At least 10 states have pending rules targeting data center electricity demand, with parameters that industry advocates acknowledge could also capture large industrial users
Decision Focus
PJM Interconnection, the 13-state grid operator covering the most concentrated data center corridor in the United States, has reported a capacity price increase from $28.92 per megawatt-day in 2024 to $329.17 per megawatt-day at current levels — a 1,038% rise that PJM attributes primarily to data center load growth outpacing generation supply. The operational signal for Global Heads of Data Center Energy is not the number itself. It is what the number is triggering: a coordinated federal and state regulatory response beginning to target data center load directly, and a political environment in which manufacturers are using the capacity price spike as the factual anchor for legislative intervention.
90-Second Brief
This week, pJM’s capacity price explosion has caused industrial electricity bills in Pennsylvania to rise 31% and in Ohio 26% as of December 2025, against a 7% average national increase, creating a visible and politically charged redistribution of grid costs onto manufacturers who share rate classifications with data centers. FERC is now proposing that companies with onsite power generation pay transmission charges on that generation as well, a rule change that directly affects data centers using behind-the-meter resources. At least 10 states have pending rules targeting data center electricity demand, with parameters that industry advocates acknowledge could also capture large industrial users. PJM was forced to take emergency load-curtailment steps to prevent rolling blackouts as peak demand hit a new record, adding grid reliability pressure to what was already a cost-redistribution fight.
What Is Really Happening?
The structural problem is that PJM’s capacity market is functioning as designed — pricing scarcity — but the scarcity itself was created by load growth that outpaced generation development. PJM’s spokesperson stated directly that data centers “can be built faster than the generation needed to serve them, driving up demand faster than supply.” The capacity charge, which typically represents roughly 10% of residential bills but can represent up to 30% of industrial bills, became the mechanism through which cost redistribution turned visible and politically actionable.
What changed this cycle is that the cost signal produced specific, auditable numbers at the factory level. A brick manufacturer in Ohio watched its monthly capacity charge rise from $1,600 to $12,000. A plastics manufacturer in Pennsylvania and Ohio saw annual capacity charges climb from $200,000 to $1.2 million. These figures are the factual evidence base that manufacturer trade groups are now presenting to FERC, state public utility commissions, and congressional offices. The White House has responded by hosting a “ratepayer protection pledge” signing with technology companies and directing that new power plants in the PJM region be funded by data center developers rather than ratepayers broadly.
The classification issue compounds the problem. Manufacturers and data centers currently sit in the same large-load rate class within PJM. Any rule designed to assign higher cost responsibility to data centers must either create a new classification or risk applying the same treatment to manufacturers — which is precisely what at least 10 pending state rules are struggling to resolve.
Why It Matters for Global Heads of Data Center Energy
The direct exposure runs across three operating fronts. First, the FERC proposal to extend transmission charges to onsite generation alters the economics of behind-the-meter strategy. Data centers with diesel backup generation, on-site solar, or co-located BESS that currently pay transmission charges only on grid-drawn power would face a recalculated cost basis under the proposed rule. The final methodology remains unresolved, but the direction is confirmed.
Second, state-level demand management rules — at least 10 currently pending — represent a patchwork of potential load restrictions, demand response mandates, and cost-allocation methodologies. Each state in the PJM footprint may adopt a different approach. For a portfolio operating across Virginia, Ohio, Pennsylvania, New Jersey, and Illinois, this creates a multi-jurisdiction compliance and cost-forecasting problem that cannot be solved at the project level.
Third, the political dynamic has shifted the negotiating posture of utilities and grid operators. PJM’s emergency load-curtailment action during peak demand is already being cited in regulatory proceedings as evidence that data center growth is creating grid reliability risk, not merely cost redistribution. If that framing consolidates in federal proceedings, the result could be interconnection restrictions or load-growth caps in specific zones rather than cost reallocation alone.
Forward View
If the FERC transmission charge proposal advances to a final rule, the immediate effect is a revaluation of behind-the-meter investment cases across the PJM footprint. Projects approved on the basis of avoided transmission costs need to be re-modeled before the rule is finalized. The second front is state classification: if even two or three states successfully create a separate rate class for data centers, that approach is likely to accelerate in the remaining PJM states and potentially in MISO and ERCOT. The third front is demand response. The emergency curtailment event at PJM creates a factual basis for mandatory demand response programs for large loads — a requirement that data center operators currently treat as voluntary.
What Is Still Uncertain
The FERC transmission charge proposal has not reached a final rule, and its scope — which loads it applies to, how onsite generation is defined, and whether co-located generation is treated differently from backup generation — has not been confirmed. The 10 pending state rules vary in structure, and it is not yet clear which will distinguish data centers from industrial manufacturers and which will apply the same treatment to both. The degree to which the White House “ratepayer protection pledge” carries binding cost-allocation implications for data center operators, versus functioning as a voluntary commitment, is also not established in the source record. Whether PJM will propose structural changes to its capacity market pricing mechanism in response to political pressure remains an open question with significant downstream implications for forward cost modeling.
One Question for Your Team
Given that FERC is proposing to extend transmission charges to onsite generation, and that at least 10 states have pending rules that may reclassify your load: has your energy cost model for PJM-footprint assets been stress-tested against both of those rule changes activating simultaneously, and do you have a position ready for the FERC comment period?
Sources
- Spokesman — Big Tech data centers are driving up power bills at America’s Rust Belt factories (Link)
