Assessing Electricity Resource Adequacy in Pennsylvania and PJM
Pennsylvania’s electricity system is experiencing a tightening of supply and demand. There is high load growth in the state, largely due to data center development, while supply is also constrained as aging generators retire and new resources deployment is hindered. These two simultaneous trends cause high electricity prices, and they also expose other challenges: recent assessments from grid operator PJM and the North American Electric Reliability Corporation (NERC), the national body charged with regulating electric system reliability have both flagged that PJM’s system is increasingly stressed, with looming reliability concerns.
In the context of ongoing policy proposals, Pennsylvania PUC commissioned an independent analysis aimed at developing load projections and evaluating resource adequacy under a set of likely possible futures. Synapse Energy Economics led a team using capacity expansion and resource adequacy scenario modeling to (a) develop projections of future resources through the long term (2040) in light of different load projections and (b) assess system reliability performance through the near term (2030) to better understand the kinds of stresses Pennsylvania’s electric system and the wider PJM grid face.
The team modeled three scenarios and found the following: in both the Reference scenario and the High Load And Low Supply scenario, we find that from 2027 through 2030, the PJM planning criterion for resource adequacy (loss of load expectation or “LOLE” of 0.1) is not met in PJM. Meanwhile, in the No New Data Centers scenario, LOLE is below this 0.1 threshold.
In the Reference scenario, which relies on PJM’s 2026 Load Forecast, modeled 2030 LOLE is 0.59 or nearly six times worse than the PJM planning criterion. This result is mainly due to large-load additions stemming from data center additions, which are projected to surge in 2030. In 2030, the High Load and Low Supply scenario, a worst-case future of higher-than-expected load additions and constrained resource deployment, has an LOLE of 13.20. This LOLE value is over 100 times worse than PJM’s planning criterion and indicates an expectation of more than 13 days with loss of load events per year. As in the Reference scenario, these reliability issues are largely due to surging data center additions. In the High Load forecast, data centers are projected to contribute over 200 TWh of additional new load between 2029 and 2030 PJM-wide. (For reference, the entire state of Pennsylvania’s 2025 electricity demand was about 150 TWh). In this High Load and Low Supply scenario, in-state generators are unable to meet between 2 and 19 percent of PJM-wide electricity between 2031 and 2040.