[00:00:00] Speaker 02:
We have six cases on our calendar this morning.

[00:00:04] Speaker 02:
Two of them are submitted on the briefs and will therefore not be argued.

[00:00:08] Speaker 02:
Our first case is Stupp Corporation et al.

[00:00:14] Speaker 02:
versus the United States in Hyundai, Steele, 2023, 1663.

[00:00:25] Speaker 02:
Mr. Winton.

[00:00:34] Speaker 03:
Good morning.

[00:00:35] Speaker 03:
Good morning.

[00:00:35] Speaker 03:
May it please the court, I'm Jeffrey Winton of Winton and Chapman, here today on behalf of Defendant Appellant, say, of Steel Corporation.

[00:00:43] Speaker 03:
This is our second time before the court on this case.

[00:00:46] Speaker 03:
In the first go-around, the court held that when commerce applies a mathematical tool like Cohen's D, it has to apply it in a manner consistent with the tool's assumptions.

[00:00:57] Speaker 03:
And the court said if it doesn't, then that might transform what might be

[00:01:02] Speaker 03:
conservative cut off into a meaningless comparator.

[00:01:06] Speaker 03:
And so the court remanded this case back to Commerce to explain why it was appropriate, either why the assumptions underlying Cohen-Stewart met in this case, or why Commerce could ignore those assumptions.

[00:01:20] Speaker 03:
In its remand determination, Commerce conducted a limited review, only addressing the articles that the court had cited.

[00:01:28] Speaker 03:
The Commerce asked us to submit those articles on the record.

[00:01:31] Speaker 03:
They did not give us an opportunity to expand the record.

[00:01:34] Speaker 03:
They did give the other parties an opportunity to rebut the articles we had submitted.

[00:01:40] Speaker 03:
And Commerce then reviewed the articles and concluded that none of those articles really addressed what Commerce was doing in its differential pricing analysis.

[00:01:50] Speaker 03:
The articles were addressing the use of Cohen's D in normal statistical practice

[00:01:55] Speaker 03:
in analyzing samples and commerce said we're not analyzing samples so we can disregard what's in those articles.

[00:02:03] Speaker 03:
The conclusion is that commerce's use of Cohen's theory is not consistent with normal statistical practice.

[00:02:10] Speaker 03:
Any doubt you might have

[00:02:12] Speaker 03:
would be removed by a report that Professor Hedge submitted subsequently in another case.

[00:02:19] Speaker 03:
And I know it's not on the record here.

[00:02:22] Speaker 03:
But if we ever were to come back again, he submitted in all the cases now.

[00:02:26] Speaker 03:
And so if this court is going to say only commerce could decide based only on the

[00:02:32] Speaker 03:
articles that were cited, you will then have a subsequent case where you'll have to deal with Professor Hedges' report.

[00:02:38] Speaker 03:
Professor Hedges is a big deal in effect size.

[00:02:41] Speaker 03:
His Hedges G is as famous as Cohen's D. It's very similar.

[00:02:46] Speaker 03:
It uses a different denominator.

[00:02:48] Speaker 03:
And in fact, Professor Hedges was so upset by what Commerce did,

[00:02:51] Speaker 03:
He subsequently wrote an article which he submitted in the Journal of Educational and Psychological Measurement explaining why the use of Cullen's deal when the data was not normal or did not have equal variances.

[00:03:05] Speaker 04:
Can I ask you, it seems to me that commerce has a lot of discretion here.

[00:03:10] Speaker 04:
They have to act in a reasonable

[00:03:12] Speaker 04:
And I recognize in the remand, we said some things about what we were looking for in terms of how they were exercising their discretion.

[00:03:21] Speaker 04:
But I think the question is, how much discretion do they have left after the remand?

[00:03:27] Speaker 04:
So for example, you write at page 44 of your brief that before commerce can rely on a rule of thumb, it must demonstrate that the phenomenon from which the rule of thumb was derived has some rational

[00:03:39] Speaker 04:
relationship to the situation for which commerce is applying it.

[00:03:42] Speaker 04:
That makes common sense to me, but isn't that commerce's call to make?

[00:03:47] Speaker 04:
What authority would you cite that says commerce has to, if it uses the rule of thumb, use it in a way consistent with how the rule of thumb was devised?

[00:03:57] Speaker 03:
Well, it's not that it has to use a proof.

[00:03:59] Speaker 03:
So there are two issues here.

[00:04:01] Speaker 03:
One is commerce's position was, we're using this normal statistical test, and therefore it's reasonable because we're just following statistical practice.

[00:04:09] Speaker 03:
And that's not true.

[00:04:10] Speaker 05:
The second is commerce says, and I think this quote... You mean that's not true because the prerequisites for application are not met.

[00:04:17] Speaker 03:
Right, exactly.

[00:04:18] Speaker 04:
So let me just make sure I understand from that what you're saying is,

[00:04:22] Speaker 04:
Maybe there's not a blanket rule that commerce has to use a statistical test in the way that the statistician who developed it used it or developed it.

[00:04:33] Speaker 04:
But if they say that's what they're doing, then they have to do what they say they're doing.

[00:04:38] Speaker 03:
Well, there are two issues.

[00:04:39] Speaker 03:
And I think the trade court said, OK, I don't think they're doing what Professor Cohen said to do.

[00:04:46] Speaker 03:
But it looks reasonable to me.

[00:04:49] Speaker 03:
And the issue is not.

[00:04:51] Speaker 03:
They're not doing what Professor Cohen said.

[00:04:53] Speaker 03:
So the question is, is this reasonable?

[00:04:56] Speaker 03:
Is it a reasonable methodology?

[00:04:58] Speaker 03:
And commerce has discretion, but it still has to be reasonable.

[00:05:01] Speaker 03:
They can't use something that's not reasonable.

[00:05:04] Speaker 03:
What we would argue is, when the conditions that are specified by Professor Cohen are not satisfied, the Cohen's D test basically gives you a random result.

[00:05:16] Speaker 03:
Um, Judge Kelly at the, at the trade court seemed impressed that commerce submitted statistics.

[00:05:21] Speaker 03:
Well, this only affects 20% of the cases.

[00:05:25] Speaker 03:
And so we shouldn't be too worried about it.

[00:05:27] Speaker 03:
But the issue is, is it getting the right 20%?

[00:05:29] Speaker 03:
If I told you that 20% of the motorists on I-95 were speeding, and so the police decided let's pull one over, one out of five motorists over at random, you say that's not right because you pull one out of five motorists over at random.

[00:05:45] Speaker 03:
You're not getting, even one out of five of those will be someone who is speeding.

[00:05:48] Speaker 03:
Four out of five won't be people who are speeding.

[00:05:51] Speaker 03:
You're getting the wrong, you're giving tickets to the wrong people.

[00:05:54] Speaker 03:
And that's the concern here.

[00:05:56] Speaker 03:
When you're not using Cohen's D properly, you get basically random results, which the court recognized in its previous decision when it said that you're getting meaningless comparator.

[00:06:08] Speaker 03:
I'm not sure which it was.

[00:06:11] Speaker 03:
And so you're left with, is this something that's reasonable?

[00:06:16] Speaker 03:
Now, commerce would tell you it's reasonable because it's based on real world observations.

[00:06:22] Speaker 03:
It's based on the heights of teenage girls.

[00:06:25] Speaker 04:
And again, should I understand your argument is they don't necessarily have to defend what they're doing based on its comparability to real world data.

[00:06:37] Speaker 04:
But if they choose to give that as the basis for why what they're doing is reasonable, then that ought to be true.

[00:06:43] Speaker 04:
Is that your position?

[00:06:44] Speaker 03:
Yes.

[00:06:44] Speaker 03:
Yes.

[00:06:45] Speaker 03:
There's two positions.

[00:06:46] Speaker 03:
One is, if they were using a statistical test in the proper way, if the conditions for using Coen's D were met in this case, I would say, I don't have an argument.

[00:06:56] Speaker 03:
They used Coen's D properly.

[00:06:58] Speaker 03:
But they didn't.

[00:06:59] Speaker 03:
The second is, OK, is it still reasonable for them to do this?

[00:07:04] Speaker 03:
You know, is it reasonable to say a measure based on the heights of teenage girls and the IQs of different students tells you something about whether there's a pattern in our prices?

[00:07:14] Speaker 03:
And of course, the commerce has defined pattern, and we agree with this, as something that wouldn't arise by chance.

[00:07:21] Speaker 03:
So the question is, OK, can you use the heights of teenage girls or the IQs?

[00:07:26] Speaker 03:
The problem is both the heights of girls, heights of everybody, and IQs are normally distributed data.

[00:07:32] Speaker 03:
Actually, it satisfies the requirements for using Cohen's D. So saying we're relying on real world data doesn't move you anywhere from, but you have to show the requirements for using Cohen's D are satisfied.

[00:07:44] Speaker 02:
Counsel, I think there's another case relating to Cohen's D pending before this court.

[00:07:50] Speaker 02:
I think it might have been argued in January.

[00:07:52] Speaker 03:
Yeah, Marmon.

[00:07:53] Speaker 02:
Marmon, yes.

[00:07:56] Speaker 02:
Relevant is that case to this case, or because they deal with different facts, is it not that relevant?

[00:08:05] Speaker 03:
I would say I think it's relevant.

[00:08:11] Speaker 03:
I was listening to the argument the other day.

[00:08:13] Speaker 02:
I would think.

[00:08:15] Speaker 03:
There's a lot of relevance.

[00:08:17] Speaker 03:
I don't 100% agree with the appellants in that case, but I think largely I agree with the appellants in that case.

[00:08:24] Speaker 05:
How do you differ from that?

[00:08:25] Speaker 03:
I think the concept of effect size, if applied properly, does what commerce wants to do.

[00:08:44] Speaker 03:
The issue here is you see a difference in our prices.

[00:08:48] Speaker 03:
is that difference in prices something that just happened because you know at random prices vary or is it something that's a pattern something that didn't happen by chance and effect size is a measure can be used as a measure of that but you have to use effect size properly I'm not sure that you know the Marmon appellates would agree with me on that but I think largely they do they focus on things like overlap and such which I think is

[00:09:17] Speaker 03:
Relevant but not critical here.

[00:09:22] Speaker 04:
We remanded, as you pointed out when you started, for a further explanation.

[00:09:26] Speaker 04:
What's your best example of how, in your view, commerce failed to give us the explanation we were looking for?

[00:09:34] Speaker 03:
Well, it's a long decision, but when you go through it, it's explaining why, and I should have this height in front of me and I apologize, but they go through and they say, you know, these are addressing a different situation than what we're addressing.

[00:09:54] Speaker 03:
And I agree, you know, commerce is using this in a way that statisticians don't use co-instate.

[00:10:02] Speaker 03:
That is clear.

[00:10:03] Speaker 05:
They focus very extensively on the distinction that they draw, Commerce, between use of a sample versus use of the full population.

[00:10:13] Speaker 05:
Right.

[00:10:14] Speaker 05:
Now your answer to that, I take it, is that Coins D and the prerequisites for the application of Coins D apply whether you're talking about a sample or a full population.

[00:10:25] Speaker 03:
Yes, I think that's right.

[00:10:28] Speaker 05:
But if you can expand on that.

[00:10:29] Speaker 03:
Sure.

[00:10:30] Speaker 05:
That's such an important fact.

[00:10:33] Speaker 03:
And I apologize, because this requires me to get into little weeds of statistics.

[00:10:36] Speaker 03:
And I know none of us are statisticians.

[00:10:39] Speaker 03:
A normal distribution is defined entirely by two items, the mean, the average point, and the standard deviation.

[00:10:48] Speaker 03:
And Cohen said the standard deviation should be the same.

[00:10:53] Speaker 03:
So you have two distributions that look exactly the same.

[00:10:57] Speaker 03:
The only difference between them is the average, the mean.

[00:11:00] Speaker 03:
If you think of heights of 13-year-old girls, the average height of a 13-year-old girl is five foot two.

[00:11:06] Speaker 03:
The average height of a 17-year-old girl is five foot four.

[00:11:11] Speaker 03:
There are two curves, and they look exactly the same, but one is shifted to the right by two inches.

[00:11:16] Speaker 03:
That's what Cohen's D is, you know, the normal distribution, and that you could have the entire population, because the features of a normal distribution don't depend on whether it's a sample or a population.

[00:11:29] Speaker 03:
It depends on that it's a normal distribution.

[00:11:31] Speaker 03:
That's what it means.

[00:11:33] Speaker 03:
And you can say, you know, Professor Cohen actually says, look, if it's two inches, we could just say two inches.

[00:11:38] Speaker 03:
Everybody understands what two inches are.

[00:11:41] Speaker 03:
But in other contexts, we don't know what the difference means.

[00:11:45] Speaker 03:
And so we're going to express that 2 inches, not as inches, but in terms of standard deviations.

[00:11:50] Speaker 03:
And so if the standard deviation of our 13-year-old girls and our 17-year-old girls is 2.5 inches, and the difference is 2 inches, we can calculate a D of 0.8.

[00:12:00] Speaker 03:
But that only makes it, otherwise, the two curves are exactly the same.

[00:12:05] Speaker 03:
One is just shifted over by 2 inches or by 0.8 standard deviations.

[00:12:10] Speaker 03:
And you can describe the difference between those distributions simply by saying that.

[00:12:14] Speaker 03:
One is 0.8 standard deviations shifted from the other.

[00:12:18] Speaker 03:
If you don't know that the data is normal, what if it's like bimodal, where it's got two peaks?

[00:12:24] Speaker 03:
Or what if it's just a straight distribution?

[00:12:27] Speaker 03:
There's an equal number of cells at every price from the minimum to maximum, so it looks like this.

[00:12:33] Speaker 03:
Well, then you can't, or what if it's a different standard deviation?

[00:12:36] Speaker 03:
One looks like this, and one looks like this, or something

[00:12:39] Speaker 03:
different.

[00:12:40] Speaker 03:
Then you can't describe that difference simply by saying, oh, all we care about is the difference in the average.

[00:12:46] Speaker 03:
Right?

[00:12:47] Speaker 03:
You can't say, okay, yes, there's a two-inch difference in the average, but what does that mean?

[00:12:52] Speaker 03:
It's not the same as if the two curves are the same.

[00:12:55] Speaker 03:
What if the 17-year-old girls are all either five foot eight or five feet tall and nobody's in between?

[00:13:04] Speaker 03:
But this 13-year-old girl is a normal, a nice curve.

[00:13:10] Speaker 03:
You can't just say the average tells us what's different.

[00:13:13] Speaker 03:
You have to look into it in more detail.

[00:13:14] Speaker 03:
There are other things going on there.

[00:13:16] Speaker 03:
And that's basic statistics.

[00:13:19] Speaker 03:
There's a whole branch of statistics to deal with situations where you don't have a normal distribution.

[00:13:26] Speaker 03:
But it has nothing to do with samples and populations.

[00:13:29] Speaker 03:
It has to do with whether the data is normally

[00:13:36] Speaker 02:
Oh, thanks.

[00:13:38] Speaker 02:
You can save it or continue.

[00:13:40] Speaker 03:
I always feel that I should leave you wanting more, so let me save two minutes and allow the other side to speak.

[00:13:49] Speaker 02:
Thank you.

[00:13:51] Speaker 02:
Mr. Kepora, you're going to take 12 minutes.

[00:14:11] Speaker 01:
Good morning, Your Honors, and may it please the Court.

[00:14:13] Speaker 01:
Robert Capura on behalf of the United States.

[00:14:17] Speaker 01:
Following this Court's partial remand in its prior decision, Commerce addressed the concerns raised by the Court in its earlier decision.

[00:14:24] Speaker 01:
Specifically, Commerce addressed the points regarding comparison groups that were not normally distributed, having small number of data points, or having disparate variances.

[00:14:35] Speaker 01:
Now we recognize the Court's concerns on these issues, but

[00:14:40] Speaker 01:
as commerce demonstrated in its remand determination, the Cohen's d equation is a robust mathematical tool used for the purpose of telling the difference between two groups.

[00:14:53] Speaker 01:
And I want to point out something here at the top, which is that the Cohen's d itself is one part of the differential pricing analysis.

[00:15:04] Speaker 01:
The Cohen's d coefficient does not tell you whether or not there's a pattern.

[00:15:09] Speaker 01:
It's just the first step in that process.

[00:15:12] Speaker 01:
Before commerce will determine that there's a pattern, they look to the ratio test, which this court already addressed in its prior decisions.

[00:15:21] Speaker 01:
So again, as already has been addressed here, commerce walked through the different academic literature that was highlighted in the court's previous decision.

[00:15:33] Speaker 01:
And it's already been stated.

[00:15:35] Speaker 01:
that the response to that is the concerns that are raised in those various papers deal with a statistical difference, not a practical difference.

[00:15:46] Speaker 01:
And the point here is that the academic literature on this is silent.

[00:15:51] Speaker 01:
as to the question of, can you use Coen's D on a population that is not normally distributed or has these other issues?

[00:16:00] Speaker 05:
Let me ask you a question about the point you made just a moment ago about saying, well, there are other tests to the ratio test and so forth that are used here in addition to Coen's D. If we were to say that Coen's D is simply

[00:16:19] Speaker 05:
inapplicable to the circumstances such as in this case.

[00:16:24] Speaker 05:
Do you think that the judgment in this case from commerce could be sustained without any regard for co-instate?

[00:16:36] Speaker 01:
I'm not sure I'm following.

[00:16:38] Speaker 01:
When you say that Cohen's D is inapplicable, I'm not sure I'm following.

[00:16:40] Speaker 05:
No, that's right.

[00:16:41] Speaker 05:
In other words, we say, well, simply under circumstances such as this, you cannot rely on Cohen's D. Could you then say, ah, but we have two other tests which demonstrate to us, to our satisfaction, that there is dumping in this case?

[00:16:58] Speaker 05:
Now, would that be a position that you think commerce could defensively take in this case?

[00:17:06] Speaker 01:
Possibly, Your Honor, I think it would depend on the contours of the court's decision if it were to go that way.

[00:17:12] Speaker 01:
I think that certainly we always turn back to the statute here, which is directing commerce to do this.

[00:17:19] Speaker 01:
Commerce has an instruction from Congress that they are to look for these patterns and shift to an alternative comparison methodology if they find a pattern.

[00:17:33] Speaker 01:
if the court were to find that, I don't think commerce would move away from a practice of looking for patterns or potentially applying a different methodology.

[00:17:43] Speaker 01:
And as far as the rest of the differential pricing analysis, yes, it's possible to continue using those.

[00:17:52] Speaker 01:
The meaningful difference test, the ratio test, these things have already been affirmed.

[00:17:57] Speaker 01:
If the court takes issue with the Cohen's D itself,

[00:18:02] Speaker 05:
it's possible for commerce to determine another way to show this difference another way based on this record in other words based on what commerce has done so far with respect to the other two tests or based on going back and starting over

[00:18:20] Speaker 01:
The answer to the first question is yes, they can do it on this record, because what we're talking about here is a methodology.

[00:18:25] Speaker 05:
Now again, depending on what the court says about the Cohen's d. Well, let's suppose hypothetically the court says Cohen's d, in circumstances such as this, is simply not reliable enough to be given any credence.

[00:18:41] Speaker 05:
And therefore, you're down to the two remaining tests.

[00:18:44] Speaker 05:
And my question really is,

[00:18:47] Speaker 05:
Can you say, well, those tests independently establish everything we need to find in order to find dumping in this case?

[00:18:55] Speaker 05:
Or would you have to go back and say, well, we'll approach this in another way, or we'll use a different test?

[00:19:02] Speaker 01:
I think you would have to.

[00:19:07] Speaker 01:
come up, they would have to come up with an explanation for how they were showing the difference, right?

[00:19:13] Speaker 01:
Because that's all Cohen Z is doing.

[00:19:16] Speaker 01:
It's just saying, okay, we've got a test group, which is the group we're analyzing for whether or not there's targeted dumping in the control group, which is the rest of the sales, and they run this multiple times as they look through, whether it's regions, purchasers, time periods.

[00:19:33] Speaker 01:
There has to be some methodology for determining whether or not the test and control groups are different.

[00:19:40] Speaker 01:
They're using Cohen's D to do that.

[00:19:43] Speaker 01:
And I think there's one other thing that this leads me to here that's been raised in this case.

[00:19:49] Speaker 01:
It's this idea that somehow Cohen's D is producing false results or false positives or something to that effect.

[00:20:00] Speaker 01:
Again, a lot of those comments come from places in the academic literature where they're talking about these statistical results.

[00:20:11] Speaker 01:
If you have a sample that's not reflective of its population, then yes, comparing samples is meaningless.

[00:20:18] Speaker 01:
You're not going to get a result that tells you anything about the populations.

[00:20:22] Speaker 01:
What commerce has said is we don't need to worry about that.

[00:20:26] Speaker 01:
because we have the full population.

[00:20:29] Speaker 01:
And I'd also like to use as an example here the court's own hypothetical from its previous decision where it raised a concern about essentially extreme results coming from the Cohen's D test, right, where you have a test group and a control group where the sales data are clustered very closely together and the values between the two groups are close.

[00:20:55] Speaker 01:
Commerce ran through this hypothetical in its remand determination, and they pushed it to the absolute extremes, where you have a test group, all sales are at $100, and the test group and the control group is all at $101.

[00:21:10] Speaker 01:
So you have zero variance.

[00:21:13] Speaker 01:
In that instance, the Cohen's D coefficient goes to infinite, and that sounds like an extreme result, but it's not a false result.

[00:21:21] Speaker 01:
because all that that's telling us is it's a mathematical determination that these two groups are different.

[00:21:29] Speaker 01:
There's not a single price in those two groups that is different.

[00:21:33] Speaker 01:
So that's all the test is doing, and then we go through...

[00:21:39] Speaker 01:
Excuse me.

[00:21:39] Speaker 01:
Yes, not a single price.

[00:21:40] Speaker 01:
That's the same.

[00:21:41] Speaker 01:
I'm sorry.

[00:21:41] Speaker 01:
I misspoke.

[00:21:43] Speaker 01:
Now, they're close together, right?

[00:21:45] Speaker 01:
But there's no overlap between them.

[00:21:48] Speaker 01:
And so then we move on to these other tests.

[00:21:51] Speaker 01:
And that's where commerce explained that in that example,

[00:21:55] Speaker 01:
you could not shift to an alternative methodology because the meaningful difference test would show that you wouldn't cross the de minimis threshold, they're too close together, the test and control groups.

[00:22:07] Speaker 01:
So again, all the Cohen's D test is doing is saying whether or not these two things are different and then we proceed to the rest of analysis to determine whether or not that matters.

[00:22:19] Speaker 04:
Am I right, the rest of analysis, the second step we call the ratio test?

[00:22:23] Speaker 01:
That's correct.

[00:22:23] Speaker 04:
isn't that inherently tied to Cohen's D because it's a measure of how many times you pass or how much data passes Cohen's D?

[00:22:35] Speaker 01:
It is tied to Cohen's D, Your Honor, but there's a key point that I want to make here that I think there may have been some confusion about earlier.

[00:22:42] Speaker 01:
The ratio test is based on value of U.S.

[00:22:45] Speaker 01:
sales, and the reason why that's important, and the trial court highlighted this, is that

[00:22:52] Speaker 01:
let's say you have a hundred million dollars in U.S.

[00:22:54] Speaker 01:
sales and you run the Cohen's D test on two million dollars worth of sales.

[00:22:59] Speaker 01:
It doesn't matter what happens in the Cohen's D test because in order to even consider an alternative methodology, you have to have more than 33 percent of total sales crossing that threshold.

[00:23:13] Speaker 04:
So 33 million in your example?

[00:23:15] Speaker 04:
At least, yes.

[00:23:16] Speaker 04:
That is helpful, but going back to where Judge Bryson was exploring with you

[00:23:21] Speaker 04:
What would we do if Cohen's D were not available here?

[00:23:26] Speaker 04:
Could we really do the ratio test?

[00:23:28] Speaker 04:
If we threw out Cohen's D, how do we know what percentage of sales pass the Cohen's D test if we're not using Cohen's D?

[00:23:38] Speaker 04:
That is, wouldn't the second step in the DPA also fall?

[00:23:42] Speaker 01:
It wouldn't fall in the sense that it is no longer valid, because I think what you could do, in answer to Judge Bryson's question, is you could develop a different methodology for determining whether or not the groups are different.

[00:24:01] Speaker 01:
The, you know, commerce is using this as a consistent mathematical methodology.

[00:24:06] Speaker 01:
They'll apply it in every time they run the differential pricing because, again, we can't look at groups of sales and just say, well, these things look like they're pretty close or there's a lot of overlap.

[00:24:20] Speaker 04:
The things we would be comparing at step two of a modified DPA couldn't be the number of sales that pass Cohen's D if Cohen's D is not being used at step one.

[00:24:31] Speaker 01:
Right.

[00:24:31] Speaker 01:
It could be sales that pass a different test.

[00:24:35] Speaker 04:
Whatever that test were to be.

[00:24:36] Speaker 01:
That's correct.

[00:24:38] Speaker 04:
I just want to understand what Commerce thinks is the scope of what it can do within the unit that would constitute reasonableness.

[00:24:49] Speaker 04:
If we were to conclude that the way you're using Cohen's D, which is a statistical test at the end of the day, is not statistically sound,

[00:25:01] Speaker 04:
does commerce think that its discretion allows it to use a statistical test in a non-statistically sound manner?

[00:25:09] Speaker 04:
Or would you agree that would be unreasonable?

[00:25:16] Speaker 01:
Well, I think if the court were to hold that, then yes.

[00:25:19] Speaker 01:
Again, I think the thing here, though, is that

[00:25:24] Speaker 01:
You know, Commerce has explained that the Cohen's D is being used for this specific purpose, and despite all the discussion about the academic literature, there is nothing there on this record that says that it's inappropriate to use Cohen in this manner.

[00:25:45] Speaker 01:
There's nothing that says that explicitly that it's okay, but what we're saying is the academic literature is silent.

[00:25:53] Speaker 01:
And so commerce's explanation as to how and why it's using this particular equation crosses that reasonableness threshold.

[00:26:03] Speaker 04:
If we are persuaded that on this record what you're doing is not statistically sound, can we nonetheless say, well, you have authority to do something that's not statistically sound?

[00:26:18] Speaker 04:
Yes, Your Honor, I would say that... Are you asking us to do that if we were to reach that conclusion?

[00:26:25] Speaker 01:
If you were to reach that conclusion, yes, Your Honor, because the equation, like I said, it's being used for this singular purpose, and every time we've gone through every different hypothetical that's presented,

[00:26:40] Speaker 01:
it works out.

[00:26:41] Speaker 01:
There hasn't been a single thing that's been presented.

[00:26:44] Speaker 01:
Even the hypothetical that Appellant's presented in their brief doesn't show that the Coen's D is not robust enough to deal with these issues and determine whether or not there's a difference between the two groups.

[00:27:02] Speaker 01:
Again, I understand

[00:27:04] Speaker 01:
It might sound strange to say that, but really, like I said, what commerce is using this particular equation for is the singular purpose.

[00:27:14] Speaker 02:
Counsel, I think you need to yield to Mr. Garrett.

[00:27:18] Speaker 01:
Yes, Your Honor.

[00:27:19] Speaker 01:
Thank you.

[00:27:34] Speaker 00:
Good morning.

[00:27:34] Speaker 00:
May it please the court, I am Jeff Garrish of Chagrin Associates on behalf of Wells Buntubior.

[00:27:39] Speaker 00:
The core issue here is whether commerce's methodology is reasonable, whether it provides a reasonable means of effectuating the statutory text and purpose of determining whether there are significant price differences among customers, regions, or time periods in order to unmask targeted dumping.

[00:27:58] Speaker 00:
The methodology that Commerce uses here is an adaptation or variation of the Cohen's D test.

[00:28:04] Speaker 00:
They're not using the Cohen's D test as Professor Cohen and others have used it for purposes of conducting behavioral research experiments.

[00:28:16] Speaker 00:
Congress at some point concluded that they would attach the coincident name to their calculations, but I think it's created a lot of confusion and, of course, a lot of litigation.

[00:28:26] Speaker 00:
What Congress has actually done is to adapt the test for its own purposes as part of an integrated methodology for determining whether targeted dumping is occurring.

[00:28:36] Speaker 00:
The test they're applying is not unlike the other tests they've used for targeted dumping in that it uses differences in average prices and standard deviations.

[00:28:47] Speaker 00:
Commerce has also adopted this test because it provides effect size thresholds as operational definitions derived from real world observations that are easy to understand and that are not based on statistical analysis.

[00:29:03] Speaker 00:
So what has happened here is that Commerce has decided in its expert judgment

[00:29:08] Speaker 00:
to use the Cohen's D formula and the most conservative effect size threshold corresponding to a grossly perceptible difference to identify whether price differences are significant for its own purposes in determining whether targeted dumping is occurring.

[00:29:24] Speaker 00:
In applying the test in this way, commerce is not beholden to conform to what the statistical literature says about the correct application for conducting behavioral science experiments.

[00:29:37] Speaker 04:
As you suggested, it seems to me commerce decided to call it a version of Cohen's D, defends its use of it based on Cohen and the work that Cohen did.

[00:29:48] Speaker 04:
And therefore, it seems only fair that we would assess how what commerce is doing compares to what the statistical literature would say you should do with Cohen's d. Isn't that a fair analysis of where we are?

[00:30:03] Speaker 00:
Well, I do think it has created confusion at the fact that they've used the Cohen's D name.

[00:30:07] Speaker 00:
And I think they've done that for certain reasons.

[00:30:09] Speaker 00:
But again, it's a test that's like the tests they've used in the past.

[00:30:15] Speaker 00:
It uses standard deviations in average price differences.

[00:30:18] Speaker 00:
And it's akin to the rules of thumb they have used in other contexts.

[00:30:25] Speaker 00:
And of course, this court has upheld those rules of thumb.

[00:30:28] Speaker 00:
in the ratio test, the 33% and 66% rules of thumb, and then the 25% rule of thumb that Commerce uses for the meaningful difference analysis.

[00:30:42] Speaker 00:
And this Court has upheld all those rules of thumb.

[00:30:45] Speaker 00:
And these are all reasonable cutoffs

[00:30:49] Speaker 00:
for Commerce to use, and that they developed using their expertise to do what the statute directs them to do, which is to address.

[00:30:59] Speaker 02:
Do you have a final summary thought?

[00:31:02] Speaker 00:
Well, again, I don't think Commerce is beholden to conform to what the statistical literature says on this.

[00:31:07] Speaker 00:
And the statistical literature cannot be the barometer for what is reasonable under the statute.

[00:31:14] Speaker 00:
What Commerce has done under the statute is reasonable, and I think this Court should uphold it.

[00:31:20] Speaker 02:
Mr. Winton has some rebuttal time.

[00:31:25] Speaker 02:
We'll give you two minutes.

[00:31:28] Speaker 03:
Good to see you again.

[00:31:32] Speaker 03:
In our briefs to the court below, and we reference this in our reply brief at page 18, we looked at the effect of exchange rates on the DPA analysis.

[00:31:43] Speaker 03:
And exchange rates are not under our control and, in fact,

[00:31:46] Speaker 03:
academic literature says exchange rates are random.

[00:31:49] Speaker 03:
We did the calculations and it turns out that just the variations in exchange rates would lead to a finding that we passed Cohen's D, that we had more than 66% over the ratio test and that, you know, I mean that difference alone, this is

[00:32:06] Speaker 03:
What the evidence shows and the academic literature shows, when the requirements set forth by Professor Cohen are not satisfied, the Cohen's D test gives you a number that's random.

[00:32:18] Speaker 03:
You know, it could be high, it could be low, it's not a meaningful number.

[00:32:22] Speaker 03:
Counting up random numbers and saying, well, 60, you know, in this case, I roll the dice and I get, you know,

[00:32:30] Speaker 03:
this many and one-third of the value when I rolled the dice was below so it's less than one-third passed so we use average average or more than two-thirds passed so I use average transaction but counting up you're counting up random results and that's what makes this unreasonable.

[00:32:49] Speaker 03:
Commerce hasn't shown

[00:32:51] Speaker 03:
that the results of the DPA, if Cohen's D, when the conditions set forth by Professor Cohen are not satisfied, he hasn't shown that those are meaningful results.

[00:33:01] Speaker 03:
And they haven't shown that 0.8 is a meaningful comparison number when the criteria required by Professor Cohen are not satisfied.

[00:33:11] Speaker 03:
And it's important to note, Professor Cohen didn't say, I have one effect size measure.

[00:33:16] Speaker 03:
He actually had eight.

[00:33:18] Speaker 03:
He said, for use in different circumstances,

[00:33:21] Speaker 03:
The D statistic is used in conjunction with a statistical test called a t-test, which requires normal distribution, equal variance, sufficient number of data points.

[00:33:32] Speaker 03:
Other situations, he has other measures of effect size.

[00:33:35] Speaker 03:
There's nothing that says D is the end-all and be-all of effect size and the only number that you can use.

[00:33:41] Speaker 03:
And in this case, the evidence shows

[00:33:44] Speaker 03:
That the results are random they're affected simply random variations can lead to positive Results and can affect the dumping calculation and that makes it unreasonable Thank you very much.

[00:33:55] Speaker 02:
Thank you to both council the cases submitted and I can assure you we will not be rolling dice