67 million Americans are waiting on a power-grid decision that just moved to the states

The country's largest power grid has spent years clearing a backlog of clean-power projects waiting to plug in. The bottleneck is no longer at the queue. PJM Interconnection, which runs the wholesale power market for 67 million customers across 13 mid-Atlantic and Midwest states, completed its first reformed cluster study in 2025. The result: more than 50 gigawatts of solar, wind, and battery projects now hold interconnection agreements, the only pool that can plausibly come online before 2030.

The next binding question is not approvals. It is whether those projects are actually built in time to matter.

PJM operates the country's biggest competitive capacity auction, the mechanism that pays generators to be available to meet future demand. Over the past two auctions, that auction hit a record $16.4 billion in costs, more than eight times prior levels. Those costs flow into utility bills for households in Virginia, Maryland, Pennsylvania, Ohio, and the rest of the PJM footprint. The capacity crunch is one of the main reasons ratepayers in PJM states have seen bills climb.

PJM's reform is a bet that the market can pick the mix once the queue moves. The new cluster study process reviews groups of projects together, rather than the old first-come, first-served review that left some applications waiting since 2018. Approval is the easy half of the problem. A 2025 Lawrence Berkeley National Laboratory analysis of projects that signed interconnection agreements between 2000 and 2022 found that 41% of them eventually withdrew. Some were speculative, some blocked by network upgrade costs, some lost their offtake. The historical baseline is that "approved" is not "built."

The new cluster pool is roughly 50 GW of solar, wind, and battery capacity. Sitting alongside it is just under 5 GW of gas-fired plants and a roughly equivalent amount of batteries that are approved but not yet built. The contrast matters. Solar and wind have low capacity value in PJM's market design because they generate when the wind blows and the sun shines, not necessarily at peak demand, so they contribute little to relieving the capacity auction crunch. Batteries and gas do. And batteries can be built about two years faster than a new gas plant, while gas turbine backlogs now push new builds into a five-to-seven-year window, according to NRDC and Aurora Energy Research.

PJM's proposed capacity regime imposes a 2032 deadline for resources to be in service. Only projects with interconnection agreements now, or in the early 2027 cycle, can realistically hit that window, Aurora's Julia Hoos told Canary Media. Everything after that is effectively a different problem.

The execution risk sits at state utility commissions and the utilities themselves. Network upgrades, including new transmission lines, transformers, and substation work needed to actually connect approved projects, are a leading cause of delays. Utilities in PJM do not get to roll interconnection-driven upgrades into rate base, the regulatory mechanism that lets them recover costs from ratepayers and earn a return, according to a Johns Hopkins, Aurora, and Halcyon analysis. Without that recovery, utilities deprioritize the work. The projects are approved on paper, but the wires to connect them are not funded.

An RMI report on PJM's "speed-to-power" problem recommends letting utilities recover interconnection-driven network upgrade costs in rate base. That is a state-level decision, made by public utility commissions in each of PJM's 13 states, often in coordination with governors and legislatures. It is the lever the source says actually has to move.

Permitting is the other state-level friction. Ohio is moving to restrict wind and solar siting even as neighboring states talk about interconnection reform. Rural PJM states that host the land for wind and solar farms are also the places where local opposition has been loudest. A 50 GW paper queue does not turn into 50 GW of electrons if the towns where the projects would be sited say no.

GridLab's analysis of eastern battery storage markets notes that PJM has barely built any grid batteries to date, lagging far behind California (CAISO) and Texas (ERCOT). The 50 GW figure is a starting condition, not a forecast. The question is whether state utility commissions, utilities, and permitting authorities can close the execution gap before the 2032 capacity deadline, and before data-center demand growth makes the capacity auction even more expensive.