Affordability and Grid Optimization in the DMV: The Important Context Heading Into July 9

DMV

Jul 3

Grant van Wyngaarden, Bandorie

The DMV is entering an energy affordability inflection point. In two years, PJM capacity costs have risen by close to an order of magnitude, turning an invisible wholesale mechanism into a direct driver of customer bills. The 2025/2026 auction cleared at $269.92 per MW-day across most of the footprint, up from $28.92 the year before, with the BGE zone at $466.35. The next auction, for 2026/2027, hit the footprint price cap of $329.17, with Dominion's Virginia zone at $444.26. PJM's total capacity cost jumped from roughly $2.2 billion to $14.7 billion in a single cycle, and that bill is now reaching ratepayers, regulators, and large-load customers at once.

That is why AEG's Q3 theme, Affordability and Grid Optimization, fits this moment. What follows are the issues worth having in front of every participant before July 9. Viewed from the supply chain that actually builds and equips the grid, the transformers, switchgear, and batteries, and the crews that install them, the connective tissue is clear: underneath the affordability debate sits a delivery and coordination problem. The region has committed to serve far more load than it can currently build, equip, or interconnect, and price is the symptom.

The important topics at a glance

The capacity-price shock and the new affordability politics. Record PJM auction results are flowing into DMV bills, and Maryland's governor is pressing PJM directly.
Who pays for the data-center boom. Virginia and Maryland have both moved to a separate rate class for hyperscale load, reshaping cost allocation.
The equipment supply chain is a real constraint. Transformer lead times of two years and longer have replaced capital and permitting as the gating factor on grid growth.
Interconnection reform and a queue that dwarfs the need. PJM's reformed, first-ready process drew 811 projects and 220 GW, far more than the region can absorb.
Optimizing the grid we already have. Grid-enhancing technologies and demand flexibility are the cheapest, fastest capacity available.
Storage as the marginal resource, and the supply risk it carries. Batteries increasingly bridge the capacity gap, yet they are the thinnest voice in the DMV room.

These six issues look distinct, but they are tightly coupled. A decision in one, a cost-allocation rule, an interconnection reform, an equipment lead time, directly shapes the others, which makes July 9 more useful as a single systems discussion than as six separate ones.

1. The capacity-price shock and the new affordability politics

A quiet wholesale market is now front-page news across the DMV. Much of the BGE-zone increase traces to reliability-must-run contracts for the Brandon Shores and Wagner plants near Baltimore, which PJM is paying to keep online until new transmission arrives around 2028. Maryland's Office of People's Counsel warns the results could raise residential bills by 2 to 24 percent depending on zone, and Governor Moore has addressed PJM's annual meeting directly to demand ratepayer protections. The deeper signal is simple: when a capacity price runs from $28.92 to the cap in two years, the market is saying new supply cannot be built fast enough to keep pace with committed load. The debate is increasingly political, but the constraint is physical.

For the room: is the capacity-price spike best treated as a market-design problem to be repriced, or a supply problem that only new build, and a faster path to energizing projects, can solve?

2. Who pays for the data-center boom

Northern Virginia is the largest data-center market on the planet, carrying a large share of global internet traffic across more than 4,900 MW of commissioned power. Dominion says orders in hand could double Virginia's data-center capacity by 2028 and approach 10 GW by 2035, lifting its summer peak from about 17,000 MW toward 29,000 MW and pushing its 2025 to 2029 capital plan near $50 billion. The affordability question is no longer whether to build, but who pays.

Both states have answered. In November 2025 Virginia's State Corporation Commission approved a new GS-5 rate class for loads above 25 MW, effective January 2027, requiring hyperscale customers to cover at least 85 percent of contracted distribution and transmission demand and 60 percent of generation demand, plus upfront collateral, and ordered Dominion to propose a fairer cost-allocation method. Maryland's 2025 Next Generation Energy Act made it the first state to require a separate data-center rate schedule, alongside $200 million in rebates. Cost allocation is where the region's two priorities collide: the economic development hyperscale investment brings, and protection of ratepayers who did not create the load. It also determines how quickly and confidently the long-lead equipment behind all of it can be ordered.

For the room: should hyperscale load fully internalize its grid costs, or is some socialization unavoidable, and at what point does cost-allocation policy start to slow the development the region wants?

3. The equipment supply chain is a real constraint

A large power transformer that took 7 to 14 months to source before the pandemic now runs 18 to 30 months, with Wood Mackenzie's mid-2025 survey putting standard units at 128 weeks and some orders at four years. The national supply deficit sits near 30 percent, and most of the roughly $2 billion in new North American capacity does not arrive until 2028. Equipment availability, not capital and not permitting, has become the gating constraint on grid growth: a project can be approved and financed and still sit unenergized because the long-lead iron is not on the dock. The deeper exposure is geopolitical, since transformers depend on grain-oriented electrical steel and a supplier base concentrated outside the United States, and scarcity feeds straight back into bills as competing buyers drive prices up and schedules out.

For the room: does the region keep competing for scarce equipment company by company, or is there a case for aggregating equipment orders across utilities and large loads?

4. Interconnection reform and a queue that dwarfs the need

PJM has moved from first-come to first-ready, first-served, and its first reformed cycle drew 811 projects totaling 220 GW, including more than 67 GW of storage. Against roughly 30 GW of net demand growth expected between 2024 and 2030, the lesson is that a queue is not a pipeline: much of that 220 GW is phantom capacity that will never be built, and the scarce commodity is bankable capacity with committed equipment and firm offtake. PJM's proposed Expedited Interconnection Track for large new loads aims to move the deliverable projects faster, and the reform rewards readiness, which in practice means whoever locked in their transformers, switchgear, and battery supply early.

For the room: what would it take to convert phantom queue positions into deliverable megawatts on a timeline that matches the region's load growth?

5. Optimizing the grid we already have

If new capacity at cap prices is the expensive answer, getting more from existing wires is the cheap one, and the fastest. In March 2026 PJM began using ambient-adjusted ratings, which can lift line capacity 15 to 40 percent, and grid-enhancing technologies more broadly, dynamic line rating, high-temperature reconductoring, and power-flow control, unlock capacity without new right-of-way or multi-year builds. An RMI analysis found roughly 95 such projects across PJM could deliver on the order of $1 billion per year in savings. Demand flexibility is the companion lever, with lead times in months rather than the years a transformer now demands. If these options are cheaper and faster, the binding constraint is not technical feasibility but institutional: utilities earn on capital projects, not on optimizing what they already own. As a result, the lowest-cost solutions are often the least likely to be pursued first.

For the room: what change in rate design or planning practice would make the cheapest, fastest capacity the default option rather than the exception?

6. Storage as the marginal resource, and the supply risk it carries

Battery storage increasingly bridges the capacity gap in PJM, yet it is the thinnest part of the DMV conversation. The absence is structural: storage developers are a younger, fragmented, merchant-heavy segment outside the utility and advisory networks these convenings usually draw on, so the resource carrying the most weight on the margin is the least represented in the room. It matters because PJM's treatment of storage, its accreditation, dispatch, and hybrid rules, is still maturing, and a storage RFP today is a bet on cell-chemistry availability, on shifting domestic-content and trade rules, and on a delivery window that has to line up with an interconnection date the developer does not fully control. That is not a financing problem. It is a sourcing and contract-management problem, and the kind of risk that gets transferred badly when the parties have not coordinated up front.

For the room: how should the DMV bring storage developers and suppliers into the planning conversations they are currently outside of?

Helpful focus for July 9

The common thread is coordination. The DMV is three jurisdictions, the District, Maryland, and Virginia, sharing one grid and one strained equipment supply chain, with three different definitions of success. Affordability will not be solved in any single lane; it is a sequencing problem across utilities, large load, OEMs, EPCs, developers, and regulators, and right now no one owns the sequence. Coordination is not failing because the problem is unclear. It is failing because responsibility is diffuse.

Article contributed by AEG partner Bandorie, a clean energy project procurement consulting firm. Written by Grant van Wyngaarden, lead consultant at Bandorie, whose career has been focused on sourcing the equipment and EPC services behind major energy infrastructure, including at Ørsted as Head of Offshore Procurement, Americas. Grant supported AEG in 2017 and 2018, most prominently as Director of Memberships.