​Two competing dynamics set the stage for near-term battery growth in PJM. Surging data center load and a wave of thermal retirements open a firm-capacity gap that the interconnection queue cannot close before 2033. TB4 spreads more than double from $53k/MW-year in 2026 to $112k by 2030, the highest of any Eastern Interconnect ISO, with around-the-clock (ATC) prices peaking at $87/MWh in 2032.

From 2033 onward, the system rebalances as new gas additions catch up with demand. Spreads settle into a structurally lower equilibrium near $69k/MW-year through 2049, with Dominion holding the East premium thanks to sustained data center growth.

​This article covers Modo Energy's PJM fundamental market outlook for Q3 2026.

​Key takeaways

• PJM TB4 spreads peak at $112k/MW-year in 2030 and ATCs at $87/MWh in 2032, both settling to ~$69k and ~$59/MWh by 2049. Projects coming online between 2028 and 2032 capture the highest-value.
• Coincident peak load steps from 173.7 GW in 2029 to 190.8 GW in 2030, 8.8x the median annual peak growth across the 2026–2049 forecast horizon. Dominion data centers account for most of the increment.
• Renewables and storage make up 89% of PJM's interconnection queue, all with low CAFs, set against 9.7 GW of expected thermal retirements by 2030.
• Modo expects 139 GW of new gas by 2049, alongside 147 GW of solar and onshore wind. PJM's most recent forecasted CAFs show annual declines of 4-8 percentage points for wind and storage, locking in continued gas dependence and shaping future capacity-market economics.
• Dominion TB4 settles at $116k/MW-year in 2049, still PJM's highest zone and only 4% below its 2030 peak. AEP and other Western zones collapse to roughly $59k.

​Peak demand in PJM jumps 17 GW in 2030 then collapses

​PJM has the biggest near-term load step in the Eastern Interconnect. Coincident peak grows from 173.7 GW in 2029 to 190.8 GW in 2030 . That single-year jump of 17.1 GW is 8.8x the median annual peak growth across the 2026–2049 forecast horizon.

​Most of the 2030 increment lands in Dominion (Northern Virginia). Secondary cohorts follow in 2031 (+9.1 GW), 2036 (+10.7 GW), and 2041 (+10.7 GW). From 2033 onward, peak growth collapses to roughly 1 GW per year as the initial data center buildout completes.

​For deeper analysis of PJM's data center load trajectory, see Modo Energy's PJM load forecast: data centers through 2046.

​System tightness is expected to worsen in the near-term

​System tightness is expected to continue from 2026 through 2030 as load growth is further exacerbated by thermal retirements.

9.7 GW of PJM thermal capacity retires by 2030, including Brandon Shores and Herbert Wagner in already-constrained MAAC zones. Both have had deactivation dates pushed back over reliability concerns.

MISO retirements compound the tightness given PJM's historical net export to MISO.

Relief from the interconnection queue is short-lived. Expected system additions through 2030 are not expected to offset the combined pressure of retirements and load growth. Modo's evaluation of the PJM queue additions through 2030 includes 25 GW of solar primarily in AEP, 4 GW of onshore wind mostly in PJM West. Offshore wind additions consist of only the 2.5 GW Coastal Virginia Offshore Wind (CVOW) project given the recent string of offshore wind project cancellations.

​Modo's forecasts 139 GW of new gas alongside the renewable expansion in PJM

​Cumulative new builds by 2049 reach 147 GW of solar plus onshore wind and 139 GW of new gas. Renewable buildout does not displace gas. Both grow together.

Gas accelerates fastest, reaching 51 GW by 2035, 95 GW by 2040, and 139 GW by 2049. Solar tracks a similar arc to 84 GW by 2049. Onshore wind hits 63 GW. Battery additions in the model stay small, around 11 GW by 2049, since the interconnection queue already supplies most of the storage build through 2030.

Multiple forces drive the gas trajectory. PJM recently revised its official Capacity Accreditation Factors (CAF) downward for renewables. Offshore wind projects continue to cancel. Data-center load growth keeps climbing. Absent changes to PJM's resource adequacy construct, these dynamics push the system toward more firm gas capacity.

Scarcity conditions amid rising load will challenge state environmental policy. PJM member-states have clean energy mandates like the VCEA in Virginia and CEJA in Maryland, but rising reliability concerns are pushing regulators to grant exceptions. Virginia's SCC approved the Chesterfield gas plant for Dominion in late 2025, while offshore wind cancellations in New Jersey and Maryland widen the gap further.

The within-PJM allocation shapes long-run congestion. The model sites most new gas outside Dominion, while Dominion itself absorbs most of the new solar (around 46 GW cumulative by 2049). Permitting drives where capacity lands, not load.

​PJM prices spike in 2030, then mean revert by 2033

​System ATC prices rise through the late 2020s and peak at $87/MWh in 2032, the highest in the Eastern Interconnect for that year. TB4 spreads more than double from $53k/MW-year in 2026 to $112k by 2030, putting PJM around $45k above the next-highest (ISO-NE and NYISO clustered near $68k).

The boom-bust shape is steeper for PJM than peers. TB4 rises from $101k in 2032 to $77k in 2033, a one-year drop no other Eastern ISO experiences. Gas additions catch up just as peak-load growth slows.

By 2049, PJM is the lowest-spread Eastern ISO at $69k/MW-year, below NYISO ($81k), ISO-NE ($77k), and only marginally above MISO ($51k). ATCs settle near $59/MWh, roughly 33% below their peak. PJM's structurally cheaper gas access prices through to the long-run equilibrium.

​Capacity prices in PJM reflect the same tightness

​PJM's Reliability Pricing Model (RPM) auctions cleared at successive record highs in the last two cycles: $269/MW-day system-wide for 2025/26 and $329/MW-day for 2026/27. Modo's forecast keeps capacity clearing elevated through 2032 as system-wide scarcity persists, before new gas build softens the curve.

​Dominion holds its East premium; AEP and the West fade after 2033

​PJM-wide averages mask substantial intra-RTO dispersion. Site selection within PJM matters as much as choosing PJM over peer ISOs.

The congestion story has two stages. Near-term (2026–2029), Mid-Atlantic and Western prices soften. Coastal Virginia Offshore Wind lands generation inside Dominion. RTEP 24 and 25 transmission upgrades expand import capability into MAAC and Dominion. Long-term (2030 onward), the 2030 peak-load step plus sustained Dominion data center growth reopen the East-West gap. Approved transmission fits today's peak but is overrun by the next load cohort. Preliminary RTEP 26 results target this gap, but nothing is firm as of forecast publication.

BESS site selection within PJM splits along these zonal trajectories. Dominion, BGE, PEPCO, AECO, DPL, JCPL, and PSEG form the East congestion zones with persistent long-run premiums over West. AEP, DAY, DEOK, and ATSI ride the near-term wave but fade after 2033. COMED and RECO move on a different curve that recovers as the system rebalances.

​What does this mean for developers, investors, and lenders?

​Projects reaching commercial operation in 2028–2032 capture TB4 spreads roughly 50% above the long-run PJM average alongside record capacity clearing. Later entrants earn structurally lower spreads, partially offset by falling CapEx through the 2030s.

Location matters as much as timing. Dominion's long-run premium depends on two watch items: the data center load forecast holding, and RTEP 26 transmission not arriving on the model's assumed schedule. East zones (BGE, PEPCO, AECO, DPL, JCPL, PSEG) carry the persistent premium; western zones (AEP, DAY, DEOK, ATSI) ride the near-term wave but fade after 2033.

This article is part of a quarterly series based on Modo Energy's PJM forecast updates. The next installment dives into battery-specific investment cases, breaking out vintage and location.

​Methodology and data sources for the full PJM forecast are available on Modo Energy's documentation site.