VALLEY FORGE, Pa. — PJM’s Markets and Reliability Committee endorsed two proposals to revise the RTO’s effective load carrying capability (ELCC) formula to add two new generation categories and limit the penalties resources face if their accreditation declines between a Base Residual Auction (BRA) and Incremental Auction (IA).
The volatility of unit ratings after auctions has been a sticking point for generation owners, who say it is unfair to commit a resource in the auction only to reduce that unit’s accredited capacity (AUCAP) afterward.
And particularly so when ELCC ratings are falling due to changes in load forecasts, they argued.
The endorsed proposal, Package C, would limit the deficiency rate for a resource that has its rating reduced after being committed in the BRA to 100% of the clearing price, rather than the 120% penalty rate. Resources could still be subject to the penalty rate if they cannot meet their committed capacity because their installed capacity (ICAP) declined, such as due to unit failure, or if a planned unit does not come online according to schedule.
The proposal passed with 80% sector-weighted support, after an initial vote narrowly missed the two-thirds threshold at 66.08%.
PJM’s Pat Bruno said the proposal would retain an incentive for market sellers to avoid the deficiency by procuring additional capacity through bilateral transactions or in the IA without subjecting them to a penalty rate. Resources would also be held to their original commitment during a performance assessment interval (PAI).
He gave the example of a resource with 100 MW of ICAP that is committed at 90 MW in the BRA. If its AUCAP were to fall to 80 MW in an IA, it would be assessed a 10-MW deficiency charge at the clearing price unless it procures additional capacity. If that unit were to output at 80 MW during a PAI, without having procured capacity to make up the shortfall, it would be assessed a 10-MW nonperformance charge.
The main motion, Package B, would have frozen resources’ ELCC ratings and AUCAP at the values used in the BRA. While resource ratings would not be changed, PJM would continue to update the installed reserve margin (IRM) and forecast pool requirement (FPR) values, necessitating that PJM modify its capacity buy/sell offers to work around any changes in accreditation.
The proposal was rejected by the MRC with 55% support. The two packages were nearly tied in a poll at the ELCC Senior Task Force (ELCCSTF), with Package B holding 66.5158% support and 68% preference over the status quo, while Package C received 66.5025% and 74.9% preference.
Load-serving entities, consumer advocates and Independent Market Monitor Joe Bowring argued that would shift all the risk of changing ratings to load, whereas Package C would more equitably split the risk between market sellers and buyers.
Bowring said it shouldn’t be any surprise that ratings can change between BRAs and IAs — it happened with the prior EFORd model as well, but ELCC is more volatile. The difference with both proposals, he argues, is they would inappropriately shift some or all of that risk to load, when it should remain with market sellers, who are capable of mitigating their risk by maintaining high performance when called upon.
‘Emblematic’ Debate
Several market sellers questioned Bowring and PJM on whether they can adjust their offers to reflect the risk of their ratings changing after an auction, noting that under EFORd, they were able to vary the amount of capacity they offered within a band defined by their annual and 5-year average forced outage values. Bruno responded that the ELCCSTF discussed whether that risk could be included in sellers’ capacity performance quantifiable risk (CPQR) values, but that did not make it into the proposal.
Vitol’s Jason Barker questioned whether generators can mitigate the risk by ensuring their units perform well because the class-based approach to accreditation means even a unit with perfect output when called upon can have its rating impacted by similar resources.
Barker also questioned PJM’s ability to identify whether changes in ELCC ratings stem from resource performance or a change in the load forecast. He suggested PJM should procure more capacity if the demand side is responsible for the increased risk but should reduce ratings if sellers are driving the risk.
“This debate is emblematic of problems with ELCC,” Barker said, adding it is creating unfair outcomes for either load or sellers no matter which approach is selected.
Bruno said PJM previously explored but found it could lead to convoluted outcomes, such as scenarios where seasonal risk shifts toward the summer while the ratings for solar units decline.
Calpine’s David “Scarp” Scarpignato said the implications of the proposals are very different when auctions are being held a year in advance with only one IA, versus the standard three-year, three-IA cadence. In the latter, he said there is more opportunity for large changes in the load forecast or a PAI, causing significant shifts in ELCC ratings.
The proposal to add new resource classes would establish oil combustion turbines (CTs) as their own bucket, organizing them from the miscellaneous “other unlimited resource” category, and breaking waste-to-energy as its own class from “steam.”
Bruno said PJM ran a sensitivity based on the 2025/26 IA and found waste-to-energy would have an 83% ELCC rating, while oil CTs would be around 85%. Since there is a relatively small amount of capacity offered by waste-to-energy, pulling it out is expected to have little impact on the steam class. Other unlimited resources have unit-specific analysis, so combining their ratings is expected to have minimal impact.
Bruno told RTO Insider that grouping oil CTs together as a class better captures correlated outages and increases the amount of performance data available for modeling a particular unit. Since there is a limited number of PAIs from which to draw performance modeling, he said grouping units can smooth the impact of outages that happen at a consistent rate across that class.
