Docket Nos. ER22-495-002, ER22-495-003

I dissent because I continue to believe that the seasonal capacity proposal put forth by Midcontinent Independent System Operator (MISO) fails to yield just and reasonable rates for the region.  While the proposal is a step in the right direction in that it seeks to value capacity in a manner that more closely aligns with the reliability risks the region faces, a proposal is not just and reasonable merely because it is an improvement over the status quo.[1]  The Commission’s order on rehearing, like the underlying order it modifies, glosses over MISO’s failure to adequately justify key details of its proposal.  The Commission’s failure to rigorously enforce the requirement that rates be just and reasonable is particularly troubling in this case, given that it will no doubt form the basis for future Commission reasoning on capacity accreditation matters. 

As I noted in my dissent to the underlying order,[2] MISO’s proposal contains a number of important defects that together result in a framework that fails to align capacity payments with ensuring availability at the times MISO’s system is at greatest reliability risk.  Today I highlight two of the most problematic design flaws in MISO’s proposal, which the Commission’s order approves without adequate explanation.

MISO’s proposal uses the wrong set of hours in crediting resources

MISO asserts that its new method of crediting resources aims to assign capacity value of a resource for each season based on its historical performance during the 65 hours when emergency conditions or tight operating conditions were present.[3]  But as I previously explained, MISO focuses on the wrong set of hours, thereby failing to ensure reliability at just and reasonable rates.[4]  

MISO determines the relevant hours by calculating an “operating margin,” which is resource supply minus demand at any point in time.[5]  But in determining resource supply, MISO includes resources that are not capable of responding when MISO may need them.  Specifically, it includes within the operation margin offline resources that have a lead time of 12 hours or less.[6]  But in response to MISO’s proposal, the IMM and the Clean Energy Coalition explained that emergency conditions generally develop along a significantly shorter timeframe than 12 hours, meaning that MISO will count resources as available that are not, in fact, capable of responding.[7]

In defense of its position, the only explanation MISO gave is that its choice of a 12-hour lead time was better than an alternative of 24 hours, which would have included even more resources incapable of delivering capacity when needed.[8]  But the Federal Power Act is not a Price is Right Showcase Showdown, and the fact that a proposed rate is closer than an unjust and unreasonable option does not demonstrate it to be just and reasonable.[9]  $100 for a gallon of milk is not a fair price, and the fact that $50 is a better alternative does not make it reasonable.

In requesting rehearing, the Clean Energy Coalition points out the non-responsive nature of MISO’s explanation: “Stating that 12 hours is better than 24 hours . . . does not show that 12 hours reaches an acceptable level of adequately measuring a resource’s ability to mitigate actual system risk.”[10]  Yet in today’s rehearing order, the Commission simply doubles down in accepting MISO’s non-responsive answer.[11]  It fails to engage with the arguments of the parties requesting rehearing, violating the Administrative Procedure Act’s requirement to “respond meaningfully to objections raised by a party” that are not “frivolous on their face.”[12]  

Nor is the Commission’s failure to engage with the parties’ arguments merely a procedural failure.  Approving MISO’s approach rather than insisting on a reasonably accurate proposal will result in unjust and unreasonable rates that could jeopardize grid reliability.[13]  As I explained in my dissent to the underlying order, it means that MISO may “ignore hours where offline resources with a lead time between 6 and 12 hours are comparatively abundant, when in fact uncertainties like load forecast errors, wind forecast errors, large unit forced outages, and other factors that are not known far in advance pose a reliability risk for the system for which these resources are of no use.”[14]

MISO’s proposal unjustifiably credits resources that are unable to serve the system

MISO’s approach to capacity accreditation has two steps: first, determine the hours over which to measure historical performance of resources, and next, assess the historical performance of resources in those hours.  Beyond looking to the wrong set of hours, as explained above, MISO also assesses historical performance in a manner that credits resources for capacity they are unable to provide.  Its capacity accreditation approach is even less accurate than its approach to calculating the relevant hours to measure whether a resource gets credit.  Two flaws stand out: (i) MISO provides capacity credit to offline resources with response times as slow as up to 24 hours; and (ii) it exempts from capacity accreditation outages for which it has 120 days of notice, so long as MISO confirms a “maintenance margin” in advance.

MISO credits offline resources with up to 24-hour lead times without adequate justification

Incredibly, while MISO’s only defense of using 12 hours as the lead time threshold for including resources in its calculation of operating margin is that doing so is more accurate than using a 24-hour lead time, it proposes to use the even-less-accurate 24-hour lead time when determining which resources get credit for delivering capacity. 

MISO’s purported rationale for this choice, and the Commission’s sole basis for accepting it, is that a “24-hour threshold assumes the resource could have reasonably been economically committed if needed.”[15]  But as the Clean Energy Coalition’s Request for Rehearing states, “whether a resource could have been committed if MISO knew about a reliability situation in advance is beside the point when assessing contribution to tight operating hours that arise on a shorter timeframe.”[16]  And as MISO itself admits, most such events do in fact occur after the day-ahead market clears.[17]  As the Independent Market Monitor explains, “[o]ffline resources that could theoretically have been committed the prior day, but whose lead time is too long for them to start to mitigate the reliability threat provide no value in such events.”[18]

Again, the Commission fails to engage with these arguments.[19]  While it points to “MISO’s expert opinion that offline resources with lead times of less than 24 hours could reasonably be committed” in day-ahead markets, the order fails to respond to the contention that such a commitment process cannot generally be deployed during the relevant events.[20]  It does not, for instance, contend that the Clean Energy Coalition is wrong that a shorter response time is generally needed.[21]

That the day-ahead markets are able to commit resources with a lead time of less than 24 hours is not in dispute.  That is a simple fact of how markets operate.  But the Clean Energy Coalition notes that for the relevant reliability events, that process will already have occurred.  Some generation resources will have been committed and will be running, and accordingly will receive capacity credit.  The question is what offline resources should be credited.  Clean Energy Coalition contends that MISO’s choice to credit resources that take up to 24 hours to start up will provide credit to many resources that will be unable to respond during the relevant reliability events because those occur on much shorter notice.  The Commission obscures the issue by referring to multiple day forecasts and other complicated descriptions of MISO’s capabilities,[22] but ultimately violates the Administrative Procedure Act because it does not respond to the central claim that many of the resources meeting MISO’s 24-hour threshold will not provide reliability value to the system if they are offline during the relevant RA Hours.[23]

“MISO’s expert opinion,” cited by the Commission, appears to be only that resources with a 24-hour lead time are able to start through day-ahead commitment processes if the need is known at the time those processes operate.[24]  The Commission order presents its own theory, not clearly articulated by MISO’s own filings or testimony, that because MISO has instituted a practice of issuing a Capacity Advisory when the projected system margin is less than 5% two or three days ahead of potential shortage conditions, MISO may have “additional insight . . . regarding the need to commit long lead time resources in advance of known reliability events.”[25]  But MISO does not contend that these days-in-advance projections are precise enough to predict the times of true need.  To the contrary, it states only that “[h]istorically most MaxGens have been driven by events occurring after the Day-Ahead Market clears[.]”[26]

Nor does the Commission confront the inconsistency in MISO’s approach.  As described above, MISO chose a shorter 12-hour lead time when determining what hours over which capacity accreditation should be measured because of “uncertainty that occurs after the deadline to commit longer lead resources has passed.”[27]  The parties requesting rehearing questioned why a different approach would be used for accreditation than for choosing the relevant hours over which to calculate credit, “given that both exercises are meant to examine risk in the system.”[28]  In response, the Commission e simply explains that the two methods are different, but does not explain why MISO’s use of a different method for each is appropriate,[29] in violation of the Commission’s obligation to engage in reasoned decisionmaking.[30]  Further, the Commission’s decision is internally inconsistent because the reasons for using a 12-hour lead time for determining what hours over which to measure apply equally to the choice of lead time used for measuring accrediting the capacity of resources, and the Commission has not explained why the purported rationale for 24 hours is only relevant to accreditation and not choice of hours.[31]

As with MISO’s use of a 12-hour lead time in determining which hours to focus on, the flaws in MISO’s approach to addressing historical performance are not merely procedural.  The Federal Power Act is violated along with the Administrative Procedure Act because MISO’s approach of crediting resources that cannot perform charges unjust and unreasonable rates to capacity consumers and compromises system reliability. 

MISO inappropriately exempts scheduled outages from its capacity accreditation calculations 

MISO’s approach to accreditation is also flawed because it excludes from the capacity accreditation calculation resource outages where, among other things, the relevant resource has given 120 days of notice to MISO.[32]  The Commission approves this proposal because it “encourages resource owners to give MISO more visibility and advance notice of available resources, which is needed in order to ensure resource adequacy in future periods.”[33]

The problem with MISO’s proposal, and the Commission’s logic in approving it, however, is that the proposal extracts far too high a price from capacity consumers.  MISO’s approach entirely excuses resource outages for up to 31 days in a season in which a resource clears, requiring replacement capacity from that resource only if it exceeds this threshold and is thus unavailable for more than a third of the relevant period.[34]

In characterizing Clean Energy Coalition’s concerns as “overstated,”[35] the order misses the forest for the trees.  While it clarifies that MISO will “exclude hours in which the unit is offline” from its accreditation calculations rather than crediting resources for performing during those hours,[36] the fact remains that the proposal asks customers to pay for capacity that is not available, so long as notice is given of that unavailability.  This means that if one resource is available every single day of a season, whereas another is unavailable for 31 days but available thereafter, both would be given full capacity credit. 

In other words, while the outage days are removed from the accreditation calculation, they are not removed from the calculation of the amount of capacity customers are required to pay for, so a resource on outage for up to 31 days will receive the same clearing price in a seasonal auction as one that is fully available.  “Notably missing” from the Commission’s rationale in approving MISO’s proposal, is “any rebuttal to comments by the Clean Energy Coalition and other intervenors questioning whether the capacity accreditation benefit offered to generators with planned outages is commensurate with the reliability benefit of granting exemptions to those generators.”[37]  There were myriad ways that MISO could encourage generators to provide notice of anticipated outages in a manner consistent with requirements of the Federal Power Act, but crediting a resource as though an outage has not occurred is not among them. 

MISO’s approach is unduly discriminatory because it may provide for the same capacity compensation “to resources that require maintenance every 120 days,” and those that require it “every 1200 days.”[38]  And it is not just and reasonable because it provides inordinate compensation to resources not commensurate with the benefits they provide to the system.[39]

Moreover, as emphasized in my dissent to the underlying order, MISO’s approach sends perverse incentives to capacity resources.  It discourages short to medium term maintenance for which a full 120 days notice can’t be given but a resource owner could have notified MISO in advance, because any form of notice short of 120 days receives no similar exemption.  It also “discourages non-consecutive outage days,” and arbitrarily disincentivizes any maintenance from occurring during a 120-day period following each noticed outage window (for which exemptions are not available).[40]  And because resource owners must secure replacement capacity for outages of more than 31 days in a single season, resources are encouraged “to arbitrarily stack outage requests at the last 31 days of the season leading into the first 31 days of the next,” exposing MISO to reliability risks if an unexpected weather event occurs at the beginning or tail end of a season.[41]  Today’s order provides no justification for these features of MISO’s proposal.

The Commission’s capacity accreditation decisions will bear on reliability outcomes in the future

Despite my view that MISO has not adequately justified its proposal, I share the perspective of my colleagues that it was laudable for MISO to seek to improve upon its outdated capacity accreditation framework.  As I note in my separate statement to the RENEW order issued concurrently during the Commission’s open meeting this month, it is clear that the today’s markets must be designed to address increasingly complex reliability challenges.[42]  Although MISO’s proposal fell short of the mark, this does not suggest that changes to MISO’s resources adequacy rules are not appropriate.  To the contrary, further changes appear necessary.

However, the Commission had other options to respond to MISO’s proposal, and  papering over the proposal’s clear design flaws as today’s order does will make the progression to a modern market design more difficult.  Consistent with setting the region on a path to greater reliability, the Commission could have instead crafted a narrow rejection order with clear guidance, as I argued in my prior dissent.[43] 

Alternatively, the Commission could have issued an Order to Show Cause under Federal Power Act section 206.  Such a procedural vehicle would be useful because, in putting forth its proposal, MISO and its stakeholders identify numerous flaws in the status quo in need of improvement.  An Order to Show Cause could have required MISO to explain or address those flaws in a more proactive way, without the constraints of accepting a filing as proposed—and without modification—under FPA section 205 and NRG v. FERC.[44]  In approving MISO’s proposal even though it fails to meet the FPA standard for rates to be just and reasonable and not unduly discriminatory, the Commission weakens that very standard.  Unfortunately, in so doing, the Commission compromises its own ability to solve problems using its power under FPA section 206. 

For these reasons, I respectfully dissent.

 

[1] See PJM Interconnection, L.L.C., 180 FERC ¶ 61,089, at P 47 n.111 (2022) (finding that even if PJM’s contention that its Intelligent Reserve Deployment proposal is an improvement over its current approach is correct, that does not render the proposal just and reasonable).

[2] Midcontinent Indep. Sys. Operator, Inc., 180 FERC ¶ 61,141 (2022) (Clements, Comm’r, dissenting) (Clements Aug. 2022 Dissent).

[3] MISO Transmittal at 16.

[4] Clements Aug. 2022 Dissent at P 8.

[5] See, e.g., MISO Transmittal at 44, 46.

[6] Midcontinent Indep. Sys. Operator, Inc., 180 FERC ¶ 61,141, at P 255 (2022) (August 2022 Order).

[7] See IMM Jan. 2022 Comments at 8 (“[T]he lead time should be set at a level that will only include the units that are typically available to be utilized when MISO experiences tight conditions that can lead to warnings, alerts and emergencies . . . . Therefore, we recommended that MISO adopt a lead time between 2 and 6 hours, which is the timeframe in which MISO often recognizes that emergency conditions are developing.”); Clean Energy Coalition Jan. 2022 Comments at 11 (“MISO’s approach of allowing resources with a 12-hour lead time to count as available for purposes of determining the system’s margin in a given historic hour fails to consider how unexpectedly tight hours typically arise in the range of 2 to 8 hours because reality diverges from forecasts.”).

[8] See August 2022 Order at P 255 (“MISO explains that, although it initially considered proposing a 24-hour lead time, it determined that, due to the operational concerns it faces, tight conditions are more likely to be identified 12 hours before a projected capacity shortfall than 24 hours ahead of time.”) (citing MISO Transmittal Tab C (Testimony of Shawn McFarlane), at 23-24 (McFarlane Test.)).

[9] Although even under Price Is Right rules, MISO’s proposal wouldn’t have met the mark, as it proposes a lead time threshold that exceeds the reasonable amount based on the record.  See The Price is Right Live! Stage Show, Official Rules (Fall 2022), 9, 10, 12, 15, https://priceisrightlive.com/wp-content/uploads/2022/09/TPiRL-rules-8_19_22.pdf (explaining that the contestant who bids closest to the retail price of the item without exceeding the price of the item wins the chance to advance to the next stage).

[10] Clean Energy Coalition Request for Rehearing at 16.

[11] See Majority Order at P 45 (approving MISO’s proposal based on its justification that a 12-hour lead time is a better choice, and failing to offer any other rationale in support of the proposal).  

[12] Constellation Mystic Power, LLC v. FERC, 45 F.4th 1028, 1054-55 (D.C. Cir. 2022) (citations omitted).  See also PSEG Energy Res. & Trade LLC v. FERC, 665 F.3d 203, 429-30 (D.C. Cir. 2011) (granting a petition for review because FERC failed to “answer[] objections that on their face seem legitimate” and its decision therefore could “hardly be classified as reasoned”) (citations omitted).

[13] See ISO New England, Inc., 147 FERC ¶ 61,172, at P 23 (2014) (finding capacity market rules to be unjust and unreasonable where capacity payments were not adequately aligned with a resource’s true contribution to system reliability).

[14] Clements Aug. 2022 Dissent at P 13. 

[15] MISO February Answer at 43.  See Majority Order at P 57 (accepting MISO’s proposal based on this explanation).

[16] Clean Energy Coalition Request for Rehearing at 18 (quoting Clements Aug. 2022 Dissent at P 17).

[17] Id.

[18] IMM Jan. 2022 Comments at 10.

[19] Majority Order at P 57.

[20] Id.

[21] While the Majority Order faults the Clean Energy Coalition for failing to provide “accompanying record evidence showing an analysis of typical timeframes in an emergency that would render the 12-hour lead time unjust and unreasonable,” this puts the evidentiary burden on the wrong party.  Majority Order at P 45 n. 118.  MISO has the obligation to demonstrate its proposal to be just and reasonable, and MISO has failed to present any such evidence.

[22] See Majority Order at P 57.

[23] See supra P 6 & n. 13 (“[T]he Administrative Procedure Act[]” requires the Commission to “‘respond meaningfully to objections raised by a party’ that are not ‘frivolous on their face.’” (quoting Constellation Mystic Power, LLC, 45 F.4th at 1054-55)).

[24] See McFarlane Test. at 24.

[25] Majority Order at P 57. 

[26] MISO February Answer at 43 (emphasis added).  Given this clear statement from MISO, it is not reasonable to infer that MISO can accurately project MaxGen events days in advance.  Nor is it necessarily relevant that some of the events may last multiple days, as MISO has not contended that it can accurately predict the duration of an event more than 24 hours in advance of it occurring.

[27] Majority Order at P 45 (citing McFarlane Test. at 23).  As explained above, while more accurate than the 24-hour lead time used for accrediting resources, this lead time was still unjustified in the face of evidence that the relevant reliability events occur on shorter notice.  See supra section A.

[28] Clean Energy Coalition Request for Rehearing at 19.

[29] See Majority Order at P 58 (explaining that MISO has chosen to use a method for accreditation that focuses on the lead time required for commitment in the day ahead market and a method for determining RA hours that “capture[s] periods of greatest stress,” but not explaining why having inconsistent methods is appropriate).  If resources with lead times between 12 and 24 hours are not relevant to the MISO system in its “periods of greatest stress,” it is unclear how it is appropriate to provide them with full capacity credit.

[30] The APA’s reasoned decisionmaking requirement dictates that “the duty to explain inconsistent treatment is incumbent on the agency.”  BG&E v. FERC, 954 F.3d 279, 285 (D.C. Cir 2020).

[31] An agency’s reasoning “cannot be internally inconsistent[.]”  ANR Storage Company v. FERC, 904 F.3d 1020, 1024, (D.C. Cir 2018) (citing Sierra Club v. EPA, 884 F.3d 1185, 1194–96 (D.C. Cir. 2018)).

[32] See Majority Order at PP 54-56 (approving this facet of MISO’s proposal).

[33] Id. P 54.

[34] Id. P 56.

[35] Id.

[36] Id.

[37] Clean Energy Coalition Request for Rehearing at 22-23.  See Majority Order at PP 54-56 (failing to provide any argument that the reliability benefit of the notice exemption is commensurate with its cost).  The Majority Order explains that the notice provisions were inserted in response to stakeholder concerns.  Id. P 54.  But while it is unsurprising that resource owners facing reduced capacity credit due to the need for frequent maintenance would request an exemption, that does not excuse MISO’s burden to demonstrate that such an exemption is just and reasonable and not unduly discriminatory. 

[38] Clean Energy Coalition Request for Rehearing at 21.  See Alabama Elec., Inc. v. FERC, 684 F.2d 20, 27-28 & n.3 (D.C. Cir. 1982) (“While the typical complaint of unlawful rate discrimination is leveled at a rate design which assigns different rates to customer classes which are similarly situated, a single rate design may also be unlawfully discriminatory” where it creates an “undue disparity”).  See also Calpine Corp. et al. v. PJM Interconnection, L.L.C., 163 FERC ¶ 61,236, at P 68 n. 112 (2018) (“Typically, undue discrimination cases involve a seller charging a different rate to similarly-situated customers; but undue discrimination can also occur when a seller charges the same rate to differently-situated customers.”).  

[39] See ISO New England, Inc., 147 FERC ¶ 61,172 at P 23 (finding capacity market rules to be unjust and unreasonable because they failed “to provide adequate incentives for resource performance, thereby threatening reliable operation of the system and forcing consumers to pay for capacity without receiving commensurate reliability benefits”).

[40] Clements Aug. 2022 Dissent at P 23.

[41] Id. 

[42] RENEW Northeast, Inc. and the American Clean Power Association v. ISO New England Inc., 182 FERC ¶ 61,085 (2023) (Clements, Comm’r, concurring, at PP 3-4).  See also Clements Aug. 2022 Dissent at PP 3-4.

[43] Clements Aug. 2022 Dissent at PP 2, 7.  See also RENEW Northeast, Inc., 182 FERC ¶ 61,085 (Clements, Comm’r, concurring, at P 8 n.14).

[44] 862 F.3d 108 (D.C. Cir. 2017).

Contact Information


This page was last updated on February 16, 2023