The House Voted 417–3 on Data-Center Power Costs. Trump Wants the Senate to Move Next.

What changed this week — and what has not

Why a data-center rate bill belongs in a home-services publication

The five facts contractors should carry forward

• The House passed H.R. 9340 by 417–3 on September 16, 2026.
• The bill’s large-load definition targets nonresidential data-center customers with peak demand of at least 100 megawatts at a single facility.
• CBO says the bill would require state regulatory commissions to consider the federal standard; under current law commissions may adopt or reject standards after that process.
• A Senate attempt to pass the companion bill by unanimous consent was blocked September 17, and a competing Democratic proposal was blocked in turn.
• President Trump said September 18 that he was speaking with Senate Majority Leader John Thune about bringing the issue to the floor, but the measure remains pending legislation—not a new rate rule.

The data-center power fight in six numbers

House vote

417–3

H.R. 9340 passed the House on September 16.

Large-load threshold

100 MW

Peak demand at one data-center facility under CBO’s description of H.R. 9340.

2026 U.S. generation

4,368 BkWh

EIA forecast for record U.S. electricity generation in 2026.

2027 sales

4,211 BkWh

EIA forecast for U.S. electricity sales, up about 2% from 2026.

2030 data-center share

11.8%

Lawrence Berkeley National Laboratory central estimate of U.S. electricity use, with a 9.5%–15.3% scenario range.

Current status

Pending

House-passed bill; no Senate passage or presidential signature yet.

A headline that says Congress is “making data centers pay” is cleaner than the statute, but it is not precise enough. The Congressional Budget Office says H.R. 9340 would require state regulatory commissions to consider whether to adopt a federal standard for large-load customers. CBO also notes that commissions may adopt or reject federal ratemaking standards under current law.

The proposed standard itself is substantial. It focuses on recovering the full incremental cost of generation, transmission and distribution upgrades needed to serve a large-load customer, along with financial assurances before a utility proceeds with upgrades. The House bill’s threshold is also deliberately high: CBO describes the covered customer as a nonresidential data-center operator with peak demand of at least 100 megawatts at one facility.

That is why the House vote matters without being the final word. It puts a remarkably bipartisan marker around the principle that ordinary ratepayers should not casually inherit the cost of infrastructure built for hyperscale loads. But the regulatory mechanism still runs through states and utilities, and the Senate has not agreed on whether that mechanism is strong enough.

The Senate disagreement is about enforcement, not the basic problem

Husted and Heinrich are not arguing over whether large data centers can create major grid costs. Their dispute is over how forcefully federal law should allocate those costs. Husted’s Ratepayer Protection Act uses the familiar PURPA model: create a federal ratemaking standard and require state regulators and nonregulated utilities to consider it. Heinrich says that is too voluntary and pushed his GRID Savings Act as a stronger alternative that would make large-load customers pay for the facilities needed to connect them.

The Senate floor episode on September 17 showed the gap. Heinrich objected to Husted’s unanimous-consent request; Moreno then objected to Heinrich’s. As of this article’s publication, neither proposal had cleared the Senate. Reuters’ September 18 report that Trump is talking with Thune raises the odds of renewed attention, but it does not resolve the policy difference or guarantee a floor vote.

For contractors, that uncertainty is exactly why state-level tracking matters. Utility commissions are already where tariffs, special contracts, cost allocation and infrastructure approvals become real. A federal headline can set the direction; the invoice a homeowner eventually sees is still shaped by local utility and regulatory decisions.

Why this fight is arriving now: U.S. power demand is growing again

The politics are following the load. On September 9, the U.S. Energy Information Administration forecast that U.S. electricity generation would rise 2.2% to a record 4,368 billion kilowatthours in 2026 and increase another 1.7% in 2027. EIA expects electricity sales of 4,135 BkWh this year and 4,211 BkWh next year, citing data-center development and manufacturing as important drivers of commercial and industrial growth.

The longer-range data-center estimates are even larger. Lawrence Berkeley National Laboratory’s 2025 update, published in June 2026, estimates data centers could account for 11.8% of total U.S. electricity use by 2030, with scenarios ranging from 9.5% to 15.3%. Those are projections, not certainties, but they explain why regulators are wrestling with rate design before every proposed facility is energized.

A 100-megawatt customer is not another restaurant, warehouse or neighborhood. It can trigger generation procurement, transmission work, substation expansion and distribution upgrades. The policy argument is about whether the costs that are incremental to that customer should remain with that customer or diffuse into the wider rate base.

How the rate-design fight can reach a contractor P&L

The homeowner side of the story: electrification depends on trust in the monthly bill

A contractor selling a heat pump, heat-pump water heater, induction range, EV charger or larger electrical service is asking a homeowner to move more of the home’s energy use onto the electric system. That can be a good decision for comfort, performance, emissions, fuel switching or total cost. But the customer still wants to know what happens to the monthly bill.

That is where an abstract rate-design fight becomes practical. If large new industrial loads cause generation or grid investments and some of those costs flow to the general rate base, the homeowner’s electricity economics can change even though the homeowner never uses a data center. If regulators successfully ring-fence more of the incremental cost around the customer that caused it, that risk is reduced.

Contractors should not turn this into a partisan sales script or promise that a federal bill will lower electricity prices. Instead, use current local tariffs in proposals, show assumptions, and refresh operating-cost comparisons when utilities change rates. The companies that quote electrification using stale cents-per-kilowatthour assumptions will eventually lose credibility even if their installation work is excellent.

The other side: the same load growth is creating a giant electrical construction market

For electrical contractors, power-equipment suppliers and firms that work beyond the residential service panel, data-center load growth can be an opportunity as well as a ratepayer issue. Every large project has to connect to a real grid. That means utility studies, feeders, substations, protection, metering, switchgear, transformers, backup generation, commissioning and controls.

Dispatch has already documented the scale of supplier positioning around that buildout. Generac disclosed an agreement with Amazon under which initial backup-generator deliveries are expected to total $2.4 billion in 2027 and 2028. Prysmian has contracted a new low-carbon aluminum cable for an Amazon data center and is pursuing its multibillion-dollar Atkore acquisition. Cooling suppliers are buying capacity around the same AI infrastructure cycle.

The Ratepayer Protection Act adds another layer to that story: who finances the grid-side infrastructure that allows those projects to connect. A tariff that requires a hyperscale customer to post financial assurances or absorb more upgrade cost does not eliminate the work. It changes who carries the risk and how regulators approve the project economics.

Related Dispatch: grid pressure is already showing up outside the data-center debate

What contractors should monitor next

• Watch whether Senate Majority Leader John Thune schedules the Ratepayer Protection Act or a negotiated alternative; do not treat presidential support as passage.
• Track your state public utility commission for data-center tariffs, special contracts and large-load cost-allocation dockets; those proceedings can matter more locally than the federal headline.
• For HVAC and electrification proposals, keep a current utility-rate assumption library by service territory so operating-cost comparisons are based on today’s tariff, not an old national average.
• For electrical contractors, map utility capital plans, substation projects and large-load interconnection activity in markets where data-center construction is accelerating.
• Separate data-center infrastructure opportunity from residential rate exposure in forecasts; the same load growth can create commercial work while making household affordability a harder sales issue.
• Train sales teams to explain electricity economics without claiming a pending federal bill will lower a customer’s bill or stop future rate increases.
• Add a quarterly rate-case review to marketing planning in high-electrification markets, especially where heat pumps, EV charging, batteries or panel upgrades are major revenue categories.

Is the Ratepayer Protection Act law?

No. H.R. 9340 passed the House on September 16. A Senate companion exists, but a September 17 unanimous-consent attempt was blocked. Reuters reported September 18 that President Trump was speaking with Senate Majority Leader John Thune about bringing the issue to the floor. Senate passage and a presidential signature would still be required for the House legislation to become law.

Would the House bill automatically force every data center to pay every grid cost?

Not automatically. CBO describes the bill as requiring state regulatory commissions to consider whether to adopt the federal large-load standard. Under current PURPA ratemaking procedures, commissions may adopt or reject federal standards after consideration. The proposed standard itself calls for recovery of the full incremental costs of upgrades needed to serve covered large-load customers and for financial assurances, but the state consideration process is a crucial limitation.

Why should a residential HVAC or electrical contractor care?

Because utility rates are part of the customer’s household budget and part of the operating-cost math behind electrification. At the same time, data-center grid expansion is creating electrical infrastructure work. A contractor does not need to take a position on AI policy to track the rate cases and capital plans that can change local customer economics.

The bottom line

Methodology

Dispatch selected the Ratepayer Protection Act for the September 19 afternoon cycle after rechecking the current home-services, electrical, HVAC, M&A, software and regulatory news flow and confirming that the morning Solar for All and midmorning Duke Energy stories were already published. The article independently verified the 417–3 House vote with the House Energy and Commerce Committee; the bill’s 100-megawatt threshold and PURPA consideration mechanism with the Congressional Budget Office and GovInfo; the September 17 Senate floor impasse using statements from both Senator Jon Husted and Senator Martin Heinrich; and the September 18 Trump/Thune development with Reuters. EIA’s September 2026 Short-Term Energy Outlook and Lawrence Berkeley National Laboratory’s 2025 data-center energy update provide demand context. The article does not state that the bill is law, that states must adopt the proposed federal standard, or that it will produce a specific residential rate reduction. The hero is a real licensed Unsplash photograph; source, photographer, publication date and license status were checked against the original page, and it is labeled as illustrative rather than as a photograph of a named data center or current event.