# Oral argument — AG der Dillinger Huttenwerke v. United States > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A98753 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** May 7, 2025 ## Text Our last case this morning is also AG der Dillinger Huttenwerke v. United States, 2024-1498. Mr. Montalbain? Yes, sir. Good morning. May it please the court, my name is Mark Montalbain. I'm the counsel for the plaintiff appellant Dillinger Huttenwerke. We have two issues on appeal today. The primary issue and the first issue is the matching of products. In a dumping case, this is settled law. You're matching a price of a product in the home market Germany with the price of a product in the United States. And the products you're matching have to be identical because you're just trying to see what is the price discrimination based on the same products. Wasn't that issue waived? It was raised in an untimely fashion? No, Your Honor. It was raised in a timely fashion. The question about timeliness is, and when you read the issues and decision memorandum, The timeliness issue exists with respect to the second issue that you're raising. There is a question in the first issue, the model match. So in the model match, the Department of Commerce will have a comment period where they say we'll look at these 12 characteristics and this is what we propose. And then everybody has an opportunity to comment during that model match period and then the questionnaire comes out with that model match. During the model match period, we brought up the issue of steel used for transporting crude oil. So there's two types of crude oil, sour and sweet. And the sour is very corrosive. So with the corrosive, you have to make steel in a special way. You basically have to clean. Wait, I'm confused. You started out talking about the cost shifting between the prime and the non-prime. No, sir, Your Honor. That's my second issue. I'm sorry. Okay, go ahead. I'm sorry. I thought you were addressing the prime versus non-prime. That's my second issue. The prime and the non-prime was already decided by this court in Dillinger, France. So I didn't want to say a lot about that right now. I'll move it to the end of my presentation. The main thing is the matching. So during the model matching period, we discussed information about you have to distinguish between these steels used for what they call sour service. And there's two types of steel. There's the steel that goes in the pipelines, and there's the steel that goes in the tanks, the pressure vessels that hold it. So we gave information. We gave a whole presentation on how you clean the steel, how you clean the salt from phosphorus out. That's costly, the whole process. And we specifically focused on the line pipe steel. And Commerce did not make a distinction for the line pipe steel between sour service and regular service. But what they did do... When did it take so long before you requested a separate product code? We didn't take a long time. We asked for the separate product code on line pipe immediately in that phase. They didn't make a separate one for line pipe. But what they did do, if you look at the model match criteria that came out, what they did do, this is the very first quality code. You found it in appendix 376. They said for this quality code, you can add new quality codes. You're not set by this list. It says use additional number codes for each additional quality you propose. And all it says is provide a detailed narrative description, explain the differences in the additional physical characteristics, and explain what number you're giving them. So from the very beginning, there was no time limit to that. Even after that initial period, they said for this quality code, you can add new ones. And so we added a new one for pressure vessel. Pressure vessel was 760. We had 760 or 759 for the sour service. And we did the same with the line pipe. And they never said that information was late. They never said we didn't propose the right information. And you'll see that they never here asked for any sort of information on the difference between prices and costs. And the reason why that is, is we've already, with the initial questionnaire, given all the information on prices and costs for every transaction. So they know all the prices and costs for every transaction. It's just a question of, for this one transaction, whether you treat it with the general pressure vessel code, or whether you give it another code number for sour service. That's the only thing you're talking about. Otherwise, there's no information missing from the record. The 759 code. Yes. I understand, correct me if I'm wrong, that you proposed this code four months after the comment period had closed, and 45 days before the preliminary determination. Yeah, we proposed that one with our supplemental questionnaire response. But that was not untimely. There wasn't a time period for it. And Commerce never said it was untimely. All the information is on the record. They never rejected any information. They never, during the procedure, said that that was untimely. And the issue about sour service. Unlike other cases that we have where Commerce makes repeated requests for missing information. Right. In this case, they did not make a request that you correct this, or that they were going to reject that particular product code? No, the only thing they did, they said was, and this was the stated reason, for all the product codes that we added, whether they were added at the very beginning, or whether they were added with the next questionnaire response, they said, change them back to the original codes because you added ones that are not in the questionnaire. That's the only stated reason they gave. When you look back at the preliminary determination, that's the only reason. They didn't say it was untimely. They just said it's not in the questionnaire. And they didn't take this exact language that said, in the questionnaire, you can add new codes. What about for the cost of non-prime plate using sales price? Yes. So, Your Honor, the non-prime plate using sales price, this court has already gone over this in the Dillinger-Prance case. And in that case, the court made it clear what the statute says, that costs have to be based on costs because they're supposed to be an independent marker or standard to judge whether the sales are below cost or not. So if you said, well, we'll set costs at the sales price and then we'll compare it to the sales price, you get in this circular reasoning. And the court in Dillinger-Prance and also in IPSCO said this is circular reasoning, that you have to base the cost of the product on costs. And the statute also says exactly what costs are, costs of materials, costs of fabrication, costs of processing. So here, it's the same situation the court already had in Dillinger-Prance. So to me, that issue is clear from your prior precedent. For me, I would like to still talk about the other issue. What should happen in that case? Are you praying for a remand? With the non-prime plate? Yeah, let's talk about the second issue. Yes, there should be a remand. But with the idea that what we did was exactly what the court said in France, what's acceptable is we use the average actual price of the prime plate to value the non-prime plate. Because the non-prime plate has to come from the prime plate. All non-prime means is it doesn't meet the standards. So we use the actual average cost. So they should remand it to use the actual average cost. You have to use costs, you can't use a sales price. So they use the sales price of that plate. So it should be remanded and they should determine costs. But the other issue is crucial because what eventually happened in this case? Commerce rejected both of the new codes. This non-prime plate or non-prime product, it is sold, right? Yes, yes. Okay. And I'm assuming, though I don't recall that, that commerce accounts for that price in its cost of production. Does it offset that against what? Basically, because no non-prime is sold in the United States, it's not really used in their calculations. The way it adversely affected us is they took the cost of non-prime plate, they basically wrote it down to the sales price because you have to sell it at a lower price, and then they put that cost on the prime plate for the home market, and that affected our home market sales. So I do see now I'm into my bottom. I'm sorry. Is your client able to separate the cost of production for non-prime versus prime? No, because – It's part of the same production process. Exactly. I understand that, but if it's not exported to the United States, I'm beginning to wonder what you're doing in this investigation. It's technically within the scope language, but it really – the non-prime plate itself doesn't come into the calculations. So I will reserve the rest of my time. We will save it for you. Thank you. Again, good morning. So with the first issue, I think there's – it's a model match, and so an important thing to keep in mind is that in addition to Germany, there are 11 other countries that Commerce was investigating with this, and so that's why it was vital to have comments and proposed model match codes during the model match period, which Dillinger just failed to do. And, indeed, the first time it did propose that code 759 for the Sauer vessel plate was in September 20, 2016, and that's at the record 3086 through 87, and that's well after it had even proposed the other Sauer code, which was in June. And so here we just – essentially, Dillinger just has – it was late. It did not timely propose the code, and Commerce was well within its discretion to deny that additional proposed code because it fell outside the time period for even commenting on the proposed model match codes. And I think that's undisputed. And I would also say, too, that with that, I want to say I heard Mr. Modelbind mention, like, matching price and all that, except it's well accepted that for the model match, it's all about physical characteristics. And so Commerce had also explained that whether something is more expensive to make or not, that's not a physical characteristic, and so that's what Commerce was looking for when it comes to the model match criteria. And I can move on to the second issue if there are no other questions about model match. But when it comes to the prime and non-prime plate, it's undisputed that Dillinger did not track the cost of production for the non-prime plate. And as Commerce explained as one of its examples – that you can't use sales price as cost, right? That is right. And that's exactly what Commerce did here. And in Dillinger, France, it was looking at the cost of production. And then on remand in Dillinger, France, the case for that, after remand in the site for that is 651 F sub 3, 1294. Similar to what happened here for Dillinger, Germany, Commerce had reopened the record in each of these proceedings and actually invited Dillinger to put information on the record as to what the cost of production is for the non-prime plate. And Dillinger said, we don't have that information. We don't track that information. And so therefore, kind of similar to the case we just had, Commerce determined there's information that's missing from the record. Specifically, what's missing is the cost of production for the production of non-prime plate. Okay, but what they're saying here is that it's the same cost of production for prime and non-prime. There's no separation. It's the same cost. Well, so that is what they're saying, and Commerce disputed that. Commerce said, actually, our understanding is, so if you're trying to produce 100 prime plates, but you only end up with 98, and so two are non-prime plate, the cost of production for the prime plate is all 100 of those plates. And so Commerce was, again, invited Dillinger to put on the record. Okay, but that's the position that Commerce took in Dillinger, France, and that we rejected. But, Your Honor, in Dillinger, France, I think a big distinction is that Commerce was trying to track the cost of production, and this court did say that you have to, you know, again, the circular reasoning that you have to have the cost of production. But similar to Dillinger, France, as what happened here, Commerce and Dillinger said, we don't track the cost of production. Our alternative, Commerce, to you, is that for, because we don't track the non-prime plate cost of production, Dillinger's alternative, both in this case and on remand in Dillinger, France, was we propose instead that for the non-prime plate, we'll just use the, I want to say it was the average total cost of all production. But Commerce said, well, hold on a second, in your books and records, so again, we're in the universe, again, where there's missing information, and Commerce is trying to figure out the best information to plug in for that missing information. And here Commerce said, well, in your own books and records, you record the non-prime plate at its likely selling cost. Right, which was the same as in Dillinger, France, and we said that was irrelevant. But, Your Honor, the distinction again. Isn't that true that the same thing was true in Dillinger, France? But in Dillinger, France, it isn't. And the reason why is, is because Dillinger, France, there had not been, we're in a different universe. Dillinger, France was 16. In Dillinger, France, Commerce relied on the fact that in their own books and records, they listed cost at sales price, and we rejected that. That's right, for the cost of production. That's pursuant to 1677B. Except now we're in a different, not different universe. We're in a different section of the statute where now we're in 1677E where there's missing information. And so, again, the missing information is what is the cost of production for that non-prime plate. And when we're in that 1677E section of the statute, that's when Commerce is looking at, well, what's information that we can use to plug this, you know, the missing data. And, again, the proposal from Dillinger was the average cost of all production, whereas Commerce said, well, but hold on, again, in your books. So, again, we're missing information here. And so this is how it is distinguishable from this court's opinion in Dillinger, France, where for this missing information, Commerce determined that the likely selling cost, which is how Dillinger reported itself, would be the best information to use. And is there enough for the question? I see that in the Statement of Administrative Action, it says that Commerce should base its decisions on that, which is most probative of the issue and the consideration. I don't – I have a difficulty in seeing how adopting selling prices and the cost of production analysis leads you to the most probative consideration. And I can understand that concern, Your Honor. It even – it's contrary to what I understand most of our dumping law. Well, but I think, again, how this is distinguishable is that for cost of production, so 1677B, you are wanting to get an accurate cost of production. But here we are in the missing information realm. And so it really was – Dillinger would prefer that Commerce use its approach, which was the average total cost for this missing information. But Commerce determined that, again, that the likely selling price, because that's how Dillinger itself recorded the, I guess, the cost of non-prime plate, because it did not track its actual production cost, was the better information to use. And I think Commerce does have discretion to determine that that is better information to use in this instance because that information was missing. And, again, Commerce had reopened the record on remand and invited Dillinger to put any information on the record, but it just – it does not track this information. Thank you. Thank you, Ms. West. Ms. Bell. Thank you, Your Honors. May it please the Court. I would like to make a few points on both of the issues that have been raised today. First, with respect to the model match, there has been some discussion regarding the timing, but I think it's helpful to lay it out a little more clearly because it does help put this issue in focus. So as has been discussed, Dillinger did not raise the issue of the Sauer Service pressure vessel plate until their first supplemental questionnaire. This was 110 days after the deadline to submit comments on model match and only 45 days before the preliminary determination. So this was significantly late in the proceeding and raised significant concerns for Commerce regarding how to handle this and what the implications were. And in particular, I think it's helpful to look at Commerce's decision memorandum, and this is appendix page 5958 in footnote 236, where they note they recognize Dillinger saying, we might raise other issues later. They said that in their model match comments. But then the Department goes on to say, it is not clear why Dillinger would have needed to perform additional analysis unless it wanted further time to estimate the margin impact of such a proposal. As noted above, Dillinger only discussed pricing data associated with its proposed product characteristic changes in the context of the overall impact on Dillinger's weighted average margin. So Commerce had concerns both in terms of the timing, how close this was raised to the preliminary determination, but also why Dillinger had sat on this information and not raised it earlier. Relatedly, Counsel for Dillinger notes that Commerce provided in the initial questionnaire the opportunity for parties to propose additional quality codes. But that was the initial questionnaire. Dillinger did not propose this in response to the initial questionnaire. They proposed it for the first time in response to the supplemental questionnaire that did not provide the same instructions. Finally, on this point, an issue was raised as to whether or not Commerce had the obligation to request clarification or additional information that may be missing. But this is not simply an issue where there was a deficiency in what they proposed, and Commerce did have that. Was the information untimely because it was provided in the first supplemental questionnaire? Correct, Your Honor. Well, there was not a specific statute of regulations saying this is the exact deadline you have to do it. How can it be untimely if they respond to a questionnaire? Well, with respect to reason— Why have supplemental questionnaires if you're not going to use the data that a party responds to? Of course, Your Honor. The information that Dillinger proposed here was not in response to a question. This was not Commerce saying, do you have any other quality codes you would like to propose? And Commerce and Dillinger said, yes, here they are. The question in which Dillinger provided this information was in response to a question that said, for everything that you reported as code 771, which was the transport plate issue, re-report that as 770. So this question did not open the door for Dillinger or request or provide any ability for Dillinger to now propose a new quality code, which is what they did here. So this was not in response to—it was in response to a question Commerce posed, but it was not responsive to Commerce's question and was therefore untimely. What about the use of sales price? Thank you, Your Honor. So on the non-prime issue, there are a few things I want to talk about, namely with respect to Dillinger France, which has obviously been discussed at length already. And I know this has been discussed, but just to lay it out a bit more clearly, in Dillinger France, the court was looking at the proper development of cost of production pursuant to 1967B. I'm sorry, 1677B, excuse me. Here, for the case before the court here, we are operating under a different statute, 1677E. So whether or not Commerce's approach was consistent with— Well, it's still cost of production, right? Yes and no. Commerce is filling in a gap of cost of production. It's the same cost of production question. And the question is whether you could use sales price when we said you couldn't use sales price. Correct, Your Honor. So the question here is Commerce does not have actual sales price. So Commerce—I'm sorry, does not have actual cost price. Commerce does not have the information that would allow it to calculate cost of production consistent with 1677B, which is why we're in E-World. And I think what's helpful, because we did discuss the subsequent proceedings in Dillinger France, which is it went back down to the CIT, and as was discussed, Commerce reopened the record, requested that Dillinger France provide this information. They could not do so, so Commerce applied facts available and took the same approach they took here. And before the lower court, this was looked at because the issue had been raised of, is this consistent with the Dillinger France decision? And Judge Katz— Well, they provided cost of production information. They said it's the same for prime and non-prime plate. And what we have is Commerce disagrees with that and says it's not the same cost of production for prime and non-prime plate, really seeming to want to undo our decision in Dillinger France. I think this gets to exactly what cost of production we're looking at. So what Commerce needs under 1677B is cost of production for specific products, the control number of specific products. And everybody agrees we don't have that. So on a product-specific basis, there's going to be differences in cost of production from product to product, and I don't think that would be disputed. And so we cannot get to those specific costs because we do not have the information with regard to the non-prime that allows for a determination of what those specific product costs are. And so that's where I think there's a difference between saying, well, you might start out with the same cost, but we don't know how that filters through to the specific products. And I see I'm running out of time. So if I may just very briefly direct the Court to the lower court's decision in Dillinger France, their discussion of NANYA, the Federal Circuit's decision in NANYA, which was dealing with adverse facts available, not facts available. But the Court did note that the Federal Circuit concluded, quote, the statute simply does not require Commerce to select facts that align with the standards articulated in other statutes and regulations. And the lower court in Dillinger France had pointed to that as to why there was no conflict with the Federal Circuit's decision in Dillinger France. Thank you, Your Honor. Thank you, counsel. Mr. Montalbain has four plus minutes. Yes, Your Honor. So back to the Sour Service Plate. So with the Sour Service Plate issue, we're just talking about the form of the plate, whether you're talking line pipe, whether you're talking pressure vessel. It's the exact same plate. What eventually happened is Commerce on remand has conceded on line pipe plate, yes, we will have two numbers, even though in the beginning, in their original investigation, they rejected both of them. They later on said that this case is just like the Bowler case, and therefore we will recognize this extra number for the line pipe plate. Well, why doesn't that transfer to the pressure vessel plate? The only reason is the timing of the case. The judge in the lower court reached a decision on the first issue with pressure vessel plate without discussing Bowler, and Bowler is a case about Austrian plate, so it's the same group of cases. And in the Bowler case, the respondent in that case, and this was after the model match criteria period, it was after the first initial questionnaire was into the supplementals, said you should add a whole new field for grade. And the court said, yes, this is really important. You can't treat this as untimely because you have to have accurate calculations. And Judge Gordon, even though we raised it in our reply brief, in his first decision on the pressure vessel, just decided in favor of the department without discussing Bowler. Two years later, when he got to the decision on the line pipe plate, he remanded it so that Commerce could talk about how Bowler affects this case. And in that situation, the department said, yes, we find that this case is analysis to Bowler, and all the information on the record shows that sour service plate is different from regular service plate. And they accepted this separate code for the line pipe. And in our response to their draft determination, we said, well, the same thing should apply to the pressure vessel plate. And their only statement was, well, the court already decided that issue, so we don't discuss it here. So now we have a great inconsistency where sour service plate is the same, whether you put it in line pipe, whether you put it in pressure vessel, but we have one where we have a separate code for line pipe. We don't have a separate code for pressure vessel. And in Bowler, they added a whole new field after all the questionnaires were already done, even into October. And I'll also say with the timing, counsel for the government was incorrect about the date of our supplemental questionnaire response. That was in August, not September. After our supplemental response was filed, the Department of Commerce extended the prelim determination date all the way to November. They never said that our information was untimely, and they never rejected it. It's still all on the record. So this is a situation where the information is on the record. It's analogous to the situation with line pipe plate, and we would ask that the court would remand this so that Commerce could look at the Bowler case and how the Bowler case applies to this case, just like they did with the line pipe plate. Thank you to all counsel. The case is submitted. That concludes today's audience. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A98753. Public record. Not legal advice.