RASC: Proposed RBDC Opt Out Adder and Draft Tariff (RASC-2019-8) (20230808)

WPPI provides below our initial comments on MISO’s draft tariff changes to institute the RBDC and associated opt out. While MISO’s changes are a good start, we see a large number of issues that require further work, and we expect to have significant comments on the next draft as well. In this light, MISO’s proposed filing schedule appears very tight. We encourage MISO to set its schedule based on progress on resolving these issues and on drafting appropriate tariff language to address them.

I note also that we agree with the calculation elements described in David Sandefur’s feedback posted 8/17.

1. MISO proposes to introduce two new terms: ‘Initial PRMR’ and ‘Final PRMR’. This complicates and confuses the tariff much more than necessary. 
1. For example, maintaining the current Module A definition of PRMR as the amount that satisfies an LSE’s RA requirement would appear duplicative of MISO’s intended definition for Final PRMR
2. We suggest the following changes:
1. Eliminate the proposed new defined term ‘Initial PRMR’
1. This approach will require MISO to revisit its review each instance of ‘PRMR’ in Module E-1, so as to determine which should be changed to ‘Final PRMR’
2. Revise the Module A definition of PRMR to specify that it is the product of (1 + PRM) and “LSE’s forecasted Coincident Peak Demand for each Season,” as in the existing-tariff description of PRMR in 68A.7
3. Define ‘Final PRMR’ as follows, similar to MISO’s proposed Module A definition:
1. For LSEs not selecting the Reliability Based Demand Curve Opt Out, Final PRMR is the capacity [or ZRC] obligation established as a result of the PRA clearing using RBDC. For LSEs selecting the RBDC Opt Out, Final PRMR is the product of LSE’s forecasted Coincident Peak Demand for each Season and (1 + PRM + RBDC Opt Out Adder).
2. Our understanding is that MISO’s intent is not to change the PRMR applicable to LSEs opting to pay the Capacity Deficiency Charge (CDC). Accordingly, it may be appropriate to define a Final PRMR equal to PRMR (defined as described above) for obligations to be covered by paying the CDC.
4. Alternatively, PRMR for LSEs selecting the RBDC Opt Out could be defined as:
1. PRMR plus the product of the RBDC Opt Out Adder and the LSE’s forecasted Coincident Peak Demand for each Season
5. (MISO could, alternatively, eliminate the definition of ‘Final PRMR’—using ‘PRMR’ for this purpose instead—and instead establish ‘Initial PRMR’ as the product of coincident peak demand and PRM, and adjust all instances of PRMR in Module E-1 accordingly.)
1. Our view, however, is that the new concept introduced by the RBDC is, in essence, the ‘Final PRMR’ concept, and thus it would be appropriate to introduce this as a new term and leave ‘PRMR’ to mean the quantity as currently calculated, eliminating the need for ‘Initial PRMR.’
3. MISO’s proposed Module A definition of ‘RBDC Opt Out Adder’ specifies that it must be “equal to or greater than zero”
1. We would suggest that is a specification of this quantity, rather than part of the term definition, and would be more appropriately included in 69A.9.1 than in Module A
4. Section 69A describes MISO’s calculation of [Initial] PRMR. We note that correct calculation of this value requires use of the proper applicable PRM, which may be affected by state regulatory authority decisions, which in turn would likely be informed by MISO’s determination of the RBDC Opt Out Adder, which may not be established until December 15 (per 69A.9.1.k). Accordingly, in order to meaningfully preserve the rights of state regulatory authorities, MISO should ensure that [Initial] PRMR is not calculated until sufficiently after December 15 to allow state regulatory authorities to exercise their authority under 68A.1.
5. Proposed new language in section 69A suggests that there is some post-PRA showing that LSEs need to make to demonstrate that they are not short. In practice, short LSEs are simply assessed net settlement charges based on their cleared ZRCs and their obligation as determined by PRA clearing. That process appears to be already fully described in 69A.7.6, with no need for reference to some additional process in 69A.
6. MISO refers in proposed new section 69A language to “LSEs that submit ZRCs for an RBDC Opt Out and/or pay the Capacity Deficiency Charge to meet all of their [Initial] PRMR.” It appears to us that these options are mutually exclusive, and thus that the “and/or” language should be replaced with a simple “or.” Does MISO concur?
7. MISO says in proposed new section 69A language that “LSEs that submit an RBDC Opt Out may not use any of the other options to satisfy their Initial PRMR.” This appears confusing to the extent that such LSEs must satisfy their Final PRMR, as this term is defined in proposed MISO language, and not merely their Initial PRMR.
8. MISO says in proposed new section 69A language that “LSEs not submitting an RBDC Opt Out may use one or more of all other available options in combination to satisfy their Initial or Final PRMR.” We find this “Initial or Final PRMR” phrase potentially confusing and suggest replacing it with “RAR.”
9. MISO says in existing section 69A language that “An ACP will be determined through the PRA process for each LRZ and ERZ for each Season and the ACP will be used to credit ZRCs that clear in the auction and to debit LSEs for the volume of their PRMR that is procured through the auction.” This section should clarify what happens when the PRA clears such that zonal share of Final PRMR < LCR.
10. Proposed section 69A.9.1 refers repeatedly to “PRM plus RBDC Opt Out Adder.” It seems to us that this is precisely the kind of situation for which ‘Final PRMR’ is defined, and that this term provides a much more concise way of referring to the RBDC Opt Out requirements. We recognize that PRM and RBDC Opt Out adder are both expressed in percentage terms, while Final PRMR is expressed in MW terms, and believe that—while either could potentially work here—the latter actually appears more appropriate.
11. Incorporating the latest proposed changes, the language in 69A.9.1.b appears entirely duplicative of language in 69A.9.1.c. We suggest reviewing all this language with an eye towards expressing individual elements of the opt-out proposal as concisely as possible.
12. Proposed section 69A.9.1.c says that “The RBDC Opt Out must include the LSE’s seasonal forecasted Coincident Peak Demand for each LRZ….” This appears unnecessary given the existing 69A.1.1.a requirement to submit demand forecast data, which should thus already be included in MECT. Does MISO think it needs this language here for some reason?
13. At 69A.9.1 MISO proposes a role for itself to enforce restrictions that a regulator may impose on LSE selection of the RBDC Opt Out. This may be reasonable and appropriate, but MISO has yet to make this case to stakeholders. To the extent that MISO intends to retain this provision, MISO should explain its thinking explicitly at RASC.
14. Proposed paragraph 69A.9.1.j specifies the consequences for an RBDC Opt Out that does not satisfy the relevant Final PRMR
1. It notes that affected LSEs will be responsible to pay the Capacity Deficiency Charge (CDC) amount for shortfalls, where CDC is defined strictly in annual terms in 69A.10.
1. Does MISO propose to prorate the annual Capacity Deficiency Charge to apply to the individual seasons with RBDC Opt Out shortfalls? If not, why not?
2. This paragraph does not specify the consequences for failing to meet the applicable share of LCR; this appears to be something MISO should specify.
15. Proposed paragraph 69A.9.1.j refers to “extrapolation,” which is both somewhat vague and inconsistent with the average calculation that presented at the August 8 RASC meeting.
1. We recommend that MISO adopt a simple explicit methodology for calculating the RBDC Opt Out Adder. It seems reasonable to us to calculate this in terms of recent PRA results. We would prefer using an approach based on recent historical average rather than extrapolation, as we expect that year-to-year variation will tend to be relatively random rather than indicative of long-term trends, and we see potential for extrapolation approaches to increase volatility that does not serve a purpose.
2. Logically, in order to reflect overall PRA clearing behavior, any average involving clearing results from both subregions should be weighted by the respective peak-demand values of the subregions; a simple average will produce aggregate RBDC Opt Out Adder results that deviate excessively from aggregate PRA clearing results.

In the August 8, 2023, special meeting of the Resource Adequacy Subcommittee (RASC), presenters discussed MISO's proposal on Reliability-Based Demand Curve (RBDC) opt-out provisions and RBDC Opt-Out Adder (X%).  Feedback was requested on the proposed RBDC Opt-Out Adder provisions and draft Tariff language (Section 69.A.1).  

Comments are due by August 18  

DTE Feedback

DTE appreciates the ability to provide feedback on the updates to the Reliability Based Demand Curve opt-out adder and draft tariff language.

Overall, DTE is in favor of the creation of the reliability-based demand curve and encourages MISO to continue with the initiative. However, in the current state, DTE would not consider the opt out provisions proposed by MISO. The uncertainty of future accreditation values within the new SAC framework, combined with the required three-year opt-in period, and the high penalty for failing to meet the planning requirement obligations in future years, leads to significant risk that would eliminate MISO’s current opt-out proposal as a viable option.

DTE believes the changes to include the ‘X’ percent being updated yearly and the X% at which you opt-out at remaining the same over the three-year opt-out period make logical sense, but DTE encourages MISO to release more data supporting the calculated X% by overlaying system-wide and regional RBDCs over the past two planning years. DTE also encourages MISO to continue investigating ways to forecast the X% more accurately as only using 2-3 seasons of historical data does not carry much statistical significance.

While DTE understands MISO is still thinking through impact of the RBDC opt out provision, how would a RERRA administratively setting PRMR at 1 day in 10 affect RBDC development and consequently other market participants outside of that RERRA?

DTE agrees with MISO’s proposal to carry forward with the RBDC framework even as debate over the opt-out method and X% adder continue.

MidAmerican appreciates the opportunity to provide feedback on updates to the Reliability Based Demand Curve (RBDC) proposal and the opt out adder.

MidAmerican would like to thank MISO for all the work they are doing to make the resource adequacy construct consistent with MISO’s market guiding principles.

As MISO transitions its resource adequacy policies, load-serving entities like MidAmerican need time to react to, and plan for, significant new policy changes like the RBDC proposal with the opt out option. The proposed timeline for implementing the RBDC is too short. The significance of this additional capacity obligation is evidenced by the MISO opt-out proposal discussed at the July 18, 2023, RASC, where MISO estimated that the opt-out amount could be as high as 3.9% to 4%, given the last three historical planning resource auctions.  While the clearing price may be lower for the additional capacity procured, utilities nonetheless need time to react in both regulatory and long-term planning processes. MidAmerican supports an implementation date further in the future, such as the 2027-2028 planning year or a phase-in over several years such that the demand curve slope flattens over time.  

MISO should simplify the short-term capacity market design to enable load serving entities to conduct long-term planning. Predicting accreditation levels and the planning reserve margin requirement into the future is very difficult under the overall proposed design changes that include MISOs Direct Loss of Load proposal, and now the new RBDC proposal. These proposals are in addition to the seasonal aspect of the short-term market which has already been implemented. The combined effect of all three of these new or proposed aspects of the short-term capacity market is an extraordinarily complicated interaction with the long-term generation resource planning conducted by entities like MidAmerican. MidAmerican continues to urge MISO to simplify the design or provide additional “bookend” scenarios in the near-term rather than Q4 of 2024 which is MISO’s current target deliverable.

The Entergy Operating Companies ("EOCs")^[1] appreciate the opportunity to provide feedback on MISO’s Reliability Based Demand Curve proposal.

The EOCs do not support MISO’s decision to omit Entergy’s proposed AFRAP mechanism from the RBDC filing MISO will make with FERC. Given the results of the July RASC motion and the resolution passed by the Entergy Regional State Committee on August 9th, the EOCs believe that there is ample State and stakeholder support to warrant MISO including the Entergy-proposed AFRAP mechanism and the RBDC in a joint FERC filing. Furthermore, FERC has accepted similar proposals in the past. See PJM Interconnection, L.L.C., 115 FERC ¶ 61,079, order approving contested settlement, 117 FERC ¶ 61,331 (2006).

As stated in prior feedback, the EOCs do not believe that filing the RBDC with only the MISO Opt-Out proposal is acceptable for the following reasons:

• LSEs have significant uncertainty related to future SAC unit ratings, unit retirement dates, unit new build CODs, and coincident peak load forecasts. For these reasons, committing to a 3-year term to cover 100% + x% of PRMR creates significant risk that an LSE may not be able to meet the Opt-Out requirements in future years, which would result in an LSE being exposed to the 2.7 x CONE capacity deficiency charge. This risk is unreasonable and excessive, and functions as a significant deterrent to use of the Opt-Out.
• Due to the “+ X%”, electing to use the MISO Opt-Out may result in an LSE having a larger load requirement than if the LSE had not Opted-Out and had instead fully participated in the PRA. Such an outcome would be unreasonable because it would penalize and burden entities that have engaged in reasonable planning and procured long-term resources sufficient to cover their load plus reserves.

Feedback by Public Service Commission of Wisconsin (PSCW) Office of Regional Markets (ORM) staff to MISO on the RBDC following the August 8 RASC meeting

Allowing the auction to reach 4x CONE for all seasons is a policy choice that MISO should more thoroughly justify and discuss with stakeholders, and if it is not possible to have a robust discussion prior to MISO’s intended RBDC filing date, then this aspect of MISO’s plan should be tabled for a later date. MISO stated it would be unlikely for prices to hit 4x CONE, but we posit that the likelihood of capacity auction prices hitting a limit is a separate question from what the limit should be. Furthermore, as our understanding is that nothing in the tariff specifies how close to the PRM the curve may cut down, the likelihood of a season hitting the cap depends in part on how MISO draws the curve and is not just a matter of shortage.

MISO suggested that years of shortage would be treated as missing data points for the calculation of X%. We disagree with this approach. If MISO proceeds with the approach to use historical data to set X%, we suggest that MISO treat shortage year data exactly the same as non-shortage year data (i.e., not as a missing data point and not as a zero) in order to achieve comparability between X% and auction participants, while reverting to a minimum of the PRM in the case that the calculation of X% resulted in a negative number. 

MISO stated that calculating a separate North and South X% would add too much complication. Given that the regions can clear differently and have different RBDCs, we suggest that MISO find a work-around to prioritize accuracy and fairness. We see a potential issue where one region could drive up the PRM via the adder (X%) for opt out participants in another region. Adding PRM to zones with higher peak demand translates to a larger ask in terms of MW, so we see value in assigning the adder amount cautiously with that in mind.

Can MISO provide stakeholders with a specific maximum quantity for both X% and a percent above the PRM for the RBDC to clear? If not, can MISO explain why?

Can MISO clarify if there is a maximum and/or minimum price where the RBDC would cross the PRMR?

The clearing quantity above the PRM for PY23-24 summer systemwide is listed as 3.9% in the August 8 slides, which is an increase from the 3.6% listed in the July 18 slides for the same season and year. When MISO updates a previous estimate, it would be appreciated if MISO would highlight this in the slides with an explanation for why a change was made.

Could MISO provide stakeholders with an explanation in writing of the assumptions used to generate the supply and demand curves in the Monte Carlo simulations? As these assumptions are critical to drawing the curve they should be shared. We note that stakeholders have asked for this information previously.

Currently, the LCR is calculated based on the PRMR, which is determined by the LOLE and accounts for transmission capabilities and availability. Will there be any changes to this procedure under the RBDC? For example, will the LCR be adjusted based on a Final PRMR from the RBDC for auction participants? And for those participants that opt-out, will the LCR be adjusted to reflect the resulting Final PRMR from the PRM + X%? Or does MISO see value in keeping the LCR based on the PRMR (the Initial PRMR) from the LOLE study?

Members of the LSE Coalition generally support the following feedback and attached electronic comments in the margin of the respective redline draft tariff documents:

MISO should adjust its proposed filing timeline to allow all questions to be resolved and proposed tariff language to undergo a full review.  In support of a full review, MISO should provide numerical examples that demonstrate the implications for the shape of the RBDC and the economics of RBDC Opt-Out if accredited capacity supply were to continue to decrease due to proposed changes in accreditation rules or net market entry developments.  For those numerical examples, MISO should explicitly account for both the IMM’s verbal claim during the August 8 RASC meeting that the RBDC will flatten as MISO models more LOLE risk and MISO’s admonition on slide 16 of its July 11 RASC meeting presentation on accreditation reform (HERE) that “PRMR expected to increase as LOLE modeling enhancements are made to better reflect risk.”

I am happy to discuss.

David Sapper

dsapper@ces-ltd.com

In addition, members of the LSE Coalition generally support WPPI's feedback.

David Sapper

dsapper@ces-ltd.com

Please see memo sent by e-mail, and distribute same as feedback.

Thanks,

Alex Zakem

for Energy Michigan

18Aug23

The OMS Resources Work Group (OMS RWG) provides this feedback to MISO on its most recent update on the Reliability Based Demand Curve (RBDC) proposal. This feedback is from an OMS work group and does not represent a position of the OMS Board of Directors.

The OMS RWG encourages MISO to formalize its plan to evaluate and implement a more robust methodology for its annual calculation of the X% adder for entities selecting the RBDC opt-out. While a simpler approach that relies on historic information to determine X% may be acceptable as an interim step during initial implementation, a forward-looking or prospective approach may be needed to better ensure comparability between those that utilize the RBDC opt-out and those that participate in the PRA.

Comments by the Environmental Sector
Regarding MISO’s August 8, 2023 RBDC presentation 

The Environmental Sector appreciates the opportunity to provide feedback to MISO regarding the proposed RBDC Opt-Out Adder.

As an overarching comment, we support the notion provided by Alex Zakem from Energy Michigan that MISO should simplify its RBDC Opt-Out methodology as a last step, and not during interim steps during formulation of the RBDC Opt-Out process, including the X adder.

To this point, while we appreciate MISO’s attempt to simplify the formulation of the X percent adder by determining one value for the whole MISO footprint, absent a compelling justification for doing otherwise, we believe it would be more appropriate to formulate an X percent adder for each subregion in each season. Considering the transfer constraint between MISO Classic and MISO South, using a single X percent adder for the whole footprint works to only artificially inflate (or deflate) the adder used for each subregion while: i) imposing an increased and unsupported cost on ratepayers in the subregion that bound at the systemwide RBDC; and ii) decreasing the reliability value of the X percent adder in the subregion that bound at a RBDC higher than the systemwide value.

Additionally, while the August 8 presentation states that the X percent adder will initially be based on the PY22-23 to PY24-25 overlay with the RBDC, we cannot figure out how, for the PY23-24 overlay specifically, the percentages used on slide 10 for the X adder were determined. A key source of our confusion is the discrepancy that exists in the material that MISO has thus far made available.

For instance, the May 24, 2023 RBDC presentation that discusses the subregional RBDC curve overlays^[1] (the most recent presentation to go into this subject) shows on slides 21 and 25 that for the summer of PY23-24, the system-wide RBDC cleared 3.6 percent above PRM, yet on slide 10 of the August 8 presentation, the system-wide RBDC is shown to have cleared 3.9 percent above PRM.

Likewise, both MISO South and MISO Classic would have cleared at the same RBDC clearing price for fall of PY23-24 (according to MISO’s May 24 presentation at slide 22), yet on slide 10 of the August 8 presentation, MISO South and MISO Classic are shown to have cleared (with a hypothetical RBDC) beyond PRM by different percentages (3.2% vs. 3.6%) in Fall 2023. Is this because Zone 9 cleared at a higher price in the actual PRA for Fall 2023? Again, there is a discrepancy between the system-wide RBDC clearing prices when comparing the May 24 presentation (0.8 percent above PRMR for the fall) with the August 8 presentation (1 percent above PRM for the fall).

An explanation of these differences would be very helpful.

Finally, we appreciate all the hard work that MISO staff is undertaking to try to implement the RBDC and, if we are going to have an RBDC Opt-Out participation option for LSEs, we hope that our comments help ensure that this method is the most effective, efficient, and equitable RBDC Opt-Out participation option possible. However, while the Environmental Sector wants to improve whatever the final option is, our comments and appreciation of MISO's work should not be construed as supporting this as the best approach.

RDBC Design Comments

 Minnesota Power appreciates the opportunity to provide stakeholder feedback on the Reliability Based Demand Curve (RDBC) from the August 8, 2023 workshop.  It is our understanding that MISO is planning on making final tariff language revisions to file with FERC in September 2023.  Minnesota Power comments are focused on two level;  1) comments regarding the premise of the proposed methodology and 2) comments regarding a refinement to what has been proposed.

 1)      Comments regarding the premise of the proposed methodology. 

 Limits of a One Year PRA

Resource Adequacy is based on having adequate capacity and energy resources to meet the demand and energy requirements.  The Planning Resource Auction (PRA) is for one year, where there is no clarity on the expected levels of capacity and demand requirements until the PRA clears.  The decisions for securing additional resource is a much longer timeframe, so the PRA does not provide a means of determining what resources should be developed.  Load Serving Entities (LSE) that are conducting long-term planning utilize proven methodologies for developing a resource plan.    

Viability of SAC Replacement Capacity for Outages Greater Than 31 Days

The Seasonal Accredited Capacity (SAC) requires that uncleared replacement capacity resources be procured for each day of an outage longer than 31 days.   Example calculations presented for the RDBC show the PRA clearing at higher capacity levels than the 1 day in 10 year standard of reliability, which reduces the availability of uncleared capacity that can be purchased by entities with planned outages longer than 31 days.  Presentations have shown the amount of additional capacity cleared is in the range of 3%.  This will reduce the amount of uncleared capacity available for replacement capacity.  This reduction of uncleared capacity is expected to aggravate the challenge for entities to purchase replacement capacity.  As an additional point, if the market clears at levels of capacity lower than the 1 day in 10 year standard, the entity seeking to purchase replacement capacity could be subject to a fine for a condition where it isn’t possible to purchasing replacement capacity.

 Higher Capacity Costs - Expected Outcome of RDBC   

The RDBC is expected to clear at a higher level of capacity than the 1 day in 10 year criteria, which results in higher capacity requirements for an LSE.  If the PRA has adequate capacity above and beyond the 1 day in 10 year criteria level, the LSE will incur additional capacity purchase costs(or have reduced revenues of selling excess capacity to the PRA).  As an example, the market clears at a value where the effective planning reserves is 3% higher than the PRM, an entity with a peak demand of 1000 MW would be required to purchase or provide 30 MW more capacity than the 1 day in 10 year criteria. 

2)      Comments regarding a refinement to what has been proposed

Method of Setting Value of X %

The proposed methodology presented at the August 8, 2023 workshop is to use a three year average based and to exclude seasons with an X value that is not greater than zero.  The discussions at the August 8 workshop were helpful and provided indication that MISO is open to other suggestions on how to develop the value of X.     

Minnesota Power provides the following feedback for determining the value of X.

-          Affirm the use of historic values to establish the value of X.

-          Affirm the establishment of X each year to get the most recent value if an entity wishes to lock in for the three year period.

-          Affirm the ability to lock into the value of X for a three year period.

-          Suggest to use all historic values of X, including 0% for years of not having adequate capacity

-          Suggest a means of weighting the most recent values of X such as the following.

• 20% Year 1
• 30% Year 2
• 50% Year 3 (most recent year)

Southwest Louisiana EMC (SLEMCO) offers the following comments regarding MISO’s RBDC Opt-Out proposal.

Regarding 100% LSE Opt-Out Participation: SLEMCO encourages MISO to reconsider incorporating in its Opt-Out proposal the ability to nominate less than 100% of its load as part of the procedure. Adding this would improve a LSE’s ability to utilize the Opt-Out alternative more effectively. Even a high minimum Opt-Out of >85% would reduce much of the load and resource variability risk that is present in a 100% requirement.   

Calculation of historical X% PRM adder:  The calculation example for the X% adder as depicted in slide 10 of the presentation materials seems to be oversimplified. As discussed in the August 8 RASC meeting, SLEMCO agrees that two changes to the calculation methodology will yield better, and not overly complex results.

1)      The first change would be that MISO does not ignore Negative Seasonal Values that may arise in the simple averaging. To avoid any possibility of the average being less than zero, MISO could replace any negative value with zero but still include it in the averaging. This would ensure that the overall average would never go negative but still allow for it to reduce the overall X% adder more appropriately.

2)      A weighting by total LSE load of the adders calculated for the North/Central and South regions in arriving at the total X% adder would more accurately reflect the MISO-wide requirement. Alternatively, using the separate calculation of the X% adder for MISO North/Central and South individually might be even more appropriate.