Commission approved an approach that would not consider whether any of the institutional costs in the white space were disproportionately or uniquely associated. That means they leave out the larger chunks. Yeah, I do understand what you're saying. I want to just make sure. I want to follow up with one more step. As I understand your argument, you've got the attribution under A2. And you're saying in addition you have to do B. A3 plus B. Plus B. You can still end up with, having done that, let's say they did it to your satisfaction, you can still end up with some inframarginal costs that are not going to be captured either way, and they're just in the institutional category, and they will be subject to the appropriate share formula, right? Yes, Your Honor. Institutional costs are subject to the appropriate share. Yes, no doubt about that. I'm just trying to figure out, this is all I was trying to do in the very beginning. When you agree there's some inframarginal costs that will be captured neither by 2 nor B. That's correct, Your Honor. And they're simply institutional. And they will be dealt with under 3, A3. It's a sharing. Your Honor. Dealt with is not the right word. They are not ignored because in the sharing arrangement you're considering institutional costs, which would include those inframarginal costs that are not captured by A2 or B. That's correct, Your Honor. So in the world there are two buckets of costs. The A2 costs, which won't capture all of the costs that are associated with competitive products, because the tax district is competitive. Then there will be the institutional bucket, and we can see that there might be some portion in that bucket that are disproportionately associated with negatives. I'm doing it the other way in my head. I'm doing your argument to me that you've got to do 2 first, and then you do B, and you now have all of those inframarginal costs accounted for the way you think they ought to be accounted for under the statute. Now you can go ahead and figure out the appropriate share. Exactly, Your Honor. But you know what? PRC didn't do B. No, I understand. Now the last question I have is, now I understand your argument, what are the examples, your best examples, of inframarginal costs that have slipped through the cracks? I don't want to look at the chart. Just tell me what you're talking about. I like to visualize. What are you talking about specifically? Your Honor, uniquely associated with competitive products would be things like the transaction costs of setting up new accounts with e-commerce shippers. They don't send letters. They send packages. Those transaction costs are just uniquely associated with packages, but they are neither attributed in the incremental cost A2 analysis that you reviewed the last time, nor were they considered by the PRC under A3. Second example, better example, Your Honor, are larger trucks. I mean, this is very intuitive to me. I hope it is to Your Honor. The post office is out there buying billions of dollars of larger trucks. Why are they doing this? Letters and postcards, old-style mail has been decimated by e-mail and new technology. They should be buying smaller trucks and using less fuel, but instead they're buying bigger trucks and more of them. Why? Because packages take up so much cubic volume. The capital cost of those larger trucks was never attributed in at least not except potentially in a trivial amount under the A2 analysis. You're not going to find those big trucks down in the blue wrecking or the associated cost piggyback cost in the little green truck. And they're not captured under B because the agency concedes. We're not even looking there. Exactly, Your Honor. And if they're not captured under B, you mean they didn't capture them under B or the statute B doesn't apply to them? No, no. B must apply to them. Okay. I want to make certain. Yes. Our position, Your Honor. Not captured under B means they didn't capture them under B. B plainly under its tax requires a consideration of them in the A3 analysis. Why? Because a larger truck is potentially... Potentially they don't. It's a truck. They say we've got cost drivers and our calculations already figure out, look, probably when we send the truck out it's going to have a bag or two of mail on it as well as all these boxes. And so we've calculated what percentage of the cost of that truck, its size, its gas, its fuel, everything, its maintenance is associated with packages and what's with... I'm using mail as a shorthand for the non-competitive. So that's incorrect. That is incorrect. Don't be bedazzled by cost drivers. Cost drivers are just the units of measurement that determine the X axis. We know that if there are a certain number of cubic unit miles driven to deliver a certain package product like three-day delivery of a three-pound box, the cost drivers just tell you where on the cost curve you're going to attribute the price that creates the blue rectangle. Cost drivers are just a device for determining the incremental cost. It's undisputed here. When we say incremental cost, how much more does it cost us to deliver packages than if we had to drive trucks to deliver mail anyhow? But that's not what's determined there, Your Honor. Remember, A2 has been interpreted. And I commend that this Court's very careful analysis of this in the prior docket. At 890 at page 1057, you will see that there's an extremely narrow definition of what costs can go into the incremental cost analysis that generates what we're calling for shorthand the blue rectangle and the green triangle. And at 1057, at this Court's prior opinion, it shows how narrow the determination is. At 1057, the volume variable cost defined as the marginal cost of the last or cheapest cost driver unit multiplied by the total number of units accrued. So what goes into that analysis, Your Honor, may be a portion of the truck that is reliably, plausibly identified with that product and only that product. But that, by definition of design, leaves out all of the incremental cost of the larger trucks that is a common variable cost. The common variable cost of those trucks are over in the white space. They weren't attributed under A2, and they cannot be attributed under A2 because of the PRC's error in B. They didn't do the B analysis. Now, don't take this from me, Your Honor. Take it from the PRC. If you go back to the docket, your review of order number 3506, where you approved the very narrow blue rectangle, green triangle analysis, we opposed it, we lost, they won. They got this very narrow analysis. Back in that docket, if you look at order number 3506 and you go to Appendix A at page 16, you will see the PRC says that common variable costs are treated as institutional costs. The trucks we're talking about involve common variable costs. It takes more trucks to deliver more packages. It takes bigger trucks to deliver more packages, more drivers, more fuel. That's a common variable cost that reflects the cost of the few letters that are in the truck and the big packages that are in the truck, Your Honor. That was not captured in the A2 analysis by definition and design because as this Court approved the A2 analysis, you never look at the common variable cost because they're in the institutional cost docket. And that's, Your Honor, why I think it's important to go back to the words of the statute. Why doesn't A2 subsume A3? The language of the statutes are different. Reliably identified causal relationship is much narrower than uniquely or disproportionately associated with. And second, the PRC's own prior decision on which this panel, sorry, which this Court in the prior panel expressly relied. If you go to the prior decision at 1067, it isn't just the representation of counsel at argument. This Court accepted at page 1067 that the way that the panel put it is, UPS offers no reason to doubt that the Accountability Act's prohibition on cross subsidization and requirement that competitive products cover a share of institutional costs will adequately ameliorate any competitive deficit left by the Commission's approach to cost attribution. That was the premise. The premise of the prior decision was A2 captures an exceptionally small sliver of package cost. But if you increase, another way, just to turn that down, just to make sure I understand the argument, another way to give you the relief that you want is to simply increase the appropriate share of the institutional cost. Well, yes, Your Honor, but we think that No matter whether B was effectively employed or not. If you increase it, what is it, 5%? What's the percent now? The new formula generates 8.8%. So if you increase it to 14%, you wouldn't be making this argument, right? Well, Your Honor, if you want to order it increased to 30%, which is what we suggest at the PRC, we would say that. We've got the relief you want. We would, but Your Honor, we don't understand. Even though we would say, well, you know, the other way to address it is you really should have done this under B, but we'll take the 30%. That's what we, right? Well, Your Honor, Your Honor I'm just trying to make sure I've got all the nuances. That's what you really should have. Two ways to capture, if you increase the percentage, it doesn't matter whether you're screwed up under B. You're saying a bigger percentage of the institutional cost, because they're in there, the institutional cost. We agree with that, and that's what we said in our comments. I at least finally understand. The institutional cost, Your Honor, if the post office is making 30% of its revenues from packages, and the packages are going up while the letters are going down, and if the A2 process fought as we thought it was at the time, if that generates 30% of cost being attributed to packages, then doesn't it seem very odd that you would attribute 8.8% of all the rest of the cost of the post office to packages? We think that's very odd, Your Honor. And let's remember, it's not a chart. Not if they're really only about 3.4% of the stuff they deliver. I'm not sure even if you got the analysis right that they would have to increase the number, or certainly go anywhere near 30%. They could still do the analysis, but they have a lot of discretion under appropriate share. That's exactly right, Your Honor. And so as much as I would like to accept Judge Henderson's proposal that you simply order a higher share, we don't ask for that. We would like to accept it, but we want to respect the agency's expertise in the first instance to decide the appropriate share. But the one thing that is not within its authority to which you should defer is the ability to nullify A3 by ignoring B. You rarely have a case, and here I really do commend the language of the TRC itself on pages JA657 and all the other pages where we say we don't have to do the disproportionately or uniquely associated cost analysis under 3633B at all because we did it already under 3633A2. We know that cannot be true. It cannot be true because the way they prevailed on 3633A2 in the prior decision of this court was to exclude from A2 analysis many common variable costs like the cost of larger trucks that need to be at some – that are disproportionately associated with packages and yet were not considered in the analysis. So the relief we ask for, Your Honor, is vacate and remand so that institutional share, which is the output, can be calculated in accordance with the statutory requirement that that output have the input of uniquely and disproportionately associated institutional cost. And without that input, the output can't be statutorily corrected. Just to clarify my question where I had started before is even if, let's just imagine in a different world, they had captured all of the, for short handles, the reliably attributable costs under A2. Your theory is they – sorry, they got all of the costs attributable to – sorry, they got reliability. They got all of the costs. They got the whole graph. All of that is in the A2 number. Right. All right? So let's assume there's – because your argument here is they got this little, on your graph, this little portion over here. There's this big thing of whether it wasn't reliable or institutional or both or whatever, they missed all that that has to get covered under institutional cost. But imagine everything in that graph actually through some great mechanical computation by them got put into A2. The way I read B is that doesn't matter. I'm suggesting this is what B says. B says, I don't care if you already counted for it up there because some of those things you included in A2 are either uniquely or disproportionately – I mean, not all of them, but some of them are uniquely or disproportionately associated with competitive products like your big trucks. We just wouldn't be buying bigger trucks. Or we certainly wouldn't be buying the volume of bigger trucks that we are. And so we need to take that picture, that concern about unique and disproportionate costs that you have, even if you're already paying them, into deciding what your fair share of purely institutional costs, not the things we couldn't get because it wasn't reliable. You know, the postal commissioner's salary. You still should owe a bigger share there, too, because you're doing more up here under A2. Is that wrong? It's correct, Your Honor, as long as there is any residual institutional cost. So if we take your hypothetical to an extreme and every single cost under A2 that is associated with – if reliably, identifiably costs were broadly interpreted as all costs associated, including the salary, then there would be very little left of the institutional cost. I'm sort of driving this question, dividing the world into reliably attributed costs, costs that we think go with competitive products, but we can't quite – there's some there that we know go with them, but we can't reliably compute them, and then what we would in lay language call purely institutional costs, the commissioner's salary, health benefits, these types of things. Your Honor, we believe that if there is any institutional cost remaining after the A2 analysis, even if the triangle, the blue rectangle and the green triangle got bigger and the white space is shrunk, there still must be – they must perform the B analysis. Here they refuse to follow their statutory mandate to perform the B analysis. The B analysis includes unique and disproportionate costs of all unique and disproportionate costs, whether some counted already under A2 or not. Correct, Your Honor. Even if they were counted already under A2, if there's an institutional cost bucket left, there needs to be a decision of what the appropriate share is that's disproportionately or uniquely associated with competitive product costs. Now, I concede, of course, disproportionately and uniquely associated costs of competitive product isn't input. The appropriate share analysis still looks at other things, and B still looks at other things, market conditions, but you rarely have such a clear statement by an agency that we simply refused to follow our mandate under B because we assumed that everything was covered under A2, which it could not be as a matter of law, given this Court's prior decision, because A2 analysis never considers common variable costs, like water and trucks, that might be disproportionately associated with packages. And because of that, Your Honor, we think this is an easy case for contrary to law and for remand. On that front – sorry, I'm sorry. I know you're trying to wrap up, but on that front, this is not quite like other statutes because they actually – before we get to the language we've been talking about, they have the authority to modify this minimum contribution amount or eliminate it altogether. So wouldn't that allow them to go, look, here's what we think really matters for appropriate share, and that's the competition factors that we've looked at, and they had a few other factors they looked at here as well? No, it does not, Your Honor. So it is true that the agency could have decided to eliminate the institutional cost minimum contribution requirement. This was enacted a long time ago before package delivery spiked and the world of e-commerce changed the mail as we know it. And if it had turned out that packages were a bust and nobody sent packages through the postal service, then maybe there would be no need for the A3 analysis. But it can't be when package – but it's arbitrary and capricious in an age where package deliveries are 30 percent – 30 percent of a tributable cost. It's arbitrary and capricious to say that the costs are – the uniquely or disproportionately associated costs are zero. Remember, that's – it's the inexorable zero that is the problem in this case. The PRC said that the A3 number is zero because we did everything under A2. That cannot be right. And the power to eliminate might – if they said they were eliminating, then we'd be here saying that was arbitrary and capricious in the current world. But they didn't. They said, oh, yes, we considered it and we assumed that we didn't have to calculate it because we already did it. They can't have already done it, and that's the essence of their argument. Judge Henderson, you've been very indulgent. If there are no further questions, I'd like to reserve some time for rebuttal. All right. We'll give you some time. Thank you. Thank you, Mr. Sheehan. May I please report, Mike Sheehan for the Commission. UPS wants to make it seem like the agency completely ignored, in its order, some costs that were either disproportionately or uniquely associated with competitive products. That – I cannot emphasize this enough – is simply not true. What the agency actually said is when figuring out what the appropriate share of institutional costs needs to be borne by competitive products, the answer is not to raise the appropriate share like UPS wants. So we heard counsel from UPS say that the only way to do it is to have the percent input equal the percent output, 30 percent in, 30 percent out. But if you look at the statutory text of 3633B, that's not at all how the statute works. The statute – I get that you have a lot of discretion, and it's certainly not for the cost allocation approach here at all. But what they're saying is, look, you just missed the vote under B because you said we're not going to look at costs that are unique or disproportionate in B. That's not correct. Because we already looked at them under – we already factored them in under H2. That's not correct. There's a difference between looking at a cost and then setting the appropriate share to take a certain percentage of the cost. So let me just start with this example of trucks, because I think it's actually very important. So I don't think counsel from UPS can contend that the commission never thought about trucks, because everybody knows, as our cost driver example demonstrated, that in order to deliver packages or mail, you need trucks. And so the cost attribution mechanism, the price scores in A2 already take that into account. It's also taken under account in A1. Are they wrong to say that you only took it into account in the little blue and green, and you missed it here? So I want to be precise about what I mean by taking it into account. So the cost of the truck that can reliably be said to be caused by a competitive product, that product has already got to cover that price. That's right. It is true that under the commission's cost attribution methodology, we only do what is reliable, and what is reliable is the cost that would disappear if we stopped making that product, right? And so there are some portions of the truck that don't go to that particular competitive product. It doesn't make sense to make that competitive product be set there as a price score, right? But it's not true that we completely ignored the stuff on the other side. That cost did not disappear from the commission's analysis. That cost is an institutional cost. Yes, it's an institutional cost category. It is. And it's calculated under A3 is what you're going to argue. It's not just an argument. That is an accurate characterization of what the commission does. So anything that's not in this green or blue box, and we disagree with their annotations on the court's box. We are discussing this on the red brief at page 51. That gives a more accurate picture of what that graph is meant to do. Anything that's in the green or blue box, a particular competitive product has got to cover the cost score. They want you to believe, Your Honor, that everything in white goes away. And that is simply not true. Everything that is white falls into the category of institutional cost. And so the result of the formula is 8.8% of that plus, you know, all of the other space. But I think the question is, is there anything in the white that's uniquely, I think, or just maybe disproportionate would be the word for the trucks, disproportionately associated with packages. Is there anything in the white that's disproportionately associated with competitive products? Look, I have two answers to that, Your Honor. So first, under the commission's interpretation of unique and disproportionate, the answer is no. Because, you know, construing unique and disproportionate from first principles, the commission defined those terms in a way that just happens to match what is in the green and the blue but not the white. But the other more important piece of this answer, Your Honor, and this really is critical, it doesn't matter even if UPS is right about what the word unique means or what the word disproportionate means. Right, and this is important because, again, suppose that they're right, that some stuff in the white really does qualify as uniquely associated or disproportionately associated, and the commission is wrong, a premise that I do not mean to concede. It's not that what follows then is that the commission has to, you know, attribute all of that to competitive products as a whole. The appropriate share doesn't need to be that percentage. I get that it's not that mechanical, but isn't the problem here if we assume there's something in that white area that is disproportionately associated with competitive products, like bigger trucks? You just wouldn't need for- No, Your Honor, again- Just assume there's something there. You're doing more big trucks than you would if you were just doing first class mail. I know there's more. If you're just doing that, you're doing that, and it's disproportionate. You didn't factor that in. The commission didn't factor that in in determining the appropriate share in this case. Is that correct? The commission's formula doesn't require- No, no, the commission said there's nothing, because there's many factors. It's an including, so there's more than what's even listed in the statute, but it says let's look at things like the competitive environment, and you've got to do at least the including things, and things that are uniquely or disproportionately associated with competitive products. You've got to at least look at those, and maybe there's some other things you throw in. So it doesn't mean I'm not at least suggesting in this question that there's any strict formulaic response. It simply did the agency do what Congress said and said that your analysis has to include in deciding their appropriate share. You've got to factor in things that are disproportionate. That factor could be, yes, they are, but it's so marginally- I don't know if you can have marginally disproportionate. It's just over the disproportionate line, or it's outbalanced by other things, or, yeah, it's there, but the competitive environment. It's not that they couldn't explain it away and even end up in the same number. It's that they just didn't talk about it. That's not right, Caroline. Okay. It's not right because UPS wants to trade on the intuition that somehow there exists this more stuff, right, that is not already captured either under A2, which is what UPS spends all of its time talking about, or here, A1, which is more broad, and would include a lot of those other costs. This is the prohibition on process utilization, right? Let me try that. Great. You're changing the floor of the competitive product cost by doing it your way as opposed to their way. In our first case, we said there are some inframarginal costs that are not being captured under 2 with respect to competitive projects. We said, but the commission assures us it'll all clean up in the end. I took that to mean that when you get to 3, the percentage would be a different number. That's not right. But forget that. I don't know what the cleanup means unless you go to B and deal with those inframarginal costs, because if you don't deal with it, we have already said you have not captured all inframarginal costs in the way you do 2. It's all because the definition is so narrow. So there are some leftover inframarginal costs. You've just kind of thrown them in the institutional area. So unless you do 3 to their satisfaction, you've changed the floor on the competitive project cost. If you found something under B, the floor would change, because you would have what you found under 2, and you would add what you found under B. Yeah, that's right. There's one minor technical detail here, which is B, A3 is the operative language, not B. A3 sets the minimum contribution requirement. That's for all competitive products. No, no, no. I understand that. What I'm saying is to try and get the cost for a competitive product, if you leave out something, you've changed the floor, right? Correct. And if A2 doesn't, in fact, take into account everything, then the floor is wrong. No, Your Honor. A2, as this Court held, takes into account precisely those costs that can be said to be caused by the production of a product. I understand. They disagree with that. No, no, no. I think the disagreement is harder than that. They are saying, I thought you were agreeing with me, that if you leave out the analysis under B, as well as acknowledge that our Court has already said you've left out some inframarginal costs in your A2 analysis. You just have, maybe because of the way you define it. So the inframarginal costs are not fully accounted for in your A2 analysis. They're not, and the Court has said that, and they were assured by you that, well, we'll work it all out. Well, they're saying, well, one way you can work it out is do what's required under B. And they're saying you haven't done it, and so the cost for the competitive project, the floor is wrong. I understand that that's what they say there, Your Honor. Well, why is it not wrong? It's wrong, they're incorrect because, and this conflation just flows throughout the briefs and also through the first argument. Again, they're conflating, you know, whether the Commission has considered the costs with whether the appropriate share is set with a percent input and a percent output. Now, that answer, I know, doesn't answer Judge Millett's question, which is, you know, maybe, you know, it's true that percent input doesn't have to equal percent output, and the Commission still needs to think about those costs, and it didn't. And I want to turn back to that question because it's very important. But as to this question, Your Honor, it is very important to remember that nothing in Section A.3 mandates a sort of proportional allocation that UPS believes is in the statute. And the word consider simply doesn't mean the appropriate share has got to be whatever costs are uniquely or disproportionately associated as a matter of the statute. And the reason we know that is because Congress said that you can make the minimum contribution requirement zero, even if there are a whole bunch of disproportionately associated or uniquely associated costs. So that's why I want to push back against the idea that somehow there is this mandate of 30 percent in, 30 percent out, that needs to be captured either through A.2 or A.3, reading B the way they do, or both. Right? That's the premise that I'm disagreeing with. Now, I know that that doesn't get the Commission all the way there, because even if you accept that, it is true that the Commission has got to talk about all of this stuff in its order. And the answer to your question, Judge Millett, is that the other side really wants to parse costs down in the manner that I've just been describing. So the only way you can fairly say that the Commission never thought about this is if you accept that the Commission simply pretended like some portion of truck costs don't go into institutional costs at all. And that's just not correct. No, no, no, no, no. No, my question is a different one. Does the Commission cause to address that some portions of trucks uniquely or disproportionately are associated with competitive products when it was trying to figure out what the appropriate share was for competitive products? I don't know about that. No, I think it's a question. Right. And if you can tell me where they did that. Because I had read your argument, maybe it's my fault, as saying because they had thought everything was covered under A.2, they didn't need to do that under A.3-B. Maybe it was helped by the scope of the board. Are you changing that? Because that's my assumption, too. No, Your Honor. No, Your Honor. Okay. Let me try to describe the Commission's reasoning instead. Okay. The Commission started with the text of the statute. What does it mean for a cost to be uniquely associated? What does it mean for a cost to be disproportionately associated? The Commission came up with definitions of those terms. Then the Commission said, we defined those terms in a particular way. And it's, you know, and I noticed some skepticism, so I want to explain this in more detail. Whatever those terms mean, we think we've done a good enough job accounting for it, either because the costs that would fall into those categories are separately baked into the price boards under A.2 or, critically, are separately accounted for by the prohibition on cross-subsidization in A.1 or both. Now, maybe there might be some cost that... Wait, wait, wait. You ran through the A.1 too quickly. No problem. Yeah, what does that mean? That's just a self-serving conclusion. A.1 is very important. What do you mean? I know it's very important. It says A.1 prohibits the Postal Service from subsidizing the costs of its competitive products with money for market dominance. Right. So the way that the Commission tests that is they figure out how much cost is going to go away if the Postal Service just stopped producing every single competitive product that it makes. I think it makes upwards of 1,000 competitive products or something like that. So suppose the Postal Service gets out of that business entirely. How many costs would go away? And then the Postal Service must then show that all of its competitive products are recovering at least that amount. So if they're getting back all of the money that it costs to make the competitive products, then the competitive products are definitionally priced in a way that doesn't have this improper subsidization because otherwise they would be getting back less money. Does that make sense? I hear you. Great. So the way that that intersects with this case is that a lot of the costs that UPS is talking about are reflected in that calculation under A.1. So, for example, again, the question goes back to trucks, right? And, you know, obviously trucks don't just do the delivery of parcels. They also deliver letters and all of that, right? Well, what about the salary? Talk about this in terms of – I don't want to interrupt you because it's very important for me to understand this. But talk about it instead of in terms of the person who's in charge, who spends a large percentage of their time negotiating contracts with e-commerce companies. I'm going to call it mail. I know some mail is competitive, but I'm talking about first-class mail here. We're only talking about packages, shall we say. Yes. That's unique. And it's accounted for, too, right? And so the methodology that the Commission uses – You said accounted for. You mean it's already captured under A.2 or it's captured for under B? Under A.1 and A.2, Your Honor. And is it your position that once it's counted under A.1 and A.2, it doesn't have to be looked at again under B? That's how I understand the Commission's decision, and that seems to me texturally difficult. I understand why. If that's what you understand the Commission, too, has said, there would be that difficulty. But I don't think that that's the fairest reading of what the Commission did. Do you agree any costs in B means both? Absolutely. Both institutional and – so, you know, the Commission agreed with – Reliably attributed costs and – It's all – Old-person institutional costs and not reliably attributed costs. So, you know, the terms are totally different. The interpretation needs to be totally different. The textual interpretation of those terms that, you know, as explained in the brief and as cited in the order, you know, it's not that we look at this and we say that, you know, these must be equivalent for some sort of purpose. No, it's a statutory interpretation. We look at what the word means, what the word disproportionately means, and we say, like, it turns out these concepts map on. And maybe this delta will help to explain why the Commission hasn't just equated the two. Suppose that UPS were able to identify a, you know, cost that is, you know, uniquely associated or disproportionately associated that isn't adequately accounted for by the cost methodology in A1 or the cost methodology in A2. In other words, suppose that A1 is very bad and A2 is very bad, and the Commission is in the process of updating those methodologies. So, you know, the Commission knows that there's that gap because A1 and A2 just cannot capture that particular cost. If that were true, right, the Commission would, of course, consider that cost, and nothing in the Commission's interpretation of the statute would prevent the Commission from considering the cost. The problem for UPS is that every single example they give... Not that nothing would prevent it from considering. No, my question is, doesn't the statute require them to consider? Oh, exactly. But, you know, and nothing in the Commission's analysis says that they wouldn't, in other words. Right. If such a cost... But then they would. But no such cost exists, and that's the problem for UPS. There are no... So the Commission's theory is... I had understood the theory to be not that there is no such thing as a unique or disproportionate cost. I had understood it to be that there's no such thing as a unique or disproportionate cost, that we haven't already captured the cost of it under A2, and now you say A1 as well. Is that, which is correct? Does that give a second? The second is correct. But then if the statute says, I don't care. If B says, I don't care if you already counted it under A1 or A2. I'm telling you, when you compute the minimum contribution for purposes of A3, you've got to look again and say, you've got disproportionate and unique costs. Yeah, we already captured that up there, but in deciding what your appropriate share is, we're going to again factor in the fact that at least in one area or two or three areas, you're the 800-pound gorilla. And that's what I didn't see happening here. And so am I just misreading the statute, or am I misreading the commission's decision that they did do that? We disagree as to what the commission did. In our view, the commission's decision is best read to do exactly the sort of statutory analysis that Your Honor explained. I know that the other side is giving quotations from the conclusion, but there are two questions. The first question is, you know, what are these costs? And the second question is, what do we do about these costs with respect to the appropriate share? And as a first question, does the commission agree that there are some unique or disproportionate costs associated with competitive products, period? Yes, of course. It did, okay. And did it agree? Did it agree? Where did it say we're considering that in determining their appropriate share? Where are their decisions? I must not have read it correctly. Where did they say we're factoring that? If you'll give me a second, Your Honor, I want to locate the portion of the brief. Sure, it's a long decision, I know. So, the commission's discussion of these terms comes at pages 657, 658 of the appendix. And, you know, there are other places where they're coming from. 657, 658. 657, 658. And there are other places. The discussion of how these things are being factored in, so that's talking about the prior decision. 657, 658 is just talking about the heading for that is previous commission decision. Right. So, what the commission did, and so this entire section, beginning in A657 and going all the way until the end, the statutory appendix goes to 667, right? So, this is the critical, this section of the commission's order is the critical section of the commission's order. Right, on this issue. On this issue, right. And so, I hear them say, here's what our old decision did. And they're saying, but then that's what they say are not already attributed. Right, on 657 they say, don't worry, we already captured those under A2. Where do they say, but we're looking again under B to see, you know, should that influence, should the nature or the volume of these unique disproportionate costs influence their appropriate share? What does it say we looked again under B? So, that goes to the commission's discussion of whether to adopt the 30%, 30%, you know, proportional application. Right, but that's not, my argument isn't that you have to do it, you know, it's not percentage matching here. It's simply that, I'm sorry, where do they look at them and say, we looked at them and it is factored in and it gets, you know, 0.05 points or whatever and it's going to be a big computation and ends up reducing the number. Where do they come in? The way I would explain this is one, they summarize, so a lot of the commission's rationale set forth in this initial notice order, 402. The first piece of this, so 657 through 659, summarizes the commission's reasoning in that circumstance and also what the commenter said in response. And then at 659, the commission goes to UPS's statutory arguments. And so, you know, this is where UPS says, you know, like A2 and A3, I'm sorry, A2 and A1 don't deal with costs that are uniquely or disproportionately associated. They say that there are some disjunctive or, so all of UPS's arguments against the statutory interpretation that the commission gave are discussed and rebutted in that section. I don't care whether it's there or not. I guess my understanding of the statute is whether it's there or not doesn't matter. The fact that you counted it once doesn't mean you ignore it. So where does it say, we're factoring it into the appropriations? This seems to be responding to that first argument, not as to the second. Here's where we are factoring. Now, they could say there are some unique or disproportionate costs, but they're so small it just doesn't affect the appropriations. What does it say there? So this is 662 to 663. So now the commission says, so we've explained the statutory interpretation. We've explained why we disagree with UPS's. Then it says, like, disproportionately associated must be broader or it would serve no purpose. And then, you know, the commission describes all of the other factors that get to be looked at by the commission that the commission has to look at in figuring out what share would be appropriate. So the commission explains, for example, in the carryover paragraph, at the bottom of 662 and the beginning of 663, that there is considerable discretion given to the commission about what to do even, you know, assuming that there are such costs. And the next paragraph on 663 says a higher degree of flexibility is inherent in the appropriate share provisions. The relevant factors are set forth, and, you know, as noted there are more than one. Each relevant factor uses flexible language for the commission to interpret, and no specific finding related to a relevant factor would mandate a specific result with regard to the appropriate share level. And then it goes through, you know, from the bottom of 663 all the way to the end of 667. For the relevant factor at issue, UPS focuses on the meaning of disproportionately associated and responds to all of the textual arguments there. I guess this all seems to be an argument about what the statutory language means. I'm asking from the more practical point of, given our view of what the statutory language means. The commission is saying given our view, the commission's view of the statutory language, here's how we factor a unique or disproportionate cost into determining the appropriate share. Where does it say that? So, I mean, the commission says, Congress does not require costs to be found, only for the commission to consider whether any exist. Nothing prevents the commission from concluding that costs uniquely or disproportionately associated are in fact captured by the costing methodology it currently employs. And so, if I may, Your Honor, can I just explain why that responds to your question? The premise of your question is that the commission failed to even talk about this alleged missing category. Just to be clear, before you go on, not talk about, factor it in. Talk about it could be it doesn't apply. I'm saying it does apply and where did you factor it in? I think we need the same. I think we're just getting behind the word consider. For the commission to consider one of these unique or disproportionate costs, it does not mean that the commission has to do anything in particular with that cost when it comes to setting the appropriate share requirement. And I guess this is why... I think we'd have to explain why it thinks that doesn't affect the appropriate share because this is congressionally identified. And they didn't say we're not doing it here. They didn't make a decision under their authority to modify or eliminate here. No, they did. They modified the appropriate share and they used their authority. This was their authority to modify. They didn't modify the statutory rules for what you consider. If you mean modify the appropriate share, because they changed it from a 5.5 to a formula, we agree on that. But they did not, in doing so, they did not say we no longer need to consider unique or disproportionate costs. Of course not, you're right. And I guess... So where do they say? Go ahead. I shouldn't have interrupted you. I'm sorry. Maybe a quick explanation for how they factored it in. I apologize, Your Honor, for being unable to make this quite clear. It's my density. I understand the question to be, you know, where does the Commission say, in setting what share is appropriate, you know, we are going to set it at this particular number, regardless of whether there might be unique or disproportionate costs that exist that aren't captured in A2 or A1. It's difficult for me to point to an exact citation for that. I thought the Commission's position was all such costs were captured in the A2 analysis. Yes. That's why the answer is so odd. The starting premise is that A2 captures whatever B requires. No, the starting premise is that, according to the Commission's definition of the language in B, interpretation of that language, either A1 or A2 captures it. Right. And, therefore, it shouldn't affect the appropriate share? Correct. And so I'm trying to explain the therefore piece. And the Commission's therefore piece has many different parts. And, you know, with some leave of your honor, I'd like to just go through it. So, first, the Commission made a lot of very express findings about the relative health of the market and, you know, how the market is incredibly healthy. So the Commission pointed out, for example, that the three big players in the parcel delivery market have increased prices at about the same rate. This is at 530 to 531 of the appendix. The Commission pointed out that the Postal Service has a huge incentive right now to maximize the prices of its competitive products regardless of what the price boards are because it's hemorrhaging money on the market-dominant side and its ability to raise the prices of market-dominant products is, as your honor knows, quite capped. This is at 579 to 580. And, you know, the Postal Service has increased prices every year at a rate far exceeding inflation. It voluntarily exceeded the existing appropriate share requirement every year. There's no evidence that the Postal Service has ever charged less than cost. And there's a complete absence of antitrust actions levied against the Postal Service, right? All of these were factual findings. Right. And I'll go to the including as a relevant segment of the prevailing competitive conditions. And I'll go to that. Right. So then, you know, the Commission talks about this other piece, right? So the A1 and A2 and A3, right, and unique and disproportionate costs. The policy arguments animating UPS's position are that, you know, the Postal Service is this 800-pound gorilla. And so we have got to make them bear, you know, some greater portion than currently exists. In UPS's view, it's unfair that, you know, trucks bought to make some, you know, purchases are not, you know, put into the price floors for those products, right? And the Commission rejects that at every turn. The Commission says, you know, that there's no empirical support for the idea that there's any sort of relationship between institutional costs on the one hand and, you know, the Postal Service's focus on competitive products on the other hand. And so, you know, for example, they point out that, you know, most of the changes in institutional costs, the increase in institutional costs, are principally linked to factors that are unrelated to competitive products, such as Postal Service pensions, interest rate fluctuations, and other non-operational factors. That's at 675 of the appendix. And so this is where the Commission says, you know, we get UPS's view about what ought to be done with uniquely or disproportionately associated costs. We're still not going to adopt UPS's proposed solution. I get rejecting their solution. I got it. I got that. I'm just trying to – well, I think I understand your position. Just to be clear, I think you said, their position ultimately was, because these costs in our view are all captured under A1 and A2, they just don't factor into the appropriate share under the – No, the appropriate share does not need to be set at a percentage. No, no, no, but they – Whatever that number is. There's more to the world than don't consider it all. And we're not adopting your percentage. There's more they could do. And one is it could be, fine, we're going to factor that into a formula. It'll be some element in our formula. Or it could be we have looked at them and concluded that they are so small relative to everything going on here that they don't move the needle. And it sounds like you're saying that's implicit when they talk about all these other things, like really institutional costs or healthcare benefits these days. But I don't hear what – I guess I hadn't seen what they said. We've looked at them, as the statute tells us to look at them, and we've just concluded that a lot's already captured, and whatever isn't captured doesn't warrant any change to our formula because – I hope this answer is seen as – You're saying it's more piecing it all together. Piecing it together. The commission did it this way. The commission said we get that in UPS's world, except that there are these more costs beyond what they want to make to capture. This is indeed UPS's policy proposals. And I want to make clear that baked into those proposals is consideration of precisely those types of costs, the white area under the graph, the trucks and all of that. So the commission had to think about those costs in the course of assessing UPS's alternative proposals for what the appropriate share ought to be, because those alternative proposals are essentially dependent completely on these allegedly omitted costs. In other words, UPS's arguments are – They had to consider them. That was simply rejecting their formulaic approach, and that's how I had read it. It's, look, you want what they were calling essentially cost allocations, a rejected methodology. But that's not the same thing as saying we've done our job of looking at these things and it moves the needle on IOTA or it doesn't move the needle at all. That's my point. It is, Your Honor, because the reasons the commission gave for rejecting that sort of proposal was not just, you know, we don't think the statute thinks 30 percent equals 30 percent or whatever. The commission made a policy judgment, and, you know, examples of this are, you know, when we talk about proportion allocation and all of that, the citations that are given and, you know, throughout the order there's a lot of discussion of these proportional type allocations, right? The commission didn't just say we can't do that as a statutory matter. The commission said that as a matter of economic policy we don't want to do that because we don't think that it makes sense to require products to bear costs like that. Instead, it makes sense to require competitive products to collectively cover, you know, some other percentage set by this formula that reflects a whole bunch of other factors, right? Let me ask you one last thing, please. In Judge Cano's opinion, says incremental marginal costs are not fully captured in the H-1 policy. That was our conclusion. So they're somewhere unaccounted for. Yes, Your Honor, but I want to be precise about how I answer that question. You mean some of what you've been going up to now has not been decided? I just want to continue. This language is incredibly complicated, so I just want to make sure I'm using the exact words. What the court held in the previous UPS case is that for the purposes of the A-2 cost floor, right, you know, the commission has captured all of the inframarginal costs that a product has cost. It's true that under A-2 there are other inframarginal costs that the products cannot be said to have cost, but that might exist, right? And so the answer to your question is the commission didn't say those costs disappear. Where are they accounted for? My question is simple. Where are they accounted for? They are all institutional costs, Your Honor, every single one of those costs. No, no, I understand. You're saying anything. I mean, we keep going in circles, but I guess I need to make sure the circle hasn't changed. Anything I captured in 2 that you're saying is an institutional cost. Yes, I understood that. Yes, an institutional cost. That means you're putting the weight on 3. You're saying the answer is going to be in 3. We'll do our work nicely there, and so no one should complain. Is that adequate? Because we've got a problem. The court has said, and this is the part that I hear no answer to, the court said you're not capturing all inframarginal costs under A-2 that are there to be captured. You are not doing it, but we hear you, and it'll probably all work out. I honestly thought that what the court probably might have meant was that, you know, there's room under B, and some of those are, and then you gave me a very vague answer, which I don't, I don't want you to try and answer it, and I'll go back and look at the materials. You said, well, that's under A-1 somewhere. That's how we account for the inframarginal. I'm actually listening to you. You said that's where the inframarginal costs that have not been properly captured under 2, we've accounted for them somehow under A-1. I have the faintest idea what you're talking about, and I still don't get it if the statute says you have to look at B. We previously have said under 2, we made a point it's inconsistent with your premise. Everything under A-2 you said, everything required under B is captured under A-2. We previously held that's not true. We said that A-2, your A-2 analysis does not capture all inframarginal costs. It doesn't do it. So your starting assumption seems wrong. No, Your Honor. So, first, I want to make clear what the court held in your answer. I know what the court held. Tell me what your answer is. I know the court held you had not captured all inframarginal costs. But the court did not hold, and this is the reason I kind of want to push back, Your Honor. The court did not hold that everything is going to get taken out exclusively in the A-3 slash B bucket. The court's reference to the prohibition on cross-subsidization is exactly the answer that I was giving with respect to A-1. So that's just the first piece that I wanted to say. The second and, I guess, more fundamental answer to your question is that Your Honor's question assumes that what the commission did was have three categories of costs, essentially. So there would be costs attributable. Then there will be institutional costs. And then there will be- Institutional, so that's residual. Exactly. I'm worried about the A-2 and B. Right. Okay. And so every single cost, everyone that is not attributable under A-2, is an institutional cost. That is part of the equation. And some of those institutional costs may be B costs. Some of those institutional costs- May be B costs. Could be, but, you know, the court never took a position- Okay. But we're now there. B costs, and it won't matter whether you account for them. They really are B costs, right? Yes, Your Honor. We're not disagreeing at all with respect to the structure of the section. Where we are disagreeing is whether the commission has adequately accounted for them. And, you know, as I've indicated in an answer to you, there are two ways you might think about accounting for, one of which is flatly wrong. And that's the, you know, it needs to be baked into the appropriate share like the other side says it does. The correct way to think about and consider to account for is whether the commission adequately discussed it. And for the reasons I've given, you know, we think the commission's order is best interpreted as adequately explaining why notwithstanding UPS's arguments, you know, accepting the existence of these theoretical costs that are not adequately captured by A-2 or A-1, it nevertheless makes sense to set the appropriate share where it is. Because the formula from their judgment is their share of the stuff that wasn't covered, whatever wasn't covered and gets shipped down to the B, their fair share of that is covered by our formula. The appropriate share. Their appropriate share. The appropriate share is covered by the formula and their factors. Right, because if it were true that the commission zeroed it all out, right, there wouldn't be any appropriate share requirement at all. It would be zero percent. But instead, the commission's formula has resulted in an appropriate share requirement. It would be zero percent because of their findings about the competitive market. No, if UPS's characterization of the commission's decision were true, in other words, the commission ignored all of this. The only stuff that the competitive product has to bear are, you know, the costs of A-2 and then, you know, also satisfy the cross-subsidy mechanic of A-1. If that really were what the commission said, and the commission thought that that were adequate under the statute, there would be no appropriate share requirement at all because, after all, it's all captured in A-1 or A-2. Right, but the point is it's not that it didn't go, that it went away. The commission created a formula that made the appropriate share requirement go up. The real dispute in this case is that it just didn't go up as much as UPS wanted it to go up. And UPS's statutory argument is that, well, the commission didn't explain itself and, you know, disagree as to whether or not the commission did. But, you know, fundamentally, the commission's explanation is we hear UPS's arguments about the existence of all of these other costs. But it doesn't matter because given our expert assessment of the prevailing market conditions and given our expert assessment of the economic underlying UPS's proposals, we're not going to do zero percent. We're not going to do 30 percent. We're instead going to adopt a formula that reflects the changing market power and market position of the Postal Service. Thank you. Thank you. Ms. Sullivan doesn't have any time left. All right, why don't you answer any questions? Thank you for your indulgence. Just three quick points. Mr. Shee said that there were uniquely or disproportionately associated costs that the commission did consider. And then we watched as he paged through the order from pages 657 to 676, and there were no such costs. If there were such costs considered, what were they? What were the disproportionate costs of tax? They're nowhere to be found. Why is that? Because Judge Edwards and Judge Malik, you're exactly right. The commission said on 657, and it said at least a dozen other times throughout the order, the reason we didn't consider any disproportionately or uniquely associated costs under B is because we already did it under A2. But we know from this court's prior opinion that can't be true because, as Judge Edwards correctly pointed out, this court could be relieving inframarginal costs on the table, but they'll be sorted out in the A3 docket, as Mr. Shee promised us at the last argument. Well, now A3 didn't sort it out. It left the inframarginal costs that might be disproportionately associated with packages, like bigger cuts, on the table. And he says, oh, well, market conditions may take care of that because A3 is just a judgment call. But market conditions is just one of two statutorily mandated factors in B. You must consider market conditions and disproportionately or uniquely associated costs. What I say is we've got this formula here. And that was my last point, Your Honor. The formula doesn't do it either. Market conditions doesn't do it. The formula doesn't do it either because the formula, by the government's admission, has nothing in it. There isn't a single variable in here that has anything to do with disproportionately or uniquely associated costs. It's all about market conditions. So if you have no input into the formula from the statutorily required disproportionately or uniquely associated cost factor, the formula can't spit out something that conforms with B. And last, A1 doesn't solve it, Your Honor. I understand Mr. Shee's desire to turn to a different part of the statute than A3 because the Commission so flagrantly ignored its obligations under A3 as expressed in B. But A1 doesn't pertain to products that are jointly caused costs between market dominant and competitive products. And most importantly, A1 is not the Commission's answer. Do I need to invoke Chenery? If you look at page 657, the Commission said we didn't do the A3B analysis because we did it in A2. They didn't say we did it in A1, nor could they. And finally, A1 and A2 together leave on the table, unconsidered by the Commission. Nearly half cost the post office. Two-thirds of truck costs are not attributed to any product. And you're right, Judge Edward, it does affect the price floor. And the reason the price floor matters is Congress said we enacted this statute to ensure fair competition in the delivery of packages where there's finally a private competitor to the monopoly of the post office held in other domains since Benjamin Franklin. When the Congress said consider fair competition and the PRC by its own admission leaves out a statutorily required factor, that is arbitrary and capricious. And I want to be clear on the narrowness of the remand. We request Mr. Shi was incorrect to suggest that I'm asking you to deliver a proportional different result. We're not asking that. The very narrow thing we ask is to send it back to the Commission so they can perform their properly required task of deciding here are the institutional costs, let's find an appropriate share, but to get there we must have a statutorily required input of looking at the disproportionately Do you say there has to be something in the formula? No, they don't have to have a formula at all. No, but could they go back and go we looked at it and we don't think for a whole bunch of reasons that it moves our needle and we're sticking with the same formula. If they come out with that outcome we might be back saying it's arbitrary and capricious, Your Honor, but at a minimum you need to remand that they do their job and they perform the B analysis, which they didn't do because they said they've done it and they can't have done it. So we respectfully request a very narrow remand and send it back to do their job under B. Thank you, Your Honor. Thank you.

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A67922. Public record. Not legal advice.