PJM's reliability backstop procurement is moving from concept to design in response to two consecutive capacity auction shortfalls. The April 10, 2026 proposal sets an initial procurement target of approximately 15 GW across the RTO, with 84% concentrated in four transmission zones driven by data center load growth: Dominion, AEP Ohio, ComEd, and PPL.
For community solar underwriting, this is not a single retail rate story. It is at least two.
In high-allocation zones, backstop costs will flow into retail rates as meaningful uplift, translating to stronger bill credit values for CDG subscribers. In low-allocation zones like BGE and PEPCO, the direct cost burden is negligible. Those zones will still absorb a suppressive effect as roughly 15 GW of pre-funded capacity enters future Base Residual Auctions as price takers, which compresses capacity clearing prices and puts downward pressure on retail rates and bill credit values.
The same procurement produces opposite retail effects depending on where your projects sit.
This paper walks through the auction dynamics that triggered the backstop process, the cost allocation framework still under development at both the PJM and state levels, Sitara's scenario analysis for BGE residential rates, the actual April 10 target distribution, and the second-order market effects that will shape retail trajectories across PJM through the end of the decade.
Two failed auctions and the backstop trigger
The December 2025 Base Residual Auction for the 2027/28 delivery year cleared at PJM's administrative price cap of $333/MW-day and still came up 6,623 MW short of the reliability requirement. It was the second consecutive auction shortfall.
PJM disclosed in post-auction filings that the 2027/28 auction would have cleared at approximately $529/MW-day without the price cap. That number matters. It tells you the market is signaling capacity scarcity that the administrative mechanism is preventing from clearing.
Under PJM tariff provisions, a third consecutive shortfall triggers a formal backstop procurement obligation. The June 2026 auction for the 2028/29 delivery year is the next test. If it clears short, the formal trigger is met and PJM is obligated to procure capacity through a mechanism outside the normal BRA process.
PJM's board did not wait for the trigger. On January 16, 2026, the board directed staff to begin immediate design of a reliability backstop procurement regardless of the June auction outcome. That decision was driven by procurement lead time. Even if the June auction clears adequately, the 2027/28 shortfall still needs to be filled, and new generation requires interconnection study and construction timelines that extend well beyond the next delivery year.
Parallel to PJM's board action, the White House and all 13 PJM state governors issued a joint proposal in January 2026 calling for 15-year capacity contracts funded by $15 billion in new generation investment. The federal-state proposal shares the core mechanic of PJM's own design: procure new capacity outside the BRA at potentially elevated prices, then allocate those costs to load.
Both frameworks leave the cost allocation question open. That is where the retail rate story starts.
The cost allocation question
PJM's February 18, 2026 Reliability Backstop Design Working Paper laid out the options for how backstop costs would flow from the wholesale market to the retail bill. Three threshold decisions drive the retail outcome.
The first is whether costs are allocated RTO-wide or by transmission zone. An RTO-wide allocation spreads costs across more than 856,000 GWh of annual load and produces a small per-kWh adder applied uniformly. A zone-based allocation concentrates costs in the transmission zones where load growth is driving the shortfall, producing larger per-kWh adders for customers in those zones.
The second is whether zone-based allocation is weighted by total load or by the load growth contributing to the reliability requirement. The distinction matters because fourteen PJM zones carry data center load growth adjustments in the 2026 Long-Term Load Forecast, but those fourteen zones vary widely in the magnitude of that adjustment.
The third is whether costs are directly allocated to the load classes driving the growth, particularly data center load. PJM's working paper was explicit on this point: PJM does not have the ability under its tariff to allocate costs directly to specific customer classes like data centers. That allocation decision sits with state regulators, who would need to create new tariff structures that do not currently exist.
The February working paper flagged these options. The April 10, 2026 proposal moved several of them to initial design positions, but the full allocation mechanics remain under stakeholder process.
For any firm underwriting community solar economics against a 20-year retail rate trajectory, these open questions translate directly into modeling uncertainty.
Three scenarios for BGE residential rates
To bound the range of possible outcomes for BGE residential customers, Sitara modeled three illustrative scenarios using PJM's February working paper options as the framework. These scenarios were developed prior to the April 10 proposal release and are presented here to show the analytical range. The actual April 10 allocation, covered in the next section, falls well outside the middle scenario.
Scenario 1: Full socialization across PJM
Backstop costs allocated RTO-wide across the full 856,000+ GWh of annual load. The per-kWh adder works out to approximately $1.50/MWh. For a BGE residential customer consuming 12,000 kWh per year, the bill impact is roughly $18 annually. Modest at the customer level, with a relatively simple cost recovery mechanism from a tariff perspective. This scenario minimizes rate impact on CDG bill credit values.
Scenario 2: Zone-allocated to high-growth areas
Backstop costs allocated proportionally to the fourteen PJM zones carrying data center load growth adjustments in the 2026 Long-Term Load Forecast. Under an illustrative 15% BGE share of the allocated costs, the BGE residential adder works out to approximately $6.10/MWh. This scenario was developed as a middle case reflecting zone-based allocation that concentrates costs where growth is occurring. It represented the scenario most likely to materially change a 20-year BGE rate trajectory, with downstream implications for community solar bill credit values.
Scenario 3: Data center specific allocation
Backstop costs allocated directly to data center load through a new state-level tariff structure. The residential customer impact falls to approximately $0.68/MWh. This scenario is structurally different from the first two because it requires state regulatory action. It is the scenario most consistent with the policy intent expressed by state governors, but the most operationally complex to execute.
The April 10 proposal: zone-level target distribution
On April 10, 2026, PJM released its Reliability Backstop Procurement Design presentation, including a slide showing the initial procurement target by transmission zone. PJM's initial procurement target is approximately 15 GW across the RTO. Four zones carry 84% of the total.
These four zones map directly to the major data center growth corridors in PJM: Northern Virginia, central Ohio, the Chicago suburbs, and eastern Pennsylvania. The alignment is not coincidental. PJM's initial target distribution reflects the load forecast adjustments that drove the reliability shortfall in the first place.
BGE's initial allocation is 92 MW, or 0.6% of the RTO target. PEPCO's allocation is zero.
For BGE residential rates, the April 10 allocation drops the direct backstop adder from the ~$6.10/MWh illustrative zone-weighted scenario to under $0.50/MWh. The scenario space has collapsed to the low end.