Oral argument — QINGDAO TAIFA v. United States

We have two cases scheduled to be argued this morning. The first case is Appeal No. 2011-1595, QINGDAO TAIFA v. United States. Mr. Mastriani? Good morning. Good morning. May it please be a pleasure to serve under your chairmanship, Mr. Mastriani. This is a case that was challenged by TAIFA to the Court of International Trade of Commerce's determination, final determination, in an administrative review covering hand trucks from China. And that appeal to the CIT generated three remands by the CIT and then four decisions, the fourth of which is now before this Court. In the underlying proceeding, Commerce applied 100% AFA rate, otherwise known as adverse facts available, based upon TAIFA's failure to provide certain information, including factors of production regarding general purpose wheels, sold to the U.S. during the administrative review. Now, despite the fact that the incomplete data regarding wheels only affected 8% of TAIFA's U.S. sales, TAIFA has conceded the application of an AFA right to the totality of its sales. After three remands by the CIT, Commerce then calculated a 145.9% rate that was attributable to TAIFA's sales. And it did so by weight averaging the three highest model-specific margins from the original investigation. However, TAIFA had a verified rate in that investigation, a company-specific rate of 26.49%. So the AFA rate of 145.9% is 5.5 times the rate that was given to TAIFA in the original investigation, and far in excess of the other rates in the original investigation and reviews for other companies. And those rates range from 0% to 46.48%. Well, we don't have to compare against other companies where there's so much data that hasn't been produced, do we? You have to take it into account in looking at the issue, and it's an overarching issue here. And I think it's the most compelling issue, and that's the issue of commercial reality, because it is absent completely from Commerce's determination. Well, how does Commerce know what commercial reality is, given the broad scope of things that were not produced and the destruction of documents and the repeated lies to Commerce? How can they possibly know what commercial reality is? Well, commercial reality, first of all, the conduct part of this issue is nonexistent. It's not before you because of the concession that AFA can apply to TAIFA sales. So that is just an emotional appeal by the government and by Gleeson to the court. It has no place here. And it's also imploring you to punish TAIFA, and that is prohibited by the statute and by this Court's precedent. Well, certainly not punitive, but there is a deterrent factor, correct? Absolutely, there's a deterrent factor. And the deterrent factor, under this Court's precedent and rubric, should be applied to the 26.49 percent rate. There should be a deterrent, an increase applied as an incentive for the party that did not cooperate to cooperate in future proceedings. You're saying our case law requires you to start at that 26 percent and go up. That's the only mechanism that Commerce can use? It's not the only mechanism, but there's a presumption that an exporter's or manufacturer's prior margin is a valid margin for the proceeding at issue. But why is that prior margin representative of what a proper rate would have been had the parties cooperated in this investigation? That is the rate that was selected by Commerce here, so that's the one we have to work with. And they looked at that data set, and they selected three models out of 96 models. That's 3 percent of the products that were sold. It happened to be 36 percent by value, but there were 96 models sold, many of which were sold. So the sales data for those models were verified? They were, yes, Your Honor, absolutely. So the record here, the problem with the CIT's statement talking about that the record is barren and that 36 percent of sales is some type of reality, it's not some type of reality. It's part of reality. The totality of the reality before the court. But if the 36 percent is based on verified data and corroborated data, why wouldn't that not reflect reality? Because it's only part of reality. The real reality of that information was available, and it was all verified. So you cannot use part of that information. There has never been a case, ever, before the CIT, before this court, where part of a verified amount of information that had, as a result, a company-specific rate was extracted, and also in a way that takes the highest margins, and then to yield a margin, and then what happens is that you try to corroborate it with information from the review. We're dealing with an AFA situation here, and Commerce must establish a dumping margin, and it's reached back and used verified data, those corroborated. It represents 36 percent of all sales. Why does that not reflect commercial reality? Because it's only part of reality. And let me read to you what the CIT said in Type III that we believe dams, for lack of a better word, the result in this case. What the court said in Type III when it decided to look at the 227.73 percent rate that Commerce had to come up with, where they had not obeyed the remand to look into this government control issue, the court decided to make a pronouncement on that rate in order to dissuade Commerce from going back to it yet again. And here's what the court said, and I'm going to read to you in its entirety. It's at page 1386 of that decision. Based on the results of the investigation and reviews cited by the parties, the court doubts this AFA rate reflects reality. The issue is, is the AFA rate so far from what has been demonstrated by actual rates that it must be rejected as in Gallant Ocean, or is it an adequately supported rate as in KYD? First of all, this is not an actual rate. It is not derived from an actual rate calculated for anyone by anyone. This rate is for a portion of types of sales. Commerce sometimes uses a portion of sales to corroborate an overall rate based on facts available. It does not appear to have done that here. The problem is the methodology. But the same judge who rejected that because she felt that Commerce didn't present enough then turned around and said that the alternative rate was supported and that there was sufficient evidence that it was a valid rate and not something that was aberrational or punitive. So, I mean, I hear what you say about what she said in Type S3, but at best she said, I just doubt it. And then she turned around and made a specific finding as to the next rate that Commerce came up with. That doesn't make it appropriate, Ron, because what the court did was it rejected, resoundingly, the 227.73% rate based on 12% of sales. It then decided to accept 36% of sales. But the same analysis that the court did with regard to 227 has to apply to the 145.9. All the court says is that 36% of sales is a large enough proportion to demonstrate some form of commercial reality when the record is otherwise barren. We're not here to talk about some form of reality. We're here to talk about reality. Well, we're not dictated by the CIT's legal analysis. We're dictated by our court's legal analysis. And there is case law that says even the highest rate can be appropriate. We can choose the single highest rate in some circumstances. Yes, you can, but it's an overall rate. That wasn't done here, because the highest rate is 46.48%. And that was rejected. No, I'm not saying the highest rate from an alternative competitor or an alternative chosen rate. I'm talking about the highest rate with respect to your own product. Right, I understand what you're saying. You're saying from a single transaction. Yeah, using the 12%. Well, Your Honor, there's no way to corroborate that, because the problem is you're taking a snippet, and the most adverse snippet from a universe of information that was all verified to yield a specific number. You keep saying it was all verified, but it was all verified based on records that Commerce said it didn't have faith in, correct? No, I'm sorry, Your Honor. We're talking about the original investigation. That's what is being used here. The information that yielded the 145.9% was taken from the original investigation where TIFA cooperated, it was verified, and it yielded a company-specific rate for which it posted a cash deposit on future import. So that's pure and inviolate. And what happened here was that the top three transactions were taken to yield 145.9% rate, which is part of a weighted average data set that we all know yielded a verified 26.49% rate, and then to somehow corroborate. And corroboration, the Congress and this Court have said, is there to ensure reasonable and accurate dumping margins and to prevent Commerce from overreaching reality. And it clearly overreached reality here because the reality, and unless I'm hallucinating here, is 26.49% rate. And the fact that the CIT in TIFA IV said that there were rates in the review that's under question where the information was rejected somehow corroborated the 145.9% rate. That's illogical also because that Court said that that information was reliable and did not reflect commercial reality. And in fact, if you look at what happened in preliminary review, there was a 3.82% margin calculated. Now, 8% of the sales had problems, but I don't see how you could ever get from 3.82% margin, take 8% of the sales, give it the most adverse inference you want, how you could ever get to 145.9%. The point is we are playing in the field of the original investigation with regard to this AFA rate. And it doesn't comport with the statute. It doesn't comport with this Court's precedent in DiCecco, in Gallin Ocean, or KYD. Would you like to reserve the balance of your time? Yes, thank you very much. All right, very good. Mr. Panzara, good morning. Good morning. May it please the Court. This Court should affirm the trial court's decision sustaining Thomas's calculation of adverse facts available in this case because that determination was supported by substantial evidence on the record and also comports with the statute, this Court's precedent, and past practice. What principle distinction is there between the method of the trial approved by the Court of International Trade this time around with respect to the 145.9% rate and the earlier computation by Commerce of the 227.73% rate? We understand that the Court of International Trade expressed doubt regarding the previously calculated rate, the 220 plus. I understand that. My question is what principle difference is there in the methodology used the first time around and the methodology used the second time around? Now, I appreciate that the numbers are different. Twelve percent of products the first time, 36% the second time. Right. There was a much more searching analysis, and of course the basis of the rate that was calculated that is the subject of the final judgment was based on a much higher volume of sales transactions. And Commerce explained why it was selecting those sales for purposes of calculating ASA. So the principle distinction is that there was a higher volume of sales the second time around than in the sampling for the first time around? With respect to the selection, and Commerce also provided additional analysis with respect to corroboration as to why it could not use the overall calculated rates that were verified in the original investigation. Commerce is not endorsing the cross-appeal and asking that we adopt the 227. They're asking that we stick with the 145. Is that right? That's correct. I'm only authorized to defend the final judgment and the basis of that is the 145.90. We're not taking a position with respect to Commerce. Now Commerce did the calculation and adopted the 145% rate under protest. Is that right? The Department of Commerce protested the rate. It explained that it felt that the original rate that it had applied was justified under the circumstances. Has Commerce now abandoned that earlier rate? Commerce is not taking a position and the United States is not taking a position with respect to their earlier rates. But you did initially calculate the 227 rate, correct? Yes, the Department of Commerce originally calculated that rate, considered that it was justified given the doubts that were expressed by the Court of International Trade in remanding that, Commerce did a much more searching analysis of its rate and provided additional explanations for why the 145.90% rate comported with the statute. So you disagree with Mr. Matriani when he says that the 145 and the 227 have to rise or fall together because the logic of them is essentially the same? That's correct. If you read the lower court's final decision, she recognizes that this approach, using the highest transaction-specific margin in the manner that Commerce did, is less than ideal. And we recognize that it's less than ideal. The normal practice is to use the prior highest calculated weighted average margin. We couldn't do that in this case because those rates that were on the record were A, derived from cooperating respondents, and B, were far below evidence of actual dumping, transaction-specific dumping that was much higher. So Commerce knew that it couldn't go below those prior calculated rates. The normal practice would have been to use the prior highest calculated rate, but in this case that was not available. If that's not available to Commerce, then it will look to transaction-specific rates, and it did so in this case. It recognizes that it's less than ideal, but because it's corroborated based on the nature of the data, that it was relatively recent, that it was a very high volume representative number of sales, it was TIFA's actual data that we're using. So to say that it's not commercial reality distorts the standard. Commerce is not recalculating what the actual rate would have been in this case. It's trying to approximate what an actual rate would have been with a built-in increase in keeping with its statutory mandate to have a determined rate. Where does the line between deterrence and punishment, where do we draw that line? How do we draw that line? The court in KYD explained that if Commerce is complying with the statute, then it's non-punitive. Where it becomes punitive would be if it's an aberrational sale or something that doesn't represent or reflect what was actually happening in the marketplace. If Commerce is using actual sales based by the respondent, and those sales are relatively recent and represent a significant volume, then Commerce has provided sufficient corroboration, and by meeting that corroboration requirement, the court is assured that the rate is not punitive, that it is merely for deterrent effect. You could look at one non-cooperating entity and use 60% of their sales and come up with a lower number, and another non-cooperating entity and use 36% of their sales and come up with a higher number, and Commerce can just do that without having to respond to whether or not there's a punitive element to it? Well, in comporting with the statute, and by corroborating the rate that has been selected, Commerce ensures that it is not punitive. If it's not corroborated, then the court raises a red flag as to whether it's punitive, and that was the problem in Gallon Ocean, was that Commerce had used the petition rate, even though other evidence on the record suggested that that rate was not reliable. The principal concern for corroboration is reliability and relevance, and within that relevance determination is the commercial reality concept that was at issue in Gallon Ocean, and because Commerce did not use, in Gallon Ocean, it used the petition rate that didn't reflect actual sales, that was problematic, and it raised a red flag in the court's mind. In this case, we don't have that problem because we're using the actual sales of Qingdao Taifa in a recent proceeding, and we're using a substantial volume of those sales. So to say that it's not... Your opponent is arguing that instead of going back and selecting the three models upon which the AFA rate was based, and using the same reasoning for using those three models, that what Commerce should have done is used the entire database or the rate that was established during the less than fair value investigation, because all of that data was verified. And why would Commerce go back and just selectively pull out three models when you have an entire database of all sales that has been verified? Why is the latter not more reflective of commercial reality than the former? Your Honor, I see I'm out of time with Lee, but I'll answer your question. You can answer the question. The reason is that the previously calculated rate upon which Appellant is relying was calculated under a circumstance of cooperation. It was not in a situation like in KYD where there was an AFA rate that carried forward and that presumption kicked in because if we're going to continue to apply AFA, then we can use that previous AFA rate. So this answers Giorgio Maldi's question, because the rate in the original investigation was much lower than the AFA rate that was established in the third review, so maybe that's the delta for punishing for non-cooperation? What rate did they get in the original investigation? 26%. 26%. So then it's 120%, we could say, that is the rate that was tacked on for non-cooperation? It would be arbitrary for Commerce to apply just blindly some multiplier, and that's not what Commerce did in this case. Commerce looked at the data and tried to approximate some actual rate with a built-in increase, and it did that by looking at the actual sales and selecting a significant volume of those sales in order to satisfy the requirement that it provide some deterrent effect to ensure future compliance. If Commerce had blindly applied what the previous rate was, it wouldn't be complying with the statute for purposes of deterrent effect, because it would be rewarding them for destroying evidence and things like that by allowing them to keep a rate that was derived at their operation. But they concede that there should be a margin, right? I'm sorry, I didn't hear you. They concede there should be a deterrent margin. They just say the margin you chose is too high. It's unclear whether appellants are arguing for a particular rate. They're saying that we should have started from that 26%. The question is, well, what relevance does that 26% have For Commerce's purposes, because that rate was derived at through cooperation, the relevance is that we know that an AFA rate cannot be that rate or below. But isn't that the best evidence of what a rate would have been had they cooperated? This Court has affirmed Commerce's practice of using prior calculated rates for purposes of AFA, but that's not to say that Commerce must always do so, and this Court has never held that Commerce is always bound to use that prior calculated rate. There's a great deal of discretion for Commerce in selecting the rate and corroborating that rate and ensuring that it complies with the statutory directive to ensure that future compliance. All right, thank you very much. Thank you, Your Honor. You've run over a bit. We won't penalize Gleeson for that, and we'll give Mr. Mastriani a little extra time when it's his turn. Mr. Chaffee, good morning. Good morning. I think this really has to start with the checkup. It says that Commerce is in the best position, based on its expertise, based on its knowledge of the individual respondent, to choose what is the appropriate non-cooperating rate. Well, it's not a motion that we're appealing here. It is that Commerce looked at this particular respondent and said, I'm sorry, you gave us misleading information. You told us that you shipped hand trucks without wheels. We found that verification. That's not true. You gave us corporate documents that were false, that also were not true. And then at verification, you can see how TIFA acted. They lied again. They ended up ripping up documents. They ended up also putting documents in their pocket. So, again, Commerce is in the position here. If you look at your precedents here, and we're looking at approximately five. You have DiCecco and you have Gallant, and then there's PAM, KYD, and Washington International. You can also pull in Tai Chen as well. I think it's relevant here. PAM referred to it. In DiCecco and Gallant, they talked about a petition rate, something that had absolutely nothing to do with the respondent. That's why they rejected it. And PAM, KYD, and especially Washington International, which is you affirm per curiam, it's the most relevant case here. In that particular case, Sue Jo, similar to TIFA here, did not cooperate. You argue correctly that more than one rate could be deemed reasonable by the court, so that Commerce could come in with a particular rate or they could come in with a slightly different rate, and we could find that both are reasonable. In your case, you've got a situation where Commerce is arguing that the 145 is reasonable. If we agree with Commerce that it was in their discretion to establish that rate, doesn't that end the inquiry? Yes, I think it does end the inquiry. But they first determined that the 2.27 rate was. Does it end the inquiry as far as your cross-appeal? I'm sorry, I don't know. If we affirm the 145 rate, does that end the inquiry as to the 227 rate, the applicability of the 227 rate? Well, I think that in this particular case, my answer would have to be no. I think you actually... You want us to find that the 227 rate... I think you should find the 227 because... I think you should argue that. We'd like to hear arguments on that. The 227 rate, it comes from the same database as the 145. It's based upon 12% of the database. We're not talking a single transaction here. We're talking an average of multiple transactions that are of the similar product, an identical product, over the period of review. So it's a long... Excuse me, the period of investigation, which was six months. So it's a substantial. It's one out of every 10 products. So here, is it collaborated? Obviously, it's tied to its own data. Is it a reflection of reality? Yes, it is because, again, it's one out of 10 of the subject merchandise shipped to the United States. Is it aboriginal? Absolutely not. Again, this is actually what TIFA did. This is what it shipped to the United States. Is it punitive? No. This, again, it feels like it's a repetitive process here, but they followed the statutory basis and found that in this particular situation, it was valid. But if commerce has now abandoned that rate, rightly or wrongly, if they've abandoned it, how could we review it? They abandoned it because the court rejected it. If you look at... Right, they could have taken a direct appeal from that. That's correct. They could have. But I think in this particular case, we are appealing the Court of International Trade ruling, and in this particular case, we're saying that the Court of International Trade was wrong in rejecting the commerce appeal. But even if it was wrong, we don't have a commerce number right now that says 227, right? Again, I think you do because, again, I think it's there. But, again, we support, we think that the evidence of the record also supports the commerce election. So your argument is that the Court of International Trade erred when it rejected the 227 rate because substantial evidence supported that rate, and that rate, therefore, should apply. That is correct. Even though commerce has now abandoned that and has now chosen a different rate, which arguably is maybe also supported by substantial evidence. I think that there is a continuum of documents that are, excuse me, rates that are supported by substantial evidence. It's similar to what you had in the Washington International case in which commerce had selections. They could have selected the old rate for the respondent, they could have selected a rate for another respondent, or they could have selected the calculated rate which came from data from that particular respondent during the review and during the earlier review. They decided to select the higher rate, I think. Let me ask you this. In your view, what is the scope of commerce's authority in responding to a remand from the Court of International Trade? Is commerce free to ignore a remand and do a recalculation and come up with a new, arguably reasonable assessment based on substantial evidence? Or are they limited to the extent to which they have flexibility in doing anything other than responding to the remand? I think that they have to respond to the remand. I think in this particular case the judge said, I do not believe that the 227 rate is in accordance with law, and therefore they could not look at the 227 rate again and give it to the court. They were limited in that. The judge had cut off the continuum, even though, again, I believe it was supported by substantial evidence, and said, I'm sorry, you can't use that rate. You must be limited to something less than that rate. Because it seems to me that that has an effect on whether commerce is free to just pick whichever reasonable rate, whichever rate they argue is supported by substantial evidence. They've got to follow the instructions of the Court of International Trade on remand. They do have to follow the Court of International Trade instructions on remand, and again, that limited, and the limitation that was placed upon the court was a limitation that I believe was beyond the court's authority in this particular situation. That suggests to me, if I'm right, that suggests to me that if substantial evidence supported the 227 rate and the Court of International Trade was wrong in rejecting that rate, then our decision would be to correct that error and have that 227 rate apply. That is our cross-appeal, Your Honor. All right. Very good. Thank you very much. Mr. Mastriani, we'll give you an extra two minutes. Thank you, Your Honor. As the government just conceded, there's no difference in the methodology between the calculation of the 227 rate and the 145 percent rate. And that rate, the 145.9 percent rate, is not a company-specific rate. It is a model-specific rate. And the government also stated that it was a less-than-ideal methodology that was used to calculate that rate. Now, what was used to corroborate that rate were higher margins of 145.9 percent in the review, which information was rejected by the Court of International Trade as not being reliable or realistic. But remember that that information yielded a weighted average margin of 3.82 percent. So you're going to have maybe 200 percent rates, and you're going to have plus 200 percent rates. And when you put them all together in the computer, it gives you 3.82 percent. So let's keep that in mind, because what happened here was the corroboration, in quotes, that was done, doesn't comport with the statute, and it doesn't comport with precedent. And this Court said in DiCecco, and I think it bears referring to, when all facts point to a rate well before the AFA rate that's assigned, in this case 145.9 percent, to allow imposition of a higher rate renders the corroboration requirement of Section 1677EC meaningless. By requiring corroboration of adverse rates, Congress intended that such rates should be reasonable and have some basis in reality. DiCecco also said that Commerce must estimate a reasonably accurate rate. That wasn't done here. And also what wasn't done here was that Commerce did not follow this Court's presumption, and the presumption that Commerce has followed before, and it's recognized in KYD that this is a very important, valid presumption, and that presumption is that an exporter's prior margin continues to be valid if the exporter fails to cooperate in the subsequent proceeding. That is exactly the case here. That rate is 26.49 percent. Nothing more, nothing less. All that type of ask, since that's the rate that was selected, is that it be used as a baseline, and then added to it what Judge Rayner referred to, that's the delta, you add a deterrent, an increase as a deterrent to induce compliance and cooperation in the future. That's what the statute requires, that's what Commerce wanted, that's what this Court has required, and that's what Commerce has to do. Its discretion is conceded, but the discretion is not unbounded, and here it was abused at best. Do you agree with Mr. Jaffe on the cross-appeal that the issue of the propriety of the 227 rate is properly before us? It is before you by Gleason, because it didn't have the opportunity at the time to appeal. One could argue because there was a remand back to the Department of Commerce, so Gleason then waited. But Commerce ultimately abandoned that rate and chose an alternative rate rather than trying to continue to justify that rate. Does that change things? Well, I think you would trample on the rights of the petitioners, the domestic industry, to be able to appeal. If they couldn't appeal at that time, they should be allowed to appeal at the time that an appeal becomes ripe, which is when the CIT issues a final, okay, it's over, here's a decision, and anybody agreed by it can now appeal it. And obviously TIFA was agreed by it, and Gleason feels it was agreed by it. So let's go back to the issue. You're not arguing that the CIT erred or that Commerce erred by not going back and adopting the margin rate established for your company in the initial investigation. That's not what you're arguing. No, no. So you're saying that's a starting point. That's a starting point, right. So after that, whatever is added on or tacked on is the deterrent rate. That's correct. So what is it about the record order, the decision of the CIT, that lacks substantial evidence that the delta that we're looking at is not an adequate deterrent or it's too much deterrent? Well, first of all, the CIT didn't use that methodology. It didn't make that statement. Commerce did. Well, I don't think Commerce did because it didn't start with 26.49%. Yeah, but they can't pull a rate out of the air. I mean, they have to establish it somehow. So they went back and they got verifiable data that had been submitted in the initial investigation upon which you rely on that 26%. They used that as a starting point, But, Your Honor, Commerce did not take 26.49% and then add the 120-some-odd percent. What it did was it took three line items out of 96 and said, okay, this is 36% by value and we're going to use this. So there was no addition as a deterrent. And I would just say this. But going to Judge Reyna's point, certainly Commerce was well aware of the earlier 26% calculation and they were aware of that as a base, if you will, even though they didn't expressly set forth a particular methodology like that. Oh, absolutely. What Commerce said was very telling. Commerce said that using that rate or any of the other company-specific rates would reward TIFA. So it was punitive. The whole thinking and the drive behind that was to be punitive. And you agree with that, right? I mean, if they used a 26%, that would not have been punitive. No, no. And we agree that to that we would expect, of course, that under the statute and law that to that rate would be added an addition. It would not be quintupled, however. Quintupling it would be, I mean, probably violate the Eighth Amendment. It would be cruel and unusual punishment. So unless you have any more questions, I have nothing to add. Other than just to say, by the way, that Todd Chen was not a corroboration case. And also in the Washington insurance case, the AFA rate that was selected was below the highest rate calculated in proceedings that dealt with that pumping. Thank you. Thank you very much.