Investor-Owned Utilities Plan $1.4 Trillion in Capex Through 2030, With Commercial Rate Classes Absorbing the Load

America’s investor-owned utilities have filed capital plans totaling $1.4 trillion through 2030, a 27 percent increase over the same forecast made one year ago. The step-up is driven primarily by data-center load growth, and the revenue requirement will land disproportionately on commercial and industrial customers, because residential rate protections are already politically entrenched in most state public utility commissions.

That is the argument PowerLines and Charles Hua make in Latitude Media this week, and the numbers behind it are consequential for any commercial building connected to a grid touched by PJM, CAISO, NYISO, MISO, or ERCOT.

The forecast jump. The $1.4 trillion figure aggregates five-year capital plans filed by investor-owned utilities across the country. Last year’s comparable aggregate sat near $1.1 trillion. The $300 billion delta in a single planning cycle is the largest year-over-year revision to utility capex forecasts in the modern rate-case era, and roughly half of it traces directly to data-center-driven generation, transmission, and distribution upgrades.

Tapestry, Alphabet’s moonshot grid unit, is separately running automated interconnection review at PJM to work through 200 gigawatts of queued projects. That queue dynamics problem is upstream of the capex number: whatever clears interconnection gets built, and whatever gets built gets recovered.

The rate-class allocation question. State utility commissions decide how capital costs are allocated across customer classes. Residential customers typically hold the political leverage, because they vote, they organize, and they have explicit protection statutes in many states. Commercial and industrial customers, by contrast, are allocated the remainder after residential protections and low-income riders are netted out.

For a $1.4 trillion capital stack, the residual line is where C&I sits. PowerLines argues that is precisely what is happening: residential bills are being held close to inflation through explicit commission orders in California, New York, Maryland, Illinois, and Massachusetts, while C&I demand charges, transmission riders, and capacity pass-throughs are escalating at multiples of the consumer price index.

PJM as the visible test case. PJM’s 2025–2026 capacity auction cleared at the $333 per MW-day cap, a tenfold increase over prior-cycle prices. Those capacity payments flow through to utility retail tariffs primarily via the capacity-related charges that appear on commercial bills. Data-center load accounted for roughly 40 percent of PJM’s $16.4 billion capacity cost base at auction, and PJM’s backstop procurement for an additional 14.9 gigawatts of data-center load is expected to be finalized by FERC in June.

The Heatmap investigation published this week documents parallel fights in MISO and ERCOT over who pays for the transmission upgrades data centers require. The regulatory trajectory in each RTO is different, but the allocation pattern is convergent: incumbent C&I customers are absorbing cost shifts that originated with new data-center interconnection requests.

Maryland’s response. Maryland lawmakers passed the RELIEF Act earlier this month, which bars data-center grid infrastructure costs from flowing into general ratepayer bills and forces the load to fund its own interconnection. If similar legislation spreads, it reduces the C&I cost-shift magnitude in those states. If it does not, the $1.4 trillion capex stack continues to land on commercial rate classes by default.

Demand charges within the allocation. The portion of C&I bills most exposed to capex absorption is not the energy charge. It is the demand charge and the capacity-related riders, which recover fixed costs based on peak kW or coincident peak contribution. As utilities add generation, transmission, and distribution assets, those fixed-cost buckets grow. The demand-charge denominator rises; the demand-charge kW-trigger threshold stays the same. Commercial buildings with 50 kW, 200 kW, or 1 MW peaks pay more each year even if their consumption patterns do not change.

Colorado Springs Utilities, which concluded its 2026 rate case on April 1, offers a precise example. The utility measures commercial demand in 15-minute intervals for any account above 10 kW, with demand charges now embedded structurally in the rate design. That is a municipal utility, not an investor-owned one, but the rate-design template is moving in the same direction across IOUs: shorter measurement intervals, lower trigger thresholds, higher per-kW demand rates.

The hedge the capex story creates. A building owner whose utility bill is driven 50 to 70 percent by demand charges, and whose utility is filing rate cases seeking recovery of capex that will grow those charges further, has a mechanical hedge available in on-site storage that clips peak demand. The hedge does not require predicting data-center load growth or reading transmission-allocation orders. It operates on the demand-charge formula directly.

This is why the commercial storage market has continued to grow through a period of compressed utility-scale project financing, tariff whiplash, and integrator bankruptcies. The structural driver is not policy and not technology. It is rate-base inflation landing on a customer class that has limited political protection and a physical tool, namely stored kilowatts, that can bypass the rate structure on site.

Five-year trajectory. If utility capex continues at the current growth rate, the 2030 aggregate could exceed $1.8 trillion by the time the next forecast cycle completes. The regulatory fights underway, Maryland’s RELIEF Act, Michigan’s megawatt-for-megawatt matching requirement, Virginia’s 20.78 GW storage mandate, all modify but do not reverse the allocation pattern. The commercial rate class remains the residual beneficiary of grid buildout and the residual payer of it.

Sources