[00:00:00] Speaker 01:
Case number 22-1335 EDA.

[00:00:04] Speaker 01:
Energy Arkansas LLC EDA petitioners versus Federal Energy Regulatory Commission.

[00:00:09] Speaker 01:
Ms.

[00:00:10] Speaker 01:
Quinn Barabanov for the petitioners.

[00:00:11] Speaker 02:
Ms.

[00:00:12] Speaker 02:
Pacella for the respondents.

[00:00:16] Speaker 02:
Good morning and may it please the court.

[00:00:18] Speaker 02:
I'm Jennifer Quinn Barabanov and I'm here on behalf of the appellants, the energy operating companies,

[00:00:23] Speaker 02:
which are engaged in the generation and sale of electricity that serves more than three million customers in Arkansas, Louisiana, Mississippi, and Texas.

[00:00:34] Speaker 02:
They belong to an organization known as MISO, which operates power markets in the central United States.

[00:00:45] Speaker 02:
This appeal involves first approval of rules that implement a change in MISO's power capacity market.

[00:00:50] Speaker 02:
from an annual construct to a seasonal one.

[00:00:54] Speaker 02:
The stated purpose of these changes was to enhance reliability, which is a laudable goal.

[00:01:00] Speaker 02:
But in this case, FERC failed to engage in the recent decision-making required by the Administrative Procedure Act.

[00:01:08] Speaker 03:
And just to be clear, because there was a lot of discussion about preservation of issues,

[00:01:15] Speaker 03:
Is your appeal dealing with the 31 day rule, the 120 day rule, and then the capacity accreditation?

[00:01:21] Speaker 03:
Is there anything beyond that?

[00:01:23] Speaker 02:
No, it's those three issues, Your Honor.

[00:01:24] Speaker 02:
I think the 120 day rules is pretty adequately addressed in our brief, so I'm really going to focus today on the first two, the 31 day rule and the

[00:01:33] Speaker 02:
methodology and just also I'd like to reserve three minutes for rebuttal if I may.

[00:01:39] Speaker 02:
Okay, so here FERC engaged in a very conclusory analysis that failed to engage with the arguments that Entergy made and its conclusion that the proposed changes enhanced reliability were in the end.

[00:01:52] Speaker 02:
not reasonably based on any evidence in the record.

[00:01:56] Speaker 02:
In fact, these changes are going to increase costs for EnterG's customers without any corresponding reliability benefits.

[00:02:03] Speaker 02:
And in some cases, they are actually inconsistent with the stated reliability goal.

[00:02:08] Speaker 02:
For these reasons, FERC's order should be reversed and remanded.

[00:02:12] Speaker 03:
Did you offer any alternatives?

[00:02:14] Speaker 03:
I mean, we're talking about numbers and reliability seems to be a fair goal.

[00:02:19] Speaker 03:
But did you offer any alternatives to the

[00:02:21] Speaker 03:
31-day versus 120-day notice that is backed up by any research or other data?

[00:02:27] Speaker 02:
Well, there was data in the record that showed that the 31-day limit had a particular impact on nuclear resources, for example.

[00:02:35] Speaker 02:
So there was that.

[00:02:37] Speaker 02:
But FERC has an obligation, as this court held in the border decision, that when it imposes a quantitative requirement, it has to articulate what the reasons are that justify that requirement and why the one that's approved

[00:02:49] Speaker 02:
is reasonable as compared to ones that are higher or that are lower.

[00:02:53] Speaker 02:
And we certainly did advocate for a longer time period, although not a specific amount.

[00:02:59] Speaker 03:
But I guess what I'm asking is, is your position that you should just attack whatever FERC has put out as its premise as opposed to offering alternatives?

[00:03:07] Speaker 03:
I'm not suggesting that you had to.

[00:03:09] Speaker 03:
But at the end of the day, we're looking at just and reasonable based on what it has presented.

[00:03:14] Speaker 03:
But if you haven't given us something else to consider, then it's harder to

[00:03:20] Speaker 03:
kind of contemplate that there is nothing else or there is something else.

[00:03:25] Speaker 02:
Yes, your honor.

[00:03:26] Speaker 02:
But at the same time, the rule that's before you for review today is the 31 day rule.

[00:03:31] Speaker 02:
And one aspect of that in particular is very hard to square with FERC's own prior approved definitions, most notably the definition of maintenance margin.

[00:03:43] Speaker 02:
Now, Entergy raised the issue

[00:03:45] Speaker 02:
that the 31 day rule being applied to impose a penalty or replacement obligation.

[00:03:53] Speaker 02:
in periods of positive maintenance margin is actually at odds with FERC's approved definition of maintenance margin, which is defined as the projected megawatts of additional generation that can be taken out of service for planned maintenance within MISO during a given period, and this is the key language here, without impacting the adequacy of generation of supply.

[00:04:17] Speaker 02:
So by definition, then, when there's positive maintenance margin, there isn't a basis

[00:04:21] Speaker 02:
believe that there's an increased reliability risk for having a longer outage.

[00:04:27] Speaker 04:
So they're doing four seasons a year.

[00:04:31] Speaker 04:
So that means each season is three months.

[00:04:36] Speaker 04:
And what FERC said is, and FERC has had, sorry, MISO has, sorry, I keep saying MISO like this.

[00:04:45] Speaker 04:
MISO has had a lot of, has had reliability issues.

[00:04:51] Speaker 04:
That's why it was changing all of this.

[00:04:53] Speaker 04:
There's no question that they needed to change things to ensure better reliability.

[00:04:59] Speaker 04:
Why, well, if I could hire to do a summer job for three months,

[00:05:05] Speaker 04:
June, July, and August.

[00:05:08] Speaker 04:
And I get paid in advance to work three months.

[00:05:13] Speaker 04:
And I say, thank you very much.

[00:05:14] Speaker 04:
Now, I'm going to be out sick for bodily repairs for June and part of July, but I'm keeping the money.

[00:05:26] Speaker 04:
And the employer said, no, we're not going to let you do that.

[00:05:31] Speaker 04:
Would that be unreasonable if they've only hired me for three months and I'm planning to be out for more than a third of that?

[00:05:37] Speaker 04:
Would that be unreasonable?

[00:05:41] Speaker 02:
The lens that that reasonableness has to be evaluated through is the stated purpose of the change, which is to enhance reliability.

[00:05:49] Speaker 04:
Sure, and the reliability includes I need to have my employees here to do the work.

[00:05:55] Speaker 04:
The reason 31 Days is there is that's a third

[00:05:59] Speaker 04:
of the season and asking people who have committed to provide energy to be there for two thirds of the season that they have committed for having been paid for all three months seems eminently reasonable to me.

[00:06:19] Speaker 04:
So what am I missing?

[00:06:21] Speaker 02:
Your honor, because there's no basis in the record to conclude that the 31 days

[00:06:27] Speaker 02:
that the days that the generator will be unavailable are days when it's necessary.

[00:06:34] Speaker 02:
First of all, because of the maintenance margin definition, which FERC did not explain.

[00:06:39] Speaker 04:
So they want to get paid for more than 31 days without generating energy during those, say, 40 days.

[00:06:47] Speaker 04:
They think getting paid for not generating energy for 31 days is not enough.

[00:06:53] Speaker 04:
It needs to be 40 days.

[00:06:54] Speaker 04:
This is all scheduled maintenance.

[00:06:55] Speaker 04:
This is an emergency step.

[00:06:56] Speaker 02:
Yes, Your Honor, but the point is that that may be a fairness issue, but it's a reliability issue.

[00:07:02] Speaker 04:
It is reliability, because just like with my employer, if my employee that I've paid is not going to be, I'm laid out the money and that person's not going to be there during my busy summer season, that's reliability for the MISO grid as a whole.

[00:07:18] Speaker 02:
But that presupposes that those 31 days when you're taking a planned outage,

[00:07:23] Speaker 02:
correspond to a period of increased reliability risk, which is not true because of maintenance margin.

[00:07:28] Speaker 02:
And then also there's no basis in the record to show that that 31 days is tied to the times of greatest need.

[00:07:34] Speaker 02:
In fact, there's evidence in the record that shows that that 31 day rule is in fact inconsistent with reliability in some respects, because the strategies- That's not even a 31 day rule, it's a 62 day rule, because they say you can spread it over two seasons.

[00:07:48] Speaker 02:
Yes, Your Honor, but that suggestion that FERCS makes actually underscores the arbitrariness of this rule, because it's not that- Are there any repair, these are planned repairs, is there any evidence of record that says that 62 days is not enough time for planned repairs?

[00:08:06] Speaker 04:
Your Honor, the fact that there are 62, no, there isn't- So there's no evidence on that, so that can be done so to the extent

[00:08:14] Speaker 04:
repairs are part of reliability, this rule fully addresses that.

[00:08:18] Speaker 02:
But there is evidence in the record, Your Honor, that not allowing units to take longer outages actually can be inconsistent with reliability in two respects.

[00:08:28] Speaker 02:
One, by discouraging prudent maintenance.

[00:08:31] Speaker 04:
And second of all, by encouraging- Not discouraging if you can do 62 days.

[00:08:34] Speaker 02:
How is it discouraged?

[00:08:36] Speaker 02:
But that the fact that the government suggests 62 days, first of all, everyone, you know, there are questions about whether or not everyone will be able to do their maintenance and across seas.

[00:08:45] Speaker 04:
Is there any evidence in the record?

[00:08:47] Speaker 04:
That's what I asked you.

[00:08:48] Speaker 04:
I thought I started with that.

[00:08:48] Speaker 04:
Is there any evidence in the record that the 62 day window that's provided if you go across two seasons is not sufficient for all this maintenance that you say is related to reliability?

[00:09:00] Speaker 04:
I thought you said, no, there's not evidence of that in the record.

[00:09:03] Speaker 02:
No, Your Honor, but there is evidence that it could actually undermine whether or not particular units can do their maintenance in that period of time.

[00:09:09] Speaker 02:
There's data in there about the typical outages, and there is data that nuclear units in particular.

[00:09:14] Speaker 04:
Is there any data about typical outages or nuclear or steam turbine plant outages that exceeds a 62-day maintenance window?

[00:09:26] Speaker 02:
Your Honor, I don't know whether there's evidence of any, but the point is that there are other knockouts.

[00:09:33] Speaker 02:
But there are other impacts, Your Honor, for concedes that people may drop out of the market, which will decrease the amount of capacity that's available and increase its cost.

[00:09:41] Speaker 02:
And they also concede that the rule may encourage inefficient maintenance, which could be at odds with reliability in the long run.

[00:09:48] Speaker 02:
The seasonal capacity is not an end in itself.

[00:09:51] Speaker 02:
It's supposed to accomplish the role.

[00:09:52] Speaker 04:
So they had to balance sort of pros and cons.

[00:09:54] Speaker 02:
Correct, Your Honor, and they didn't explain.

[00:09:55] Speaker 04:
How do they misbalance the pros and cons?

[00:09:57] Speaker 02:
They didn't explain how to square the rule with their approved definition of maintenance margin.

[00:10:04] Speaker 02:
That's by definition arbitrary and capricious.

[00:10:07] Speaker 02:
You can't have a definition that you've approved that says that it's okay to take units out of service and then

[00:10:14] Speaker 02:
ignore that in your reasoning to say that this additional requirement is necessary.

[00:10:20] Speaker 03:
What is your concern about what precedent this sets in terms of future capacitive determinations?

[00:10:29] Speaker 02:
Your honor, energy doesn't oppose the goal here.

[00:10:33] Speaker 02:
Reliability is a laudable goal.

[00:10:35] Speaker 02:
I think the question is just whether or not the agency

[00:10:38] Speaker 02:
Fulfilled its obligation to offer a reasonable reason analysis and we don't think that that was true here.

[00:10:43] Speaker 03:
The reason I ask that question is because you say this goes against prior for precedent.

[00:10:48] Speaker 03:
So I'm asking you what do you believe this precedent sets.

[00:10:51] Speaker 02:
I think this precedent is really would be about what is what is present on this record your honor in terms of the explanation that's provided.

[00:10:58] Speaker 02:
It's not some per se necessarily- What FERC precedent is it going against?

[00:11:03] Speaker 02:
Well, I may have misspoke about the precedent.

[00:11:05] Speaker 02:
It's just the fact that FERC has this approved definition.

[00:11:08] Speaker 02:
And what it has done, its reasoning is inconsistent with that prior approved definition of maintenance margin.

[00:11:14] Speaker 02:
So turning to the SAC methodology itself, when fully implemented, the SAC methodology, the Seasonable Adjusted Capacity methodology for accrediting resources,

[00:11:25] Speaker 02:
is arbitrary and capricious for two reasons.

[00:11:29] Speaker 02:
It is unreasonable because there is no basis in the record for FERC to have found that it's predictive.

[00:11:35] Speaker 02:
And it involves an unreasonable amount of volatility.

[00:11:38] Speaker 02:
So with respect to the first issue, whether or not it's predicted, the methodology suffers from a critical problem, which is that it very heavily weights a very small sample of hours.

[00:11:50] Speaker 04:
When it is finally impotent… Can you raise the sample size issue in your rehearing petition?

[00:11:55] Speaker 02:
Yes, Your Honor, I believe we did.

[00:11:57] Speaker 02:
I can get you the specific site to that on rebuttal, Your Honor, in terms of the page site.

[00:12:06] Speaker 02:
But I believe it was in our June appendix at page 785 to 789.

[00:12:14] Speaker 02:
We said that it was an unreasonably volatile and a poor predictor of availability.

[00:12:19] Speaker 02:
So basically, I'm sorry.

[00:12:21] Speaker 04:
When you come up on rebuttal, please do tell me where you talk about sample size in your re-hearing petition.

[00:12:26] Speaker 02:
Okay, that may be a shorthand for a way to refer to it.

[00:12:29] Speaker 02:
We certainly did criticize the 60- What is your shorthand?

[00:12:31] Speaker 02:
The sample size phraseology, Your Honor, but we did criticize the 65 hours, adding up to a total of 90 hours.

[00:12:38] Speaker 02:
We did criticize the 3% of hours as opposed to thousands of hours.

[00:12:41] Speaker 04:
I understand, but that's not the same thing as an argument about sample size.

[00:12:45] Speaker 02:
Okay, Your Honor, I understood it to be the same.

[00:12:48] Speaker 02:
If there's something about it you think to be different.

[00:12:51] Speaker 00:
Maybe this will help.

[00:12:52] Speaker 00:
So I thought there were two different issues.

[00:12:57] Speaker 00:
One was the argument I think you

[00:13:00] Speaker 00:
I thought you preserved was saying that the tier two hours are arbitrarily small.

[00:13:09] Speaker 02:
Correct.

[00:13:10] Speaker 00:
There's a different issue that worried me that I don't think is preserved, which is when MISO was asked to show reliability, they came back with, I'm going to try to find it here,

[00:13:30] Speaker 02:
The 11 days, Your Honor?

[00:13:32] Speaker 00:
The 521, the 11 days with two regents.

[00:13:37] Speaker 00:
And that's a different issue.

[00:13:38] Speaker 00:
It is.

[00:13:38] Speaker 00:
I thought that one was not preserved.

[00:13:42] Speaker 02:
Right.

[00:13:42] Speaker 02:
So Judge Millett, I was referring, when I said sample size, I was referring to the 65 hours, the high, the tight margin hours that are weighted when it's ultimately implemented.

[00:13:54] Speaker 02:
80% of the accreditation is based on 3% of the hours.

[00:14:00] Speaker 02:
And the remaining 20% is based on the remaining thousands of hours.

[00:14:04] Speaker 00:
The second point I mentioned, I don't think is preserved.

[00:14:09] Speaker 00:
So if you disagree, I'll give you a chance to point me to something on that.

[00:14:14] Speaker 02:
Right.

[00:14:14] Speaker 02:
So I think the argument related to the 11 days is not about sort of as a sample size, because I think it's kind of hard to understand.

[00:14:22] Speaker 02:
There's some ambiguity as to precisely what that data is.

[00:14:25] Speaker 02:
But with that said, I think we did preserve the argument that the problem with the data that MISO provided is that it is aggregated data.

[00:14:35] Speaker 02:
And because it is aggregated regional data, it cannot provide a basis to validate a methodology for accrediting individual resources.

[00:14:47] Speaker 02:
It's a total mismatch.

[00:14:49] Speaker 02:
And so it provides no support.

[00:14:51] Speaker 02:
And that is true both in terms of whether the methodology is predictive and also in terms of whether or not the level of volatility is unreasonable.

[00:15:01] Speaker 00:
I mean, the RTO's concern is whether things work on a system-wide level.

[00:15:09] Speaker 02:
Yes, your honor, but as we pointed out in our brief, we're dealing with load serving entities here who have to plan around these accreditations and some of them, particularly ones that are smaller and may not have diverse fleets of which several categories that several of the energy operating.

[00:15:28] Speaker 02:
committees fall into are going to really have trouble adjusting to this volatility, in particular because the methodology is flawed.

[00:15:36] Speaker 02:
You can't plan around an incentive that is essentially generating, based on a methodology that is essentially generating random results because of that very small 65 hour over three seasons.

[00:15:52] Speaker 02:
methodology.

[00:15:53] Speaker 02:
It's about the number of hours and the weighting of them.

[00:15:56] Speaker 00:
But on the 65 hours is small.

[00:16:04] Speaker 00:
It's 3%.

[00:16:05] Speaker 00:
3% of the hours get 80% of the weight.

[00:16:09] Speaker 02:
Yes, Your Honor.

[00:16:10] Speaker 00:
But, I mean, those are the 3% of hours that they care most about.

[00:16:15] Speaker 00:
Those are the hours when there's a reliability problem.

[00:16:18] Speaker 02:
Yes, but the question is whether those hours during a season are predictive of what's going to happen going forward.

[00:16:25] Speaker 02:
And really, the point is, is there a reason basis in the record for Birx's condition?

[00:16:31] Speaker 04:
Do you think it should be 80 days, or do you think it should be a smaller number of days?

[00:16:35] Speaker 02:
Your Honor, it should be enough to compress the volatility to a reasonable amount, for one thing.

[00:16:45] Speaker 02:
which FERC didn't demonstrate what it did.

[00:16:47] Speaker 02:
And then second of all, there should be some basis for determining that it's predictive.

[00:16:52] Speaker 02:
And it should be predictive of what it sets out to do, which is a credit individual resources.

[00:16:57] Speaker 02:
It's not about system-wide predictions.

[00:17:01] Speaker 04:
They say for stock sales, prior performance is no guarantee of future performance.

[00:17:08] Speaker 04:
it may be in this context that they really have to sort of learn by experience.

[00:17:14] Speaker 04:
And so if they, why is it unreasonable to draw upon where there have been, they say, look, we're gonna look at the times when there have been the real problems.

[00:17:26] Speaker 04:
And that's not all 90 days or 91 or two days of our season.

[00:17:32] Speaker 04:
It is, we're gonna focus on the 65 days.

[00:17:37] Speaker 02:
Your honor, but there's no basis in the record to say that these historical data are going to be predictive of what happens in the future.

[00:17:43] Speaker 04:
For example, so now... Are they supposed to... Is your position that they should go back to the forced outage data that they did on their prior?

[00:17:50] Speaker 04:
Not necessarily.

[00:17:52] Speaker 04:
So you don't know whether the numbers should be bigger or smaller.

[00:17:58] Speaker 04:
I haven't heard you say you have a position on that.

[00:18:01] Speaker 04:
And you say, it's fine for them not to go back to the old scheme.

[00:18:06] Speaker 04:
And I'm not telling you there's a number bigger or smaller.

[00:18:09] Speaker 04:
But then they can't rely on history of past outages.

[00:18:12] Speaker 04:
I'm not sure what it is that you think would be reasonable for them to do.

[00:18:16] Speaker 02:
Your honor, in terms of the bigger or smaller number, I think that might have been some confusion as to which set we were talking about.

[00:18:22] Speaker 02:
So with respect to the 65 hours.

[00:18:24] Speaker 02:
I'm talking about 65 days.

[00:18:25] Speaker 02:
65 hours.

[00:18:25] Speaker 02:
Yes, your honor.

[00:18:27] Speaker 02:
No, that's OK.

[00:18:27] Speaker 02:
Thank you.

[00:18:28] Speaker 02:
So yeah, I mean, we think that a bigger set or a longer historical window where maybe the season shouldn't, you know, maybe the seasonal methodology, that's not something we challenge.

[00:18:39] Speaker 02:
Obviously, you know, maybe the waiting should be different.

[00:18:42] Speaker 02:
I mean, FERC hasn't articulated a reasoned basis for why those particular that particular number of hours or why that waiting is necessarily appropriate.

[00:18:53] Speaker 02:
And that's really where we take where we take issue with it.

[00:18:56] Speaker 04:
So you want, did you propose to FERC a different number, I keep thinking days, hours, or different waiting?

[00:19:07] Speaker 04:
And they are, because they are, by the way, they sort of are integrating this with different, you know, I guess the first three years of compliance.

[00:19:15] Speaker 04:
Yes, Your Honor, but the argument that FERC makes to defend that, which is that, you know, you can adjust based on the phase-in, doesn't justify the- I understand that, but I understand there were other statistics there, and I didn't know if you had argued to FERC that, you know, that thing you're doing the first year, that different percentiles between the two tiers should be what you're doing long-term, or are you simply

[00:19:40] Speaker 04:
Saying that what you did is wrong, but you didn't offer FERC a position on what percentiles.

[00:19:45] Speaker 04:
I think what you're referring to weighting, you mean the different percentiles for tier one and tier two?

[00:19:49] Speaker 04:
Correct.

[00:19:50] Speaker 04:
So different percentiles.

[00:19:51] Speaker 04:
If you didn't offer that, and you didn't offer a different number of hours?

[00:19:57] Speaker 02:
No, Your Honor.

[00:19:58] Speaker 02:
I don't believe we did.

[00:19:59] Speaker 02:
But the purpose for today is what we did raise these concerns about the methodology.

[00:20:03] Speaker 02:
And under the APA, FERC was obligated to respond to them and offered a recent analysis.

[00:20:08] Speaker 02:
And it did not.

[00:20:10] Speaker 00:
On the point whether there's evidence that the method is predictive, you're running a little bit uphill because they get a fair degree of deference for fact-finding or predictive judgment and such.

[00:20:29] Speaker 00:
Back to the chart I mentioned at 521 and 522,

[00:20:38] Speaker 00:
I can't pretend to understand all of it, but it does seem to be some evidence that in particular regions.

[00:20:54] Speaker 00:
Yeah, I'll let you in particular reads particular seasons in particular regions when there were these reliability crunches.

[00:21:05] Speaker 00:
the ISAC methodology worked better than the UCAP methodology.

[00:21:12] Speaker 00:
So am I mischaracterizing?

[00:21:15] Speaker 00:
Am I misunderstanding what that chart is showing?

[00:21:18] Speaker 02:
That is what the chart shows, Your Honor, but there are two issues with the chart.

[00:21:22] Speaker 02:
One is, and the primary one, is that it doesn't validate what this methodology is supposed to do.

[00:21:28] Speaker 00:
There was no- Then they have some evidence of predictiveness

[00:21:34] Speaker 00:
And you might have had an argument that these aren't enough entries or they're cherry picked or who knows, but you didn't make that argument.

[00:21:46] Speaker 02:
But Your Honor, the point here is that it doesn't, it doesn't, FERC has made no connection between aggregated data and the validity of this method for accrediting individual resources.

[00:21:56] Speaker 02:
We did raise a concern that there was no data provided that related to the

[00:22:00] Speaker 02:
whether this methodology was reasonably accurate in doing what it was designed to do, which was to accredit individual resources.

[00:22:09] Speaker 02:
And there was evidence in the record to show that at the market participant level, which again is still aggregated data, not at the individual resource data level, but that volatility could be as much as 20% year over year.

[00:22:23] Speaker 02:
And FERC concedes that the methodology will increase volatility

[00:22:28] Speaker 02:
at the individual resource level, but it doesn't.

[00:22:32] Speaker 04:
They calculated that volatility would be much lower.

[00:22:36] Speaker 04:
than the example you thought that a worst case scenario that you thought could happen.

[00:22:41] Speaker 02:
So they did, Your Honor.

[00:22:43] Speaker 02:
But again, that was MISO wide.

[00:22:46] Speaker 02:
So fully aggregated data, the data that we cited to for the 20% figure was more granular but still aggregated data.

[00:22:56] Speaker 02:
And that volatility, FERC acknowledges that volatility is an issue and that it comes with costs.

[00:23:02] Speaker 02:
And the question, really,

[00:23:03] Speaker 02:
This has FERC engaged in a rationalized reason and explanation for why utilities should incur these costs based on a lot of methodology.

[00:23:13] Speaker 02:
Your honor, here they haven't provided any explanation for why the amount here is reasonable.

[00:23:20] Speaker 04:
So it's not inherently unreasonable.

[00:23:22] Speaker 04:
You just want more explanation for why they're willing to tolerate this level of volatility

[00:23:28] Speaker 04:
balanced against the other goals they're trying to achieve.

[00:23:30] Speaker 02:
What we would like to see, Your Honor, is an analysis of what the volatility is at the individual resource level, and then to have FERC pass on that.

[00:23:38] Speaker 04:
So then, is it the error?

[00:23:41] Speaker 04:
We read the record as of having looked at some individual levels.

[00:23:44] Speaker 04:
That's one thing.

[00:23:45] Speaker 04:
On the other hand, so your position is it's just inherently unreasonable for FERC to have done this only at the sort of- Aggregate.

[00:23:53] Speaker 04:
Big rise level.

[00:23:55] Speaker 02:
Correct, Your Honor.

[00:23:55] Speaker 04:
And the best authority for that is?

[00:23:57] Speaker 02:
Uh, it is the, uh, the key span case, your honor, uh, which basically, uh, lays out that, uh, when FERC, um, you know, makes a decision, particularly in that case about accreditation, uh, based on a limited data set, it has to show, uh, why, uh, that, uh, was, uh, reasonable and it has to show that it was consistent, uh, with promoting the stated goal of reliability.

[00:24:24] Speaker 00:
Why does volatility matter apart from accuracy?

[00:24:32] Speaker 00:
The method is accurate and one thing they said on volatility was that the swings reflect, they think the method is accurate and therefore the swings reflect underlying supply fundamentals.

[00:24:50] Speaker 00:
Well, and I take it you disagree with that premise about accuracy, but.

[00:24:56] Speaker 00:
What what's the volatility concern above and beyond whether the method?

[00:25:01] Speaker 02:
Well, two things, your honor.

[00:25:02] Speaker 02:
First of all, historic fundamentals is an interesting term because in our view.

[00:25:08] Speaker 02:
There shouldn't be these swings in accreditation if the operating characteristics of the unit remain the same.

[00:25:15] Speaker 02:
There may be things that happen, some of which are the result of happenstance, something breaks, there's a, you know,

[00:25:22] Speaker 02:
a max gen event that is caused by an outage somewhere else.

[00:25:27] Speaker 02:
But the costs of the volatility are several.

[00:25:32] Speaker 02:
One is that it makes it extremely, the primary one is that it forces utilities to incur costs to accommodate that volatility.

[00:25:43] Speaker 02:
You know, and in that it's going to force them to incurred expenses either by going to the market or by building additional capacity based on a methodology that is flawed and the whole notion that they're going to be able to plan around this and predict the accreditation, which is what first suggests.

[00:26:05] Speaker 02:
is just implausible.

[00:26:06] Speaker 02:
If the methodology is fundamentally flawed, it is not sending accurate signals that people can respond to.

[00:26:17] Speaker 04:
We will give you a couple minutes for rebuttal.

[00:26:18] Speaker 02:
Thank you, Your Honor.

[00:26:26] Speaker 01:
Good morning, Your Honors.

[00:26:27] Speaker 01:
Beth Pacella for the Commission.

[00:26:29] Speaker 01:
I think I'm going to start with the argument that the petitioners actually made

[00:26:34] Speaker 01:
in their opening brief regarding individual analysis regarding accreditation and whether it's predictive.

[00:26:44] Speaker 01:
Their argument actually was at page 38 of their opening brief.

[00:26:52] Speaker 01:
For one thing, MISO's quantitative analysis does not look at the performance of individual resources, the appropriate test of whether the seasonal

[00:27:01] Speaker 01:
accreditation is an improvement over the prior system, UCAP, is whether the methodology is better or worse at predicting.

[00:27:10] Speaker 01:
And that's not the issue before the court.

[00:27:12] Speaker 01:
It wasn't the issue before the commission, and that's not the issue before the court.

[00:27:15] Speaker 01:
The issue is whether this proposal, this Section 205 proposal, is just unreasonable.

[00:27:21] Speaker 01:
So whether it's better or more or less predictive than before doesn't matter.

[00:27:27] Speaker 01:
So the individual issue was only raised in that context

[00:27:30] Speaker 01:
It was not in a different context.

[00:27:32] Speaker 01:
And so there's no answer in my brief to that because it wasn't raised to the commission and it wasn't raised in its opening brief.

[00:27:40] Speaker 01:
And as the commission found quite reasonably that there's not undue volatility here.

[00:27:46] Speaker 01:
The commission acknowledged that it's expected that accreditation volatility at the resource level will be higher than aggregate.

[00:27:53] Speaker 01:
level of volatility, but the commission found for several reasons that any volatility at the individual level would not be unreasonable.

[00:28:02] Speaker 01:
As the court already mentioned, any individual volatility would just reflect the availability in the past three years, historical availability.

[00:28:12] Speaker 01:
Secondarily, the commission extensively discussed the mitigation measures that would decrease any volatility that would occur.

[00:28:20] Speaker 01:
several aspects of the proposal that will mitigate volatility, and petitioners don't argue here that they're not, and I'd be happy to talk about each of them if you'd like.

[00:28:31] Speaker 01:
The third thing the Commission said was there are substantial benefits to the proposal, and fourth, that the benefits outweigh any volatility.

[00:28:40] Speaker 01:
So it's not that the Commission didn't recognize that there would be volatility, but the Commission recognized that the volatility would in the first place be appropriate, it would be extensively mitigated, and

[00:28:50] Speaker 01:
that the benefits of this proposal are incredibly important, critical reliability benefits were necessary if that volatility occurred.

[00:29:00] Speaker 00:
Can we go back to accuracy for a minute?

[00:29:04] Speaker 00:
Sure, absolutely, Your Honor.

[00:29:05] Speaker 00:
It seems like that's a critical issue in the system where 3% of the hours drive 80% of the calculation.

[00:29:17] Speaker 00:
Sure.

[00:29:17] Speaker 00:
The commission itself

[00:29:20] Speaker 00:
raised that question in the, I forgot what it's called, the compliance order.

[00:29:26] Speaker 00:
And Misa came back with this chart that I mentioned, which I have to say it looks somewhat flimsy.

[00:29:42] Speaker 00:
So what that chart is?

[00:29:45] Speaker 00:
11 days, two items, no indication that this is a representative, these data points are representative.

[00:29:55] Speaker 00:
How much confidence can we have in that?

[00:29:57] Speaker 01:
Tremendous confidence because as the Independent Market Monitor explained, this is compelling evidence.

[00:30:04] Speaker 01:
It is the Mid-Continent MISO took this

[00:30:11] Speaker 01:
It's exact methodology and it applied it to two historical circumstances of maximum generation events.

[00:30:19] Speaker 01:
And it showed that applying it in these two seasons, it had a deviation of only 0.8% in the summer season and minus 1.2% in the winter season.

[00:30:32] Speaker 01:
So this exact methodology has been proven to actually work.

[00:30:36] Speaker 01:
And nobody argued, particularly petitioners, that more season

[00:30:40] Speaker 01:
you know, more analysis should have been done, more extensive analysis should have been done.

[00:30:45] Speaker 00:
I take the forfeiture point.

[00:30:47] Speaker 00:
Maybe that's all you need, but sorry.

[00:30:52] Speaker 00:
How do we know that these two are sufficiently robust and not cherry picked?

[00:31:01] Speaker 01:
Um, so again, nobody argued that.

[00:31:05] Speaker 01:
So I don't know what the commission would have said in response to that.

[00:31:07] Speaker 01:
I'm sorry, your honor.

[00:31:08] Speaker 01:
But this commission's second point, um, which, you know, in the hearing order, um, j eight 15 paragraph 26 is that this methodology also considers all reasons, unlike the prior methodology.

[00:31:21] Speaker 01:
And so that in and of itself, it'll make it innately more innately predictive because it considers all reasons for, um,

[00:31:28] Speaker 01:
for having an outage not being available.

[00:31:32] Speaker 03:
And what's the precedent that this opinion is going to set when there are other RTOs that wish to transition to seasonal maintenance or adequacy construct?

[00:31:41] Speaker 01:
So I think that just like in any circumstance, if any other region wants to come in and do a similar thing and move to seasonal, if they find that the record supports their need to do so, like it did here,

[00:31:54] Speaker 01:
then, you know, you need to have real quantitative evidence to support the predictiveness of this.

[00:32:01] Speaker 01:
And you need and you have to show that it's just unreasonable.

[00:32:04] Speaker 01:
I mean, every circumstance can be different.

[00:32:06] Speaker 01:
Every region is different.

[00:32:08] Speaker 01:
It's not I don't know that another region have to do their own analyses.

[00:32:13] Speaker 01:
and have their own proof.

[00:32:14] Speaker 03:
Well, you've got Commissioner Clemens dissent, you know, on various factors.

[00:32:20] Speaker 03:
And it seems to go to that.

[00:32:22] Speaker 03:
Do you need to move beyond more than just the status quo?

[00:32:25] Speaker 03:
You know, is this just an incremental change?

[00:32:28] Speaker 03:
And then we see how it works out.

[00:32:29] Speaker 03:
And then we're right back here again.

[00:32:31] Speaker 01:
I think most of the dissenting commissioners concerns are the issues that weren't preserved here.

[00:32:41] Speaker 01:
So that's

[00:32:42] Speaker 01:
just a short answer, but I am happy to just talk in general.

[00:32:44] Speaker 01:
You know, the commission is allowed to do something that is predictive and then it's set forth this information report requirement for the very specific purpose of let's make sure that this is working.

[00:32:57] Speaker 01:
So there will be a check on this just like so responsibly the commission did that.

[00:33:03] Speaker 01:
It didn't just say we found this just unreasonable and we're done.

[00:33:06] Speaker 01:
It will be looking at this information report after the first three years of this methodology being used, and it will have that opportunity, and so will the public have that opportunity to determine whether things are sufficient.

[00:33:19] Speaker 04:
When FERC isn't making this up, it's presented to it by the independent system operator based on what they think their grid needs.

[00:33:28] Speaker 04:
And I suppose other ISOs may have different proposals and everyone may watch and learn from this.

[00:33:35] Speaker 01:
That's exactly right, Your Honor.

[00:33:36] Speaker 01:
That's right.

[00:33:36] Speaker 01:
So the regional operator is an expert in its own region.

[00:33:42] Speaker 01:
FERC is an expert on top of that and really

[00:33:47] Speaker 01:
petitioners don't present much.

[00:33:48] Speaker 00:
And the market monitor thought you didn't go far enough.

[00:33:52] Speaker 01:
Right.

[00:33:53] Speaker 01:
That's right, Your Honor.

[00:33:54] Speaker 03:
And what's your position with respect to criticisms regarding, like, for example, PGM, there was a much longer ramp-up period.

[00:34:04] Speaker 01:
So, and again, that's an issue that isn't an intervener-only issue, so that's not properly before the court, but I'm still happy to talk about it.

[00:34:12] Speaker 01:
The commission explained the difference.

[00:34:16] Speaker 01:
did explain the difference between those cases.

[00:34:20] Speaker 01:
The first one, the reliability pricing model order.

[00:34:25] Speaker 01:
There, the four-year transition period was to allow a brand new market.

[00:34:31] Speaker 01:
And so that's just one year longer, and it's an entirely new market versus just a changed portion of this market.

[00:34:39] Speaker 01:
And the second case, the capacity performance resource order.

[00:34:43] Speaker 01:
There, there was a proposed five-year transition period.

[00:34:48] Speaker 01:
But the difference here is the record establishes an urgent need to implement this in the near term.

[00:34:57] Speaker 01:
So allowing a five-year transition period of this waiting to get to 80-20 really is a viability issue.

[00:35:06] Speaker 01:
And it doesn't provide the incentives.

[00:35:08] Speaker 01:
The whole point of the 80-20 waiting is to give

[00:35:13] Speaker 01:
market participants incentive to have their capacity that they commit to the market and are paid to provide actually be available.

[00:35:20] Speaker 01:
I mean, that's the whole point of having capacity market is about reliability and making sure that there are, that it's available.

[00:35:30] Speaker 01:
As this court explained, excuse me, as the court explained, in some case, I apologize, the whole purpose of it is

[00:35:42] Speaker 01:
is that, and it's in my brief.

[00:35:44] Speaker 01:
And I think there's just one more thing I think the court understands my point.

[00:35:48] Speaker 01:
I just wanted to talk about maintenance margin for a moment.

[00:35:52] Speaker 01:
You know, the commission explained and petitioners do not rebut the fact that maintenance margin does not provide a complete picture of system reliability.

[00:36:02] Speaker 01:
And that's because it considers only planned

[00:36:05] Speaker 01:
not any other type of outage.

[00:36:06] Speaker 01:
And the fact is, if parties are waiting, that's a big point about why you need 120 days notice.

[00:36:13] Speaker 01:
Because if somebody gives enough notice, then it's part of the maintenance margin analysis.

[00:36:17] Speaker 01:
So maintenance margin doesn't tell you, it only tells you what the ISO knows is already has been announced as a planned outage.

[00:36:28] Speaker 01:
But if people have later planned outages, that's not part of the maintenance margin analysis on top of

[00:36:34] Speaker 01:
everything else the commission said.

[00:36:37] Speaker 04:
There does seem some tension, though, with rigid rules around maintenance and going offline for care and tending to these really important facilities.

[00:36:52] Speaker 04:
You don't want to encourage, you don't want to incentivize them to keep operating if they really need to be doing some maintenance.

[00:37:00] Speaker 04:
It's fine balance.

[00:37:03] Speaker 04:
here and as I understand the plan, there's a couple, the theory is they can always go ahead and do their maintenance.

[00:37:11] Speaker 04:
They just might either have to, there are different options.

[00:37:14] Speaker 04:
One of them is to opt out of the capacity market.

[00:37:18] Speaker 04:
That's right, Your Honor.

[00:37:19] Speaker 04:
Does that mean just not bidding in the first place or is it something you do after you've already bid in?

[00:37:26] Speaker 01:
You don't commit yourself to the market.

[00:37:27] Speaker 01:
You don't offer your capacity to the market.

[00:37:29] Speaker 04:
You can't commit and then somehow get back out.

[00:37:31] Speaker 01:
it's just not joining the market.

[00:37:33] Speaker 01:
And that is a pro-reliability thing.

[00:37:37] Speaker 01:
The commission, the point of this proposal is not to have the most potentially available capacity in the market, it's to have more certainly available capacity in the market.

[00:37:46] Speaker 04:
Well, they don't know far enough in advance to opt out for an entire season of capacity, which seems to be a very, very difficult decision for them to make, obviously.

[00:37:56] Speaker 04:
And they wanted to,

[00:37:58] Speaker 04:
something came up and they really thought they should deal with it now, but it was going to take more than 31 days.

[00:38:04] Speaker 04:
They need to deal with it sort of mid-season.

[00:38:06] Speaker 04:
The plan says that they can acquire replacement capacity.

[00:38:11] Speaker 04:
Can you just explain to me how that happens?

[00:38:14] Speaker 04:
Do they buy it off one of the other, I don't know, a day-end market?

[00:38:19] Speaker 04:
Or how do they get this replacement capacity?

[00:38:21] Speaker 04:
Because they themselves are replacement capacity.

[00:38:23] Speaker 01:
The commission spoke to that in the first order.

[00:38:27] Speaker 01:
and in the hearing order as well.

[00:38:29] Speaker 01:
Those with a large number of resources can rely on their fleet to provide that capacity.

[00:38:35] Speaker 04:
No, I was talking about would they acquire replacement capacity from some other?

[00:38:38] Speaker 01:
Well, this is how they do it.

[00:38:40] Speaker 01:
Sometimes you can just self-supply.

[00:38:42] Speaker 01:
Okay.

[00:38:43] Speaker 01:
So with resources that have smaller, you know, entities that have smaller resources, they do it through bilateral agreements and the capacity auction.

[00:38:52] Speaker 01:
And the commission explained, and the record showed,

[00:38:55] Speaker 01:
that excess capacity is expected to be available because as Mr. McFarland's evidence showed, his testimony showed, under the proposal, capacity reserve requirements are going to be generally lower in non-summer months because it's now not going to be the annual one-day peak summer time.

[00:39:16] Speaker 01:
Each season will have its own capacity reserve requirements.

[00:39:22] Speaker 01:
So they'll be lower and there's likely to be excess capacity for those seasons that entities will be willing to sell.

[00:39:29] Speaker 01:
And the evidence actually showed that the move to seasonal construct will result in nearly double the amount of uncleared zonal resource credits in non-summer season.

[00:39:41] Speaker 01:
So there will be liquidity in the market for replacement capacity.

[00:39:48] Speaker 04:
Thank you very much.

[00:39:49] Speaker 04:
Thank you.

[00:39:52] Speaker 04:
Okay, it's been Barabanov.

[00:39:56] Speaker 04:
Did I say that correctly?

[00:39:56] Speaker 02:
Yes, just a few brief questions in rebuttal, Your Honor.

[00:40:00] Speaker 02:
I'm going to give you two minutes.

[00:40:02] Speaker 02:
I just wanted to say that.

[00:40:03] Speaker 02:
Thank you.

[00:40:04] Speaker 02:
I apologize.

[00:40:05] Speaker 02:
No problem.

[00:40:06] Speaker 02:
Okay, so going back to your question, Judge Millett, and one that you raised, Judge Childs, about whether or not there were alternatives proposed, I just wanted to point out that this was a 205 proceeding in which FERC was supposed to just either approve or disapprove of the MISO proposal.

[00:40:22] Speaker 02:
the comments were made in that context.

[00:40:26] Speaker 02:
Second point, Your Honor, is that, again, with respect to the definition of a maintenance margin, that there really isn't an adequate explanation for why that is not sufficient here based on FERC's own definition, either in their briefs or in the decision we're here today.

[00:40:43] Speaker 04:
Because by definition, it's not our explanation was it doesn't work.

[00:40:46] Speaker 04:
One, if we don't know well in advance who's going to be using it.

[00:40:50] Speaker 04:
Because everyone could go, aha.

[00:40:52] Speaker 04:
You could have some large percentage of the capacity for any given season say, aha, we're going down in July for maintenance.

[00:41:00] Speaker 04:
That's our best time to do this maintenance.

[00:41:03] Speaker 04:
But if maintenance one doesn't know what the other is doing and suddenly they've got a large percentage of the capacity offline all at once.

[00:41:11] Speaker 04:
But if each one individually looked at maintenance margin, not knowing what any other supplier was going to be doing.

[00:41:17] Speaker 02:
Isn't that the problem?

[00:41:18] Speaker 02:
Well, maintenance margin does take into account what other people are doing.

[00:41:21] Speaker 04:
If you don't challenge 120-day advance notice, but if they don't have enough advance notice, then if everyone shows up on the same day thinking nobody else was going to be doing it, it's going to be a problem.

[00:41:32] Speaker 04:
It's not going to work.

[00:41:34] Speaker 02:
Your Honor, the definition says that it will not impact the adequacy of supply when there's positive maintenance margin.

[00:41:41] Speaker 04:
They have to know well in advance who is planning to do their maintenance

[00:41:47] Speaker 04:
the first week of July.

[00:41:49] Speaker 04:
And I assume that suppliers don't all talk to each other and say, hey, I'm going to take the first week.

[00:41:55] Speaker 04:
You want to take the second week.

[00:41:56] Speaker 04:
I assume they just simply lob in their requests.

[00:41:59] Speaker 02:
Well, yeah.

[00:42:00] Speaker 02:
And maintenance margin is adjusted based on those requests that come in.

[00:42:04] Speaker 04:
But if they all come in simultaneously, what happens?

[00:42:07] Speaker 02:
Then they need to be vetted and approved based on what the margin is going to be.

[00:42:12] Speaker 02:
And that changes over time.

[00:42:15] Speaker 02:
Going also to the SAC methodology.

[00:42:17] Speaker 02:
What about surprise forced outages?

[00:42:20] Speaker 04:
Well, that also goes to- There's a lot of balancing they have to do here, and it's a lot to say that, and the old scheme was where they left it up to the suppliers to make their own decisions about this.

[00:42:33] Speaker 04:
They only considered forced outages, and that's what was not working for them.

[00:42:37] Speaker 04:
So I'm just having a little trouble understanding what's so wrong about them imposing some more constraints and rules about the process as long as

[00:42:47] Speaker 04:
anybody who needs to do repairs can do them whenever they need to do them through these alternative methods or spreading it over two seasons.

[00:42:53] Speaker 04:
That's what I'm having trouble understanding.

[00:42:56] Speaker 02:
Your honor, there's nothing wrong with sort of imposing some constraints per se, but they should be consistent with the definitions that FERC has already previously improved.

[00:43:04] Speaker 04:
They had maintenance margin in the old scheme, and that wasn't enough.

[00:43:08] Speaker 04:
Maintenance margin is an invention of this scheme, right?

[00:43:10] Speaker 04:
They had that last time, that wasn't enough to protect them.

[00:43:14] Speaker 02:
Your honor, but they have to square their reasoning to say it doesn't provide a complete picture.

[00:43:18] Speaker 02:
It doesn't support penalizing generators for engaging in prudent maintenance.

[00:43:22] Speaker 02:
Because in some cases, the record is clear that longer than 31 days can be prudent.

[00:43:27] Speaker 02:
And just to make a couple really quick points on the methodology, your honor, the point that now the 65 hours accounts for forced outages as well as planned outages and that that makes it better,

[00:43:43] Speaker 02:
that doesn't change the fundamental problem, really, right?

[00:43:46] Speaker 02:
Which is that there is this very small sample size of the 65 hours a season and heavy weighting.

[00:43:52] Speaker 02:
And so what's considered doesn't really change that fundamental problem and that the results are going to be plagued by randomness.

[00:44:01] Speaker 02:
Thank you.

[00:44:02] Speaker 04:
Any questions?

[00:44:03] Speaker 04:
Thank you very much for your time, Council.