Legal Case Summary

cases number 13, 1391, Mueller, commercial, the main coagency United States, Mr. Baseford. Thank you, Yarnis. Good morning and may it please the court. Yo, hi, Baseford, and I'm here today on behalf of Mueller commercial and Southland Pipe Nipple Company. Mueller commercial is a producer of pipe fittings in Mexico and also produced is a reseller of the pipe that was the subject of the administrative review below. It is undisputed that Mueller commercial is all fully cooperated during the proceeding, answering all the questions that the department asked and participating in a one week long on-site verification. Despite that full cooperation, commerce applied an adverse inference in the calculations of Mueller's anti-dumping margin. So let me tell you what the problem that I see is, I'm sympathetic to the notion that cooperating parties shouldn't have their dumping margins increased as a result of adverse facts available. But what the government argues, as I understand it, is that there is a necessary deterrence here, that this is unlike Shanzu, where there was no relationship between the cooperating party and the AFA party, that there is a relationship here and that your client had some leverage to force the non-acorporating party to cooperate and that there was on top of that a risk that if the government did not use AFA under these circumstances, that there would be a tendency to re-root the exports so that the high dumping margin, the 48% or whatever it was, could be avoided and the exports could go through Mueller having a much lower dumping margin. So those are the two problems that I see the two potential distinctions from Shanzu. Could you address those? Absolutely, I'd be glad to. And let me address the second one first. In the final result that the Department of Commerce published, any pipe that was exported by turnium would be subject to the 48% rate and any pipe that was exported by Mueller, that was produced by turnium, would be subject to the 19% rate. So there's absolutely nothing unusual about the fact that a producer's pipe is subject to different cash deposit rates or ultimate dumping margin. All right, so what I think commerce is saying is that under these circumstances, there would be an incentive for turnium to stop its own imports in the United States and instead to funnel them through Mueller to avoid the high AFA rate of 48%. Right, and the Department has tools available to address that, which is a knowledge test

. And that's what they found. If turnium has knowledge at a particular cost sale to Mueller is going to the United States, then that sale is subject to turniums, right? However, if turnium has no knowledge that that particular sale to Mueller will be sent to the United States and Mueller exports some pipe that was produced by turnium to the United States, it's Mueller's rate by the pipe. That's the way the anti-cumpling system is set up. If commerce has a concern that that reseller policy or that reseller methodology can result in people shifting, then that was concerned to be addressed up the street. How often does that happen? And in the Indian anti-dumping proceedings, that commerce says, oh no, this is a really a turnium export and it's just funneled through Mueller. And that evasion rationale, how often does that come up? But you see, it's not evasion. It's exactly what commerce found and said in Tong Mung, which is that a situation where a producer sells to multiple resellers and each reseller has its own rate. There's no evasion because one reseller sells to the United States versus another one. I mean, this comes up, I would say often, if there's a whole policy at the Department of Commerce, they've developed this knowledge test to determine who is the exporter. And that's the critical question because dumping margins are calculated on the exporter. And for Mueller's margin, they looked at Mueller's US sales and compared them to Mueller's home market sale. So there can, and that is what dumping is, it's the comparison of your US price to your foreign market price in this case. The attorney and whatever's turning themselves in the home market is irrelevant to this calculation. And so this notion that they're trying to close a loophole or deal with evasion, it's simply not true because had they found during the review that Mueller was merely a conduit for turnium, then turniums rate would have applied because turnium had knowledge that the product was going to the US. What they found during the review is that turniums, that Mueller for further processes some attorney in pipe, it turns some attorney pipe into non-subject merchandise, it sells some of that pipe in the home market and some of it is clearly exported to the United States. So for any given sale, turniums does not know that the pipe itself and the Mueller is going to the United States and thus turnium is not the exported to the United States

. And I want to clarify something, turnium and Mueller are unrelated as defined by the statute and commerce has never put forth an argument that they're related. They're not affiliated under the statute and they interact with each other at arm's length. What about the other argument that even though there's an arm's length relationship, as you say, that still a purchaser has some leverage or with a supplier to get the supplier to cooperate. And turnium did cooperate. If they didn't cooperate fully, but they answered the original question or that was issued to them, they didn't think that's a full answer to the question. No, but if I may, turnium put forth hundreds of pages of information on the record with respect to its cost. And the limited hold that the department found was that that cost of information wasn't product-specific enough. That the department wanted product specificity on five characteristics and turnium failed to do so. I understand. And what commerce is saying is that Mueller, because it's purchasing from turnium, has some leverage over turnium to force them to fully cooperate with commerce. What's the answer to that? The answer is that there was no finding that Mueller failed to cooperate to the best of its ability. Everybody agrees with that. Right, so that there was, if that's the approach they wanted to take, that Mueller somehow didn't do enough in the underlying review, they'd fall way shorter that they expressly found that Mueller did everything that fully cooperated. Except to stop purchasing from turnium, because turnium was not providing a full information required for..

. That's the point of the leverage, not that you didn't independently cooperate, but that as between the regulatory authorities here and you, you're in a much better position to get turnium to do what our... To engage in full cooperation with our regulatory. Well, of course administrative reviews that the Department of Commerce are retrospective. And so by the time the administrative review has commenced, the entire period that is covered by that review has come to a conclusion. So Mueller did encourage turnium to participate and turnium as well as Mueller's other unaffiliated supplier submitted information. That the information was not perfect. No one disputes. There was a hole in the record, a small hole. Hey, can I return to that? Just get some clarification about the knowledge test. You can specify. So, the podium's termium said... There's this 20-something percent gap in dumping rates between the resellum Mueller and our own

. So we're going to shift a bunch of our supply that we ordinarily would have imported directly to sell to Mueller. We don't know where any given piece of it is gonna go. The council should not be expressing opinions about the questions by nodding. Please don't do that. Anyway, I think you're probably getting the questions. With the knowledge test be satisfied or dubiously satisfied or what if turnium did a kind of large scale shift but had no idea where any individual pipe or segment was gonna go. So you're on in that hypothetical because the knowledge test is not part of the appeal but I understand where you're getting at. The way the department applies the knowledge test is on a transaction-specific basis. So they would test for any given transaction at that time did the producer know that the product they were selling to the reseller was going to be resold as subject merchandise. That I guess is what feels like it makes it a not fully satisfactory answer to the underlying policy just concerned about use of Mueller by turnium to evade its 48% rate. But again, I mean, it's the crux of the department's analysis is who is the exporter. Once they determine who the exporter is then that exporter gets its own dumping margin and the dumping liability for the importer is based on who that exporter was. So again, if that's a concern that the department has in the macro then they need to go to Congress to fix it or adopt a new policy. But their normal course is to define who the exporter is based on the knowledge test. And in this case, they clearly found that Mueller was the exporter. So there's no shifting

. And we think from the preliminary results, Mueller's margin was around 5% and then the final result, it was 19%. The differential that exists now in the final results that they published is greater from 19% to 48%. Then the differential would be if you fixed it in the sense that if you didn't allow an adverse inference to be used in calculating a margin for a fully cooperative respondent. Okay, thank you from the other side. Thank you, reserve my time for the bottle, please. Mr. Adleship. May I please the court? To begin with the questions from Judge Dyke and Judge Toronto regarding the knowledge test that was raised in my friend and colleagues' reply brief, the knowledge test, as was conceded, applies on a transaction-specific basis. And so it does not solve in and of itself the problem of potential circumvention of the statute. What does transaction mean in that sense? In other words, the question that Commerce asks under its reseller policy is whether the producer knows that a specific sale of goods, a specific transaction, whether that is bound for the United States through a reseller. So as in this case, Turnium knows full well after years of doing business with Mueller that Mueller is reselling Turnium products in the United States. But it can close advice to the fact that any particular sale, it doesn't know whether that's going to the United States. But there's no question that Commerce has the power to deal with the evasion. And we specifically recognize that in Taiwan, right? If there is evasion. That's certainly a policy judgment that Commerce should can and should take into account when interpreting the story of, but that's not an answer to my question. My question is Commerce can, if Turnium starts using Mueller as a conduit for its own sales in order to avoid its AFA rate, Commerce has tools to deal with that, right? That is a limited solution to a difficult problem

. I don't know, but it has tools to deal with it, right? I mean, you may think this is a better way to go about it, but it has the authority to deal with it, right? You're on that authority is not sufficient. Yes, that you have the authority. There's no question the knowledge has to exist. But how it's problematic, because how does Commerce assess the knowledge of a party that refuses to cooperate altogether? It's difficult. Well, but Mueller is cooperating. He certainly can ask Mueller whether these things are being followed as an evasion through it, right? There doesn't need to be an intent to evade the laws. Turnium can simply close its eyes to wherever its products are going, and simply sell all of its products to Mueller, knowing full well that many of them, or even most of them, are bound for the United States, but not knowing specifically about a particular shipment. And so the... So, I mean, do you think that the knowledge test ends? Would or would not be potentially applicable in supposed to following facts. Turnium in one year is importing X amount to the United States. In the next year, it drops down to year zero, and the same volume has now been shifted to selling to Mueller, and that amount is going to be... Turnium has been selling to Mueller before, and maybe 30% or something comes into the United States, and the other 70% is going to other countries

. Would a knowledge inference be possible simply by looking up the numbers, the shift? You're honest, both as possible, but I can't speculate about those particular circumstances that really aren't at issue here. What Commerce did was it interpreted its facts available authority under 1677E to fill the information gap that was left by Turnium's missing cost of production data. This data was very important because to calculate Mueller's dumping margin accurately as the statute requires, Commerce needed this cost data from Turnium, which was the producer of the products. Well, what should Mueller have done, or what could Mueller have done that it didn't do to force more cooperation from Turnium? The construction of the statute at issue here, and Commerce's election from the facts available, doesn't turn... No, you're not answering my question. Please answer my question. What else could Mueller have done to force Turnium to cooperate that it didn't do? You're on our own, not aware of anything that they could have done short of threatening not to do business with them in the future, in the future. And it's important that we recognize that this case involves the problems of not only missing information but adverse inferences against an interested party that failed to cooperate. The missing information was Turnium's missing cost. So Commerce had to determine what is it know about Turnium in order to select the best information on the record to fill the gap left by the Turnium's missing cost. Well, Commerce already knew that Turnium had failed to cooperate in multiple consecutive reviews, had a high AFA rate, and still refused to cooperate. And also, despite the fact that Turnium's predecessor had provided the exact kind of information that Commerce had sought here regarding cost of production, Turnium declined to produce that information with the calculation that's being made here only affects Mueller. It doesn't affect Turnium, right? The cost of production information is being sought for the purpose of calculating an accurate dumping margin for Mueller, yes. And it has nothing to do with Turnium's rate

. That's right, that's a separate issue because Turnium failed to cooperate in providing information relevant to the calculation of its dumping margin, thus leading to the 48% AFA rate. But separate and apart from that, Commerce selected Turnium as a cost of production respondent in this proceeding to provide cost of production information relevant to Mueller's dumping margin. So why wouldn't it have been more accurate to take all, instead of cherry picking the three highest tuna sales, that is where there was a departure from the sales price and cost of production. Why not take all, is it 26 instances in which the tuna cost of production were below the sales price? Well, without getting into the specific numbers, which are the sales, which are confidential, in order to select the most probative data. Commerce chose those three tuna transactions because based on what it knew about Turnium, those were the most probative. You have a spectrum of tuna. But isn't the case that those three were chosen because for deterrence purposes and not because they were for the most accurate? It was both, Your Honor, they were also the most accurate. The spectrum of tuna's costs provide a range. And the question was, which of tuna's costs were most probative of what was missing? Turnium's missing costs. So we had to make a judgment about which was the most probative and Commerce in making that judgment considered what it knew about Turnium. You're going to get Commerce say that those were chosen because they were the most probative. I've read the record pretty carefully. And I see Commerce saying they were chosen because for deterrence purposes, not because they were the most accurate. Where did Commerce say that that was the most accurate? You're off. We are going to be cite in the brief numerous instances in the IND memo and the FAM memo that Commerce, which is give me one where it says these are being chosen for accuracy rather than the term. Sure

. Even in part for accuracy. Even in part for accuracy. Yes. In the IND memo, JAA 44, A 44, the selective facts available are intended to produce an accurate non-punitive dumping margin for Mueller. And also on the same page, JAA 44, the IND memo, that it states that Turnium has withheld, I'm sorry. Application of an adverse inference only to the missing cost of production information that Turnium has withheld is a reasonable and limited inference based on the information on the record that ensures Turnium does not benefit from its failure to cooperate and also avoids an inaccurate dumping margin for Mueller. You see, we can't know for sure if Mueller's margin would have been higher if Turnium had actually cooperated. Yeah, but what you say on page 50, for example, is we find the other two-nicost differentials are not sufficiently high as to evacuate the purposes of facts available rule. High M, you not find that any of these differentials are high enough to encourage participation in future segments of the proceeding, which sounds like rejecting the other data because it wouldn't accomplish the deterrence rationale, rather than rejecting the other data because it would not be accurate. You're on our, in this particular case, commerce considered all of the statutory goals, fairness, accuracy, encouraging cooperation with data requests as well as the turns, and concluded that each of those statutory goals would be a good idea to do that. But that's a general answer. But the quote that I read you says, we can't use this other data because it's not sufficiently deterrent. It doesn't say we're not, we can't use this other data because it would lead to an inaccurate result. Am I right? No, you're right. I think, I believe commerce also stated that the other tune of data was less probative of Mueller's, of the missing Turnium cost as well. Where does it say that? Excuse me, bear with me, just one more round

. Commerce explained in the, in the AFAMMO, J-A-50, that it had a fine turnium, a total AFA rate of 48% in the prior review, and this had an induced turnium to cooperate during the current review, leading commerce to conclude that turnium had refused to cooperate, in this case, to provide its missing cost data because its data would have been less favorable. So what commerce tried to do was to avoid the unfairness of Mueller receiving an artificially low rate just because turnium had refused to cooperate. It's part of the overall decision was consideration of all of the statutory goals. I do sure wish to save time for the other defendants, or can you tell us more to elaborate on these points, which obviously are concerning us. I would, I would, I would like to just wrap up quickly so that I allow my colleagues to, I'm using their time, but it's up to you as to how you use it. Okay, well, respectfully, your honor, I had agreed to give them the time. So in summary, commerce's decision making here, as the trial court held, was thorough, thoughtful, logical and complete, and we respectfully request that the court affirm the trial court's well-reasoned judgment that commerce's decision is both lawful and reasonable. Thank you. All right, thank you. Please, please, Shider. May I please the court also be very grateful for every brief. First, there was a question, does the adverse inference that commerce applied, in this case, did it affect only Mueller? And the answer to that is no. It affected Turniam. Turniam had a very direct interest in achieving as low, as possible, a dumping margin for Mueller, as it could. And the reason for that is that the lower Mueller's dumping margin, the greater the access to Turniam would have to the US. Mark, is there a question of motivation? Is there a question of what happened? Yes, but the point is that, as soon as a higher margin for Mueller is implemented, that does not only have an effect on Mueller, it also has an effect on Turniam. But it has no effect in terms of the calculation of any dumping margin applicable to Turniam. And not to Turniam's direct exports. No. But the other point that I would like to make is that on the knowledge test, the department's policy is very clear that generalized knowledge that some sales are going to go to the US market is not enough to attribute the sales to the manufacturer. They then become the sales of the reseller. And in addition to that, I would just like to say that we discussed this at length, and our briefed case tracks very closely the situation in the KYD case. And finally, if I may, I would just like to say a very brief word about the fine furniture case, which is a related case that is currently before the court. This case provides an even stronger basis for the use of adverse facts than fine furniture. In fine furniture, the unity of interest between... I have a plastic with this case. Very good. I think that..

. But it has no effect in terms of the calculation of any dumping margin applicable to Turniam. And not to Turniam's direct exports. No. But the other point that I would like to make is that on the knowledge test, the department's policy is very clear that generalized knowledge that some sales are going to go to the US market is not enough to attribute the sales to the manufacturer. They then become the sales of the reseller. And in addition to that, I would just like to say that we discussed this at length, and our briefed case tracks very closely the situation in the KYD case. And finally, if I may, I would just like to say a very brief word about the fine furniture case, which is a related case that is currently before the court. This case provides an even stronger basis for the use of adverse facts than fine furniture. In fine furniture, the unity of interest between... I have a plastic with this case. Very good. I think that... Can I just ask you the question that I think we were discussing? Sure. At the end of Mr. Adelschick's argument, start at the point at which the commerce says, we don't have certain information from Turniam. We're going to look at Tuna. And now it looks at Tuna, and it says it's got this range of different costs. And one thing it thinks is, it would be a good idea to choose the worst of those because that would induce better behavior. Put that aside. Is there anything else that commerce said in choosing among the range of Tuna costs to indicate that we think the ones that we chose with these three are actually a better indication of the missing Termium costs, cost information than the ones we didn't choose. Well, I think commerce was trying to do two things. Number one, they had no information about Turniam's missing costs, so they were trying to plug that hole, and they also needed to come up with some deterrence, and that was what they were looking for. They saw that buy. I think what you just said is not responsive at all to my question. I apologize. Is there anything besides deterrence that commerce said to indicate that the three Tuna costs they chose that they thought that those were more accurate on any basis than the ones they did not choose? I'm not aware of anything other than commerce's general statement that they were attempting to come up with an accurate margin for Mueller because they didn't have Turniam's costs. And the other thing that I want to mention is that in the prior review, Turniam had an adverse fact available rate of 48%, and Mueller knew about that rate, and was certainly in a position to choose not to have Turniam's as its supplier. Thank you, Your Honor

. Can I just ask you the question that I think we were discussing? Sure. At the end of Mr. Adelschick's argument, start at the point at which the commerce says, we don't have certain information from Turniam. We're going to look at Tuna. And now it looks at Tuna, and it says it's got this range of different costs. And one thing it thinks is, it would be a good idea to choose the worst of those because that would induce better behavior. Put that aside. Is there anything else that commerce said in choosing among the range of Tuna costs to indicate that we think the ones that we chose with these three are actually a better indication of the missing Termium costs, cost information than the ones we didn't choose. Well, I think commerce was trying to do two things. Number one, they had no information about Turniam's missing costs, so they were trying to plug that hole, and they also needed to come up with some deterrence, and that was what they were looking for. They saw that buy. I think what you just said is not responsive at all to my question. I apologize. Is there anything besides deterrence that commerce said to indicate that the three Tuna costs they chose that they thought that those were more accurate on any basis than the ones they did not choose? I'm not aware of anything other than commerce's general statement that they were attempting to come up with an accurate margin for Mueller because they didn't have Turniam's costs. And the other thing that I want to mention is that in the prior review, Turniam had an adverse fact available rate of 48%, and Mueller knew about that rate, and was certainly in a position to choose not to have Turniam's as its supplier. Thank you, Your Honor. Thank you, Ms. Schneider. Good morning, Your Honor. May it please the Court of Roger Schetter? We have a certain dependent interveners. We'll rest the base on our briefs. I would make one point that has been made by my co-counsel thus far. Judge Dikiro for this Court in SKF that the statute clearly gave commerce the authority to utilize suppliers records for cost to produce a transaction when looking at an exporter. I would submit to this Court. Yes, it is because I would submit to this Court that if you overturn the lower court's decision in this case, then essentially you will be eviscerating your decision in SKF. But there will be no following that. Let me explain it very quickly. What happened here is commerce shows three transactions from TUNA, which were the least favorable to Mueller and ignored, I guess it's 23 others or whatever the number is. So nobody saying they can't use the supplier cost. The question is which transactions should they be looking at? Can they cherry pick the transactions to use the three worst ones? And they can to draw an adverse inference. And I submit to this Court that if you say that commerce cannot draw an adverse inference under the statute to the interests of the foreign supplier that doesn't cooperate, then in review, after review, case after case, those foreign suppliers will say that my cost of production is higher than my selling price to the exporter. I will be advantaged because the Court has said it can't use an adverse interest to me, the supplier, if I don't supply my cost

. Thank you, Ms. Schneider. Good morning, Your Honor. May it please the Court of Roger Schetter? We have a certain dependent interveners. We'll rest the base on our briefs. I would make one point that has been made by my co-counsel thus far. Judge Dikiro for this Court in SKF that the statute clearly gave commerce the authority to utilize suppliers records for cost to produce a transaction when looking at an exporter. I would submit to this Court. Yes, it is because I would submit to this Court that if you overturn the lower court's decision in this case, then essentially you will be eviscerating your decision in SKF. But there will be no following that. Let me explain it very quickly. What happened here is commerce shows three transactions from TUNA, which were the least favorable to Mueller and ignored, I guess it's 23 others or whatever the number is. So nobody saying they can't use the supplier cost. The question is which transactions should they be looking at? Can they cherry pick the transactions to use the three worst ones? And they can to draw an adverse inference. And I submit to this Court that if you say that commerce cannot draw an adverse inference under the statute to the interests of the foreign supplier that doesn't cooperate, then in review, after review, case after case, those foreign suppliers will say that my cost of production is higher than my selling price to the exporter. I will be advantaged because the Court has said it can't use an adverse interest to me, the supplier, if I don't supply my cost. There will be no reason for those foreign suppliers. The adverse inference against the supplier. In fact, the attorney here got a 48% dumping rate as a result of AFA, right? And nobody's challenging that. Yeah, but not as to the exports made by Mueller of TUNA's products. If commerce cannot utilize something adverse in determining turniums costs, then Mueller will get a lower rate as Judge Shlant to was asking earlier. It's impossible under the knowledge test under his hypothetical. Let's say TUNA has been selling 25,000 tons directly to the US and gets hit with a 48% duty. And it previously sold 50,000 tons to Mueller, which x4 to 25,000 tons. And the next year it sells 75,000 tons to Mueller, which x4 is 50,000 tons. But it says we don't know among those 75,000 tons, which of those are going to be the 50,000 tons. I can tell you one of the knowledge test. There is no way that commerce can say, well, we look at the total numbers when we impute knowledge. They're required by their own practice for knowledge to be specific to the transactions. They can't apply a generalized test of looking at total exports. So this case is very important to commerce's ability to require that foreign suppliers cooperate in investigations by supplying cost of production information when an exporter is exporting their products. This is a critical case

. There will be no reason for those foreign suppliers. The adverse inference against the supplier. In fact, the attorney here got a 48% dumping rate as a result of AFA, right? And nobody's challenging that. Yeah, but not as to the exports made by Mueller of TUNA's products. If commerce cannot utilize something adverse in determining turniums costs, then Mueller will get a lower rate as Judge Shlant to was asking earlier. It's impossible under the knowledge test under his hypothetical. Let's say TUNA has been selling 25,000 tons directly to the US and gets hit with a 48% duty. And it previously sold 50,000 tons to Mueller, which x4 to 25,000 tons. And the next year it sells 75,000 tons to Mueller, which x4 is 50,000 tons. But it says we don't know among those 75,000 tons, which of those are going to be the 50,000 tons. I can tell you one of the knowledge test. There is no way that commerce can say, well, we look at the total numbers when we impute knowledge. They're required by their own practice for knowledge to be specific to the transactions. They can't apply a generalized test of looking at total exports. So this case is very important to commerce's ability to require that foreign suppliers cooperate in investigations by supplying cost of production information when an exporter is exporting their products. This is a critical case. It goes far beyond the facts of this case and is very important to commerce's ability to obtain that information in the future. Can I just ask them the risks of sounding like a broken record? Are you in the same page as I'm inferring from your colleagues that this rest just on deterrence? Yes. I believe this rest upon deterrence against turnium for not providing the information that was requested. Okay. Thank you, Mueller. Thank you very much. First, I'm perplexed by a 30 percent differential between 48 percent and 19 percent is less problematic than if they didn't use adverse inference and we got a 5 percent margin. So that would only be a reduction of 15 percentage points. So this talked about the version and all of that, I think really is just smoke. Commerce has clear discretion under its reseller policy. We use the knowledge set the shorthand and I apologize for that. It's not actual knowledge. What commerce has found is that the producer has to have actual knowledge or reason to know. And commerce certainly has the tools available to review information. But it's still in some sense transaction specific. Because that's the way commerce has decided it wants to apply it

. It goes far beyond the facts of this case and is very important to commerce's ability to obtain that information in the future. Can I just ask them the risks of sounding like a broken record? Are you in the same page as I'm inferring from your colleagues that this rest just on deterrence? Yes. I believe this rest upon deterrence against turnium for not providing the information that was requested. Okay. Thank you, Mueller. Thank you very much. First, I'm perplexed by a 30 percent differential between 48 percent and 19 percent is less problematic than if they didn't use adverse inference and we got a 5 percent margin. So that would only be a reduction of 15 percentage points. So this talked about the version and all of that, I think really is just smoke. Commerce has clear discretion under its reseller policy. We use the knowledge set the shorthand and I apologize for that. It's not actual knowledge. What commerce has found is that the producer has to have actual knowledge or reason to know. And commerce certainly has the tools available to review information. But it's still in some sense transaction specific. Because that's the way commerce has decided it wants to apply it. Absolutely. And in this case, what we have is a lack of accuracy and a fundamental substantial evidence problem. Because commerce states that it applied a neutral, non-cunitive, non-adverse inference to fill the gap in Mueller's calculation. And at the same time, as Judge Dyke you identified, you're saying the methodology to fill that gap, which they claim was neutral, was adverse. And they clearly state that the methodology they pick, they cherry pick three transactions in order to come up with an adverse number. That is a fundamental tension that cannot be reconciled. And it is fundamentally unfair that a fully cooperative responding like Mueller faced the consequences for that state step. If commerce had concerns about the information, the attenium submitted, attenium was a mandatory respondent, they had a direct vehicle to go after them. That's why this case is very similar to SKF. In SKF, this court acknowledged the lower courts finding that it is unfair to use an adverse inference against a fully cooperative responded due to the non-cooperation of an unaffiliated supplier. In that case, the unaffiliated supplier was not even a party before commerce. Here commerce had a direct mechanism to go after the party that they want to induce cooperation from. And this notion that the attenium hasn't cooperated in the past, that's true, but in the preceding review. But in the review that is at the bar, they provided cost information. It was imperfect cost information. They submitted three times the information department and they answered two sets of follow-up questions

. Absolutely. And in this case, what we have is a lack of accuracy and a fundamental substantial evidence problem. Because commerce states that it applied a neutral, non-cunitive, non-adverse inference to fill the gap in Mueller's calculation. And at the same time, as Judge Dyke you identified, you're saying the methodology to fill that gap, which they claim was neutral, was adverse. And they clearly state that the methodology they pick, they cherry pick three transactions in order to come up with an adverse number. That is a fundamental tension that cannot be reconciled. And it is fundamentally unfair that a fully cooperative responding like Mueller faced the consequences for that state step. If commerce had concerns about the information, the attenium submitted, attenium was a mandatory respondent, they had a direct vehicle to go after them. That's why this case is very similar to SKF. In SKF, this court acknowledged the lower courts finding that it is unfair to use an adverse inference against a fully cooperative responded due to the non-cooperation of an unaffiliated supplier. In that case, the unaffiliated supplier was not even a party before commerce. Here commerce had a direct mechanism to go after the party that they want to induce cooperation from. And this notion that the attenium hasn't cooperated in the past, that's true, but in the preceding review. But in the review that is at the bar, they provided cost information. It was imperfect cost information. They submitted three times the information department and they answered two sets of follow-up questions. There is no doubt that there is information about atteniums cost on this record. It's not product specific, and we understand that. So what does commerce do? Does they look at tuna's cost and say, what are the differences between the products? No. Do they look at Mueller's acquisition costs, which are non-questionably in arms-length transactions and say, okay, Mueller, when you're at arms-less with turnium, what's the differential between the products you buy? No. They cherry pick three transactions, a minuscule amount, I can't say the number because it's confidential, and apply that aboriginal difference to all of Mueller's costs. By the way, the cost that they say they don't want to use in the calculation. So the irony here is they say, we will not use Mueller's acquisition's cost because we want the suppliers' cost. The supplier provides the cost, albeit imperfect. And then because of the imperfection, we're not going to use any of the suppliers' costs. We're just going to use the cost that we said we didn't want to use, which was Mueller's cost. And I submit that that is a fundamental substantial evidence problem. I would also say that there have been cases where the department has looked at product shifting and volumes as between people to find circumvention. I think there was a case involving sweaters from Taiwan that the department used that approach. So it's not that the department is without tools to address what they purport to be their concern. But it is fundamentally unfair to subject Mueller to an adverse inference. And the mere fact that they say they did not change the reality of the situation

. There is no doubt that there is information about atteniums cost on this record. It's not product specific, and we understand that. So what does commerce do? Does they look at tuna's cost and say, what are the differences between the products? No. Do they look at Mueller's acquisition costs, which are non-questionably in arms-length transactions and say, okay, Mueller, when you're at arms-less with turnium, what's the differential between the products you buy? No. They cherry pick three transactions, a minuscule amount, I can't say the number because it's confidential, and apply that aboriginal difference to all of Mueller's costs. By the way, the cost that they say they don't want to use in the calculation. So the irony here is they say, we will not use Mueller's acquisition's cost because we want the suppliers' cost. The supplier provides the cost, albeit imperfect. And then because of the imperfection, we're not going to use any of the suppliers' costs. We're just going to use the cost that we said we didn't want to use, which was Mueller's cost. And I submit that that is a fundamental substantial evidence problem. I would also say that there have been cases where the department has looked at product shifting and volumes as between people to find circumvention. I think there was a case involving sweaters from Taiwan that the department used that approach. So it's not that the department is without tools to address what they purport to be their concern. But it is fundamentally unfair to subject Mueller to an adverse inference. And the mere fact that they say they did not change the reality of the situation. Because Mueller's dumping margin calculation was based on costs that were inflated by an adverse inference. And those weren't turniums costs. There were Mueller's costs that were submitted, which the department asked multiple questions about, which Mueller responded and were never found to be inaccurate. Can you just remember recite? And I think the word, again, because I know you've done it, but maybe just a tiny bit slower, even though your time is about to run out. If they, if in looking at TUNA's costs, they hadn't been interested in deterrence. What should they have done in selecting what they thought the best analogy would be? Yeah, I think there's two things I could have done. They look, they look at all of TUNA's cost of production compared it to all of the products that Mueller purchased from TUNA and the aggregate basis and say on average, did Mueller pay more or less than the TUNA's cost of production? And if, as we submit in our brief, the record shows that Mueller paid more than what the cost of production was for TUNA, which makes sense in an arms length transaction, then there's no need to adjust Mueller's acquisition costs for the product that's purchased from TUNA. And another option would have been to look at TUNA's cost and the relative cost differences between products and perhaps use that. I think that on the record, there are multiple neutral facts available. Did you also submit to Commerce argument evidence that you as purchaser understood that these TUNA products were the ones most close to these TUNA and ones? No, absolutely not. Well, let me put that, when you meant these products, if you meant the top three that they selected. No, no, if that's the range of 23 plus 3 if that's the one. Yeah, yeah. So the information that's provided both by TUNA and TUNA, and by Mueller, too, describes the product. So that Commerce has on the record the physical characteristics of what was purchased from TUNA, the physical characteristics of what was purchased from TUNA, and could have done more even-handed and neutral comparison and not cherry-picked three products purely because they were the highest products. That's the reason they picked it

. Because Mueller's dumping margin calculation was based on costs that were inflated by an adverse inference. And those weren't turniums costs. There were Mueller's costs that were submitted, which the department asked multiple questions about, which Mueller responded and were never found to be inaccurate. Can you just remember recite? And I think the word, again, because I know you've done it, but maybe just a tiny bit slower, even though your time is about to run out. If they, if in looking at TUNA's costs, they hadn't been interested in deterrence. What should they have done in selecting what they thought the best analogy would be? Yeah, I think there's two things I could have done. They look, they look at all of TUNA's cost of production compared it to all of the products that Mueller purchased from TUNA and the aggregate basis and say on average, did Mueller pay more or less than the TUNA's cost of production? And if, as we submit in our brief, the record shows that Mueller paid more than what the cost of production was for TUNA, which makes sense in an arms length transaction, then there's no need to adjust Mueller's acquisition costs for the product that's purchased from TUNA. And another option would have been to look at TUNA's cost and the relative cost differences between products and perhaps use that. I think that on the record, there are multiple neutral facts available. Did you also submit to Commerce argument evidence that you as purchaser understood that these TUNA products were the ones most close to these TUNA and ones? No, absolutely not. Well, let me put that, when you meant these products, if you meant the top three that they selected. No, no, if that's the range of 23 plus 3 if that's the one. Yeah, yeah. So the information that's provided both by TUNA and TUNA, and by Mueller, too, describes the product. So that Commerce has on the record the physical characteristics of what was purchased from TUNA, the physical characteristics of what was purchased from TUNA, and could have done more even-handed and neutral comparison and not cherry-picked three products purely because they were the highest products. That's the reason they picked it. It's not fairness. It's not accuracy. It's they picked the three products that generated the highest margin for it. See my time. Any more questions? Any more questions? Okay. Thank you very much. Thank all of you. The case is checking on this submission.

cases number 13, 1391, Mueller, commercial, the main coagency United States, Mr. Baseford. Thank you, Yarnis. Good morning and may it please the court. Yo, hi, Baseford, and I'm here today on behalf of Mueller commercial and Southland Pipe Nipple Company. Mueller commercial is a producer of pipe fittings in Mexico and also produced is a reseller of the pipe that was the subject of the administrative review below. It is undisputed that Mueller commercial is all fully cooperated during the proceeding, answering all the questions that the department asked and participating in a one week long on-site verification. Despite that full cooperation, commerce applied an adverse inference in the calculations of Mueller's anti-dumping margin. So let me tell you what the problem that I see is, I'm sympathetic to the notion that cooperating parties shouldn't have their dumping margins increased as a result of adverse facts available. But what the government argues, as I understand it, is that there is a necessary deterrence here, that this is unlike Shanzu, where there was no relationship between the cooperating party and the AFA party, that there is a relationship here and that your client had some leverage to force the non-acorporating party to cooperate and that there was on top of that a risk that if the government did not use AFA under these circumstances, that there would be a tendency to re-root the exports so that the high dumping margin, the 48% or whatever it was, could be avoided and the exports could go through Mueller having a much lower dumping margin. So those are the two problems that I see the two potential distinctions from Shanzu. Could you address those? Absolutely, I'd be glad to. And let me address the second one first. In the final result that the Department of Commerce published, any pipe that was exported by turnium would be subject to the 48% rate and any pipe that was exported by Mueller, that was produced by turnium, would be subject to the 19% rate. So there's absolutely nothing unusual about the fact that a producer's pipe is subject to different cash deposit rates or ultimate dumping margin. All right, so what I think commerce is saying is that under these circumstances, there would be an incentive for turnium to stop its own imports in the United States and instead to funnel them through Mueller to avoid the high AFA rate of 48%. Right, and the Department has tools available to address that, which is a knowledge test. And that's what they found. If turnium has knowledge at a particular cost sale to Mueller is going to the United States, then that sale is subject to turniums, right? However, if turnium has no knowledge that that particular sale to Mueller will be sent to the United States and Mueller exports some pipe that was produced by turnium to the United States, it's Mueller's rate by the pipe. That's the way the anti-cumpling system is set up. If commerce has a concern that that reseller policy or that reseller methodology can result in people shifting, then that was concerned to be addressed up the street. How often does that happen? And in the Indian anti-dumping proceedings, that commerce says, oh no, this is a really a turnium export and it's just funneled through Mueller. And that evasion rationale, how often does that come up? But you see, it's not evasion. It's exactly what commerce found and said in Tong Mung, which is that a situation where a producer sells to multiple resellers and each reseller has its own rate. There's no evasion because one reseller sells to the United States versus another one. I mean, this comes up, I would say often, if there's a whole policy at the Department of Commerce, they've developed this knowledge test to determine who is the exporter. And that's the critical question because dumping margins are calculated on the exporter. And for Mueller's margin, they looked at Mueller's US sales and compared them to Mueller's home market sale. So there can, and that is what dumping is, it's the comparison of your US price to your foreign market price in this case. The attorney and whatever's turning themselves in the home market is irrelevant to this calculation. And so this notion that they're trying to close a loophole or deal with evasion, it's simply not true because had they found during the review that Mueller was merely a conduit for turnium, then turniums rate would have applied because turnium had knowledge that the product was going to the US. What they found during the review is that turniums, that Mueller for further processes some attorney in pipe, it turns some attorney pipe into non-subject merchandise, it sells some of that pipe in the home market and some of it is clearly exported to the United States. So for any given sale, turniums does not know that the pipe itself and the Mueller is going to the United States and thus turnium is not the exported to the United States. And I want to clarify something, turnium and Mueller are unrelated as defined by the statute and commerce has never put forth an argument that they're related. They're not affiliated under the statute and they interact with each other at arm's length. What about the other argument that even though there's an arm's length relationship, as you say, that still a purchaser has some leverage or with a supplier to get the supplier to cooperate. And turnium did cooperate. If they didn't cooperate fully, but they answered the original question or that was issued to them, they didn't think that's a full answer to the question. No, but if I may, turnium put forth hundreds of pages of information on the record with respect to its cost. And the limited hold that the department found was that that cost of information wasn't product-specific enough. That the department wanted product specificity on five characteristics and turnium failed to do so. I understand. And what commerce is saying is that Mueller, because it's purchasing from turnium, has some leverage over turnium to force them to fully cooperate with commerce. What's the answer to that? The answer is that there was no finding that Mueller failed to cooperate to the best of its ability. Everybody agrees with that. Right, so that there was, if that's the approach they wanted to take, that Mueller somehow didn't do enough in the underlying review, they'd fall way shorter that they expressly found that Mueller did everything that fully cooperated. Except to stop purchasing from turnium, because turnium was not providing a full information required for... That's the point of the leverage, not that you didn't independently cooperate, but that as between the regulatory authorities here and you, you're in a much better position to get turnium to do what our... To engage in full cooperation with our regulatory. Well, of course administrative reviews that the Department of Commerce are retrospective. And so by the time the administrative review has commenced, the entire period that is covered by that review has come to a conclusion. So Mueller did encourage turnium to participate and turnium as well as Mueller's other unaffiliated supplier submitted information. That the information was not perfect. No one disputes. There was a hole in the record, a small hole. Hey, can I return to that? Just get some clarification about the knowledge test. You can specify. So, the podium's termium said... There's this 20-something percent gap in dumping rates between the resellum Mueller and our own. So we're going to shift a bunch of our supply that we ordinarily would have imported directly to sell to Mueller. We don't know where any given piece of it is gonna go. The council should not be expressing opinions about the questions by nodding. Please don't do that. Anyway, I think you're probably getting the questions. With the knowledge test be satisfied or dubiously satisfied or what if turnium did a kind of large scale shift but had no idea where any individual pipe or segment was gonna go. So you're on in that hypothetical because the knowledge test is not part of the appeal but I understand where you're getting at. The way the department applies the knowledge test is on a transaction-specific basis. So they would test for any given transaction at that time did the producer know that the product they were selling to the reseller was going to be resold as subject merchandise. That I guess is what feels like it makes it a not fully satisfactory answer to the underlying policy just concerned about use of Mueller by turnium to evade its 48% rate. But again, I mean, it's the crux of the department's analysis is who is the exporter. Once they determine who the exporter is then that exporter gets its own dumping margin and the dumping liability for the importer is based on who that exporter was. So again, if that's a concern that the department has in the macro then they need to go to Congress to fix it or adopt a new policy. But their normal course is to define who the exporter is based on the knowledge test. And in this case, they clearly found that Mueller was the exporter. So there's no shifting. And we think from the preliminary results, Mueller's margin was around 5% and then the final result, it was 19%. The differential that exists now in the final results that they published is greater from 19% to 48%. Then the differential would be if you fixed it in the sense that if you didn't allow an adverse inference to be used in calculating a margin for a fully cooperative respondent. Okay, thank you from the other side. Thank you, reserve my time for the bottle, please. Mr. Adleship. May I please the court? To begin with the questions from Judge Dyke and Judge Toronto regarding the knowledge test that was raised in my friend and colleagues' reply brief, the knowledge test, as was conceded, applies on a transaction-specific basis. And so it does not solve in and of itself the problem of potential circumvention of the statute. What does transaction mean in that sense? In other words, the question that Commerce asks under its reseller policy is whether the producer knows that a specific sale of goods, a specific transaction, whether that is bound for the United States through a reseller. So as in this case, Turnium knows full well after years of doing business with Mueller that Mueller is reselling Turnium products in the United States. But it can close advice to the fact that any particular sale, it doesn't know whether that's going to the United States. But there's no question that Commerce has the power to deal with the evasion. And we specifically recognize that in Taiwan, right? If there is evasion. That's certainly a policy judgment that Commerce should can and should take into account when interpreting the story of, but that's not an answer to my question. My question is Commerce can, if Turnium starts using Mueller as a conduit for its own sales in order to avoid its AFA rate, Commerce has tools to deal with that, right? That is a limited solution to a difficult problem. I don't know, but it has tools to deal with it, right? I mean, you may think this is a better way to go about it, but it has the authority to deal with it, right? You're on that authority is not sufficient. Yes, that you have the authority. There's no question the knowledge has to exist. But how it's problematic, because how does Commerce assess the knowledge of a party that refuses to cooperate altogether? It's difficult. Well, but Mueller is cooperating. He certainly can ask Mueller whether these things are being followed as an evasion through it, right? There doesn't need to be an intent to evade the laws. Turnium can simply close its eyes to wherever its products are going, and simply sell all of its products to Mueller, knowing full well that many of them, or even most of them, are bound for the United States, but not knowing specifically about a particular shipment. And so the... So, I mean, do you think that the knowledge test ends? Would or would not be potentially applicable in supposed to following facts. Turnium in one year is importing X amount to the United States. In the next year, it drops down to year zero, and the same volume has now been shifted to selling to Mueller, and that amount is going to be... Turnium has been selling to Mueller before, and maybe 30% or something comes into the United States, and the other 70% is going to other countries. Would a knowledge inference be possible simply by looking up the numbers, the shift? You're honest, both as possible, but I can't speculate about those particular circumstances that really aren't at issue here. What Commerce did was it interpreted its facts available authority under 1677E to fill the information gap that was left by Turnium's missing cost of production data. This data was very important because to calculate Mueller's dumping margin accurately as the statute requires, Commerce needed this cost data from Turnium, which was the producer of the products. Well, what should Mueller have done, or what could Mueller have done that it didn't do to force more cooperation from Turnium? The construction of the statute at issue here, and Commerce's election from the facts available, doesn't turn... No, you're not answering my question. Please answer my question. What else could Mueller have done to force Turnium to cooperate that it didn't do? You're on our own, not aware of anything that they could have done short of threatening not to do business with them in the future, in the future. And it's important that we recognize that this case involves the problems of not only missing information but adverse inferences against an interested party that failed to cooperate. The missing information was Turnium's missing cost. So Commerce had to determine what is it know about Turnium in order to select the best information on the record to fill the gap left by the Turnium's missing cost. Well, Commerce already knew that Turnium had failed to cooperate in multiple consecutive reviews, had a high AFA rate, and still refused to cooperate. And also, despite the fact that Turnium's predecessor had provided the exact kind of information that Commerce had sought here regarding cost of production, Turnium declined to produce that information with the calculation that's being made here only affects Mueller. It doesn't affect Turnium, right? The cost of production information is being sought for the purpose of calculating an accurate dumping margin for Mueller, yes. And it has nothing to do with Turnium's rate. That's right, that's a separate issue because Turnium failed to cooperate in providing information relevant to the calculation of its dumping margin, thus leading to the 48% AFA rate. But separate and apart from that, Commerce selected Turnium as a cost of production respondent in this proceeding to provide cost of production information relevant to Mueller's dumping margin. So why wouldn't it have been more accurate to take all, instead of cherry picking the three highest tuna sales, that is where there was a departure from the sales price and cost of production. Why not take all, is it 26 instances in which the tuna cost of production were below the sales price? Well, without getting into the specific numbers, which are the sales, which are confidential, in order to select the most probative data. Commerce chose those three tuna transactions because based on what it knew about Turnium, those were the most probative. You have a spectrum of tuna. But isn't the case that those three were chosen because for deterrence purposes and not because they were for the most accurate? It was both, Your Honor, they were also the most accurate. The spectrum of tuna's costs provide a range. And the question was, which of tuna's costs were most probative of what was missing? Turnium's missing costs. So we had to make a judgment about which was the most probative and Commerce in making that judgment considered what it knew about Turnium. You're going to get Commerce say that those were chosen because they were the most probative. I've read the record pretty carefully. And I see Commerce saying they were chosen because for deterrence purposes, not because they were the most accurate. Where did Commerce say that that was the most accurate? You're off. We are going to be cite in the brief numerous instances in the IND memo and the FAM memo that Commerce, which is give me one where it says these are being chosen for accuracy rather than the term. Sure. Even in part for accuracy. Even in part for accuracy. Yes. In the IND memo, JAA 44, A 44, the selective facts available are intended to produce an accurate non-punitive dumping margin for Mueller. And also on the same page, JAA 44, the IND memo, that it states that Turnium has withheld, I'm sorry. Application of an adverse inference only to the missing cost of production information that Turnium has withheld is a reasonable and limited inference based on the information on the record that ensures Turnium does not benefit from its failure to cooperate and also avoids an inaccurate dumping margin for Mueller. You see, we can't know for sure if Mueller's margin would have been higher if Turnium had actually cooperated. Yeah, but what you say on page 50, for example, is we find the other two-nicost differentials are not sufficiently high as to evacuate the purposes of facts available rule. High M, you not find that any of these differentials are high enough to encourage participation in future segments of the proceeding, which sounds like rejecting the other data because it wouldn't accomplish the deterrence rationale, rather than rejecting the other data because it would not be accurate. You're on our, in this particular case, commerce considered all of the statutory goals, fairness, accuracy, encouraging cooperation with data requests as well as the turns, and concluded that each of those statutory goals would be a good idea to do that. But that's a general answer. But the quote that I read you says, we can't use this other data because it's not sufficiently deterrent. It doesn't say we're not, we can't use this other data because it would lead to an inaccurate result. Am I right? No, you're right. I think, I believe commerce also stated that the other tune of data was less probative of Mueller's, of the missing Turnium cost as well. Where does it say that? Excuse me, bear with me, just one more round. Commerce explained in the, in the AFAMMO, J-A-50, that it had a fine turnium, a total AFA rate of 48% in the prior review, and this had an induced turnium to cooperate during the current review, leading commerce to conclude that turnium had refused to cooperate, in this case, to provide its missing cost data because its data would have been less favorable. So what commerce tried to do was to avoid the unfairness of Mueller receiving an artificially low rate just because turnium had refused to cooperate. It's part of the overall decision was consideration of all of the statutory goals. I do sure wish to save time for the other defendants, or can you tell us more to elaborate on these points, which obviously are concerning us. I would, I would, I would like to just wrap up quickly so that I allow my colleagues to, I'm using their time, but it's up to you as to how you use it. Okay, well, respectfully, your honor, I had agreed to give them the time. So in summary, commerce's decision making here, as the trial court held, was thorough, thoughtful, logical and complete, and we respectfully request that the court affirm the trial court's well-reasoned judgment that commerce's decision is both lawful and reasonable. Thank you. All right, thank you. Please, please, Shider. May I please the court also be very grateful for every brief. First, there was a question, does the adverse inference that commerce applied, in this case, did it affect only Mueller? And the answer to that is no. It affected Turniam. Turniam had a very direct interest in achieving as low, as possible, a dumping margin for Mueller, as it could. And the reason for that is that the lower Mueller's dumping margin, the greater the access to Turniam would have to the US. Mark, is there a question of motivation? Is there a question of what happened? Yes, but the point is that, as soon as a higher margin for Mueller is implemented, that does not only have an effect on Mueller, it also has an effect on Turniam. But it has no effect in terms of the calculation of any dumping margin applicable to Turniam. And not to Turniam's direct exports. No. But the other point that I would like to make is that on the knowledge test, the department's policy is very clear that generalized knowledge that some sales are going to go to the US market is not enough to attribute the sales to the manufacturer. They then become the sales of the reseller. And in addition to that, I would just like to say that we discussed this at length, and our briefed case tracks very closely the situation in the KYD case. And finally, if I may, I would just like to say a very brief word about the fine furniture case, which is a related case that is currently before the court. This case provides an even stronger basis for the use of adverse facts than fine furniture. In fine furniture, the unity of interest between... I have a plastic with this case. Very good. I think that... Can I just ask you the question that I think we were discussing? Sure. At the end of Mr. Adelschick's argument, start at the point at which the commerce says, we don't have certain information from Turniam. We're going to look at Tuna. And now it looks at Tuna, and it says it's got this range of different costs. And one thing it thinks is, it would be a good idea to choose the worst of those because that would induce better behavior. Put that aside. Is there anything else that commerce said in choosing among the range of Tuna costs to indicate that we think the ones that we chose with these three are actually a better indication of the missing Termium costs, cost information than the ones we didn't choose. Well, I think commerce was trying to do two things. Number one, they had no information about Turniam's missing costs, so they were trying to plug that hole, and they also needed to come up with some deterrence, and that was what they were looking for. They saw that buy. I think what you just said is not responsive at all to my question. I apologize. Is there anything besides deterrence that commerce said to indicate that the three Tuna costs they chose that they thought that those were more accurate on any basis than the ones they did not choose? I'm not aware of anything other than commerce's general statement that they were attempting to come up with an accurate margin for Mueller because they didn't have Turniam's costs. And the other thing that I want to mention is that in the prior review, Turniam had an adverse fact available rate of 48%, and Mueller knew about that rate, and was certainly in a position to choose not to have Turniam's as its supplier. Thank you, Your Honor. Thank you, Ms. Schneider. Good morning, Your Honor. May it please the Court of Roger Schetter? We have a certain dependent interveners. We'll rest the base on our briefs. I would make one point that has been made by my co-counsel thus far. Judge Dikiro for this Court in SKF that the statute clearly gave commerce the authority to utilize suppliers records for cost to produce a transaction when looking at an exporter. I would submit to this Court. Yes, it is because I would submit to this Court that if you overturn the lower court's decision in this case, then essentially you will be eviscerating your decision in SKF. But there will be no following that. Let me explain it very quickly. What happened here is commerce shows three transactions from TUNA, which were the least favorable to Mueller and ignored, I guess it's 23 others or whatever the number is. So nobody saying they can't use the supplier cost. The question is which transactions should they be looking at? Can they cherry pick the transactions to use the three worst ones? And they can to draw an adverse inference. And I submit to this Court that if you say that commerce cannot draw an adverse inference under the statute to the interests of the foreign supplier that doesn't cooperate, then in review, after review, case after case, those foreign suppliers will say that my cost of production is higher than my selling price to the exporter. I will be advantaged because the Court has said it can't use an adverse interest to me, the supplier, if I don't supply my cost. There will be no reason for those foreign suppliers. The adverse inference against the supplier. In fact, the attorney here got a 48% dumping rate as a result of AFA, right? And nobody's challenging that. Yeah, but not as to the exports made by Mueller of TUNA's products. If commerce cannot utilize something adverse in determining turniums costs, then Mueller will get a lower rate as Judge Shlant to was asking earlier. It's impossible under the knowledge test under his hypothetical. Let's say TUNA has been selling 25,000 tons directly to the US and gets hit with a 48% duty. And it previously sold 50,000 tons to Mueller, which x4 to 25,000 tons. And the next year it sells 75,000 tons to Mueller, which x4 is 50,000 tons. But it says we don't know among those 75,000 tons, which of those are going to be the 50,000 tons. I can tell you one of the knowledge test. There is no way that commerce can say, well, we look at the total numbers when we impute knowledge. They're required by their own practice for knowledge to be specific to the transactions. They can't apply a generalized test of looking at total exports. So this case is very important to commerce's ability to require that foreign suppliers cooperate in investigations by supplying cost of production information when an exporter is exporting their products. This is a critical case. It goes far beyond the facts of this case and is very important to commerce's ability to obtain that information in the future. Can I just ask them the risks of sounding like a broken record? Are you in the same page as I'm inferring from your colleagues that this rest just on deterrence? Yes. I believe this rest upon deterrence against turnium for not providing the information that was requested. Okay. Thank you, Mueller. Thank you very much. First, I'm perplexed by a 30 percent differential between 48 percent and 19 percent is less problematic than if they didn't use adverse inference and we got a 5 percent margin. So that would only be a reduction of 15 percentage points. So this talked about the version and all of that, I think really is just smoke. Commerce has clear discretion under its reseller policy. We use the knowledge set the shorthand and I apologize for that. It's not actual knowledge. What commerce has found is that the producer has to have actual knowledge or reason to know. And commerce certainly has the tools available to review information. But it's still in some sense transaction specific. Because that's the way commerce has decided it wants to apply it. Absolutely. And in this case, what we have is a lack of accuracy and a fundamental substantial evidence problem. Because commerce states that it applied a neutral, non-cunitive, non-adverse inference to fill the gap in Mueller's calculation. And at the same time, as Judge Dyke you identified, you're saying the methodology to fill that gap, which they claim was neutral, was adverse. And they clearly state that the methodology they pick, they cherry pick three transactions in order to come up with an adverse number. That is a fundamental tension that cannot be reconciled. And it is fundamentally unfair that a fully cooperative responding like Mueller faced the consequences for that state step. If commerce had concerns about the information, the attenium submitted, attenium was a mandatory respondent, they had a direct vehicle to go after them. That's why this case is very similar to SKF. In SKF, this court acknowledged the lower courts finding that it is unfair to use an adverse inference against a fully cooperative responded due to the non-cooperation of an unaffiliated supplier. In that case, the unaffiliated supplier was not even a party before commerce. Here commerce had a direct mechanism to go after the party that they want to induce cooperation from. And this notion that the attenium hasn't cooperated in the past, that's true, but in the preceding review. But in the review that is at the bar, they provided cost information. It was imperfect cost information. They submitted three times the information department and they answered two sets of follow-up questions. There is no doubt that there is information about atteniums cost on this record. It's not product specific, and we understand that. So what does commerce do? Does they look at tuna's cost and say, what are the differences between the products? No. Do they look at Mueller's acquisition costs, which are non-questionably in arms-length transactions and say, okay, Mueller, when you're at arms-less with turnium, what's the differential between the products you buy? No. They cherry pick three transactions, a minuscule amount, I can't say the number because it's confidential, and apply that aboriginal difference to all of Mueller's costs. By the way, the cost that they say they don't want to use in the calculation. So the irony here is they say, we will not use Mueller's acquisition's cost because we want the suppliers' cost. The supplier provides the cost, albeit imperfect. And then because of the imperfection, we're not going to use any of the suppliers' costs. We're just going to use the cost that we said we didn't want to use, which was Mueller's cost. And I submit that that is a fundamental substantial evidence problem. I would also say that there have been cases where the department has looked at product shifting and volumes as between people to find circumvention. I think there was a case involving sweaters from Taiwan that the department used that approach. So it's not that the department is without tools to address what they purport to be their concern. But it is fundamentally unfair to subject Mueller to an adverse inference. And the mere fact that they say they did not change the reality of the situation. Because Mueller's dumping margin calculation was based on costs that were inflated by an adverse inference. And those weren't turniums costs. There were Mueller's costs that were submitted, which the department asked multiple questions about, which Mueller responded and were never found to be inaccurate. Can you just remember recite? And I think the word, again, because I know you've done it, but maybe just a tiny bit slower, even though your time is about to run out. If they, if in looking at TUNA's costs, they hadn't been interested in deterrence. What should they have done in selecting what they thought the best analogy would be? Yeah, I think there's two things I could have done. They look, they look at all of TUNA's cost of production compared it to all of the products that Mueller purchased from TUNA and the aggregate basis and say on average, did Mueller pay more or less than the TUNA's cost of production? And if, as we submit in our brief, the record shows that Mueller paid more than what the cost of production was for TUNA, which makes sense in an arms length transaction, then there's no need to adjust Mueller's acquisition costs for the product that's purchased from TUNA. And another option would have been to look at TUNA's cost and the relative cost differences between products and perhaps use that. I think that on the record, there are multiple neutral facts available. Did you also submit to Commerce argument evidence that you as purchaser understood that these TUNA products were the ones most close to these TUNA and ones? No, absolutely not. Well, let me put that, when you meant these products, if you meant the top three that they selected. No, no, if that's the range of 23 plus 3 if that's the one. Yeah, yeah. So the information that's provided both by TUNA and TUNA, and by Mueller, too, describes the product. So that Commerce has on the record the physical characteristics of what was purchased from TUNA, the physical characteristics of what was purchased from TUNA, and could have done more even-handed and neutral comparison and not cherry-picked three products purely because they were the highest products. That's the reason they picked it. It's not fairness. It's not accuracy. It's they picked the three products that generated the highest margin for it. See my time. Any more questions? Any more questions? Okay. Thank you very much. Thank all of you. The case is checking on this submission