RASC: RBDC Proposal and Design Elements (RASC-2019-8) (20230228-0301)

Arkansas Public Service Commission’s Feedback 3/24/23

Feedback Request: MISO is requesting feedback on the reliability-based demand curve proposal and design elements by March 24, 2023

Feedback Summary:

As described below, the Arkansas Public Service Commission (APSC) has concerns about the development of the downward sloping demand curve (now referred to as the Reliability Based Demand Curve or RBDC).  If MISO insists on proposing the RBDC for use in the Planning Resource Auction (PRA), MISO must include a real Opt-Out Provision as described below.  MISO’s AFRAP and the current FRAP do not qualify.

THE RBDC SHOULD APPROXIMATELY REPRESENT CONSUMER VALUATION OF RELIABILITY

While the APSC is not opposed in principle to a downward-sloping demand curve, the APSC has some practical concerns about the development of the RBDC.  First, the RBDC should approximate as closely as possible the marginal value of reliability.  It should not just be an “administratively”- determined demand curve. 

Second, use of an “anchor point” based on the Net Cone is problematic.  The purpose of the RBDC is to reflect consumers’ valuation of reliability in MISO. The Net Cone is a cost-based measure that has nothing to do with consumer valuations of reliability. 

Third, MISO has not provided sufficient data support for the scalar value used in the shift from the Marginal Reliability Impact Curves (“MRI”) to the RBDC.  That formula is:

(MWH/MW-yr.)*(1 yr./365) * ($/MWH) = $MW-day

where $/MWH is the scalar value apparently based on Value of Lost Load (“VOLL”).  Is there academic support, or survey support for the VOLL value(s)?  Does it change over the seasons, or every year?  This scalar value is of major import since slight changes in that scalar value can result in large changes in in the Clearing Price in the Planning Reserve Auction because of the inelastic nature of the Supply Curve.

Fourth, MISO refers to a preference for convex demand curves because it reflects the incremental reliability to consumers of additional capacity.  However, note that decreasing marginal value of reliability is reflected in concave, straight-line, and convex demand curves. If the data show that the RBDC is convex, and representative of consumer valuations, that is fine.  However, there should be no a priori justification of a convex RBDC.

Unless, the issues are adequately addressed, the APSC will likely not support a downward-sloping demand curve.

OPT-OUT IS REQUIRED IF THE RBDC IS IMPLEMENTED

If MISO insists on proceeding with RBDC implementation, an Opt-Out Option is required. An Opt-Out Option would recognize that a Load Serving Entity (LSE) will satisfy its PRMR with generation it owns or has purchased bilaterally, and it will insulate an LSE from any (i) exposure to PRA pricing, and (ii) change to its capacity obligation (PRMR) determined according to NERC’s 1 in 10 standard.

MISO’s AFRAP option does not meet the requirements of an Opt-Out Option. Neither AFRAP nor the existing FRAP allow an LSE to avoid increases in its capacity obligation under a RBDC. Combined with the impacts from MISO’s SAC proposal, the increase in capacity requirements would impose significant negative cost consequences on many customers, without sufficient demonstrated benefit.

MISO must recognize that any support previously provided by OMS for the RBDC/DSDC was expressly conditioned on incorporation of an Opt-Out Option.  Suggestions that the Opt-Out Option was not clearly defined at that time are irrelevant.  The concept of Opt-Out is to be insulated from the effects or held-harmless from the repercussions of the mechanism to which the Opt-Out applies. In this case, it means being held-harmless from additional costs and changes to existing capacity requirements.

Providing the Opt-Out Option will ensure that those states/LSEs that choose to participate in the PRA with a RBDC are free to do so, and those that choose not to participate need not.  If states/LSEs agree that the PRA is beneficial, they will likely participate.

Submitted on behalf of East Texas Electric Cooperative, Inc. (ETEC):

ETEC appreciates the opportunity to submit the following comments and requests to MISO regarding the Reliability-Based Demand Curve (RBDC):

ETEC is concerned about the rapid pace of change in MISO’s resource adequacy reforms and the lack of transparency behind MISO’s reform designs. Since September 2022, ETEC has exerted significant resources managing its implementation of MISO’s seasonal and resource accreditation changes. Whereas ETEC usually is prepared for the Planning Resource Auction (PRA) well in advance, ETEC is still contending with issues that arose due to MISO’s evolution of the rules and its inability to effectively convey those changes to all stakeholders. In addition, MISO has had its own implementation struggles that will now result in the PRA being delayed. Accordingly, ETEC strongly believes that MISO and its stakeholders need to see and understand the results of 2023-24 PRA and have an opportunity to digest those results/outcomes before further major reforms are implemented. MISO’s proposed reforms, including the RBDC, should be analyzed based on the behavior and action of market participants in the 2023-24 PRA. Further reform cannot be properly designed by MISO or evaluated by stakeholders without a complete understanding of the impacts from the initial reforms.

ETEC also believes that MISO needs to provide considerably more information to allow for meaningful stakeholder participation. To date, MISO has provided preliminary graphical representations of RBDCs with high level commentary of their derivation. ETEC would like to see MISO develop and provide a design document that details MISO’s design process and derivation of RBDCs along with workpapers that provide supporting detailed information and calculations.

ETEC has these other specific comments related to the RBDC:

• Under the RBDC, load-serving entities (including ETEC) would be subject to more volatility and uncertainty with their capacity requirements. With its limitations, the Advanced FRAP does not practically mitigate exposure to this volatility.
• ETEC does not seek to extract Net CONE from the MISO wholesale markets. Rather, ETEC seeks to meet its load requirements at a stable and reasonable cost. MISO’s ‘sustainability’ principled basis for the RBDC appears misaligned with the regulatory context of MISO’s load-serving entities.

AMP and WPPI support the following feedback requests.

I'm happy to discuss.

David Sapper

dsapper@ces-ltd.com

Annual Offer

• On its face, it is curious that any “marginal” cleared unit would require uplift in any given season.  Please explain the extent to which that is due to MISO allowing annual offers to be “marginal” in MW clearing but not in pricing.

• Example 6 uses “equivalent seasonal offer” prices that are seemingly irrational (or at least very simplistic and non-strategic).

• Please explain why uplift should ever be needed if clearing prices were truly socially efficient in the presence of a sloped RBDC and in the (presumed) absence of the current shortage and near-shortage pricing rules.

• Please prove that there is no “equivalent seasonal offer” price at which R3 could have cleared in the same manner as in Example 5.

• Does the IMM agree with the way MISO constructed Example 6?

• Please explain how the combination of 1) zero MW clearing solutions for R3 in any given season and 2) non-zero MW clearing solutions in any other season would satisfy MISO’s stated anonymous-utility-provided rationale for annual offers that some resource owners want to clear non-zero MW amounts in each season.

• Please provide examples of capacity production costs that are truly and necessarily incurred on an annual basis.

• Please specify if the identified truly annual costs are accrued over a calendar year, planning year, accounting year, or some other 12-month period.

• As Example 6 demonstrates, MISO and the IMM will lack equivalent seasonal offer prices for resources that submit annual offers.  This lack of counterfactual seasonal offers appears to create a situation in which, for any given PRA, it will never be clear if allowing annual offers would be more socially efficient than disallowing them.

• Please explain how MISO can assure stakeholders that the PRA solution (including uplift) in which any annual offer clears will always have lower annual costs than if annual offers were disallowed.

• Please explain to what extent MISO can assure stakeholders that the PRA solution (including uplift) in which any annual offer clears will have lower seasonal costs than if annual offers were disallowed.

RBDC

• Regarding the development of RBDC curves from MRI curves on slide 10 of MISO’s March 1 RASC presentation, please provide details on specific Monte Carlo simulations / scenarios that were performed on supply curves and potential RBDC curves to support the achievement of Net CONE in long-run equilibrium analyses.

• Please provide all relevant work papers.

• In the Monte Carlo modeling, to what extent will MISO assume changes in inframarginal energy and operating reserve (EOR) revenue affect shifts in modeled PRA supply?

• In the Monte Carlo modeling, what determines the minimum and maximum amounts of modeled PRA supply? 

• Do MISO and Brattle agree that the resulting RBDC (i.e., the PRA demand curve that most likely would provide prototype resource Net CONE pricing over time if a resource like the prototype resource were marginal) expands as more supply is modeled? 

• That is, do MISO and Brattle agree that the amount of supply that is modeled affects the volumetric range of the resulting RBDC?

• For the purpose of developing sub-regional RBDCs, how will MISO model supply from the other sub-region or from External Resource Zones (ERZs)? 

• What assumptions about intra-regional (Midwest-to-South, South-to-Midwest) transfer capability will MISO and Brattle make?

• What assumptions about ERZ export limits will MISO and Brattle make?

• Please explain how the graphic in on slide 11 of the March 1 RASC presentation comports with the description of analysis on slide 12? 

• On slide 11, the Monte Carlo supply curves seem to be defined by different $/MWd scalar multipliers.  That does not seem to match the “Year-to-Year Variability in Supply” explained on slide 12.  Please explain how slide 12 describes slide 11 graphics.

• How can MISO and Brattle model AFRAP participation without MISO even providing a complete AFRAP proposal?

• How do MISO and Brattle use historical data to model AFRAP participation?  Do they assume AFRAP participation will match historical FRAP participation in MW terms?

• How does the use of historical data account for the monotonically declining reserve margins that MISO, PJM, and SPP are projecting? 

• In light of the declining reserve margins and thus declining total potential Planning Resource supply, will the MISO-Brattle approach artificially inflate the RBDC needed to produce Net CONE pricing? 

• Do MISO and Brattle want to use historical data for ease of analysis? 

• Did FERC approval of the ISO-NE and PJM studies or results hinge on the use of historical data?

• Were respective regional capacity reserve margins projected to decline over the next 5 years when FERC approved the ISO-NE and PJM studies or results? 

• Regarding the preliminary system-wide RBDCs on slide 13 of MISO’s March 1 presentation, what LOLE targets were used for each season?

• The system-wide RBDCs on slide 13 are, by simple observation, not “reflective of the MRI curves” on slide 15. 

• MISO explained privately that is due to data presentation challenges.  Please elaborate.

• Please explain how any MRI could, and also why any MRI curve used to define RBDCs should, exhibit buyer expectation of increasing returns to scale in any relevant range of output (i.e., ignoring initial increasing returns in production).

• Does the PRA demand curve represent the aggregate Planning Resource demand of actual LSEs in the MISO region?

• If not, should it?

• If so, would the resulting-regional (Midwest versus South) RBDCs express differences in actual LSE demand for Planning Resources across sub-regions? 

• Do MISO and Brattle think all MISO LSEs value the marginal reliability impact of Planning Resources proportionally (based on LSE load size) the same at all reserve levels?

• Should the PRA design be consistent with and facilitate the resource planning processes and preferences of LSEs in the MISO region?

• Should the PRA demand curve be defined by a Planning Resource (or ZRC) product that comports with those LSE resource planning processes and preferences?

• Should the Planning Resource (or ZRC) product reflect LSEs’ resource planning objective to serve their respective gross loads as forecasted within some reasonably bounded range?

• Should the PRA demand curve reflect preferences of LSEs and their respective RERRAs in the region to mitigate steep scarcity pricing and to mitigate clearing quantity risk beyond some target reliability level of Planning Reserves (e.g., 0.1 LOLE)?  That is, should the PRA demand curve should be bounded in terms of price under scarcity conditions and bounded in terms of quantity under surplus conditions relative to the target MW amount?

• Will MISO update RBDC parameters on some regular schedule, or according to any particular internal process or stakeholder process? 

Net CONE

• Regarding the Net CONE timing example on slide 28 of the March 1 RASC presentation, MISO states it is not proposing changes to reference technology (CT) to determine CONE. Were changes to reference technology considered?

• Has MISO considered using a composite prototype resource (e.g., 70% thermal generator/30% battery storage) on which to base the CONE and the inframarginal EOR revenue for netting?

• How will MISO model or calculate inframarginal EOR revenues?

• How often would MISO update Net CONE values?

Michigan Public Power Agency (MPPA) supports the comments of WEC Energy Group on MISO's RBDC Proposal and Design Elements.

Tom Weeks, Wholesale Markets Lead, MPPA

Illinois staff continues to support MISO’s work on the reliability-based demand curve (RBDC). The RBDC more properly values capacity and avoids volatile year-to-year swings in the capacity market, sending better price signals to keep economic resources online and attract new investment in the region. While not the sole reform needed, the RBDC is a crucial piece of maintaining reliability in coming years.

Illinois staff acknowledges that adopting an RBDC might make the PRA less attractive to some states. The ongoing work on the PRA should focus on all of the market design questions that may make participation in the market more attractive for states and LSEs. Some opt-out measure for the RBDC is reasonable. However, any LSE that opts out should not be able to take advantage of the PRA once a zone’s resource needs are met. Reasonable opt outs should also come with strong penalties in the event that the entity falls short in meeting its planned targets.. ICC staff would like to see MISO continue to adjust the market, to make it as attractive to state and LSEs as possible in order to incent participation.

ICC staff would also like a better understanding of why as more states adopt the AFRAP, the RBDC becomes steeper until it approaches vertical. Does this reflect a change in risk in the underlying MRI curve because of a decline in the amount of capacity participating in the auction? Put another way, why do increasing opt-outs require a change to the curve?

Finally, we should look to lessons learned in other jurisdictions. The complexity and gamesmanship present in the PJM VRR Curve process and its quadrennial reviews should be avoided. Reference units should be considered and the most reasonable adopted, perhaps on a zonal level. For instance, a new CT is unlikely to be built in MISO Zone 4. And, continued analysis should be provided as to how many LSEs need participate in order to achieve the intended goals. The continued conversations about the RBDC should focus at least as much on these market design issues and the value proposition for states and their LSEs as it has on how not to participate.

-          Simulating the LOL MWh is the basis for the demand curve shifting, so the same analytical challenges apply to this issue as raised in Direct LOL.

-          Resource planning that doesn’t provide surety of the PRMR is not conducive of a stable planning process.  This is counter-intuitive for investments of the magnitude required for planning to meet the PRMR. 

• The amount of demand curve shape driven by the LOL MWh isn’t clear at this point , but it appears that it could be a sanity check that in the end the slope of the demand curve might be steep enough to reduce the concerns of this being a moving target.
• Does it actually make intuitive sense that the demand curve moves with varying levels of actual PMR of having more resources?  How much would it move, and how is this slope align with any level of analytical scrutiny?

-          The premise that the RBDC would provide a meaningful price signal to avoid the market clearing at CONE is challenged by a number of points

• One year PRA doesn’t provide adequate investment signal
• Limited market signal capability for one capacity clearing price for all types of capacity and market participants.
• This is not needed in States that have a regulated integrated resource plan where capacity resources additions are made based on their customer needs.  It introduces additional financial risk for customers to achieve a price signal that is not needed.

Concept of a “residuals market” aligning with WEC resolution from Feb 28-Mar 1 RASC – pricing is useful for efficiently clearing long/short positions for parties – vs. having a market that can establish capacity value. 

-          FRAP concerns – not being able to protect customers

-          Long-term planning of needing to know the total planning reserve margins – sloped demand curve doesn’t provide certainty

-          AFRAP provides a 3 year locking mechanism doesn’t provide the level of assurance on long-term resource planning that typically has timeframes of 10-15 years.

Comments

of the

Association of Businesses Advocating Tariff Equity (ABATE),

Illinois Industrial Energy Consumers (IIEC),

Louisiana Energy Users Group (LEUG),

Texas Industrial Energy Consumers (TIEC),

Coalition of MISO Transmission Customers (CMTC),

Midwest Industrial Customers (MIC),

and

NIPSCO Large Customer Group (NLCG)[1]

Regarding

RASC: Reliability Based Demand Curve (RBDC) Proposal and Design Elements

(RASC-2019-8) (20230228-0301)

March 24, 2023

ABATE, IIEC, LEUG, TIEC, CMTC and MIC, as representatives of the End-Use Customers (EUC) Sector, and NLCG appreciate this opportunity to provide comments to MISO.

While we appreciate MISO’s work to date with respect to exploring the potential use of a downward sloping demand curve within the MISO PRA, the EUC Sector is greatly concerned that MISO is plunging into the market design without developing the white papers and analyses need to fully explain and support the perceived need for the proposal, the underlying theory behind the proposal, the various alternatives to the proposal that are available, and why the proposal is superior to those alternatives.  In addition, MISO needs to produce analysis that shows that the proposal does not introduce adverse side effects or other harm that is more problematic than the issue the proposal is trying to address.  Finally, we are concerned that MISO continues to have a proposed filing date at FERC for its proposal of the 2^nd Quarter of 2023.  Accordingly, the EUC Sector urges and requests that MISO revise its work timeline to allow sufficient analysis, development and stakeholder review.  Further, if MISO ultimately files a proposal with FERC, it should not seek implementation in the PRA prior to the 2025/2026 Planning Year.

For background, at the January 17-18, 2023 and February 28, 2023 - March 2, 2023 meetings of the MISO Resource Adequacy Subcommittee (RASC), MISO provided additional information with respect to its proposal to modify its Planning Resource Auction (PRA) under its current season resource adequacy construct to introduce the use of a convex downward sloping demand curve, which MISO refers to as a Reliability Based Demand Curve (RBDC), that is intended by MISO to support a seasonal long run equilibrium capacity market price equal the Net Cost of New Entry (CONE).  MISO’s proposal would also introduce co-optimization in the solution of the four seasonal auctions conducted under the current MISO PRA process.

The new information MISO has provided in the last two MISO RASC meetings has documented how MISO has designed its preliminary proposed market-wide RBDC based on a Marginal Reliability Impact (MRI) curve methodology.  MISO has also shared its preliminary results of that design, which are shown below:

[Note: The figure showing MISO's preliminary market-wide RBDC results is shown in the PDF version of these comments that has been separately e-mailed to MISO Stakeholder Relations.] 

In addition, MISO has shared its preliminary proposed approach to calculate Net CONE, which has yielded a preliminary 2023/2024 Planning Year value of approximately $150 per MW-day annually, or $600 per MW-day for a single season.  This preliminary Net CONE value was used by MISO in the Monte Carlo analysis that MISO performed to produce the above preliminary proposed seasonal market-wide RBDCs.  Finally, MISO has presented its preliminary proposal with respect to how it proposes to use co-optimization in the solution of the four seasonal auctions that are conducted under the PRA.

As we indicated in our December 28, 2022 comments to MISO:

“MISO’s proposal could potentially lead to earlier new market entry by merchant capacity resources by offering a higher and more stable Auction Clearing Price (ACP) than can be provided under the current PRA.  This could potentially reduce the risk of the PRA not acquiring sufficient capacity for MISO to on average meet MISO’s Loss of Load Expectation (LOLE) target of no more than one firm load curtailment day in ten years. 

However, it is important to note that no analysis has been presented at this point that conclusively demonstrates that adoption of MISO’s proposal would in fact produce that benefit and, if the expected new market entry does not occur, the outcome of MISO’s proposal could simply be to increase the market price for capacity for buyers within the MISO footprint without providing a commensurate benefit to those buyers.  Given this, the EUC Sector recommends that great care be taken by both MISO and its stakeholders in considering MISO’s proposal and whether it will actually produce the benefits that MISO and the MISO Independent Market Monitor (IMM) expect to be provided from the proposal.”  

We also indicated in those December 28, 2022 comments:

 “… the EUC Sector is greatly concerned that it appears that MISO has assumed that its proposal would be effective without conducting any analysis to evaluate whether its proposal would in fact be reasonably expected to do so.  The expected effectiveness of MISO’s proposal should not be assumed.  The expected effectiveness of the proposal needs to be reasonably demonstrated through analysis before a decision is made to file the proposal at FERC.”

“Without supporting analysis and the receipt of feedback from stakeholders, MISO appears to have decided in its proposal to use a convex downward sloping demand like that of ISO New England rather than a linear downward sloping curve such as that used by the New York ISO and PJM.  The choice of the specific type of downward sloping demand curve should only be made after an analysis of the impact each type of curve is completed and stakeholders have been provided an opportunity to provide feedback on that analysis and the selection of the downward sloping demand curve type.”

“Pursuant to Section 6w of Michigan Public Act 341 of 2016, the Michigan Public Service Commission (MPSC) has implemented an annual capacity demonstration process under which all electric suppliers in the State of Michigan are required to make an annual demonstration to the MPSC that they have acquired sufficient capacity to serve 95% of their expected PRMR in the MISO Planning Year commencing three years after the end of the current MISO Planning Year.  MISO has not addressed how its convex downward sloping demand curve proposal will interact with the MPSC capacity demonstration provisions and whether it would be compatible with them.”

“MISO has proposed a very aggressive timeline under which it would file its convex downward sloping demand curve proposal with FERC by the end of the 2^nd quarter of 2023 for implementation in the PRA for the MISO 2024/2025 Planning Year, which will be conducted in April 2024. 

Given the number of questions that need to be addressed with respect to MISO’s proposal and the fact that MISO has yet to conduct its PRA without issue with its new seasonal resource adequacy and availability-based capacity accreditation provisions in place, we believe MISO’s proposed timeline is unrealistic. 

The proposed timeline should be revised to allow sufficient time to allow sufficient analysis, development and stakeholder review of the proposal before it is filed at FERC assuming that analysis, development and stakeholder support ultimately supports such a filing.  In addition, to allow the necessary time for the foregoing and to ensure any issues with the implementation of MISO’s seasonal resource adequacy and availability-based capacity accreditation construct are fully resolved, if MISO’s proposal is ultimately filed at FERC, it should not be implemented in the PRA prior to the 2025/2026 Planning Year.“

These comments and concerns remain unaddressed by MISO and will need to be addressed by MISO.

Thank you for providing us an opportunity to provide these comments.  If it would be of help, we would be glad to discuss any of the above comments further with MISO and other stakeholders.  Please do not hesitate to contact any of the following representatives:

Jim Dauphinais

Brubaker & Associates, Inc.

(Consultants to ABATE, IIEC, LEUG, NLCG and TIEC)

(636) 898-6725

jdauphinais@consultbai.com

Ali Al-Jabir

Brubaker & Associates, Inc.

(Consultants to ABATE, IIEC, LEUG, NLCG and TIEC)

(361) 994-1767

aaljabir@consultbai.com

Kevin Murray

Ken Stark

McNees Wallace & Nurick LLC (for CMTC)

(614) 719-2844

murraykm@mcneeslaw.com

kstark@mcneeslaw.com

Kavita Maini

KM Energy Consulting, LLC (Consultants to MIC)

(262) 646-3981

kmaini@wi.rr.com

The Entergy Operating Companies ("EOCs")^[1] appreciate the opportunity to provide feedback on MISO’s Reliability Based Demand Curve proposal. The Entergy Operating Companies generally support MISO implementing a reliability-based demand curve (RBDC) in the PRA to help form price signals that more reasonably reflect the supply and demand and needs of the market. The RBDC alone is not sufficient to ensure reliability and ensure that each LSE procures reasonable long-term resources to meet its load obligations; rather, at a minimum, appropriate regulation of resource planning and a reasonable minimum capacity obligation are also needed.  However, the RBDC, if properly implemented, would improve the MISO PRA and provide helpful incentives, particularly to LSEs not subject to state regulation and to merchant generation owners, to help achieve reasonable market outcomes consistent with maintaining reliability over the long-term.

Formulation of Demand Curve

The EOCs request that MISO provide additional detail in the form of workpapers demonstrating how the Marginal Reliability Impact curves are translated into Reliability Based Demand Curves through Monte Carlo Analysis to support recovery of Net CONE.

Advanced FRAP (AFRAP)

The EOCs support a FRAP opt-out mechanism in the MISO PRA that allows LSEs to meet their reliability needs without being subject to the uncertainty and load obligation increase imposed by a sloped demand curve.  LSEs that are subject to retail regulation and that engage in reasonable resource planning to serve the large majority of their loads with long-term resources are entitled to the optionality that a reasonable FRAP opt-out mechanism would provide.  Such a mechanism also helps preserve the jurisdiction of state regulators, acknowledges the discretion of regulated LSEs to determine what types and quantities of resources are appropriate to meet the needs of the customers whose interests they have a duty to protect, and provides the right incentives and price signals to ensure other LSEs engage in reasonable resource planning.  An LSE that has procured enough generation to meet the overwhelming majority of its customers’ needs at the 0.1 LOLE planning reserve margin for a multi-year period should not be exposed to the impacts of a sloped demand curve and be forced to procure extra capacity. Nor should such an LSE that chooses the AFRAP option be required to hold a capacity buffer if they wish to sell excess capacity.

As presently designed, AFRAP is not a plausible or attractive option because of the combination of the following four provisions; (1) an LSE must cover 100% of their PRMR, (2) the planning reserve margin (PRM) is uncertain across the three-year term and is higher than the 0.1 LOLE PRM, (3) the penalty for not meeting the AFRAP obligation is set at the capacity deficiency charge (2.7 x CONE) which is significantly higher than the highest possible auction clearing price, and (4) an LSE’s ability to sell excess capacity is limited. Please see the EOCs’ January RASC feedback detailing proposed changes for how AFRAP could be improved to address the EOCs’ concerns while preserving the intended benefits of the RBDC.

MISO AFRAP Compared to PJM FRR

The PJM FRR option upon which the AFRAP is modeled only requires LSEs to self-provide an amount equal to the pre-determined planning reserve margin.  In this way, PJM’s “fixed” resource requirement (FRR) stands in contrast to the “variable” resource requirement (VRR, which is PJM’s name for its sloped demand curve).  The purpose of PJM’s FRR opt-out is to allow LSEs to avoid the incremental procurement associated with the sloped demand curve.

While the PJM FRR option does require LSEs to hold a buffer of capacity above the planning reserve margin if they wish to make sales, the purpose of that buffer is to cover the additional uncertainty associated with the multi-year load forecast and forward auction design.  Because MISO has a prompt auction, there is no need for such a buffer.[2]

The Environmental Sector submits the following comments, questions, and requests in response to MISO’s February 28, 2023 presentation on the Reliability-Based Demand Curve(s) (the "presentation").

AFRAP

Regarding the proposed Advanced Fixed Resource Adequacy Plan (AFRAP), Slide 7 of the most recent presentation indicated that MISO is still evaluating all feedback and in conversations with OMS. We look forward to continuing the discussion on AFRAP at the next RASC meeting.

MRI Curves

Slide 9 of the presentation indicates that MISO is referencing its Marginal Reliability Impact (MRI) Curves against Peak Load. Since MISO is looking to shift its resource accreditation methodology to one that looks less at peak load and more at the riskiest hours, does MISO foresee recalculating the MRI Curves for riskiest hours too?

Seasonal Co-optimization and Annual Offers (“Social Welfare”)

Slide 21 of the presentation indicates that for the purposes of the PRA, MISO equates the maximization of social welfare with the minimization of overall production costs. We understand that “social welfare” is a common term used in classical economics, but we suggest that its utility in this context should be reexamined. From a purely economic perspective that only examines direct, known market prices, this terminology may be correct, but it is also true that when accounting for all the social costs that a given resource may create, the minimization of overall production costs does not result in the maximization of social welfare.

As it applies to seasonal co-optimization, we believe that the uplift provisions will result in the subsidization of polluting resources, thereby incentivizing further development or continued operation of such polluting resources, and thus exacerbating the social harm committed by those resources. As such, if the PRA objective function is to “maximize social welfare” in light of social costs, then the PRA objective function fails.

To this extent, we ask that MISO either remove the reference to the maximization of social welfare or that MISO incorporates social costs within its PRA objective function to account for social costs within seasonal co-optimization. Our preference would be that MISO choose the latter option, which we believe is feasible since MISO already does this in other internal processes. One such example is the incorporation of the social cost of carbon when calculating the benefit metrics of LRTP portfolios. But to the extent that MISO does not see such incorporation as appropriate, we suggest that MISO remove the reference to maximization of social welfare altogether.

Annual Commitment Examples

Thank you for providing the supplemental material on March 16, which includes additional examples of how resources may or may not clear under various scenarios mixing seasonal and annual offers.

Of the example scenarios provided in the supplemental material, we see that the only example showing necessary uplift for resources providing annual offers is when that resource is marginal across all seasons (example 5). Are there other scenarios when an annual offer is extra-marginal in at least one season but still clears based on the annual planning year, thus also requiring uplift for that resource in at least one season? If so, please provide an example of this scenario.

Use of Monte Carlo simulations

Our experience in other RTOs with sloped demand curves has shown that a Monte Carlo analysis can overstate risk, and thus may lead to recommending an overly conservative demand curve. This is mainly due to the manner in which Monte Carlo scenarios can overstate the "shocks" that a large system may experience and ignore some volatility-dampening dynamics such as the fact that decisions about new entry and retirements will be informed by market participants’ anticipations of other entry and exit, as well as reacting to load growth. In using a Monte Carlo approach, it will be critical for MISO to adopt realistic assumptions and supply and demand variability, so as not to present an inflated sense of risk that could bias the curves toward over-procurement of capacity and prices higher than necessary for reliability.

Advanced CT Continuing as Reference Technology

We were happy to see that MISO alluded to the prospect of the reference technology used in calculating Cost of New Entry (CONE) and Net CONE changing in the future.^[1] MISO has stated that changes to the reference technology are not the subject of the current RBDC effort, but we believe that such changes should be part of the current RBDC effort because of the substantial consequences that the choice of reference technology will entail for calculating CONE and Net CONE. It is our hope that MISO takes this prospect seriously as other RTOs have or are moving away from advanced CT as their reference technology.^[2] We also believe that reconsideration of the reference technology would include factors such as whether the reference technology is likely to be built, whether its costs and revenues can be estimated with a reasonable degree of accuracy, and whether using that resource as the reference technology is likely to produce prices high enough to meet the reliability standard but not so high as to add unnecessary costs. As such, the reference technology shouldn’t just be the technology with the lowest upfront cost, but the one most likely to be placed into service.

Use of 3-year average historical to calculate inframarginal rents for Net CONE

MISO proposes to use 3-year average historical data to calculate the inframarginal rents for Net CONE, but then to possibly scale this data to reflect forward conditions.^[3]  It is unclear why MISO has chosen this approach, rather than simply doing a forward-looking estimate of energy and ancillary service net revenues. As MISO acknowledges, the choice of a gas-fired reference resource means that inframarginal rent is highly correlated with gas prices.  Gas prices, and therefore energy prices, are projected to be very different in the next few years than they have been in the recent past.  MISO suggests that it may account for this by using a scaling factor, but has thus far been noncommittal on this point.  Underestimating future inframarginal rents would lead to higher capacity prices than are needed to attract new entry and retain resources.  MISO should examine the use of forward energy and ancillary services revenues in order to give a more accurate sense of the true net cost of new entry that resources would look to recover through the PRA in the commitment year.  PJM recently switched to a forward-looking methodology for the energy and ancillary service revenues used in its Net CONE calculations for similar reasons.  MISO should evaluate the possibility of a true forward offset, rather than relying primarily on historic data, and offer further detail and comparison to stakeholders on this approach versus the scaling factor approach it is considering.

The OMS Resources Work Group (OMS RWG) appreciates the opportunity to provide feedback to MISO on the Reliability-Based Demand Curve (RBDC) proposal and design elements. This feedback is from an OMS work group and does not represent a position of the OMS Board of Directors.

Marginal Reliability Impact (MRI) Curves

The inclusion of the graphs depicting the seasonal demand curves within the RBDC materials helped enhance the OMS RWG’s understanding of key concepts and design changes from the current PRA based on a vertical demand curve. To that end, the OMS RWG requests that MISO provide the same graphs, depicted on slide 13 of the RBDC material, at a larger scale with granular numerical scaling and an overlay of anticipated supply curves based on either historical data or forward projections. If the graphs on slide 13 are the actual results of a simulation, and not offered for purely illustrative purposes, the OMS RWG would like to understand why MISO’s design of the curve appears so steep. Too steep of a demand curve would lead to high levels of volatility in the equilibrium price year-to-year. Relatedly, the OMS RWG would appreciate seeing MISO’s estimation of potential auction clearing prices and clearing quantities by season following the close of this year’s PRA. The draft estimates MISO shared for summer have been helpful, but a look at fall, winter, and spring are needed for a more complete understanding of the RBDC’s impact.

The OMS RWG is concerned that MISO’s use of the Expected Unserved Energy (EUE) metric in establishing the MRI curves without changing how the Planning Reserve Margin Requirement is determined will result in unnecessary variability in the year-to-year requirement. In OMS’s Position Statement regarding Consideration of a Revised Demand Curve, the OMS Board expressly requested that any changes to the demand curve must be durable to reduce the need for future design changes. MISO could accomplish this objective by establishing a stable EUE metric that will not need to be adjusted every year. Additionally, there is a limit for the value of reliability, and the design of the demand curve must reflect that reality. Otherwise, as the IMM has stated, the incremental value of reliability continues ad infinitum.

The OMS RWG understands that MISO has engaged with the Brattle Group to assist in the design of the MRI-curves. The OMS RWG requests that MISO provide the study details that were used in designing these curves.

Seasonal Co-optimization with Annual Commitment Model

Like those raised during the RASC by several stakeholders, the OMS RWG has concerns about MISO’s proposed annual commitment model. The OMS RWG understands that MISO proposed the annual commitment model to provide Market Participants additional optionality, but MISO’s model appears to interject unnecessary complications and variability into a construct that’s already experiencing substantial change. Would MISO be willing to state which, if any, Market Participants requested such an offering? Knowing which participant requested such an action and why could lead to better understanding of the reasoning thereof.

The OMS RWG also supports further discussion and explanation regarding who pays for uplift provided to those resources that clear annually. Would LSEs that opted out of the PRA be subject to uplift? If an annual commitment model is implemented, the OMS RWG questions why a similar model isn’t available for resources to offer in two or three seasons.

Net Cost of New Entry (CONE)

While the OMS RWG generally supports the methodology MISO proposes for the Net CONE calculation, we request that MISO explore its continued reliance on a single reference technology (peak CT unit) in determining CONE. Given the increasingly dynamic nature of the resource portfolio, it appears to be the time to consider multiple technologies.

Lastly, the OMS RWG continues to reiterate the seventh bullet point on OMS’s Position Statement on Consideration of a Revised Demand Curve, which reads:

Revising the PRA’s demand curve is just one tool to improve the RAC. MISO must continue to work on enhancements to its energy and ancillary services markets and coordinate those market reforms with the RAC.

It is imperative that MISO work on enhancements to the energy and ancillary services markets, including enhancing scarcity pricing and reevaluating the ORDC and VOLL, in addition to working on all of the above because implementing an RBDC alone will not be sufficient to incentivize new generation. 

Mississippi and Louisiana Public Service Commissions’ Feedback 3/22/23

Feedback Request: MISO is requesting feedback on the reliability-based demand curve proposal and design elements by March 24, 2023

Feedback Summary:

As described below, the Louisiana Public Service Commission (LPSC) and Mississippi Public Service Commission (MPSC) (the Commissions) oppose the downward sloping demand curve (now referred to as the Reliability Based Demand Curve or RBDC).  If MISO insists on proposing the RBDC for use in the Planning Resource Auction (PRA), MISO must include a real Opt-Out Provision as described below.  MISO’s AFRAP and the current FRAP do not qualify.

RBDC WILL BE INEFFECTIVE

The Commissions have repeatedly urged MISO not to proceed with a downward sloping demand curve, most recently the RBDC, because MISO has failed to offer sufficient analysis or support for the claims that the RBDC would (i) promote a delay in retirement of thermal generation serving load in MISO, (ii) promote investment in the types of new generation resources needed to operate the system (e.g., storage or thermal resources with the required attributes), or (iii) increase reliability.

First, thermal generators are retiring or suspending operations in MISO largely to advance state and utility policy goals or comply with increasingly stringent environmental regulations.  If a state is closing a coal generator to avoid the cost of installing pollution control equipment or to shift to renewable resources, it is highly unlikely that a utility would extend operations in the hope of receiving incremental revenue in a capacity auction. And, if these units are in rate base, the owners already recover a return on investment in that resource. Policy, not revenue, primarily drives these retirement and suspension decisions.

Stakeholders have requested that the IMM and MISO provide specific examples where additional capacity revenue would have extended the operating life of a thermal plant slated for retirement; they have not.  Neither MISO nor the IMM have provided sufficient information to demonstrate that the RBDC would delay thermal generation retirement. Their proposal rests entirely on supposed economic theory.

Second, additional capacity auction revenue will not provide a sufficient revenue stream to persuade independent, non-utility developers to invest in thermal units likely needed for reliable operations.  Current federal subsidies do not favor gas or coal generation development; they primarily are focused on advancing wind, solar, and storage facilities, and other clean energy technologies. As is evidenced by the MISO interconnection queue and MISO’s LRTP futures, developers need no other financial incentives to invest in renewable generation. The only parties likely to invest in thermal generation, if any, will be vertically integrated utilities who can recover their investment through rate base.

Third, there is no evidence that paying additional compensation to generator owners that are long on capacity will increase reliability. “Resource Adequacy” is a requirement, not a market.  It identifies a minimum amount of capacity that must be constructed to serve load at its peak with a stated reserve margin. Once that generation is constructed and operational, the requirement is met.  If, for arguments sake, anyone believes that having excess generation constructed improves reliability, that still does not justify payment of additional revenue in the PRA. Why?  Because that surplus generation is already constructed; it is already available. Even if it is not subject to Must Offer requirements because it did not clear in the PRA, it is available if MISO calls on it for emergency energy.  And, as the IMM has noted more than once, even generators that do not clear in the PRA may be subject to economic withholding repercussions if they do not participate in the energy markets.  The argument - - that surplus generation beyond the Planning Reserve Margin (PRM) provides additional reliability and should be compensated through the PRA - - is unsupported.

OPT-OUT IS REQUIRED IF THE RBDC IS IMPLEMENTED

If MISO insists on proceeding with RBDC implementation, an Opt-Out Option is required. An Opt-Out Option would recognize that a Load Serving Entity (LSE) will satisfy its PRMR with generation it owns or has purchased bilaterally, and it will insulate an LSE from any (i) exposure to PRA pricing, and (ii) change to its capacity obligation (PRMR) determined according to NERC’s 1 in 10 standard.

MISO’s AFRAP option does not meet the requirements of an Opt-Out Option. Neither AFRAP nor the existing FRAP allow an LSE to avoid increases in its capacity obligation under a RBDC. Combined with the impacts from MISO’s SAC proposal, the increase in capacity requirements would impose significant negative cost consequences on many customers, without sufficient demonstrated benefit.

MISO must recognize that any support previously provided by OMS for the RBDC/DSDC was expressly conditioned on incorporation of an Opt-Out Option.  Suggestions that the Opt-Out Option was not clearly defined at that time are irrelevant.  The concept of Opt-Out is to be insulated from the effects or held-harmless from the repercussions of the mechanism to which the Opt-Out applies. In this case, it means being held-harmless from additional costs and changes to existing capacity requirements.

Providing the Opt-Out Option will ensure that those states/LSEs that choose to participate in the PRA with a RBDC are free to do so, and those that choose not to participate need not.  If states/LSEs agree that the PRA is beneficial, they will likely participate.

PLEASE REPLACE PRIOR FEEDBACK FROM WEC ON THIS ITEM WITH THE FOLLOWING:

WEC Energy Group supports the comments submitted by the Public Service Commission of Wisconsin (PSCW) Office of Regional Markets (ORM) Staff and MidAmerican.  WEC Energy Group has previously submitted feedback on the RBDC that mirrors their comments and observations.

If MISO continues to move forward with the RBDC proposal, we renew our request for an opt-out provision that does not contain restrictive provisions such as a three-year commitment, commitment of 100% of the LSE's obligation, and the 3-year moving average of cleared margin percentages from past PRAs.  As aptly noted in the comments of other stakeholders, MISO's RA construct was not designed as a multi-year forward capacity market; rather, the PRA is a balancing market for residual capacity that results from the lumpiness of generation (and load) retirements and additions.  The opt-out needs to preserve the fundamentals of MISO's RA construct and not include conditions.

MidAmerican appreciates the opportunity to provide feedback on the Reliability-Based Demand Curve (RBDC) proposal and design elements.  (RASC-2019-8) (20230228-0301).

MidAmerican has several concerns over the use of a Reliability-Based Demand Curve (RBDC) and the calculation of net-Cone.

Of greatest concern is that the use of the RBDC fails to honor a guiding principle of stability. MISO cannot ignore this principal. Early in the process MISO had a criterion that the planning resource process should have the ability to reasonably inform state and utility long-term resource planning processes which rely on accreditation. If the RBDC is used, load serving entities (LSEs) are forced to do resource planning with an unknown target. LSEs have no idea going into an auction if they are long, short, or flat.

The second concern is the use of Marginal Reliability Impact (MRI) curves to create the slope of the RBDC curve and using the value of lost load (presumably) to create the curve. Every LSE has their own value of lost load and MISO is trying to use a one size fits all approach where it doesn’t apply.

Finally, the use of net cone is problematic. Not only is the use of a combustion turbine to determine inframarginal rent incorrect, but the planned location within a zone also has a significant impact in determining net cone. For example, in zone 3, if a unit is built on the west side of Iowa, it will likely be receiving $10 less per MWh on an annual basis than a unit built on the east side of the state. MISO will also need to rely on the price of natural gas for this process which is volatile enough that is might make the results incorrect. Net cone also encourages entities to buy from the market. Net cone is not high enough to incent to build a unit that would be annualized cost over 20 years, so unless forced to pay net cone for 20 years, it is cheaper to pay net cone then to build generation.

In the beginning the Planning Reserve Auction (PRA) was intended to be used as a balancing market where small volumes would be cleared but it is now being advertised as a tool that should be used to make decisions on resource planning. Since the auction clears in April and the planning year starts in June it can’t be used in that manner. MISO, the regulators, and the stakeholders need to decide if this process needs to truly be a full capacity market with auctions three to five years in the future with an obligation to build or remain as a balancing market. Stakeholders are already making decisions about retirement of thermal units which has nothing to do with any projection of PRA results.

Ameren Missouri supports MISO's effort to introduce a demand curve that aligns with the current Resource Adequacy construct.  A Reliability Based Demand Curve (RBDC) will clear resources with reliability attributes in MISO's Planning Reserve Auction while remaining technology neutral.  Currently, incremental capacity is undervalued and leads to uneconomic retirements of resources needed for reliability.   

Feedback by Public Service Commission of Wisconsin (PSCW) Office of Regional Markets (ORM) Staff to the Midcontinent Independent System Operator (MISO) on the Reliability-Based Demand Curve (RBDC) proposal and design elements 

MISO has stated the that their current RBDC plan is untenable if a significant portion of participants opt out. Given the discrepancy between that aspect of MISO’s plan and the OMS statement[i] which conditions support for the RBDC on an opt out, we encourage MISO to shift away from designing specific RBDC elements and instead to enter a more open-ended, exploratory phase where objectives of the RBDC and the capacity market may be considered. It may be that there are design elements of a revised RBDC plan that could satisfy the underlying concerns of states desiring an opt out. 

For example, one concern is that although the current proposal would more appropriately compensate resources for capacity when they incidentally have excess, there is not necessarily a mechanism to clearly incentivize new generation as resources that intentionally build out to the RBDC (e.g., planning to build to meet an additional two to three percent of capacity beyond the PRMR) could incur losses if those resources had a higher actual Net CONE than MISO’s target Net CONE. Utilities may have a higher Net Cone because of their resource type, the age of the resource, the scale of their resource, etc. Furthermore, recouping the cost of building via Net CONE is achieved over 20 years of payments, which presents risk as over that time supply, technology, and market rules will change.  Moreover, if the hope is that merchant generation will be built in LRZs where generation is not owned by vertically integrated utilities, it is not at all clear that the RBDC will send the price or consistency signals that merchant generators would likely need to facilitate their investment.

If MISO does consider moving toward a more exploratory phase, some potential ideas we hope MISO would consider include: 

• If there are ways to more clearly incentivize resources in a geographically targeted way 
• Whether zone-specific curves could result in solutions that better achieve resource adequacy goals for each zone or state 
• If there are ways to make MISO’s targets (e.g., CONE or Net CONE) more competitive with alternative investment opportunities such building in PJM