PPL Electric Utilities, the Pennsylvania utility serving roughly 1.4 million customers, is publicly arguing that more data centers on its grid can lower costs for residential ratepayers rather than raise them – provided the new loads pay for the infrastructure they trigger. That position, laid out by PPL President Christine Martin in a recent Power Perspectives interview on EnergyCentral, inverts the popular assumption that hyperscale computing is a pure cost burden on households. The stakes are concrete: PJM demand forecasts have swung from roughly flat growth a decade ago to projections of multiple gigawatts of new load, and how regulators allocate the resulting grid costs – to the data center or to the general rate base – will determine whether residential bills fall, rise, or hold steady through the 2030s.
The fixed-cost math that makes big loads attractive
The core logic utilities are now pushing is simple arithmetic. A grid’s fixed costs – transmission lines, substations, transformers, control systems – do not shrink when demand is flat. Those costs are recovered through volumetric rates, meaning every kilowatt-hour sold carries a share of the fixed bill. When a utility adds a 500 MW data center that runs at high utilization around the clock, it spreads those fixed costs across far more kilowatt-hours. All else equal, the per-unit cost to every other customer falls.
That logic held comfortably for decades, when load growth was a given and utilities planned around 2-3% annual increases. It broke down after roughly 2010, when efficiency gains, rooftop solar, and deindustrialization flattened demand across much of the country. Utilities spent fifteen years planning for a world where electricity sales barely grew. The data center boom – accelerated by AI compute demand – has reversed that planning assumption faster than most utilities could adapt.
Martin’s central point in the interview is that PPL is not merely accommodating this new load but actively preparing for unprecedented demand growth. That preparation cuts across load forecasting, transmission planning, and customer communication. The “unprecedented” framing matters: PJM, the grid operator covering PPL’s territory, has seen its load forecast revisions accelerate sharply in recent years, and interconnection queues across the region are now dominated by large loads rather than just generation.
The affordability question is where the math gets contested. Bringing a data center onto the grid does lower average costs if the customer pays its full marginal cost – including the dedicated substation, the upgraded transmission path, and the capacity needed to serve it. The risk is that utilities socialize those costs into the general rate base, effectively asking residential customers to subsidize the buildout that serves a single hyperscale tenant. That is the crux of the debate Martin is addressing.
The transparency piece of her message is not a soft skill but a regulatory necessity. Utilities that announce gigawatt-scale data center deals without explaining the cost allocation to the public invite exactly the kind of rate case challenges that end with costs being reallocated or projects being delayed. PPL’s bet is that early, clear communication builds the political cover needed to secure the rate treatment that makes the fixed-cost math work for everyone.
Why the load-growth reversal is reshaping utility economics
The shift from flat to surging demand is not unique to PPL. Across PJM, MISO, and ERCOT, utilities are revising load forecasts upward at a pace not seen since the 1990s. The drivers are concentrated: AI data centers, semiconductor fabs, and electrification of transport and industry. The common thread is that these are large, lumpy loads that arrive in gigawatt increments rather than the distributed growth utilities historically modeled.
That lumpiness creates a planning problem. A utility cannot build a 500 MW transmission path incrementally; it has to commit to the full upgrade years before the customer’s first megawatt is served. That front-loaded capital spending raises the stakes on every interconnection agreement. If the data center commits to a long-term contract and pays for the dedicated assets, the utility’s shareholders and other customers are protected. If the data center can walk away or renegotiate, the costs fall back on the general rate base.
This is where the connection to broader sector trends becomes clear. The same demand surge is driving a parallel scramble on the generation side. PJM’s capacity market has seen prices spike as reserve margins tightened, and the region is now wrestling with how to add dispatchable capacity fast enough to serve loads that want firm, round-the-clock power. Data centers’ preference for 24/7 clean energy has pushed interest toward nuclear, geothermal, and long-duration storage – technologies that were marginal in utility planning just a few years ago.
Data center load also behaves differently from the industrial loads utilities historically served. A steel mill or chemical plant has variable output tied to commodity prices; a data center runs at near-constant high utilization and demands power quality that older industrial customers never required. That firmness makes data centers exceptionally valuable as cost-spreaders – but it also means they cannot be curtailed or shed without breaking service commitments, which puts a hard floor under the capacity the utility must procure.
By comparison, the last time PJM faced comparable load growth, the generation buildout was led by natural gas combined-cycle plants that could be permitted and built in under three years. Today’s preferred resources – advanced nuclear, enhanced geothermal, long-duration storage – have longer lead times and unproven supply chains. That mismatch between load arrival and generation availability is arguably the single biggest structural risk in the current demand surge, and it is a risk that plays out differently for every stakeholder.
There is also a regulatory dimension worth flagging. FERC’s Order 2023 overhauled interconnection procedures to clear generation queues faster, but it did not address the cost allocation question for new load. That question is being fought out in state commissions, where utilities are proposing rate designs – demand charges, minimum bills, and dedicated cost-recovery riders – that determine who ultimately pays for the grid upgrades. The outcome of those proceedings, not the headline announcements about gigawatt-scale data center campuses, will decide the affordability question Martin is raising.
Who this affects
- Utility planners: Your load forecasting methodology needs to handle lumpy, gigawatt-scale additions without spreading their costs into the general rate base. Model both the “data center pays its own way” scenario and the “costs are socialized” scenario, and present both to regulators before the first interconnection agreement is signed.
- Generation and storage developers: The data center demand surge is creating a market for firm, 24/7 clean power that most existing resources cannot serve. Projects with long-duration storage or firm clean attributes will command premium pricing in PJM’s capacity market, but only if they can clear interconnection before the load arrives.
- Policy analysts and regulators: The affordability outcome hinges on cost allocation rules you set now. If you allow dedicated data center infrastructure into the general rate base, residential bills will subsidize AI compute. If you require full marginal cost recovery from the new load, you protect households but risk losing the economic development prize.
- Investors: Utilities with strong balance sheets and transparent cost-recovery mechanisms for large loads are better positioned than peers who socialize costs and invite regulatory pushback. Watch how each utility structures its data center contracts and whether those contracts survive a rate case challenge.
What to watch next
- PPL’s next integrated resource plan or load forecast update, which should quantify how much of its projected demand growth comes from data centers versus other load classes.
- State commission rulings in Pennsylvania and other PJM states on cost allocation for large-load grid upgrades – these decisions will set the template for the region.
- PJM’s next capacity auction results and reserve margin projections, which will show whether generation buildout is keeping pace with the load forecasts utilities are now filing.
- Whether PPL or other utilities file rate cases that include dedicated data center cost-recovery riders, and how intervenors respond.
Bottom line
The data center boom is not a burden utilities are being forced to absorb – it is an opportunity they are actively courting, because large, high-utilization loads are the most efficient way to spread fixed grid costs. Whether that opportunity becomes a benefit for residential ratepayers or a cross-subsidy depends entirely on cost allocation decisions being made in state commission proceedings over the next two to three years. The utilities that communicate those tradeoffs transparently – as Martin is doing – are the ones most likely to get the regulatory outcomes they need.
Read the full report at EnergyCentral.
Note: facts and figures attributed above to reflect that outlet's original reporting. Broader context, cross-sector connections, and forward-looking scenarios reflect independent analysis by our editorial team.
About this article: Drafted by Energy Ai with AI-assisted research and writing based on public reporting, then reviewed under our editorial process before publication.
Leave a Reply