Oral argument — YANGZHOU BESTPAK v. United States

Department of Commerce found that the anti-dumping rate for our client Bestpak, a fully cooperative respondent free of Chinese government control, should be based on the simple average of the zero rate for YAMA, the only company individually investigated, and a 245% rate for the China-wide entity that was based on total adverse facts available, a simple average of those two rates. We believe this decision is not reasonable, contrary to judicial precedent, not sanctioned by the statutory language of the Statement of Administrative Action, not justified by the problem that there was a limited amount of information in the record, and not justified by the Department of Commerce's AUV margin analysis. What is the judicial precedent that it's contrary to? Well, the judicial precedent, the case was decided in our brief, Your Honor. The most recent case is the CHANGZHOU case decided by this court recently, which you can distinguish on the facts, but if you look at the language of CHANGZHOU, you look at what the court was talking about, this CHANGZHOU court was saying the AFA rates are disfavored by the statute. They're saying that fairness or accuracy is the overriding purpose of the anti-dumping statute when calculating a rate for a cooperating respondent. And then the CHANGZHOU court recently said, I'm going to quote, AFA rates are disfavored by the statute. Indeed, the only AFA rates contemplated under the section, the simple average methodology, are those determined for individually investigated parties. Such rates are generally based on verified data and bear a reasonable relationship to the party's actual business practices. In this case, Your Honor, the rate of 245% has absolutely no rational relationship to best practice business practice. If you look at the record, if you look at the comparison between what YAMA is like, YAMA had a zero rate, and 245% China-wide adverse facts available rate, we think it's absolutely clear, Your Honor, that our company is closer to YAMA and has absolutely nothing to do with the China-wide rate. We fully cooperated like YAMA, we were free of government control like YAMA, we submitted all data as requested, we certified to the accuracy of the data, we made our deadlines, we completed our questionnaire responses, we did absolutely nothing wrong. You're asking the courts, though, to decide on a record that you didn't make in the agency. I'm sorry, Your Honor? You've got to exhaust your administrative remedies. I believe we absolutely exhausted our remedies, Your Honor. We made our case brief to the agency, we made these arguments in our case brief to the agency, we argued that when the Department of Commerce is averaging a China-wide rate and a zero rate is absolutely contrary to law, and that our... Well, in the absence of one invoice you're talking about, there really was nothing, and the invoice was not before Commerce in this aspect of the case. Yes, the Court of International Trade said the invoice was not before the agency, we didn't exhaust. We disagree on that point, but even if you take that invoice out of the record, if you look at the record even without that invoice, which we believe completely supports our case, the Department of Commerce still should not be allowed to use a China-wide adverse facts government-controlled rate, punitive rate, to assess duty on our fully cooperative client, fully cooperative company. And what's really disturbing to us, Your Honor, the Yama rate, the zero rate, we first realized the zero rate would take place, the Department of Commerce's preliminary determination. If the Yama rate would have been 3%, there would have been no question that that would have been representative of BestPak's rate, BestPak would have had a 3% rate. But what happens, because the Yama rate is zero, then the Department of Commerce decides to take the zero and the 245%, and use a simple average of those two rates, and apply it to our company. How do we know that? You said with some certitude that if it had been a rate of 3%, then your client would have gotten... such as BestPak, which have established they're free of government control in all situations, unless the mandatory respondent rates are zero rates, or based on adverse facts available. We have the unusual situation, which was not our doing, which was based solely on the Department of Commerce decisions, that they selected two mandatory respondents, one mandatory respondent received a zero rate, the other mandatory respondent, they said... I understand there's frustration, there's a bad feeling about how things went down here, or what went down, I mean that frustration was expressed by the CIT. We believe it's illegal, Your Honor. I'm sorry? Yes, Your Honor. The CIT expressed the frustration, you called it unfortunate and even frustrating. But the fact remains that, and initially, in her initial view, I think the CIT even mentioned that, sort of blamed, or shifted the burden to Commerce, saying that it puts itself in a precarious situation by only taking two respondents. But let's assume that we think that was a rational or reasonable decision, given the play here, that they tipped these two respondents. It's not their fault that the rates turned out to be the rates that they were. So if we assume that it was reasonable under these circumstances to only tip these two mandatory respondents, it's not the government's fault that the rate came back at zero percent. So what was it to do then, if it found itself with these two rates, zero and the AFA? The government had a chance to do something correct, Your Honor. They had a chance to make a reasonable determination with the zero and AFA where we lied. We're closer to the zero. Doing a simple average of the zero and AFA is totally, in our opinion, absurd and unfair. But is there anything that you can point to in the statute or otherwise that precludes them doing that? To use the China-wide rate. We believe the statutory language precludes the China-wide rate. To use the China-wide rate. If you look at the statute, they say if you have a zero and an AFA situation, they have to use a reasonable method. And we're saying, you know, that is per se unreasonable when applied to a fully cooperative respondent who's free of Chinese government control. And then the statute says you could use the averaging, the estimated weighted average dumping margin for the exporters, producers individually examined, individually investigated. So you're allowed to use an averaging, but it has to be of companies that are individually investigated. Here, Yama was individually investigated, but the China-wide entity is not individually investigated. And there's case after case that before the CIT and recently in this court saying you can apply a punitive rate for a company such as BestPak that fully collaborates. How does Gallant Ocean apply in this circumstance? We believe that the principle of Gallant Ocean absolutely applies. The language of Gallant Ocean applies. The facts may be a little bit different because I believe in Gallant Ocean, the rate may not have been corroborated. And here, the rate was barely corroborated. But the principle, when you have such an incredibly punitive rate of 245% to take a simple average of that rate and apply it, you know, to our company who's more similar to Yama, there's something terribly wrong and unfair and punitive. And again, as the courts have said over and over again, this purpose of the statute is not supposed to punish a fully cooperative respondent. We are being punished for something totally, totally beyond our control. You know, in this case, petitioners say we should have done something more. The government says we should have done something more. This problem doesn't come up until the preliminary determination because nobody knows that Yama's going to get a zero. And that's the problem, how it came up before the Department of Commerce. Again, if Yama gets 3% or anything above 2%, no problem. We use the Yama rate. But because Yama got a zero, which couldn't have been contemplated, the Department of Commerce had to do, you know, they had a problem. They had to do something differently. And in our case before the department, we said, you know, pick another mandatory respondent at this time. And we said, you know, try to use other information in the record. But the Department of Commerce tells us it's too late because it's after the preliminary determination. They're not suggesting for your client zero was the only possible rate they could have picked. And you're objecting to the midpoint number that they picked here, right? We would like zero, you're right. Be honest. We're arguing for zero. But we could say they could have used another method that would have gotten us close to the Yama rate. The problem is, what are the rational methods that they could have come up with closer? So you would have been up here if they had picked 50% and argued, well, no, that was an arbitrary number. Where did that come from? The midpoint test has some theoretical rationale. The midpoint test is based on a simple average of Yama, who has over 10,000 sales, 600 different products, 130,000 kilograms sold to the United States. And this China-wide entity is based on one offer of one product of minimal, minimal quantity. Am I wrong? I mean, if they had picked something closer but not at zero, you would have been up here making an arbitrary, unreasonable argument. And if the number was picked out of thin air, which it possibly may have been, you would have had a chance. I mean, where do they come from? Your Honor, we have several points, Your Honor. The first point is we believe as a matter of law the Department of Commerce can never use a China-wide AFA rate as an average. You know, just as a matter of law, because it's a government-controlled entity. And there's a lot of case law saying you can't take the rate for a government-controlled entity and apply it to a company who's not subjected to government control. Well, they did apply that rate, right? They applied the average. It was just as bad. They applied a simple average. But secondly, if you don't agree with us on this matter of law, and we're very firm we believe this is a matter of law, you could go to the facts in the case and say, what would have been a reasonable solution? We could think of, you know, thinking here now, we could think of reasonable solutions. They could have used some type of weighted average. They said the, for example, they corroborated the rate for the Chinese entity by saying it's approximately 4% of Yaba's sales, where you could take a 4% and 100% weight average. They could have asked for additional information. We gave, we said, look, you know, come back. We'll give you more information. They could have done all these things. But they did something that was terribly arbitrary and terribly unfair, and it wasn't necessary. Would you like to save the rest of your rebuttal? Yes, please, Your Honor. Thank you. The overriding purpose is to make this as accurate as possible. How can you justify this as accurate? Your Honor, this is justified under the statute as a reasonable method when commerce has only de minimis and facts available. How can it be accurate when you have one data point, zero, that was actually computed? The other one is, of course, a punitive measure for adverse facts. Respectfully, Your Honor, we submit it's not punitive. That's not permitted under the AFA statute. It is a rate. The AFA rate was chosen based upon a petition but was corroborated through transaction-specific margins from Yama's data that ranged business proprietary information, so I won't say the numbers. But we're looking for economic reality here, some kind of commercial reasonableness. How can you do a substantive evidence review when you only have one benchmark? The substantial evidence… You have to have something else, don't you? You could have gone out and gotten someone else through a mandatory procedure. You didn't. Well, the average unit value analysis that commerce conducted was to address exactly this point of whether there is substantial evidence to support this separate rate. Based upon the AUV analysis… There's no evidence. You've just averaged a zero and a 250% and come to 125%. How is that accurate? There is not much evidence on the record of what BestPacks margin would actually be. It is true that this is an imperfect analysis, but the average unit value… It's imperfect, but why couldn't you have just asked for another participant, got a few more data points? One can't be accurate. Again, Your Honor, there's not just one data point. There are two. They were averaged. Yes, it's a simple average, but that is commerce's permittivity. We both recognize that these are two entirely separate situations. One is a China-wide government-controlled entity, the other one is not. The one that was not, more similar to the BestPacks here, got a zero, and yet you're saying it's accurate to give their closest similar company over 100%? Again, the average unit value analysis derives from the information from the quantity and value submitted by BestPack and that were available on the record for Yama and for Ningbo Jintian, suggested that in fact BestPacks margin may in fact be somewhere in between the two, in between Yama and Ningbo Jintian. May, may, may, may. Why don't you find out? You have a chance to. Just ask for somebody else. Get some more data. Accuracy is what you're supposed to be doing. You are amounting to being punitive, which you admit is not what the statute allows. But we would also submit that we're not being punitive here. You're guessing, and the guess is according to most neutral examinations of the record, is far in excess of what BestPack would have fallen into. The invoice gives us at least some indication that that's true. The commerce determined that the information on the record, although again there is not much information on the record, suggested that there was enough evidence to say that the separate rate was appropriate. That meets the substantial evidence review, whether if we looked at it... It has to be reasonable and accurate. I have trouble with both reasonable and accurate. It doesn't seem to be either. The statute for the all others rate... You're so thin here, how can you suggest that you've got enough evidence to say these people deserve a doubling of their price, which of course will charge to the American consumer most of the time, because you didn't get enough data. Commerce got the data that is required and permitted under the law. They chose two mandatory respondents. As soon as Gentian drops out, why don't you get another mandatory respondent? Because now we know you only have one data point now that is going to give you... The other one isn't going to give you anything other than adverse facts. There's no requirement that commerce decide to change mandatory respondents when one doesn't cooperate. How can you be reasonable and accurate under these circumstances? The statute explicitly contemplates that it would be reasonable to have only zero and ASA rates in some circumstances, and that the reasonable thing to do in that circumstance is to average them together. That is what commerce did here. It is using a methodology explicitly contemplated and permitted by Congress, both in the statute itself and under the... Can I just ask you a different variation? I mean, it's on point. But as I pointed out to your friend, the CIT in the initial opinion observed that commerce put itself in a precarious situation when it selected only two mandatory respondents in this instance. Was she right? The... I'm sorry, was she right? That commerce was... Is that comment accurate? It is... That you put yourself into a precarious situation by only selecting two mandatory respondents? It is certainly true that when you have fewer mandatory respondents, then there will be fewer data points at the end of the investigation. That does not mean that commerce was going to end up in a situation where it couldn't choose an all-others rate. And in fact, it did so here. It still had two data points to average together. That's all it was ever going to have to average together. And the statute... In your brief, you say, at page 10 of your brief, you say further, that it did not have sufficient time to obtain, evaluate, and employ additional factual information from separate rate companies, as suggested by BESPAC. Did you put on evidence of that? I mean, is that based on some statement you made, or is that attorney argument, or is that some evidence you put into the record establishing that? Commerce determined that it would only use the two mandatory respondents and would not seek additional information. And it did say, and I don't have a specific page citation for it, but I can get it for you, Your Honor. Well, you cited 68, which I hadn't looked at. That it did not have that sufficient time. Commerce is hampered by limited resources, as the rest of the government is, and it determined that it would use the mandatory respondents that it chose under section 1677F1, as permitted by that statute. I'm using the methodology there to choose mandatory respondents. Well, is your friend right in terms of his outline of events as they occurred? In other words, if the other guy had had 3%, this would have been entirely different, and he would have had the 3% rate. It was only because of the zero, which was not necessarily something that was contemplated at the initial stages? The statute 1673DC5A says that in the usual circumstance you will take an average number excepting the zeroes and AFAs. But then section 5B says you only have zeroes and AFAs. So if there hadn't been a zero. So if there had been something other than zeroes and AFAs, yes, the required method would be to... And the reason the statute does that is because they figure the AFA and the zero may be inaccurate. They want some accuracy, right? And so the idea is let's get close to figures we actually have. You don't have any figures here, do you, except for a zero? Again, Commerce used the average unit values looking back to the information submitted by BESPAC, the separate rate respondent, the information that that company did submit, which is what BESPAC urged Commerce to use to figure out a rate. Commerce looked at that average unit value and compared it to Yama's, compared it to Ningbo Jinchan's, and said, look, BESPAC is pretty much right in the middle. That means that this average doesn't seem reasonable. What do you mean? They reached the conclusion that BESPAC was really right in the middle? What does that mean? The average unit values that were calculated from BESPAC's submit on remand. This is what Commerce did to... Well, that number is right in the middle if you're a mathematician or a real person saying if you have this number and this number, this number is right in the middle. I'm looking for a more thorough, fulsome rationale from the Commerce Department other than the obvious statement that 125 is the middle number between zero and 250%. The average unit value is not the rate, that's not the margin used. That's the value, the price basically. And the information, obviously Commerce can't figure out normal value and construct specific margins for these companies. That's why we're in the situation we're in. But what information it did have was these average unit values, which is an imperfect proxy for what margins might be, but was some information that Commerce did have and what Commerce did use to substantiate its separate rate. Did you want to give Mr. Doris some time? Yes, Your Honor. Thank you. May it please the Court, I just wanted to correct a few errors that were made in terms of terminology in this case. It must be remembered that Commerce was very forward looking in the way they set up the statute. They said, first we're going to allow there to be individual investigations of different companies, then it said there may be cases where there's too many individual respondents to investigate, so we're going to say you can narrow the investigation. But would you narrow 19 down to 2, you lose 1, and you end up with this situation with a zero and an AFA? Well, that's also by the statute, Your Honor. The statute contemplates that both of those circumstances are kind of unusual, and yet you put them together, you didn't, but the Commerce Department put those into the same equation. There's a penalty of everybody who's not, who wasn't given a chance to give special information. Of course, they were given a chance. There ought to be a voluntary respondent, Your Honor. And that was another thing to correct. You don't wait until the prelim to find out what the rate is because you're gambling then to see whether the rate's going to help you or hurt you. As soon as they jumped out, they could have become a voluntary respondent too. So they had the opportunity. Or the Commerce Department could have gone and got additional. That's not part of the Commerce Department's practice. They go by the statute which says they make their selection at the beginning, based on the volume under the statute. And there's nothing that forbids them from doing something to get an accurate rate. I don't think the word accurate is in the statute. And they're the ones, see, that knows that Yama's looking like a zero. No one else knows that because they don't have all the figures. Commerce is the one that knows Jinten just dropped out and Yama's looking like a zero. Whoa, we've got a real unusual circumstance here. Let's get some accuracy. And they didn't do that, did they? Well, two things. I don't think there is a requirement for accuracy in the statute, Your Honor. Well, that's actually the Roan opinion that we said in reviewing these. We said the overriding purpose of the statute is to compute anti-dumping as accurate as possible. That's binding precedent on you and them. The statute for the separate rate says that they have to estimate what the separate rate respondents will get. When you have, in this case, 12 different separate rate respondents. As accurate as possible. That's what the standard is for reasonableness under our Roan opinion. But with 12 different respondents that are separate rate respondents, you can't investigate every one of the 12 respondents given their limited resources. And I don't think anybody suggested that. What everyone is looking at is how you get something that's reasonably accurate when you have one drop out and go to adverse facts only, and the other you can see is going to be a zero. At that point, why don't you take some step to correct? And there was no step taken to correct. How can that be reasonable? I think at the time the department didn't know that YAML was going to get a zero. They didn't know how close it was going to be. They had already made their selections of respondents. Questionnaires were already sent. It takes time. They were under very strict statutory time limits. Actually, if you look at the record closely, there were some facts on YAML that could have given them a 1%. They gave them a zero, creating the problem again. I thought that was the difference in the CBD case and not the AD case. And the other about the emphasizing the fact that this was an individually investigated respondent, Ningo Jimpin was. Thank you, Mr. Doris. Mr. Marsha. Thank you, Your Honor. Can I just ask you a quick question? I don't want to take up all your time, but why didn't you ask for a voluntary respondent status? Your Honor, this is the beginning of the case. The beginning of the case, we are not one of the largest companies, largest exporters. Department of Commerce picks two mandatory respondents. We're willing to live with the fact that if those two mandatory respondents don't both stay in the case, or if one stays in the case, the rate is going to be above the minimums. In the normal case, the rate is going to be above 2%. This is just the way you normally go. In that scenario, to be a voluntary respondent requires an incredible amount of work right from the beginning. We're willing to take a representative rate from all the mandatory respondents. So why don't you have to live with that now? I mean, you rolled the dice, you took a gamble, and things didn't turn out that way. Because it's a very, very unusual circumstance, and we didn't take a gamble. We took a gamble. If we had been a voluntary respondent, yes, we would have been in all the way. But that doesn't justify what the Department of Commerce does. There's no justification for the Department of Commerce acting totally unreasonably, in our opinion, contrary to law, just because we didn't say we didn't request voluntary respondent status. So in that scenario, every single company in every single case is going to have to request voluntary respondent status right at the beginning of the case. And you're going to have hundreds and hundreds of requests. The Department of Commerce routinely denies these requests because they say they don't have the resources. But every company would have to make the request because of the unusual circumstance that it may end up that the company who's left standing gets a zero. It's not the way it works in the real world, where you have the possibility of dozens and dozens or hundreds of potential voluntary respondents. Companies don't do that. They go and they say, okay, you have two companies. Somebody's going to end up standing above the minimus. We'll go there because that's the general rule. The general rule is that the mandatory respondents represent the industry. It's a representative rate. And I think the government agreed because the statute says it's a 3% rate or above a 2% rate. It's representative because it's zero. The statute says you shouldn't use it. Now, the fact that it's zero, I don't think the statute's saying it may not be inaccurate. I think the problem with the zero rate, saying it could be a zero rate is a zero company gets out of the case. So it's excluded from the order. I think that's one of the reasons you have the zero rate not being used in the average. But you don't use the zero and you don't use the adverse facts available, but you still have to do something reasonable to get a rate that makes sense. And I believe the statute's written on a reasonable standard to look at a market economy case for individually investigated companies. A China-wide entity is not individually investigated, so we're saying it's a matter of law. We'd like the court to say this is a matter of law. You can never use a China-wide rate in this average. The statute talks about an individually investigated company. And then the Statement of Administrative Action, which puts some meat on it, it says the expected method in this case will be to weight average the zero and de minimis margins and the margins determined pursuant to facts available. To weight average. The Department of Commerce used a simple average and it's a tremendous difference. If you would have weight averaged in this case, you'd have minimal, minimal quantities from this China-wide entity that they used to calculate that 240% average. The corroboration was based on less than 4% of Yang's sales. If you weight averaged, you could come up with something reasonable. And the Statement of Administrative Action says... Final thought, Mr. Marshak. Thank you, Your Honor. Thank you, Mr. Marshak.