PJM has withdrawn its non-capacity backed load (NCBL) proposal, shifting the focus of its solution for rising large load additions (LLAs) to creating a parallel resource interconnection queue, reworking price-responsive demand (PRD) and providing more insight into the load forecasting process for state utility commissions. (See PJM Revises Non-capacity Backed Load Proposal.)

The changes were presented Oct. 1 to stakeholders as part of PJM’s Critical Issue Fast Path (CIFP) process addressing LLAs, now in its second phase, in which design components are fine-tuned before being bundled into comprehensive solutions during phase 3. Another phase 2 meeting is scheduled for Oct. 14 with 11 proposal sponsors set to present.

PJM’s proposed expedited interconnection track (EIT) aims to create a pathway for resources capable of quickly entering service to receive a generator interconnection agreement through a 10-month study process. Applicants would be required to pay a nonrefundable study deposit starting at $500,000 and a $10,000/MW readiness deposit, as well as commit to being in service within three years of requesting an EIT study, though output may be limited if network upgrades are not complete by then. PJM Vice President of Planning Jason Connell said the EIT is envisioned as a permanent addition to the RTO’s interconnection processes.

If a resource does not enter service within three years, it would forfeit the readiness deposit and be subject to the same penalties for breaches of project milestones in the standard interconnection process.

All fuel types would be permitted, but projects would have to be at least 500 MW to participate and only 10 applications would be approved annually. The studies would be conducted according to when they were requested, and network upgrade costs would be assigned individually. No changes would be permitted in site control or attributes such as fuel type, nameplate capacity or equipment type.

Applications would be required to receive sponsorship from the state in which the resource would be located, which Connell said is intended to provide a degree of buy-in and reduce the odds that a project might receive expedited treatment from PJM only to become mired in siting and permitting challenges.

Connell said PJM decided on the 500-MW requirement by determining it would meet the amount of annual load growth expected while limiting the impact to projects in the standard interconnection queue. If a smaller requirement and larger number of applications were allowed, PJM found that would extend the amount of time needed to complete interconnection studies and defeat the purpose of an expedited pathway, he said.

Grant Glazer of MN8 Energy questioned if PJM would consider allowing a portfolio of projects to be included as one application to reach the 500-MW requirement. He said projects with a lower voltage and smaller nameplate capacity would be faster to develop and could provide a more economic form of capacity than larger resources.

PJM’s Tim Horger said EIT studies would use the latest system model case, and the upgrades they’re assigned would be added to the modeling for the next queue cycle. For any projects submitted while Transition Cycle 2 is ongoing, the latest model for that cluster would be used, and the resulting network upgrades would be added to the modeling for Cycle 1.

Adrien Ford, Constellation Energy’s vice president of wholesale market development, questioned if PJM would consider shrinking the 500-MW threshold, saying there are 300-MW uprate projects to nuclear units that could take advantage of the process.

Connell said PJM did not focus on facilitating uprates, as there are already opportunities for their studies to be accelerated.

Unpopular NCBL Dropped

The NCBL concept would have required participating large loads to forgo the guarantee of capacity, exempt them from paying for the service and removed that load from the capacity market.

It would have been triggered if the amount of forecast supply in a Base Residual Auction (BRA) fell short of the amount of expected demand.

The mandatory variant of the proposal received the greatest backlash from stakeholders, who argued it would make the PJM region unattractive for data center developers and undermine market signals. Opponents also argued that making the model voluntary would not solve jurisdictional issues around the RTO defining the retail service consumers could receive.

PJM sought to address the jurisdictional challenges by shifting the responsibility for assigning NCBL status to customers onto electric distribution companies and load-serving entities. It would have determined the RTO-wide amount of NCBL that would be needed to meet the reliability requirement in an auction and allocated portions to zones according to the amount of planned large loads forecast.

Claire Lang-Ree, an advocate with the Natural Resources Defense Council, said it was unlikely that resources utilizing EIT would be able to enter service before 2030, and thus would be unable to help with high capacity prices until after 2033. She questioned whether PJM’s revised proposal could deliver the same reliability as the mandatory NCBL concept.

Old Dominion Electric Cooperative’s Mike Cocco said removing NCBL from PJM’s proposal eliminates the original’s core design component from a reliability perspective. He said the load growth in PJM is unprecedented, and there needs to be a way to ensure it can be integrated reliably without impacting existing consumers, which NCBL would have accomplished. He suggested that changes to the manual load shed procedures could provide a similar benefit, but these decisions need to be part of the centralized CIFP solution, as the issue will only become more contentious if stakeholders wait to negotiate until after the capacity auction.

Horger said PJM is considering changes to manual load shed, but those will likely come outside the CIFP process.

Additional Changes to CIFP Proposal

Instead of the NCBL construct, Horger said PJM is now proposing changes to PRD to encourage flexibility from large loads.

The dynamic retail rate for PRD would be replaced with an energy market price, with the strike price serving as the offer. Horger said the change would make PRD function similar to a voluntary NCBL construct.

PJM is also proposing changes to its load forecasting process to add a step in which state utility commissions could review and provide feedback on the LLAs submitted by utilities under their jurisdiction.

Entities submitting LLAs would be required to ask the customer requesting service whether they are considering multiple sites for their projects and provide that response to PJM. Horger said the change is intended to identify instances where several utilities are projecting load growth for a data center that will ultimately only be built in one location.

Commitments to procure a minimum amount of capacity for planned large load customers are also being considered.

PJM Executive Vice President of Market Services and Strategy Stu Bresler said the RTO is open to exploring a model for long-term capacity procurement, either as part of its CIFP proposal or through subsequent stakeholder processes. He noted that the reliability backstop auction provides for some of that capability already, albeit following three years of the capacity market falling short of the reliability requirement and FERC approval of its implementation.

Advanced Power Proposes Higher Maximum Price

A design component from Advanced Power would double the maximum price of an Incremental Auction (IA) if the corresponding BRA clears short of the reliability requirement and use the increased ceiling for the subsequent BRA if the higher price is needed to clear enough capacity.

Ron Paryl, vice president of markets and risk management for Advanced, said this would create an additional opportunity for demand response to resolve the shortfall, while also allowing the auction to be responsive to updates to load forecasts and provide price discovery for the value of capacity. It would also avoid discrimination between consumers and allow those most price-sensitive to avoid high capacity costs, he said.

Advanced also proposed to lock resources’ effective load-carrying capability ratings if they would fall between a BRA and corresponding IAs, preventing sellers in the BRA from having to procure additional capacity to cover their commitment, particularly if prices increase above the original maximum price under the company’s first two components. If ELCC ratings increase, the resource owner would be able to bid that additional capability into the auction.

The potential for changes to the load forecast to shift resources’ ELCC ratings was seen in discussions around how to apply the 2025 load forecast to the parameters for the third 2025/26 IA; the forecast led the risk profile to shift toward the winter, causing ratings for several resource classes to fall. PJM opted not to include preliminary figures from the forecast, and stakeholders voted to lower the Capacity Performance penalties resources face if they cannot meet their commitment due to falling accreditation. (See PJM Stakeholders Endorse Proposals to Rework ELCC Accreditation.)

Stakeholders questioned how DR offers are mitigated and whether the proposal would create market power concerns, while DR providers said adding reviews of offers would be complicated for aggregated resources.

Paryl said there is no requirement that DR be mitigated and so it should be able to make offers the sellers feel represent the costs for them to curtail.

Joint Proposal from Suppliers, Data Centers

A proposal from large suppliers and data centers would focus on making the forecasting of large loads more accurate and add triggers for demand and supply-side solutions. The proposal was sponsored by Calpine, Constellation, Talen Energy, Amazon, Google and Microsoft.

Large loads would be required to provide commitments, such as electric service agreements or arrangements to bring their own supply, in order to be included in the load forecast. A “reality check” would look at possible supply chain constraints, historic completion rates and other factors that could inhibit the number of projects completed. Characteristics such as ramping and utilization would also be factored in.

If a BRA falls below 98% of the reliability requirement, the demand-side solutions would be implemented immediately in that auction, starting with a voluntary large load DR model where participation is limited to a set number of hours a year, with reduced ELCC ratings. That could be followed by deployment of a new emergency procedure dispatching emergency backup generation to bring some of the large loads off the PJM system. The final step would be a curtailment of large loads participating in a voluntary model akin to NCBL.

If a shortfall persists after the demand-side options have been implemented, the proposal would see PJM solicit multiyear commitments of up to seven years, with shorter offers clearing first. Eligible resources include new and reactivated generation, existing generation with an offer cap above the top of the variable resource requirement (VRR) curve and DR. Those resources would clear at the top of the VRR curve and then enter subsequent auctions at the default gross avoidable-cost rate for their technology class minus the unit-specific estimated energy and ancillary service revenues. The clearing price the units would receive would remain the same across the duration of their commitment. The model would be in place between the 2028/29 and 2031/32 BRAs.

Constellation’s Ford said the BRA trigger criteria are important to minimize the impact to the market signals to attract long-term solutions. “We really want to avoid reliance on these potentially lower-quality products,” she said.

Enchanted Rock

A proposal from microgrid and backup power developer Enchanted Rock would establish a voluntary NCBL model in conjunction with states, EDCs and LSEs to allow large loads to be more flexible and create a pathway for them to interconnect ahead of network upgrades that might inhibit their ability to be reliably integrated onto the grid on a firm basis.

Joel Yu, Enchanted’s senior vice president of policy and external affairs, said voluntary NCBL is the best option for providing data centers with the ability to choose their level of flexibility, but there needs to be more adequate incentives on the supply side.

“If that load is making a commitment to provide flexibility via an NCBL structure or perhaps a different structure — as long as that flexibility can be modeled up front in an interconnection study process — we believe there’s an avenue for that load to access some amount of non-firm grid service on a provisional basis,” Yu said. “We’re not proposing any changes or options with respect to broader planning processes, but [it would] help to attract voluntary participation via the speed-to-power incentive.”

Additional Proposals to be Discussed Oct. 14

Several stakeholders have also submitted alternatives, to be presented during the CIFP meeting Oct. 14.

They include a joint proposal from Eolian Energy and the Brattle Group; proposals from the NRDC, Vistra and East Kentucky Power Cooperative; and a package from Johns Hopkins University associate professor Abe Silverman and Sue Glatz, principal consultant at Glatz Energy Consulting.

There will also be presentations from NOVEC, the Independent Market Monitor, Mainspring Energy, the Maryland Office of People’s Counsel and the office of Pennsylvania Gov. Josh Shapiro, but materials from these had not been posted online as of press time.

The EKPC proposal would require that large loads identify the LSE that will serve them before they can be incorporated into the load forecast and VRR curve, and institute “significant” penalties for LSEs that do not cover their own demand through owned or bilaterally contracted capacity. The penalties would only be assessed against LSEs within locational deliverability areas that are short of their reliability requirements in a BRA. Large loads would be defined as at least 50 MW.

Vistra proposed to impose penalties on any LSEs that are capacity deficient during emergency procedures in an effort to create an incentive for physical hedging and load flexibility. It includes a handful of options for how penalties could be determined.

The NRDC proposed a mandatory NCBL variant for any large loads coming online after the 2026/27 BRA that are not bringing their own generation. Large loads would also be able to gain firm service by participating as DR or PRD, or signing other loads to participate on their behalf; their curtailment risk could also be reduced by contracting with energy-only generation.

The Eolian and Brattle package would create a bilateral integration of generation portfolios and load structure for large loads to procure capacity through adjacent supply, with some backup provided by load flexibility. New resources participating would qualify for a 90-day expedited interconnection study and would not have their output derated by ELCC; instead, the owners of the resource and load would share the risk of underperformance.

The proposal from Silverman and Glatz is based on mandatory NCBL for new large loads so long as the capacity auction clears above the midpoint on the VRR curve. Another option would be to bifurcate the auction, first clearing non-LLA customers and then running a second auction for LLAs and any capacity resources that did not clear in the first run. To reduce the potential for double-counting large loads, they proposed to exclude them from the load forecast unless the relevant utility confirms that all distribution and transmission upgrades will be complete on time; the customer attests that it is not considering alternative locations for the project; and the customer can provide evidence of commercial maturity.