[00:00:00] Speaker 00: 16-1059, Association of Oil Pipes, Oil Pipelines, Petitioner versus the Federal Energy Regulation Commission, et al. [00:00:12] Speaker 00: Mr. Reed for the petitioner, Mrs. Chu for the respondent. [00:01:08] Speaker 02: Good morning. [00:01:09] Speaker 02: Good morning, Your Honor. [00:01:10] Speaker 02: Good morning to the Court. [00:01:11] Speaker 02: I'm Steve Reed, and I'm appearing on behalf of the Petitioner of the Association of Oil Pipelines. [00:01:17] Speaker 02: This is the fourth time this Court has had occasion to address the FERC's establishment and review of its index methodology for setting oil pipeline rates. [00:01:26] Speaker 02: Prior to the order under review now, the commission, with some guidance from this court, has established a stable set of principles for setting the index that has been referred to below as the con methodology. [00:01:39] Speaker 02: In the 2015 order at issue here, Burke departed from those principles in two significant ways, without either acknowledging its significant change of policy or providing a reasoned explanation for those changes. [00:01:53] Speaker 02: The two key errors here are, first, the exclusion of the middle 80% of the data on industry cost changes in calculating the index. [00:02:02] Speaker 02: And the second is the redefinition by the commission of the costs that the index is attempting to measure from actual industry cost changes to so-called recoverable cost of service data. [00:02:15] Speaker 02: Together, these two errors resulted in about a 50% lower adder to the index than one calculated under the neutral principles the Commission had previously espoused. [00:02:28] Speaker 06: I'd like to start by talking about- Well, you started off by saying that I thought that FERC didn't acknowledge the changes. [00:02:35] Speaker 02: Yes, Your Honor. [00:02:35] Speaker 06: I thought FERC did acknowledge the changes in its order. [00:02:38] Speaker 02: We don't believe so, Your Honor. [00:02:39] Speaker 02: And let me talk about why we say that. [00:02:43] Speaker 02: What FERC said with respect to the middle 50, middle 80 was that they had always intended to measure normal costs, normal cost changes in the industry. [00:02:54] Speaker 02: And that is what they said when they first set up the index in 1995. [00:02:57] Speaker 02: They talked about that. [00:02:59] Speaker 02: But subsequent to that 1995 decision in which they talk about the goal being to measure normal cost changes, quote unquote, the commission did adopt an index that was based on the middle 50 and the middle 80. [00:03:12] Speaker 02: They used it in 2003 on remand from this court. [00:03:15] Speaker 02: They used it again in 2006 in their second index review. [00:03:19] Speaker 02: And in 2006, they specifically defended [00:03:21] Speaker 02: the middle 50, middle 80 model as an effective way to, quote unquote, exclude outliers, statistical outliers. [00:03:29] Speaker 05: But not as an exclusively effective way. [00:03:33] Speaker 02: That's correct, Your Honor. [00:03:34] Speaker 05: They've been inconsistent and they've justified it on each occasion. [00:03:38] Speaker 05: They've never said that it has to be both. [00:03:43] Speaker 02: It's correct, Your Honor, they've never said it has to be both, but they did say, very explicitly in 2006, that the middle 50, middle 80, was an effective way of excluding the Statistical Appliances. [00:03:54] Speaker 05: But that's not to say the opposite. [00:03:55] Speaker 02: Well, I understand, Your Honor, but let's follow the history through. [00:03:58] Speaker 02: Now we come to the 2015 order. [00:04:02] Speaker 02: In 2010, the Commission toyed with [00:04:04] Speaker 02: middle 50, middle 80. [00:04:06] Speaker 02: They said, well, we've looked at the middle 50 and middle 80 in the past. [00:04:11] Speaker 02: We've been a little inconsistent. [00:04:12] Speaker 02: But in this case, we don't really have to consider that. [00:04:15] Speaker 02: It's unnecessary to use the middle 80 because the middle 50 gives us a very robust statistical sample with 76% of the industry barrel miles. [00:04:25] Speaker 02: 76 is pretty close to 80. [00:04:29] Speaker 02: So the decision in 2010 was the commission simply saying, [00:04:34] Speaker 02: We can decline to use the middle 80, where we recognize that the middle 50 is providing us a robust sample. [00:04:41] Speaker 02: In 2015, the commission says something very different. [00:04:44] Speaker 02: Now they're saying the middle 80 is inherently flawed, because by definition, some of the companies in the middle 80 are going to be further from the median than the companies that are in the middle 50 set. [00:04:58] Speaker 02: If there was an inherent flaw in the middle 80 that the mere fact that some of the companies would be further from the median disqualified the middle 80 from being used, that should have been apparent in 2000. [00:05:09] Speaker 03: They don't say inherent. [00:05:11] Speaker 03: They don't say inherent flaw. [00:05:13] Speaker 02: Well, they don't say it's inherent, but it is inherent, Your Honor. [00:05:15] Speaker 02: It's a practical matter. [00:05:16] Speaker 05: That's a critical. [00:05:17] Speaker 05: I had the same question my colleague did. [00:05:20] Speaker 05: I don't remember them saying it's inherently flawed. [00:05:22] Speaker 02: They didn't use the word inherent, Your Honor, but the flaw they've identified, they specifically talk about, is the data in the middle 80 is by definition further from the median than the data in the middle 50. [00:05:34] Speaker 02: That's what they say is their decision. [00:05:35] Speaker 05: The data in this case, in their view, did not satisfy what they felt the requirement should be. [00:05:50] Speaker 05: to determine which of these databases make sense. [00:05:54] Speaker 02: Well, I understand, Your Honor, that's what the Commission is presenting its decision as, and that's our point about why they're not acknowledging the change. [00:06:01] Speaker 02: The change is to go from a, you know, every period we analyze, whether the middle 50 and the middle 80 work for this particular period, that's what they've done since 2003, [00:06:11] Speaker 02: to now they're saying there is a flaw in the middle 80 that prevents us from using it. [00:06:16] Speaker 02: And that flaw is incurable. [00:06:18] Speaker 02: I mean, the middle 80 is always going to have some data in it that's further from the median than the middle 50. [00:06:24] Speaker 02: But if that was the issue, the commission would have had an issue with it in their prior cases. [00:06:28] Speaker 02: They haven't explained why it's different now in 2015, why it's anymore a problem that those [00:06:36] Speaker 02: costs are necessarily more dispersed than it was in their prior cases. [00:06:39] Speaker 03: Put aside for a moment the question of change for a moment. [00:06:44] Speaker 03: I know that's a big part of your argument, but just for the moment. [00:06:46] Speaker 03: And what is substantively or otherwise unreasonable about using the 50 rather than the 80? [00:06:56] Speaker 02: I think what's substantively questionable, Your Honor, is we're trying, the Commission has said from the outset, we're trying to measure industry-wide cost changes. [00:07:04] Speaker 02: The middle 50 in this case only contained 56% of the industry data, the barrel miles that were moved. [00:07:10] Speaker 02: That's not a particularly robust sample, despite what the commission says. [00:07:14] Speaker 02: It's certainly not the kind of sample they've looked for before. [00:07:17] Speaker 02: They've said in their cases before, more data, if it's accurate, is better. [00:07:22] Speaker 02: They said they were specifically in the 2011 rehearing order. [00:07:26] Speaker 02: rejected a proposal from the shippers that would have only included 58 percent of the barrel miles in the sample. [00:07:33] Speaker 02: That's not good enough. [00:07:34] Speaker 03: But this really gets to what is our role here, which is how can we draw a line based on barrel miles that are covered when they make an argument, again, put aside the change, that the middle 50 is going to be a better [00:07:49] Speaker 03: evaluation, avoid more of the outliers, how can we second-guess that, or how can we draw a line based on the barrel miles, do you think? [00:07:57] Speaker 02: Well, Your Honor, I don't think we're asking to draw a bright line on barrel miles. [00:08:01] Speaker 02: What we're saying is the Commission has an unexplained inconsistency, to borrow a phrase from the Brand X decision. [00:08:06] Speaker 03: Well, that's the change again, and so you're back to the change, and I understand why you're on that. [00:08:11] Speaker 02: And the change really is the heart of our argument, Your Honor, that the Commission has in the past said two very important things, two principles they apply here. [00:08:18] Speaker 02: He said in 2006, the middle 50, middle 80 is a good way, an effective way of excluding outliers. [00:08:24] Speaker 02: And they said more recently that they want more data if it's accurate. [00:08:28] Speaker 02: They haven't questioned the accuracy of the data in the middle 80 in this case. [00:08:31] Speaker 03: You keep using the phrase, and I understand why, reasoned explanation. [00:08:36] Speaker 03: But is it really, is there a possibility of a reasoned explanation to get to the 50? [00:08:42] Speaker 02: Well, in 2010, they gave, I think, a plausible explanation of why they were only using the middle 50. [00:08:47] Speaker 02: They said, well, it contains a very broad sample, 76% of the barrel miles. [00:08:52] Speaker 02: We weren't going to quibble about the difference between 76% and 80%, but now we're talking about the difference between 56%, which is barely more than half of the industry, and 80%, [00:09:03] Speaker 02: when there's no question that the data, there's no question that's been raised by the commission about the data in the Middle 80 being inaccurate. [00:09:10] Speaker 02: In fact, the commission recognized that the statistical distribution of the data in the Middle 80 was a predictable statistical path and said that generally supports the idea this is accurate data. [00:09:21] Speaker 02: It's not erratic, it's not bouncing all over the places. [00:09:24] Speaker 05: I'm complex. [00:09:26] Speaker 05: In 1994 and 2010, they rested on the Middle 50. [00:09:31] Speaker 05: And they never suggested, as far as I can see along the way up to now, that they were obliged or thought they would be obliged to consider both and measure it that way. [00:09:43] Speaker 05: And in this case, they specifically said the middle 50 more effectively than the middle 80 percentage excludes pipelines with anomalous cost changes. [00:09:51] Speaker 05: And that appears to support that conclusion. [00:09:53] Speaker 05: Why would we second-guess that? [00:09:55] Speaker 02: Your Honor, again, I think it goes to the change. [00:09:57] Speaker 02: They're now saying something that's the polar opposite. [00:10:00] Speaker 05: But that's what I'm asking you. [00:10:02] Speaker 05: I'm going to bite on your change argument, OK? [00:10:05] Speaker 05: Because I don't see it. [00:10:06] Speaker 05: They started in 94 with Middle 50 being the point of observation. [00:10:12] Speaker 05: They did it again in 2010. [00:10:14] Speaker 05: And along the way, Middle 50 was always in play. [00:10:18] Speaker 05: And they never said that Middle 50 was out of play. [00:10:22] Speaker 05: But they have done Middle 50 alone in the past, in 94 and in 2010. [00:10:27] Speaker 02: Your Honor, again I think though this goes to the question of the reasoned explanation as Judge Kavanaugh brought up. [00:10:33] Speaker 05: Well then we're finding whether or not they have satisfied us, they've explained why they're using only Middle 50. [00:10:40] Speaker 02: That is the issue precisely, Your Honor. [00:10:42] Speaker 02: And what we say is in 1995, they only had the middle 50 before them. [00:10:46] Speaker 02: No one presented the middle 80. [00:10:48] Speaker 02: And the reason the expert gave for that was that the data in 1995 was very poor and varied erratically. [00:10:54] Speaker 02: There were a lot of outliers. [00:10:56] Speaker 02: By 2003, when they established the 50-80 approach, they had a much better data source. [00:11:01] Speaker 02: Everybody acknowledged that the errors that had existed in 1995 no longer were so prevalent. [00:11:08] Speaker 02: And at that point, they made a very specific determination that the middle 50, middle 80 was an effective way to exclude those statistical outliers. [00:11:15] Speaker 05: And effectively. [00:11:16] Speaker 02: Yes. [00:11:17] Speaker 02: But they've not explained, in this case, why, if it was effective in 2006, what's different today. [00:11:23] Speaker 02: They haven't said the data is more inaccurate. [00:11:26] Speaker 05: They've pretty much acknowledged it is accurate. [00:11:28] Speaker 05: At least that's what I was reading. [00:11:30] Speaker 05: Now, you may say they really didn't. [00:11:31] Speaker 05: But I thought they explained pretty carefully that at some point, [00:11:36] Speaker 05: that the outliers were obvious here and would impair the analysis if you used Middle 80? [00:11:42] Speaker 02: Well, two points on that, Your Honor. [00:11:44] Speaker 02: First of all, their explanation for why they say there are statistical outliers in the Middle 80 that are not in the Middle 50 was simply that they're further from the median. [00:11:53] Speaker 02: Again, if that was their rationale, that would have applied equally in 2003 and 2006. [00:11:58] Speaker 02: They haven't explained what's different about that. [00:12:01] Speaker 02: The second thing they said is, well, the shippers came in and pointed out a lot of anomalies in the data. [00:12:06] Speaker 02: But the commission in the earlier part of its decision didn't buy into the shipper's point that that was anomalous data. [00:12:12] Speaker 02: And in fact, they pointed out that those supposed anomalies are spread throughout the sample, not only in the middle 80, but there are many of those examples in the middle 50. [00:12:21] Speaker 02: So again, they haven't provided a neutral principle for saying, why are we now backing off of our earlier thought that middle 80 is a good sample because it has more data that is accurate. [00:12:34] Speaker 02: in any kind of sample, and you want more data. [00:12:37] Speaker 03: So your point is that their prior decisions have effectively, in effect, set up a principle that the middle 80 should be used unless there's some special reason not to use it? [00:12:48] Speaker 02: Correct, Your Honor. [00:12:48] Speaker 02: That there's no reason not to use the middle 80 unless there's something wrong with that sample or something about the middle 50, as in 2010, that made it a more usable standard. [00:12:59] Speaker 02: They've done it by their conduct, I would say. [00:13:01] Speaker 05: I mean, I'm confirming they've never said that. [00:13:04] Speaker 05: Well, they have. [00:13:05] Speaker 05: I did, much as I hated having to do it. [00:13:08] Speaker 05: I can honestly say that. [00:13:09] Speaker 05: I did read these things. [00:13:12] Speaker 05: God bless you, Your Honor. [00:13:14] Speaker 05: My wife said, why can't we go out to dinner? [00:13:17] Speaker 05: Well, I have to leave work. [00:13:19] Speaker 05: Honestly, I can't find that statement anywhere. [00:13:21] Speaker 02: Well, I would point to the 2006 order, Your Honor, where the issue was raised specifically of whether [00:13:28] Speaker 02: The shippers had presented a methodology that they argued did exclude outliers. [00:13:34] Speaker 02: It included more of the data than the middle 50, middle 80. [00:13:37] Speaker 02: And the commission said the commission should not abandon its current methodology because the evidence of existence and proper treatment of outlier data were extensively addressed in prior commission proceedings. [00:13:51] Speaker 02: And the current methodology was specifically designed to take this matter into account. [00:13:56] Speaker 02: The commission uses the middle 50% and 80% of the relevant cost data, thus ensuring that the index is not driven by statistical outliers. [00:14:04] Speaker 02: That was a specific fact finding. [00:14:06] Speaker 02: And now the commission is saying the exact opposite thing, which is that necessarily the middle 80 is driven by statistical outliers because it's further from the median. [00:14:14] Speaker 02: It contains data that's further from the median. [00:14:16] Speaker 02: And we don't think those two findings can be reconciled [00:14:21] Speaker 02: Sub salentio. [00:14:22] Speaker 02: The commission needs to explain why was that principle clear in 2006 and suddenly one to be disregarded in 2015. [00:14:31] Speaker 02: They could have made presumably specific fact findings about the middle 80 and middle 50 and said there's something wrong with this database. [00:14:39] Speaker 02: We don't want to use it. [00:14:39] Speaker 02: But they specifically did not do that. [00:14:41] Speaker 02: They did not find it to be inaccurate. [00:14:43] Speaker 02: They didn't even buy into the shippers' argument that there were specific anomalies in it. [00:14:48] Speaker 02: They simply said, it's too spread out. [00:14:50] Speaker 02: And in saying it's too spread out, that leads to our concern that what's really going on here is they're saying, because it's so dispersed, it's going to lead to a higher index. [00:15:01] Speaker 02: And indeed, they specifically talked about it as skewed upward. [00:15:03] Speaker 03: So you see them, you see the current opinion as, in effect, saying, we're going to use the middle 50 unless there's some special reason not to use the middle 50. [00:15:12] Speaker 02: That's probably a good way of putting it, Your Honor, that they've kind of shifted the presumption from what they had in 2006 [00:15:18] Speaker 02: that the model was 50-80, and there needed to be a good reason to use 50 to one where they're saying, we're almost never going to use the middle 80. [00:15:26] Speaker 02: And inherently so, I use that word, because it's a property of the distribution of the data. [00:15:31] Speaker 02: There's no way to get around that. [00:15:32] Speaker 05: Where is the language of the 2000 opinion that acknowledges in 2006, there's a firm rule, and we are now going to take exception to that rule? [00:15:41] Speaker 05: They state the rule, and then they move and say, [00:15:49] Speaker 02: Well, Your Honor, I think the Commission says in its opinion that they had never ruled on this before. [00:15:55] Speaker 02: They simply disregard the evidence of what was in the 2006 order because they say in their opinion, yeah, we used 5080 in the past, but we never really addressed it. [00:16:05] Speaker 02: We never really thought much about it. [00:16:06] Speaker 02: Well, it's right in their opinion. [00:16:07] Speaker 02: They did think about it. [00:16:08] Speaker 02: Wait, in 2010? [00:16:09] Speaker 05: What are you talking about now? [00:16:11] Speaker 02: In 2010? [00:16:12] Speaker 05: Well, in the 2010 and the 2015 orders, they basically said... No, but what I'm saying is if 2006 was a firm principle, as you argue, [00:16:20] Speaker 05: then why did the Commission miss it in 2010? [00:16:22] Speaker 05: It was a firm principle with respect to which they should have stated it and they should have said, now we're going to have an exception here because whatever. [00:16:36] Speaker 02: Well, I think in that case, they did give a plausible explanation for why they were using the Middle 50, which was the coverage in terms of barrel money. [00:16:44] Speaker 02: That's what they rested their decision on. [00:16:46] Speaker 05: I understand that part of it. [00:16:47] Speaker 05: They haven't said anything like that here. [00:16:49] Speaker 05: No, but what I'm saying is they did not say in 2010 what you're saying, that there is a cardinal rule to be followed unless, and you've tried to [00:17:00] Speaker 05: I'm not sure what you're trying to do. [00:17:01] Speaker 05: You're trying to make 2006 more than I read it to be, and 2010 would have been the place where the Commission would have acknowledged what you're saying and said they're going to write it differently, even though there was a firm rule. [00:17:13] Speaker 02: Well, Your Honor, we didn't appeal the 2010 order, but clearly we would have had the same objection had the Commission said in 2010 we're ruling out using the Middle 80. [00:17:23] Speaker 02: All they said in that case was it's unnecessary on these specific facts to use the Middle 80. [00:17:28] Speaker 02: They didn't say or imply that they were never going to use it again. [00:17:31] Speaker 02: And here in 2015, for the first time, we're getting a decision saying there's an inherent flaw, there's this basic mathematical flaw in the Middle 80 that we can't use it. [00:17:41] Speaker 02: And that is a real reversal from what they said in 2006. [00:17:43] Speaker 02: And they not only haven't explained it, they haven't even acknowledged that it's a reversal from what they said in 2006. [00:17:49] Speaker 02: And that's our central point. [00:17:52] Speaker 02: My time has expired, and I would like to reserve a few moments for rebuttal. [00:17:57] Speaker 03: OK, thank you. [00:18:06] Speaker 01: Good morning, brothers. [00:18:14] Speaker 01: This is Etichu for the Commission. [00:18:17] Speaker 01: The oil pipeline index administered by the commission is designed to accommodate ordinary pipeline cost changes. [00:18:26] Speaker 01: It's not disputed that pipeline cost experiences diverge dramatically looking at a pipeline cost change. [00:18:34] Speaker 01: It's possible, very possible for a pipeline undergoing a major expansion to have cost changes in the billions of dollars. [00:18:43] Speaker 01: So that's why these statistical outliers are excluded [00:18:47] Speaker 01: from the pipeline index. [00:18:50] Speaker 01: The whole point is just to capture the central tendency of the data, which was recognized by the court in the original 1996 Association of Oil Pipelines decision. [00:19:02] Speaker 03: So... Doesn't the 80 exclude the outliers? [00:19:07] Speaker 01: No, not in this case. [00:19:09] Speaker 01: So from what I'm hearing from counsel, [00:19:13] Speaker 01: There is no, as Judge Edwards was observing, there is no decision by the commission in 2006 that the middle 50 and middle 80 were the norm. [00:19:24] Speaker 03: That wasn't my question. [00:19:25] Speaker 03: My question was, doesn't the middle 80 exclude the outliers? [00:19:29] Speaker 01: No, Your Honor, not in this case. [00:19:31] Speaker 03: What do you mean by not in this case? [00:19:33] Speaker 06: Why eliminate to this case? [00:19:35] Speaker 01: So it's both not in this case and also [00:19:40] Speaker 01: in general. [00:19:42] Speaker 01: It is explained in the order of paragraph 43, but also in the footnotes to page to paragraph 43, which provide the basis for the commission's finding. [00:19:54] Speaker 03: So that seems, what you just said, like a general principle that the 80 doesn't sufficiently exclude the outliers, and their point is [00:20:04] Speaker 03: that that's new and that hasn't been sufficiently elaborated upon. [00:20:12] Speaker 01: And it's not new, Your Honor. [00:20:13] Speaker 01: 2006, it wasn't really an issue. [00:20:18] Speaker 01: Both parties in 2006, both the shippers... I'm sorry to interrupt. [00:20:23] Speaker 03: You've used the AD, the Commission has used the AD in the past, correct? [00:20:27] Speaker 01: Yes, it has. [00:20:28] Speaker 01: And in general, it's been uncontested. [00:20:30] Speaker 01: It hasn't made a major difference. [00:20:32] Speaker 03: But presumably when the commission used it in the past, it thought it did a suitable job of excluding the outliers while also measuring appropriately. [00:20:42] Speaker 03: But now what you just said is that the AD does not do a great job of excluding the outliers. [00:20:51] Speaker 03: And their point is, and as you can tell by the questions to opposing counsel, they have a [00:20:57] Speaker 03: difficult questions to face too, but their point is that that's a change or new or both. [00:21:05] Speaker 01: Right, that's because they're reaching back to the 2006 order and ignoring the 2010 order, the index order immediately preceding this review, in which the commission said that the middle, including the middle 80 results in including more statistical outliers in the calculation of the index. [00:21:27] Speaker 01: And for that reason, using only the middle 50 is the most appropriate method of trimming the data [00:21:35] Speaker 01: and getting only ordinary pipeline cost changes. [00:21:39] Speaker 03: But they say that that was due to special circumstances explained by the commission in that order, whereas here [00:21:49] Speaker 03: instead of identifying something special about what's going on in the market now, that it's just a defaults, rules probably too strong, but presumption that the 50 is the right thing, the 80 is not the right measurement, and that's the way it's going to be going forward. [00:22:06] Speaker 01: I heard that argument. [00:22:07] Speaker 01: It's not just the presumption, your honor. [00:22:10] Speaker 01: In paragraph 42 and 43, JA 771, the commission [00:22:18] Speaker 01: cites back to its 2010 index review and talks about how in that review, the commission found that the middle 50%, more effectively than the middle 80%, excludes pipelines with anomalous cost changes. [00:22:33] Speaker 01: And then immediately after that, in paragraph 43, the commission says that the record in this proceeding does not provide a basis for altering that position. [00:22:44] Speaker 03: And I think what they would say in response is that that is a rewriting of what was, yeah. [00:22:53] Speaker 03: And rewriting what was said in 2010. [00:22:58] Speaker 01: No, Your Honor. [00:22:59] Speaker 01: Actually, if you flip over to JA 772, what we see here is in paragraph 44 and footnotes 83 and 85, the Commission is explaining why it's appropriate to use the middle 50 here, and that is because first, the middle 50 is [00:23:19] Speaker 01: representative robust sample of the central tendency of the pipelines in this case that's at footnote 85 where the agency says it is not the case that the middle 50 percent represents a narrow or selective sector of the industry and later in the same paragraph we explained that [00:23:40] Speaker 01: Although the barrel miles represented in this particular review represent about 56% of industry barrel miles, well, first of all, that's not a now or a selective sector of industry. [00:23:54] Speaker 01: We've used even fewer barrel miles in the past. [00:23:57] Speaker 01: But also, the difference between the 56% of barrel miles here and the 70-something percent in the last review [00:24:05] Speaker 01: is attributable to a single oil pipeline, Enbridge Lakehead, which was in the middle 50 in the last review, but not in the middle 50 in the current review. [00:24:17] Speaker 03: So in that footnote 83, though, there does seem to say why the middle 80 is specifically incorrect or [00:24:26] Speaker 03: not suitable in that case. [00:24:29] Speaker 03: Now, in your favor, in the text of the order, it does make a general comment. [00:24:34] Speaker 03: Pipelines in the middle 80 are more likely to have outlying cost changes. [00:24:39] Speaker 03: So I guess that's the general principle that you would say they're referring back to. [00:24:42] Speaker 03: But the footnote, 83, does seem to refer to specifics that make the middle 80 inappropriate there. [00:24:51] Speaker 01: That's right, Your Honor. [00:24:52] Speaker 03: And I guess what they would say is you don't have those specifics here. [00:24:55] Speaker 01: We do, your honor. [00:24:57] Speaker 01: And just to back up a little bit, the two proposals that were before the agency, the one we're most familiar with is the association's proposal to use both the middle 50 and the middle 80. [00:25:09] Speaker 01: Then on the other hand, we had shipper proposals to manually trim pipeline data, which means that under the shipper proposal, the commission would be looking at the particular circumstances of individual pipelines and taking them out. [00:25:25] Speaker 03: And you point, I might have been quoting the wrong page. [00:25:27] Speaker 03: You referred us back to the 2010, right? [00:25:31] Speaker 06: I think she's talking about the 2015 order. [00:25:34] Speaker 06: 2015 order, okay. [00:25:35] Speaker 06: Yeah, because note 83, because that was my question. [00:25:40] Speaker 06: So if note 83 deals with specifics here, then your point is that in the 2015 order, [00:25:47] Speaker 06: the commission did both the general and the specific. [00:25:50] Speaker 06: It generally pointed out that 80% incorporates outliers, and then it specifically pointed out that in the specific data set used for purposes of this five-year period, it proves the point. [00:26:00] Speaker 01: Yes, Your Honor, and I'm sorry, it's both footnotes 83 and 85. [00:26:04] Speaker 03: Right. [00:26:04] Speaker 03: In this order. [00:26:05] Speaker 01: In this order. [00:26:06] Speaker 01: In this order on review, I apologize. [00:26:08] Speaker 03: But they are saying that when you go back, sorry, I have two different orders here. [00:26:13] Speaker 03: that when you refer to the 2010 order, you're misdescribing the 2010 order. [00:26:17] Speaker 03: That was my question that I asked. [00:26:20] Speaker 01: Oh, OK, I'm sorry. [00:26:22] Speaker 01: The 2010 order. [00:26:23] Speaker 03: The 2010 index review. [00:26:25] Speaker 01: Right, provides a rationale for why it's appropriate to use the middle 50 alone, and it explains [00:26:36] Speaker 01: It provides a nice summary because it says, this is paragraph 61 of the 2010 order, even when accurate data is reported, pipelines in the middle 80 as opposed to the middle 50 are more likely to have cost changes resulting from factors particular to that pipeline, such as a rate-based expansion, plant retirement, or localized changes in supply and demand. [00:27:00] Speaker 03: So I'm saying that- So your argument is that- Sorry, let's just make sure I'm on the right thing. [00:27:05] Speaker 03: Paragraph 61, right? [00:27:07] Speaker 03: Yeah. [00:27:07] Speaker 01: Of the 2010 order. [00:27:08] Speaker 03: Yeah, okay, I'm on it. [00:27:11] Speaker 03: Go ahead, Judge. [00:27:12] Speaker 06: So your point is that in paragraph 42 of the 2015 order, there's a characterization of the 2010 order. [00:27:19] Speaker 06: And the characterization is that the 2010 order generally understood that using an 80% figure is problematic because it includes too many outliers. [00:27:30] Speaker 06: Your opposing side would respond, no, really what was going on in 2010 is that it dealt with specifics as to that data set. [00:27:37] Speaker 06: And your point is that no, when we refer back to 2010 and we described it as a general matter, if you look at paragraph 61, we were actually right in describing it as a general matter that, in fact, 2010 incorporated this general view that the high 80% figure is too wide because it has an undue capacity to introduce outliers. [00:27:56] Speaker 01: That's right, Your Honor. [00:27:57] Speaker 01: And based on our review of the record in this proceeding, there's no reason to depart from the 2010 order. [00:28:07] Speaker 01: If I may, it may be accurate. [00:28:09] Speaker 03: And the Dual 10 order went on, and just picking up in paragraphs 62 and 63 to do the, as John Srinivasan says, to do the specific problems with using the 80 there. [00:28:20] Speaker 03: And you're saying that they're doing the same here. [00:28:23] Speaker 03: Right. [00:28:26] Speaker 03: presumption that the 50 is going to be better. [00:28:30] Speaker 03: And by the way, in both 2010 and here, we've pointed out problems with the 80. [00:28:36] Speaker 03: That's what you're saying. [00:28:36] Speaker 06: Born out in the specifics. [00:28:37] Speaker 03: The general is born out in the specifics. [00:28:39] Speaker 03: Born out in paragraphs 61 or 62 and 63. [00:28:42] Speaker 03: 61 through 63 then and in paragraphs. [00:28:45] Speaker 03: Footnotes 83 and 85 now. [00:28:48] Speaker 01: Yes, that's right, Your Honor. [00:28:49] Speaker 01: And it is borne out in the 2015 review. [00:28:53] Speaker 01: And in addition, it may be helpful to take a look at the discussion in the section that addresses the shipper's manual data trimming. [00:29:03] Speaker 01: Because at paragraph 34, for example, JA 764, [00:29:09] Speaker 01: That's where some of the problems with these outline pipelines is discussed. [00:29:16] Speaker 01: And at footnotes 60 and 61, we see that by using the middle 50, [00:29:24] Speaker 01: we take care of the problems introduced by using some of these idiosyncratic pipelines that appear in the middle 80. [00:29:33] Speaker 01: So the commission is realizing that between these proposals of using a composite of the middle 50 and 80, and the proposal of looking at individual pipelines, by just using the middle 50, [00:29:47] Speaker 01: we end up with a methodology that's objective and transparent and doesn't require individual looks at pipelines for why their data is idiosyncratic. [00:30:02] Speaker 01: If I may, to the extent it may be helpful, JA54 has a chart. [00:30:10] Speaker 01: This is a chart appearing in the testimony of the association's expert. [00:30:17] Speaker 01: And this chart. [00:30:20] Speaker 03: What does that chart tell me? [00:30:23] Speaker 03: I need a translation. [00:30:24] Speaker 03: It's obvious. [00:30:25] Speaker 03: It's obvious. [00:30:25] Speaker 03: Yes. [00:30:26] Speaker 03: Just looking at that chart, it's magical. [00:30:28] Speaker 03: I studied that one, too. [00:30:32] Speaker 03: Yes. [00:30:32] Speaker 03: OK. [00:30:32] Speaker 03: What does this mean? [00:30:33] Speaker 01: Figure two. [00:30:34] Speaker 01: It may be helpful to take a look at it. [00:30:36] Speaker 01: If you take a look at figure two, you'll see that, [00:30:40] Speaker 01: These data points represent different pipelines. [00:30:45] Speaker 01: And you'll see that there's a much larger grouping within this lighter colored bands, which represents the middle 80. [00:30:52] Speaker 01: And you see when you go beyond, excuse me, the middle 50. [00:30:55] Speaker 01: You'll see that when you go beyond the middle 50 to the middle 80, that you're seeing more dispersion in the costs, which bears out the commission's conclusion that the middle 80 includes more idiosyncratic [00:31:10] Speaker 01: cost experiences. [00:31:14] Speaker 06: Yeah, that seems like it's almost mathematically guaranteed to be true, because any time you expand the frontier, you're going to have a greater capacity, at least, to sweep in idiosyncrasies. [00:31:24] Speaker 01: Right. [00:31:25] Speaker 01: That's right, Your Honor. [00:31:29] Speaker 03: OK. [00:31:30] Speaker 01: If you have no more questions, I'll see you the rest of my time. [00:31:33] Speaker 03: Thank you. [00:31:33] Speaker 03: Thank you. [00:31:47] Speaker 02: Grateful, Your Honor. [00:31:49] Speaker 02: Council for the Commission has focused us on paragraphs, I believe it's 43 and 44 of the 2015 order, and appropriately so. [00:31:59] Speaker 02: That's where the crux of this issue lies. [00:32:04] Speaker 02: First of all, I would highlight the language in the middle of paragraph 43 says, however, by definition, costs at the top or bottom of the middle 80% deviate significantly from the cost experience of other pipelines. [00:32:16] Speaker 03: Don't they say something similar? [00:32:20] Speaker 03: in 2010. [00:32:22] Speaker 02: Absolutely, Your Honor. [00:32:23] Speaker 02: And we can see that they said these same things in 2010. [00:32:26] Speaker 02: So why is it? [00:32:29] Speaker 02: But those have never been subject to this court's review. [00:32:32] Speaker 02: If they didn't give a quote. [00:32:34] Speaker 03: But it wasn't challenged. [00:32:35] Speaker 03: It wasn't challenged. [00:32:36] Speaker 03: That was the rule. [00:32:37] Speaker 03: So we did all this tracing of the paragraphs to try to figure out why it was done. [00:32:42] Speaker 03: Let me finish. [00:32:43] Speaker 03: Why it was done in 2010. [00:32:44] Speaker 03: And paragraph 61 in 2010. [00:32:48] Speaker 03: says the middle 50% more appropriately adjusts the index levels for normal cost changes as opposed to the middle 80%, which by definition includes pipelines relatively far removed from the median. [00:33:05] Speaker 02: Yes, Your Honor. [00:33:05] Speaker 03: And then in paragraph 42 and 43, in this 2015, they seem to be [00:33:14] Speaker 03: following along with what they said in 2010. [00:33:17] Speaker 03: So the question was, have they really changed from 2010? [00:33:21] Speaker 03: My question was, were they mischaracterizing in this order what they said in 2010? [00:33:26] Speaker 03: But when you trace the paragraphs back, it doesn't look that way. [00:33:30] Speaker 02: It's not our argument, Your Honor, that they are mischaracterizing what they said in 2010. [00:33:36] Speaker 02: Our point is they made the same mistake in 2010 to the extent they were not expected. [00:33:41] Speaker 03: But then how is this a change? [00:33:42] Speaker 03: This isn't a change. [00:33:43] Speaker 02: Well, it's a change from their policy in the last order from this commission that was reviewed by this Court. [00:33:49] Speaker 05: Oh, that's not the way we review. [00:33:51] Speaker 05: They're entitled to establish precedent without our blessings. [00:33:55] Speaker 02: Well, Your Honor, when they make the change, I don't think it's the issue. [00:33:58] Speaker 02: It's that they made a change. [00:33:59] Speaker 02: I disagree with that. [00:34:01] Speaker 02: I don't think that's right. [00:34:02] Speaker 06: I don't think that's right. [00:34:04] Speaker 03: Because there's an opportunity to get review the last time. [00:34:08] Speaker 03: Yeah, they've changed the policy 30 years ago, but it was unreviewed. [00:34:11] Speaker 03: You can't come here now and say, well, it was never reviewed, so. [00:34:15] Speaker 02: But we would have been somewhat anomalous to come to this court for review of that statement, which was essentially dictum in 2010, because they go on to say in the 2010 order, really, the basis of our decision here is that we've got 76% of the barrel miles. [00:34:28] Speaker 02: We don't really need the middle 80 to get a good sample. [00:34:31] Speaker 02: In effect, they were saying, we don't need to reach the question on the merits. [00:34:34] Speaker 02: Some thoughts we've expressed. [00:34:36] Speaker 02: We couldn't have really sought review of that, I don't think. [00:34:38] Speaker 02: I understand. [00:34:39] Speaker 02: I understand. [00:34:40] Speaker 03: But that's why I was asking a series of questions on put aside the change. [00:34:43] Speaker 03: Because I think ultimately, your position, I think, is [00:34:48] Speaker 03: You all should be using the middle 80, not the middle 50, because the middle 80 excludes the outliers and is a more accurate representation of what's going on. [00:35:00] Speaker 03: Is that fair? [00:35:02] Speaker 02: Yes, and let me highlight one issue that I think goes to the point of what we're making. [00:35:06] Speaker 02: The council for the commission stood up and said, look, a big problem with the middle 80 versus the middle 50 is that it's got this one big pipeline in. [00:35:15] Speaker 02: Enbridge pipelines, 15% of the barrel miles. [00:35:17] Speaker 02: That really skews things. [00:35:19] Speaker 02: This is an example of the kind of thing the commission thought skewed things. [00:35:22] Speaker 02: But if you look in their own decision, footnote 87 of the 2015 decision, [00:35:28] Speaker 02: The second sentence, they're talking about Enbridge-Lakehead, very same pipeline. [00:35:33] Speaker 02: And these pipelines have a disproportionate effect because the Kahn methodology uses a weighted average in conjunction with a simple average to measure the central tendency. [00:35:43] Speaker 02: Although the size of these pipelines makes their data particularly relevant for assessing industry-wide barrel mile cost changes, the commission goes on to say that's not a good enough reason to have two separate indexes for prudent products. [00:35:56] Speaker 02: They've clearly recognized here that Enbridge Lakehead is a good indicator of industry-wide cost changes. [00:36:02] Speaker 02: And yet, their middle 50 rule is excluding pipelines like Enbridge [00:36:08] Speaker 02: We would say arbitrarily. [00:36:09] Speaker 02: They haven't really given a reason why that's a good example of a statistical outlier. [00:36:15] Speaker 02: They've simply made the statement, it's further from the median, whether they did it in 2010 or 2015, they still have fundamental problems. [00:36:22] Speaker 02: They're saying that the only thing wrong with the middle 80 is that it's further from the median. [00:36:26] Speaker 02: If you look at the chart we were referred to, you can see the distribution. [00:36:30] Speaker 02: And the effect of choosing the middle 50 over the middle 80 will be a lower index every time. [00:36:35] Speaker 02: And that led to our concern that just as the commission did in the first index review where this court found that their methodology was designed to lead to a lower index, that's not a good enough reason for what they adopted. [00:36:49] Speaker 02: Are there no further questions? [00:36:51] Speaker 02: OK. [00:36:51] Speaker 02: Thank you. [00:36:52] Speaker 03: The case is submitted.