# Oral argument — Dorbest Limited v. United States > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A7760 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** January 6, 2010 ## Text Good morning, everyone. The panel has before it a total of five cases. Two of them will be submitted without argument on the briefs later today. Those are Appeal 7126, Warren v. Department of Veterans Affairs, and Weed v. Social Security Administration. That's Appeal 3255. On the argument list, we'll hear argument first in Appeal 1257, Dorbest Limited v. United States. Mr. Grimson, good morning to you. Welcome to the Court. Good morning, and may it please the Court. Jeffrey Grimson from Trautman Sanders for Dorbest, the Plaintiff Appellant. I'm going to spend ten minutes of my time on direct and the remaining five minutes on rebuttal, reply, and surreply. It'll be a little bit unusual, but that's the sequence that we've worked out here. Congress gave Commerce two instructions as to how to value the factors of production in non-market economy cases. First, Paragraph 1 of 19 U.S.C. 1677 B.C. provides a general instruction that Commerce shall value the factors of production, or FOP, quote, based on the best available information regarding the values of such factors in a market economy country. When does Commerce use 1, and when does it use 2? It uses 1 as a general. It's a chapeau type of a statutory provision, but Congress gave subparagraph 4 as the specific parameters to select surrogate values. Even if they're not the best information available? Well, I mean, in the conference report accompanying the 1988 Act, Congress made a clear connection between the term best available information and the two statutory factors of paragraph 4. Isn't that the issue in the case, what the correct understanding is of the interplay between those two sections? What do you think the correct interplay is? Our view is that the interplay is that Paragraph 1 is a general instruction to Commerce. That's not helpful to say that. What trumps what? Paragraph 4 trumps Paragraph 1. How do we know that? Well, if it doesn't trump Paragraph 1, then there's no reason for all these two statutory factors being set forth in Paragraph 4. Commerce could just have free reign in every case to say we're selecting the best available information without passing through the two filters of Paragraph 4. But Paragraph 4 has an escape clause to the extent possible. That's right. So is that the hint that Congress has given us about the relationship between the first paragraph and 4? Well, we view that as Congress saying to Commerce, unless it's impossible, you should stick within countries that satisfy both the tests. And here it was possible. Commerce had a value that qualified under both the statutory parameters and the various other factors that Commerce ran the wage rates through India. And it was $0.21 an hour. And Commerce threw that out and said we have to throw it out because we have this regulation. Clearly, that writes Paragraph 4 and the two filters of Paragraph 4 right out because they are not paying any attention at all to whether the surrogate data comes from a country which is a significant producer of comparable merchandise. And they go out of their way to include every country in the world, Luxembourg, Norway, Switzerland, that has data that passes their various filters. So Paragraph 4 is the more specific instruction. And we don't think that it's reasonable to say that Paragraph 1 provides a general escape clause for Commerce to use whatever it wants in every case, regardless of the two filters that Congress spent time enacting in Paragraph 4. Here it was possible for Commerce to use the $0.21 an hour figure for India. So to the extent possible, a test of Paragraph 4 is clearly met. And Commerce has said on the record of this case that the Indian value was not an anomaly. Instead of using that, they followed their very complicated regulation, which they promulgated in 1997, which required a regression analysis. They fit a line to a global… But doesn't Paragraph 4 cover everything that Paragraph 1 covers? It's more specific than Paragraph 1. Yes, it covers everything that Paragraph 1 covers, but Paragraph 4 is the more specific instruction. When you get down to actually selecting prices or costs of factors of production, you're into Paragraph 4. And Commerce made it clear and has confirmed in the legislative history that they meant that that goes hand-in-hand with the best available information. You can't write out those two statutory factors by the general best available information words in Paragraph 1. That's Commerce's position in this case. And this court has only ever dealt with Paragraph 4, the words to the extent possible in one case, and that's the Nation Ford case. And in that case, it was the flip side of this, where Commerce had a distorted domestic value and they wanted to go outside the country. And this court said that is acceptable. Here, they had an acceptable value in India, but they said they have to file the regulation, throw that value out, and go outside India, where they found, under this regression calculation, actually a value that's statistically distorted. Well, what are the effects of remedy you're asking for, other than to prevail in this case? Are you basically saying that we should hold void portions of the regulation as being fatally inconsistent, in your view, with the statutory command of Paragraph 4? Yes, that's what we're asking. We believe this regulation is facially invalid, and it's invalid. Why wasn't it challenged under the APA back in the 90s when it came out, if it's so blatantly inconsistent with the statute? Well, that's an interesting question, because this is a case of first impression. This has never been challenged before. Well, it's been challenged in the CIT. Sorry? It's been challenged in the CIT. It hasn't been challenged in this court. The validity of this regulation, this case, Dorbass's case, is the first case challenging the validity of the labor rate regulation. Usually when you say first case, you mean Allied Pacific was similarly a challenge. I guess that this case preceded it. Yes, that's correct. In that case, it's not yet final. Okay, there's a remand due in that case at the end of, I believe, at the end of February. Let me ask you this question. How many times has the CIT adverted to this issue, either in extensive discussion, as in this case in Allied Pacific, or even by reference to the census? Dorbass is the first final case out of the CIT. Final or not final, it takes a long time for some of these cases to become final. That doesn't mean that the CIT hasn't said something about the issue. To your knowledge, how many times has the CIT addressed this issue? In the Dorbass case, in the Allied Pacific case, there's a case from last June, which came after our initial brief, called Taayan Bayan. Those are three cases that they dealt with the facial validity question of this regulation. There's a couple of other cases where parties argued about the data set, a case on honey and a case on mushrooms. Those are the only three? Yes, those are the three. Let me ask you, if I can follow up on Judge Dyke's question about one and four. Setting aside two, which I have a hard time figuring out where that fits in. You say that four is more specific than one. Yet, the sentence of one that you're relying on, or that your opposing counsel is relying on, says, the valuation of factors of production shall be based on the best available information. Then four says, valuation of factors of production, in valuing factors of production, shall utilize. I don't see a difference in specificity. It says, in valuing the factors of production, under paragraph one. It's clearly a further modification of paragraph one. It is more specific. It's saying, when you find the best available information under paragraph one, valuing the factors of production, here are the two filters you need to run the data through in a non-market economy case. Why do you need one at all, if you have four, under your theory? In your view, isn't four completely replacing one, in essence? Am I missing something? Commerce relies on one when they have arguments about a choice of two surrogate values that both qualify under paragraph four. Commerce says, well, we have an obligation to select the best available information. This is how sometimes they'll pick, they'll decide to select the data that's publicly available, rather than a data that's not publicly available, or more contemporaneous, rather than other. You mean for factors of production, holding the labor rate. Yes, that's correct. Even values that satisfy all of four, commerce, when choosing between what values to use, they will often rely on authority under paragraph one, and say, we have an obligation to select the best available information. Do you think that if commerce, instead of having the regression line as the sole basis for deciding labor costs, had used the regression as a fallback, and had said, all right, here's our regression line. If we find that in a particular case, the surrogate does not give us a value that we feel confident in, that we will use, we will resort to the regression. Would that be valid in your view under the statute? Well, to go back to what commerce said when it first proposed the regulation in 1996, commerce actually intended the regression to incorporate certain of the factors of paragraph four. Yeah, I understand that. But I'm trying to get at whether you think it would be permissible for commerce to use the regression line, but just not as an exclusive or first line recourse. Well, I don't think that when you put data into the regression line that fails to meet the statutory factors of paragraph four, that you are complying with what Congress directed. The lower court, Judge Pogue, in ruling on our facial challenges, I'm getting way over my time, in ruling on the facial challenge, said that it's conceivable that commerce could be faithful to the statute and still file its regulation. And I guess what Judge Pogue was thinking there was that perhaps commerce could apply a regression to a data set that meets the factors under it. I understand what Judge Bryson is asking. He's saying suppose that the only data under four is, let's say, Indian data, and for some reason that that's an anomaly. And I know in this case it hasn't been determined to be an anomaly. I think he's asking you could you then go beyond the kinds of countries identified in paragraph four to use a regression analysis using all countries or other developed market economies? Are there circumstances in which you could use a regression analysis bringing in other countries other than those identified in four? I wouldn't read it that far, but I would say that to the extent possible language, the exception of four could kick in at some point if there's no meaningful knowledge of something from outside. And that's what the court found in Nation Court. But this is a very different situation where they had usable data but said we can't use it because our regulation forces us to go outside India to use data that has an error rate of plus or minus 40 cents, as we pointed out in our brief. I mean, this fails in our views. Chevron one on the statutory analysis we talked about, but also Chevron two. This does not lead to more accurate results. And you could see that right from the output of Commerce's regression calculus. In your view of four, what do the words to the extent possible add? They add a safety valve for an exception, and I'll agree that. If you deleted those words, when would there ever be a different result under applying four? If you deleted the words to the extent possible, when would there be a different result in applying four? Yes. Well, I mean, paragraph one, the best available information, paragraph, if they did not have this under paragraph one language in four and to the extent possible, I guess then the statute would look as if only paragraph one was there. But that's not what Congress wrote. And in legislative history, they were pretty clear that at least go hand in hand, and it's not one or the other. But going hand in hand doesn't help us. At least it doesn't help me. It's like when you say, well, four is more specific than one. Well, yeah, I agree. In a way, it is. But so what? What conclusion flows from the fact that it's perhaps more specific? I'd say the conclusion that flows is that they can't rely on the more general paragraph one to essentially ignore paragraph four. That's what they're doing in every single non-market economy case involving the labor input. Out of all these inputs, Commerce singled out labor and said to Congress, I mean, we have a better idea of how to value this, this one raw material input. And it is one of the factors of production that's enumerated in the statute, just like the rest of Commerce said, we've got a better way, a more accurate way that smooths out everything from case to case. When we say non-market economy, are we talking exclusively about China or are there other current non-market economy cases? The current two hot countries for the non-market economy cases are China and Vietnam. And there may be others involving the stands of the former Soviet Union, but there's no activity in those. This applies to every non-market economy. Well, I understand. I just wondered as a practical matter whether we're just talking about China. We're not just talking about China. And actually, if you go to a situation like Vietnam where the result of this labor rate calculation is a number in the $0.30 range, then this y-intercept of Commerce's calculation, which is $0.31, that's having a tremendous effect. They're making a global minimum wage of $0.31, and you can't say that's coming from the low-income countries. That has to be because Commerce is lumping in Luxembourg and the U.K. and Sweden and countries that are near to $40,000 a year in GNI per capita. So I'm not going to address the excise duty waiver question, but I'm happy to answer questions the court may have. Just to conclude on the labor rate, this is the one input that Commerce has singled out and decided to ignore the statute, paragraph C-4, in every single case, regardless of the quality of the data. Here they had a good piece of data from India that was not an anomaly, and they refused to use it based on an illegal regulation. I guess the real question is, what does the language, to the extent possible, mean in 4? And Commerce is constraining that to mean, well, it's not possible to use this data because it's not good enough, and so we're looking to this regression analysis, including 54 countries or whatever it is, to get better data. And the question is whether that's a reasonable construction of the statute under Chevron. Well, our position is that absolutely not. It was possible to use the Indian value. They are using it. It is in the regression calculations. But what does it mean that it's possible? You have a number. You can do the computation. That's not a very meaningful definition of possible. You have a number that's from a publicly available source, It's from the primary surrogate country that Commerce has elected. It's consistent with all the other values that Commerce is using to do the cost buildup, so there's no possibility of strains and mismatches occurring when you have labor from outside of your multiple... Your definition of when it would be impossible is when it's from a tainted source. If it's from an acceptable source, then it's possible. Yes, and that's consistent with... Even if the data were provably misleading, but accurate, it's from a good source, it's a good number, but in a larger context, Commerce says it's misleading, it would still be possible in your definition of possible, and therefore Commerce should have to stay under four and not go back to one. I mean, I think that the agency can construe the to-the-extent-possible language in a reasonable way. This is not it. This is a robotic, automatic rejection of every labor figure from the primary surrogate country in every single case. To the extent possible, yes, there could be a value that was qualified under paragraph four that Commerce still said, in this case, we can't use it. They're not conducting any case-by-case evaluation of this one piece of data in every non-market economy cases. This is applying like a machine. Once a year, this calculation comes out. Nobody understands it. That's why we're here 13 years after this was promulgated, is that this is a Frankenstein of a methodology that nobody has bothered to look into before, but it is patently illegal on its face, and it leads to inaccuracies, and we spent a lot of time at the CIT talking about the inaccuracies. For this court, I would just highlight the Y-intercept value and the P-statistic. These both are signs of a highly questionable result. All right. Thank you. Mr. Dorn? I'm going to plead the court, Joe Dorn, on behalf of Appellant AFMC. Before I begin, I have three handouts that are taken right from the record, and I thought it would be easier than having you pummel the briefs and so forth. One is a table from page 12 of our October 27th reply brief. The second is page 18 of our June 9th opening brief, and then attachment three from the second remand analysis memorandum, and if it's appropriate, I'd like to ask the deputy clerk to hand those out. Just tell us where you are. We've got the materials on the bench. I'll address the two issues raised by AFMC's appeal. First, Commerce's decision to include the financial statements of four smaller companies in the second remand results was based on substantial evidence. Its later decision to exclude those financial statements— The question is whether that's the right test. Whether you can say it was supported by substantial evidence, even though Commerce is not appealing that, or whether you have to show that the omission of the four was arbitrary and capricious. I mean, it does make a difference, and we've had this issue before about people challenging remand results where Commerce has changed its mind. Right. Well, in the first, in DoorVest 1, Judge Polk sent the case back for further explanation. He did not make a finding that substantial evidence was lacking. But in DoorVest 2, Judge Polk did find that there was a lack of substantial evidence, and that requires this court to review that determination. They know, though. Well, I don't know whether that's true or not. It may be that we read Commerce's saying in the final remand that it's not worth arguing about, or we've made a determination that we're going to go with the omission of the four, and that's satisfactory to us even though we disagree with the original remand. Right. And if Commerce were saying that, if they had said we're going to use the data without the four, then it seems to me you'd have to say that that was arbitrary and capricious rather than going back to the original remand order. Well, Your Honor, the Commerce Department, of course, said they only excluded the four under respectful protest. Right. So they didn't agree with the court. And it's our position to set forth in our brief that the decision in the third remand results under respectful protest itself is not based on substantial evidence and is also contrary to law. But suppose you're wrong on that. Suppose that Commerce was right both times. It's certainly possible for the agency to be right in terms of the substantial evidence test despite having come to different conclusions. It's your position, I take it, that because they were right the first time, in your view, should have been sustained that you don't have to prove they were wrong in the final remand results. That's correct. But they were wrong in the second remand. Now, let's assume you're wrong about that. Let's just talk about why it is that we should focus on the previous, the question of whether they were right or at least sustainable in the, what is it, second remand results. Why, in your view, is that something that we can and should look at in terms of our review at this point of the final judgment? Because the Commerce Department plainly stated in the third remand results that it was only doing so under respectful protest. The Commerce Department's not indifferent here. It said that it was making the third remand results under protest. It did not agree with the court's imposition of a requirement. Suppose they said, okay, we reconsidered this and we think that the Court of International Trade had a point. We're going to adopt their approach and not include the court. If that were the case, and if their decision was based on substantial evidence which is not contrary to law, so be it in this court. And that's common law in your view. But in our view, the decision in the third remand results was clearly contrary to law because the court's instructions, directions to Commerce were contrary to law. But what do you understand, Steve? I think the problem is what was Commerce saying when it gave up and said we're doing this under protest? Were they saying, well, it's not worth arguing about? No. What Commerce is saying is it did not agree with the idea that you needed a more rigorous analysis, some type of statistical type analysis. What the judge did was basically hang him. But isn't Commerce entitled to say we give up and this is just not worth fighting about? But they didn't. They said under respectful protest, they went down a road they did not want to go down. Do you know what that meant, that it's not worth fighting about? They could have continued to challenge it, right? They didn't have to give up, right? Right, but I think if they were indifferent, if they were saying this is an alternative way to look at it and here's how we come out, that would be one thing. But here they did it under protest. Is there any case law that holds that if Commerce says they're doing something under protest, that has some magical effect, that changes anything? I haven't seen it, Your Honor, but I would note that the Department of Commerce is not opposing our appeal on this issue. They're not supporting it either. That's correct, Your Honor. But let me point out a couple of flaws. Why shouldn't the review be what Commerce does, not what it says? Well, if Commerce does something that's only a result of an erroneous instruction from the Court of International Trade, clearly the result of that exercise is contrary to law. But you would say, to take the extreme case, if the CIT sent the case back to Commerce and said, I'm sending this back as a purely ministerial matter, but you will come down with the following result, and Commerce has been instructed to come down with this result and therefore coming down with the result, but we still think that the position before was right, you would say that the question then still is whether they were right the first time. Correct, Your Honor. So you're saying, in effect, the CIT, although perhaps in a more subtle form, has held the gun to Commerce's head and Commerce has acquiesced, but only because of the gun. Right. And here's where the Court got off track. The trial court had this view that scale economies affect SG&A ratios. It said it endured best one. Commerce has never agreed with that premise. I mean, surely, just to explain the SG&A ratio, it's the numerator is the SG&A expense. I understand all that, but I think the real question is, can Commerce say, it's just not worth arguing about this. We're going to give up. And if that's what you read as what's happened here, and it's similar to Judge Price's hypothetical, we think the Court was wrong, but it's not worth fighting about. We give up. Do we give any deference to that decision? No, Your Honor. Commerce has a statutory obligation to determine the most accurate dumping margin possible, and the accurate dumping margin is based upon the second remand results, not on the third remand results. And in the second remand results, if I can just get to the facts, Commerce made two factual findings, either of which supports the inclusion of all seven statements. First, Commerce found that, quote, given that none of the seven surrogate companies approximate the size of Dorbest, we did not find that their relative size in relation to Dorbest is a basis for the inclusion or rejection of the financial statement. And if you look at our October 27 survey reply brief at page 12, and you see the listing of the revenues of Dorbest compared to the revenues of the other seven surrogate companies, you'll see that conclusion by Commerce is exactly right. I mean, it would be one thing if you had one company that was of the same size as Dorbest, and then you had a bunch of other companies that were not the size of Dorbest. But here you had no company that came anywhere close to the size of Dorbest. So Commerce reasonably explained that size cannot be a relevant criterion for selecting surrogate statements in this case. Second, Commerce found no evidence that relative size is a primary driver in the differences in the financial ratios of the Indian surrogate companies on the record. And on page 18 of our opening brief, we lay out in the columns the revenue of each of the surrogates, the overhead ratio, the SG&A ratio, and the sum of the overhead and SG&A ratio. And if you look at that data and look at Commerce's findings, you'll finally surely see support for Commerce's conclusion. Commerce noted that the SG&A ratio for IFP, which is the top of the heap, represents the midpoint of all the SG&A ratios of all the companies, all the seven companies. And although IFP, the largest company, had revenues 100 times larger than D&D, and even as a move to infusion design, its SG&A ratio of 24% was comparable to the average of those three companies. And IFP, the largest company, had the highest overhead ratio. Ragbeer, the second largest company, had the lowest overhead ratio. Well, what's the point? The point is that there's no correlation. Looking at the data, there's no correlation between size and SG&A ratio. There's no correlation between size and overhead ratios. There's no correlation between size and the sum of overhead and SG&A ratio. And therefore? And therefore, there was no basis to exclude the so-called four smaller companies. Right. On the basis of size. And so the logic of that is that in opining to the contrary, Judge Pogue was wrong, and that just brings us right back to square one, that if Commerce felt strongly about it, that they should have maintained their position. But they didn't. They abandoned their position. They did not abandon their position. They kept their position that the better approach was to exclude all four. But under respectful protest, what they— Respectful protest doesn't mean anything. It's an empty word. They did X and they said Y. But they did X. Yeah, but the Y was contrary to law. They were not required to do some type of quantitative analysis. Commerce gave up because it says we can't do a quantitative analysis. We don't think it's required. We've never done it before. Is it arbitrary and capricious for them to give up? Yes. Why? Because it's arbitrary and capricious for the court to require them to use a test. That's an accurate question. But if you look at just the third remand results in isolation, what are the problems with it? Well, first, you've got this arbitrary decision that there are three large companies and four smaller companies. Look at the data in our table that goes through the revenues. There's no dividing point there. Well, why should we be reviewing the minutiae of the data used by an administrative agency and a court? Because you have— We're not the Commerce Department. But there has to be— We're not retrying the case here. I understand, Your Honor, but there has to be substantial evidence for the division of the three large companies from the four smaller companies in order to decide to exclude the four smaller companies. And Judge Poe didn't articulate any rationale for the division. The Commerce Department didn't agree with the division and certainly didn't rationalize any basis for the division. Norbest never came up with a reason for saying, here's where you draw the line. Why do you draw the line at this point as opposed to saying they were just— that IFP is the only large company and the other six were smaller, or that IFP and Rag Beer were the two large companies? So what do you want us to do? And what will happen next? We want you to send this case back on the grounds that the third remand results are not supported by substantial evidence. And they're also contrary to law because the court imposed a— Then we'll have a whole other round. Commerce will do something, and then it'll be litigated in the Court of International Trade, and then it'll come here in a subsequent appeal. Better to get the margin right, which is the object here. I mean, this is—the end game is to come up with an accurate margin. And the second remand results is something the Commerce Department found was accurate with respect to this issue, and there's substantial evidence for that finding. Does it have to be a perfect result? It has to be based on substantial evidence. It has to be based on the law. And the second remand results— Well, the test is best available information. Does that mean perfect? It does not mean perfect, Ron. Well, how much less than perfect can it be and still be sustainable? Well, it has to be—there has to be substantial evidence to support the dividing line. Of course, we know that. Well, here there is not any substantial evidence to support the division of the three large companies and the four smaller companies, and there's no substantial evidence to show that the size has any bearing on SG&A ratios, overhead ratios. I guess in your view, it isn't a question of no substantial evidence. In your view of the record, there's zero evidence. That's exactly right, Your Honor. Exactly right. Well, zero— The only evidence of the record is the trial court's high-priority belief that economies of scale have some effect on SG&A ratios. And that was stated in Door Best 1, and that's what got this train off the tracks from the get-go. All right. Let's hear from the government. Could I just say a word about rubberwood, and I'll eat into my rebuttal time, if I might, Your Honor. You already have to the extent of seven and a half minutes. We'll give you some time on rebuttal. Let's hear from the government. Thank you, Your Honor. Carrie Dunsmore for the United States, Your Honor. Good morning. Your Honor, returning to the question of the labor wage rate, the labor wage rate regulation used by the Department of Commerce is not facially invalid. And, in fact, it does not violate Section 19 U.S.C. 1677 B.C. 4. That provision states that commerce shall utilize, to the extent possible, the prices or costs of— Well, why wasn't it possible to use something other than its regression analysis for 54 countries? Well, Your Honor, we believe that the information was utilized. It's just that commerce went beyond that information and included other information as well. First, the regression line itself, the set of data in the regression includes information from— I understand. I understand that. But why is it not possible to come up with a number utilizing the data that's encompassed within four? Why is that not possible? Because, Your Honor, commerce has determined, first in its labor wage rate regulation, when they revisited the regulation recently, and in the context of this litigation and its remand, that that's not the best available information, that it's not— Well, put aside the best available information. Let's assume that that's not the criterion. Let's assume that the criterion is to the extent possible. I understand your argument is the best available information is also a relevant standard. Well, let's put that aside for the moment. If we were to determine that the only standard is to the extent possible, why is it not possible to use data from India and other countries falling within paragraph four instead of doing a regression analysis using 54 countries when you haven't made a determination that the Indian data is an anomaly? Well, in this case, based on the record of this case, it was demonstrated that commerce, as they had in the original labor wage rate regulation, but again, revisited it on the remand, and it's discussed in great length both in the second remand and in Judge Pogue's opinion, that that simply did not give as accurate of a rate. That the labor wage rates for the economically comparable countries were so varied, and they were so different. So your view is it's not possible to use data if there's more accurate data available from another methodology? Well, no, Your Honor. I believe that the labor wage rate regulation does use that data. It just uses additional data as well. But included within the regulation is data from economic— I understand that, but paragraph four contemplates using only the data from these comparable countries. Well, simply, I don't think it contemplates only using the data. It contemplates using the data to the extent possible. It's silent as to using additional data. Suppose you had a case, let's just say, a non-market economy, Kyrgyzstan and the market economy that is proposed as the surrogate country is Tajikistan, and we're talking about countries for which there's unimpeachably strong evidence that they are equivalent in every respect except that one is non-market and the other is market. So that there would be no argument available to commerce or anybody else that using Tajikistan labor rates would be distorted. Nonetheless, I take it, commerce would use the regression line even if that line produced a number that was very different from the labor rate in Tajikistan. Correct? It's commerce's practice to use the regression line. That's what the regulation states. However, if there is evidence, as there was in this case, as there was not in this case but there is in your hypothetical, that there was perfect data, commerce would certainly consider it. Does the regulation make clear that that is the default position? No, it doesn't make clear because commerce doesn't believe that that's likely. I'm trying to get a sense of what their methodology is. Are they committed, heart and soul, locked in on the regression line? Or is there room to say, and the regression line is one way to do it and if we're unhappy with looking at the data, we'll go to the regression line. It's more the former than the latter, is it not? It is more the former than the latter. So in my case, although you say, well, they'd be free to look at that. There's nothing in the regulation that would give them that freedom, I take it. No, there isn't. So the answer is, no, they would go with the regression line, they would ignore Tajikistan's labor rate, correct? That's what the regulation says. And that's what they would do, therefore, right? For example, in this case, Your Honor, that information was, there was the proposed notice, the information was brought to commerce. It did consider it. The regulation says the regression line and this commerce believes... But it considered it to what end? It considered it with an eye towards deciding whether they would use the surrogate country as opposed to the regression line or simply they considered it because they thought it was interesting but quite irrelevant since they used the regression line. No, I think they considered it. I think the record reflects that they considered all the options and concluded the regression line was the best available information. You mean in this particular case? In this particular case, correct. Does that mean that they are free in labor rate determinations and NME cases to use something other than the regression line under their regulations? I thought you just told me that they weren't. Well, the regulation is, they're bound by the regulation. Right. But the reality of it is your hypothetical commerce is determined from its regulation is unlikely to happen. I think that it's important to remember that the labor, the regulation... But the regulation is automatic is what you're conceding. It's automatic regardless of the circumstances of the individual case, right? That's correct, Your Honor. And the question is whether that's a valid approach. Commerce believes it is a valid approach and Judge Pogue upheld that because commerce has determined that that is the best available information. Suppose that this regression analysis approach were used for every factor of production. Could that be done? Theoretically it could, but the fact is that the information does not exist. Labor is unique in the sense that there exists a pool of information that's reliable and can be used to construct a regression analysis like this. But it sounds to me as though what commerce is saying is that we like one data set better than the other, not that the data set that would come within Paragraph 4 is necessarily inaccurate or anomalous. It's just that you can get better information using the regression analysis of the 54 countries, right? Well, I think commerce is under a statutory obligation to get better, to use the best information, and that's what C1 says. Well, so commerce's position basically depends on using the best information criterion to trump Paragraph 4, right? No, I don't think what they've done here trumps Paragraph 4. The paragraphs can be read in concert. And another thing that I think has been lost a little bit is that commerce is not comparing. By using the regression analysis, commerce isn't comparing China to Switzerland. Commerce is positing a global relationship between first national income and labor and then is using the broadest set of data to get the most accurate line. But when we look to China, for example, in this case, it's being compared against countries with similar GNI. So it's an economically comparable spot on the regression line. It's not that commerce has simply thrown C4 in the air. It's just that it's constructing it in a different method. And in constructing its global relationship, it's using information beyond the provision. But it's not disregarding economic comparability. It's just using it in a different fashion. You said that the Sections 1 and what we're calling 1 and 4 have to be read in concert. That's a nice-sounding lawyerism. But what does it really mean? Well, it means that both – I don't think commerce thinks that one trumps the other, but that its actions have to comply with both. And that's why there is an escape valve for the extent possible. Commerce must consider – There's no escape clause from the regulation that seems to require that you use the regression analysis and only the regression analysis, no matter what alternatives may exist. And the reason for that is that when promulgating the regulation, commerce studied it and believed that this will be the best available information, that it is always more accurate. Even in my case. Where, by hypothesis, it's not. In your hypothesis, Your Honor. Again, they don't believe that. I know. My hypothesis doesn't exist. But if it did, again, to be clear on this, because I want to make sure I understand your answer. To be clear, even if someone, an economist sitting in the Department of Commerce said, Whoa, this is exactly right. Tajikistan is a twin of Kyrgyzstan, and this is a perfect replication of the labor rate to the extent that we can tell. You would still go to the regression line. The regulation provides that they start with the regression line. Start with and end with. Although, again, in this case, when it was brought to court, we went beyond it and revisited it, and perhaps in that context they would revisit it. They would consider whether to abandon the regulation, not that they have capacity within the regulation to do something other than pick the point on the regression line. Right? Yes. Okay. I mean, yes. Okay. It doesn't explicitly say within the regulation you may never go outside of this, but it also doesn't have an explicit escape clause either. The regulation must represent commerce's understanding of what the statute commands. Right? Yes, sir. So as to the regression analysis, commerce is in effect saying in the regulation, in order to comply with the statute, presumably best information available, we will always use the regression analysis. Because commerce believes that that will always be the case. Commerce's belief has nothing to do with what is consistent with Congress's mandate. That's the question. Is the regulation consistent with Congress's mandate? It's not an answer to that to say, well, it's based on what commerce believes. But the regulation, as written, is consistent because, again, within the regression analysis is the information from those countries. Well, it's buried in there with 53 other things. So it's essentially neutered, negated, overpowered. Yes, it's in there technically, but it has no real meaning anymore. I mean, that's what commerce has determined. That's to the extent possible means in this instance. Because of the nature of labor, that is the extent possible that they can use these factors. But what they've determined is that it's never possible to use the data within Paragraph 4 standing alone. And that seems a bit arbitrary given Judge Price's hypothetical. So maybe the regulation's invalid. That's the problem. Why didn't the regulation say, if there's a problem with Applying 4, we'll look at regression analysis, and if that doesn't seem to make sense in the context, we'll look at something else? You guys wrote it. You could have written it to give yourself flexibility. That might be a problem. Perhaps that would... Again, it was written that way because the evidence showed that relying on single countries... Suppose, hypothetically, that we held that the regulation, particularly the part that requires reliance on the regression analysis, is inconsistent with the statute and therefore is invalid. And your colleague at the table and you would then huddle because the Secretary of Commerce would say, Well, now what do I do? And you would say, Oh, easy. Just rewrite the regulation exactly the way it was before, and then after the word regression analysis, put in a few words that say, as long as that makes sense in view of other data, and if not, use something else. We would have some hope. It's an easy fix. No? The regression, as written, we believe is valid. Your version would also be valid, Your Honor. Anything else? You're going to talk about the four small company exclusions? Commerce. I'm going to ask you a question. All right. What do you understand Commerce to have said about that? Is the Commerce statement that we're doing this under protest mean it's not worth arguing about? How are we supposed to interpret that? Commerce believes that the original determination was supported by substantial evidence, but it also believes that the court's subsequent determination was supported by substantial evidence. We didn't appeal to that issue, and we don't have... So in your view, and this may be out of the area that you've briefed, and perhaps not appropriately directed to you, but if you can give us some guidance on this, it would be helpful. In your view, in that situation, it's pretty unusual, but in that situation, would our task be to determine whether Commerce's first determination was correct, i.e., supported by substantial evidence, or would it be to determine whether Commerce's latest determination is supported by substantial evidence, so that if we found both were, what would we do? In a sense, this isn't your problem, because you're not taking a position on this, but if you have a view, it would be helpful to us. I'm not certain, Your Honor. My understanding is that traditionally when Commerce does something under protest, that's their method of preserving the issue for appeal, but again, we believe that what the court did was supported by substantial evidence, so it's unclear to me which standard you would review the second remand determination. So if you had taken an appeal, you could have said, I suppose, in that appeal, you know, under protest, we did this, and frankly, we think this is supported by substantial evidence, given what the trial court required of us, but we appealed because we think the trial court was wrong to require more than, and ultimately a different result than, was reached in the previous remand, and in that situation, you would say that our choice would be confined to deciding whether what the trial judge did was wrong in sending it back for a further remand determination, and in fact, if your earlier determination was supported by substantial evidence, we would have to affirm that, I take it? Correct, Your Honor. So the only question then is whether you advanced an appeal or not, and whether that matters for purposes of what we do. I think that's right. Okay. So when you say we do it under protest, but we're not going to pursue it, how do we interpret what Commerce said? Do we interpret Commerce as saying this isn't worth arguing about? No, Your Honor, it just, the fact that we chose not to take an appeal doesn't signal one way or the other. It doesn't signal that it's not worth arguing about? No, Your Honor, the decision to take an appeal has many factors beyond, I think. The fact that we didn't appeal doesn't mean we don't think that there are, we believe both things. Why shouldn't you be held to make an election? If you think it's wrong, fight it. If you think it's right or maybe right, then don't fight it, but you can't have it both ways. You can't say he's wrong, but I'm going to do what he orders. It just creates chaos. No one knows what to do. No one ever will straighten out the problem if you can decide one way but talk the other way and have your cake and eat it, too. But if Commerce isn't able to do that, then these cases below the trial court could go on forever. If we continue to fight the court forever, they'll just keep remanding them. At some point, we need to draw a line so that we can move on to an appellate court. Well, exactly. That's the point. So that's what Commerce's decision was, that it was important to move on and not continue to argue about this. No? Well, not to continue to argue about it in front of the trial court. And then in this case, we chose not to bring an appeal. But in other instances, the Commerce Department might do something under protest so that they could bring an affirmative appeal. Why do you assume it would be futile to fight against what you conclude is an illegal decision by a trial judge? Judge Kogue is not bound by what he decided previously if Commerce in the remand decision educates him and persuades him that he was wrong. No, I understand. Judges can learn things from Commerce just as Commerce can learn things from judges. And therefore, I don't understand why you say, well, we couldn't contest this because maybe it would just go on on infinite items. Well, yeah, maybe it wouldn't. Maybe it wouldn't. Maybe he'd change his mind. Trial judges often change their mind in the face of new analysis, new data, new arguments, new citations, and so forth. Why shouldn't you be held to make a clean choice? No, I understand your point, Your Honor. But in the context of this case, Commerce made the determination that there was more, as I believe you said earlier, that there's more than one version of the facts that can be supported by substantial evidence and chose to go with the latter. I'm not actually clear on the record exactly why they made that choice, but I'm not sure I can follow your question. Well, it certainly is the case, as you correctly point out, that it's a pretty big deal for the government to take an appeal, and frequently the government does not pursue an appeal even when it thinks its position is clearly correct and it's not something that's done lightly. I suppose that the argument would be that there ought to be room for the government to say, well, we're not going to trouble you with an appeal, but you still shouldn't treat this position as having been abandoned for purposes of analysis of what's gone on before. I agree, Your Honor, yes. But the significance of saying under protest has no meaning in this case, but it's a warning to the court and the government that in a future case you might fight and not give up for the collateral reasons that Judge Bryson alludes to. There are always factors of judgment call of whether the government will bother to appeal or not. Yes, Your Honor. So there's no effect in this case to say under protest, but it's to try to preserve against an argument in a later case that you sort of waive the right to complain. I think that's right. That's all it is. Well, not in a later case, but in a different case, we might. In this case, we've got law enforcement that did an appeal, but not... It would be later and different. Well, right, but later and different. That's right, Your Honor. But if you had said in response to the, I forget the number, the second remand order before the third remand determination, in response to that, in the third remand determination, if you had said, ah, now that Judge Boggess pointed this out to me, we now realize the error of our previous ways, we think he's absolutely right, and we now, not under protest at all, but with enthusiasm, embrace his analysis and we come to a different result, then I guess there would be no argument, would there, that we ought to then look at the second remand determination. Instead, we would be stuck with or directed to the third remand determination. Under protest, it seems to me, does it not, has the effect of saying, you really ought to focus on the second remand determination as the point at which Commerce said what it really thinks. What it's saying now is what Judge Boggess is telling it to think. I think... Is that a fair characterization? Again, I think Commerce said that both were supportive of this, but I think it is fair to characterize under protest as a signal of sorts. But does it reflect a signal that the interest of finality, of not having continued back and forth in the courts over this issue, is that Commerce has decided that it is better to put an end to this and have finality rather than to continue to argue about it. Is that what it reflects? I think it reflects that as well as what Judge Bryson is saying. I mean, I think it can say perhaps more than one thing. Okay. For the reasons stated in our brief, we ask that you would affirm the trial court will count. All right, thank you. Mr. Grimson? I think I've extended all my time. We've restored it, so you've got five minutes. Okay. Picking up on the labor rate regulation issue, the question of whether they have to apply the regulation in every case is clearly answered in Commerce's rulemaking, in the 2006 rulemaking. They say under the Department's regression methodology, the value for labor will be the same in every proceeding involving a given enemy. They are not going to go outside of the regression. I thought she admitted that. She said that they could have some flexibility. No, I don't think so. I don't think that's what she said, but anyway, go ahead. Maybe she ended up admitting it. If that's the case, forget what I just said. I agree with her, though. When Commerce promulgated the original regulation, they actually included language which could have made it at least in part consistent with the statute. They had the words in there in the middle of the regulation that the regression would be applied to countries found to be economically comparable, but they took it out. So they made it end. Put it out in the final reg back at the original issuance. That's correct, in the 97 reg. They took it out. They said we haven't changed anything, which was wrong. Then they said years later that people had an opportunity to comment on this, and only one person did. But as we said, page 45 to 50 of our brief, what the world understood was that they were actually trying to comply with the statute on promulgating this reg, and the 97 change was really something of a bait and switch. I want to deal with the rubberwood ministerial error very quickly. Mr. Dorn didn't argue this. This is the situation I mentioned before where I'm preceding him. But in brief, there is a controlling regulation on raising ministerial error allegations. Yes, but the problem is that Congress sometimes has corrected errors that were not brought up within the time limits of the regulations, and the question is has it had an inconsistent practice of correcting errors sometimes but not other times, and whether to be consistent with its past practices should have corrected the error here. I think the question is whether it was an abuse of discretion under the past And it could be. If there were an identical situation or identical situations in the past where they corrected similar errors, and they failed to explain why this was different from that, maybe it would be an abuse of discretion not to correct it. I don't think the regulation, in other words, is a complete answer to this argument. It's not a complete answer to this argument where we have an active appeal because the courts have said not withstanding the regulation, Congress, you do have the discretion. But one thing that I wanted to point out is that in the three briefs filed at the court, the petitioners don't even deal with the five-day time limit of the controlling regulation. And furthermore, they don't deal with the second reason that Congress cited below, which is letting them change and raise a new issue now, two years into the complaint, could prejudice not only Dorbust but potentially up to 200 other parties whose rates were dependent, in part, on Dorbust's rate. And that's why Congress articulated a valid reason and why, in this case, it's not an abuse of discretion for them to say, you're too late. Two years is too late. And that the Gene Onyipin case that's cited, CIT case, it's not final and cited in the petitioner's surreply, is a case that did not involve that additional finding by Congress of prejudice. I think you'll find in the petitioner's position simply that you should fix this because it's wrong. But the analysis of late ministerial error always raises this tension between finality and accuracy. And Congress told Commerce to fix these things within a reasonable time, and Commerce said five days is a reasonable time. And the courts have said Commerce has discretion to decide on individual cases whether to fix them. So Commerce did it, and its decision was based on substantial evidence, and it surely satisfies the abuse of discretion standard. In this case, we really should have three counsels here. We have discussed that this morning, whether we needed to bring in a chainsaw to maybe separate this table over here, but we worked it out. You both behaved very well. Appreciate it. Thank you. Thank you, Mr. Dorn. Thank you, Your Honor. I'll start with wage rate. What Dorbess would have this court rule is that Congress directed Commerce to use only data from economically comparable countries and only significant producer countries, even if those data, even if data from other countries would make the result more accurate. Congress could not have meant that Commerce's hands were tied to only use economically comparable data and only use significant. Why couldn't Congress have made a choice? Look, we're going to keep this simple. It won't be perfect. It'll be a little quick and dirty, but that's preferable, and therefore it's going to be a little arbitrary, but that's the way we want it. Why couldn't they have made that choice? Well, we certainly think that Paragraph 1 does trump Paragraph 4 because Paragraph 1 says shall use the best information available, and it says in countries considered to be appropriate by the administering authority, so it's up to Commerce to determine what the appropriate countries are. So that's the shall version in Paragraph 1, and then Paragraph 4 is to the extent possible. So Paragraph 1 does trump 4, and this court has basically blessed us with what 4 means. But in shape proof, this court said, in determining the valuation of factors of production, the critical question is whether the methodology used by Commerce is based on the best available information and promotes accuracy in the dumping margin. And that's exactly what Commerce has done here, and Commerce has not ignored Paragraph 4. It's addressed Paragraph 4 on at least three occasions. And this court should defer to Commerce's interpretation of the statute with respect to each of those three occasions, first in the 1996 and 1997 rulemaking. And there Commerce was explaining that it had this practice of just, you know, picking a single surrogate country, and we're going to say, well, that's the estimate of China's wage rate or Vietnam's wage rate or other. And he says that is very unpredictable. It's just a random exercise. It's unfair because there's such a large variation in wage rates for this handful of economically comparable countries. And Commerce says that's not fair. It's not predictable. It's just random. And so it came up with this approach of using regression analysis to create a variable average wage indexed to the GNI. Well, what's the interpretation of that? Is Commerce wrote a better statute than did the Congress? No, because the – They don't have the leeway to do that. No, they don't, Your Honor. They don't know what Congress wrote, whether it's wise or not wise, efficient or not efficient. But under paragraph one, it's their obligation that Congress says you shall use the best available information, and the prior approach was not the best information available. The legislative history is not all that helpful, but on this particular point it seems to me it might have something to say to us, and you're familiar with it, I'm sure, and the conference committee refers to this very issue and says the factors would be valued from the best available evidence in a market economy country or countries that it is a comparable level of economic development. That sounds like what the committee is saying they mean by best available information in one and the factors in four is that they are to be read together, i.e., best available evidence, as they said, in a market economy country at a comparable level of economic development. And the – As opposed to best available evidence standing on its own without regard to what the evidence of the market economy country that's at a comparable level would be. And this allows – this happens – that happens in this process because the broadest data set is indexed to GNI of the particular NME. So in the decision below, Commerce said the regression methodology allows the department to rely on the broadest data set possible to arrive at a wage rate that is directly tied to each NME's GNI. And Judge Polk reviewed all the alternative approaches that have been argued by Dorbess and so forth, and Commerce explained in each instance why its approach was more accurate. And Judge Polk also said that Commerce explained that economic comparability is established in the regression calculation itself through the per capita income in the NME in question, quote, which ensures that the result represents a wage rate for a country economically comparable to the NME, end quote. Because all the regression is doing is creating a statistical relationship, and then it applies the result, the coefficient, income coefficient, to China's NME to make that wage rate comparable to China. Now, India, on the other hand, you know, is just one of 61 data points, and Commerce did say it's not an anomaly in the sense that you have to throw out that data point just like you have to throw out this data point. Why doesn't the statute contemplate an individualized determination instead of an automatic rule? I don't see that, Your Honor. I know you don't see it, but why shouldn't the statute be writ as requiring some sort of individualized determination rather than an automatic rule? Well, I think it's fairly common for agencies to develop practices in rulemaking. They can develop policies and practices either in rulemaking or in individualization. It's not so common to adopt a rulemaking which says we're never going to follow the statutory approach. Well, but they didn't say that, of course, Your Honor. They said what they're saying is we're going to try to follow the prong one of the statute and come up with the best available information, and what we've been doing in the past doesn't work because it's random. It's unpredictable. You have these people just playing games saying we want country A because it happens to have an outlier low wage rate like India, and the result of the dumping market is going to depend on the selection of that single surrogate country in Commerce's. That doesn't result in the best information available. We have a much better way to do it. We've got this global data set that allows us to tie the individual NME's G and I to this income coefficient and accurately predict its wage rate. They may be right about that, but one approach would have been to go to Congress, and the approach they chose instead was the easier approach of going to the Federal Register. The question is can they do that given what looks like a fairly specific methodology that's laid out in force? But Congress did say in pronged force to the extent possible, and Commerce has explained in the 1996-1997 rulemaking proceeding, again in 2005 and 2006 when it went back and did a notice and comment proceeding, and Dorbest commented, for example, on the methodology then, and Commerce went through the same analysis and explained why this approach is far better than randomly picking an individual wage rate from an individual surrogate country given that randomly. Translation of that is just to say the methodology represented by force is stupid, that there's a better way to do it, but that doesn't mean that isn't what Congress chose. But Congress didn't choose that because in paragraph one it says you shall use the best information available, and in four it says to the extent possible. Well, unless Jim is right as I understood him in suggesting that four is a definition in that circumstance of what best information available means. So there's no conflict. There's no issue about this one from four. Four is defining for those circumstances what the best information as required by one is. That's not the approach that this court took in Nation 4. In Nation 4, this court blessed Commerce's use of world market price for a production input rather than using the price in India, the surrogate country, because that was the best available information. And here we're not saying not to use any... I thought it was the world market price for goods originating in India. No, it was the world market price for goods that were imported into India. Is that the Aniline case? Aniline, that's correct, Ron. And here, Commerce isn't saying we're not going to use any data from India or from any of the economically comparable countries. It's not saying we're not going to use any data from significant producer countries. What it's saying is we're going to use that data, but then in order to get a more accurate result to get the best information available, we're going to add to that to get a more robust statistical correlation between wage rates and GNI. And India is not a twin of China. India's gross national income is 420. We can't retry the new facts here. We've given you lots of extra time and other parties extra time, so we're going to take the appeal on your advisement. We thank all three counsel very much. Back to you, Ron. Appreciate it. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A7760. Public record. Not legal advice.