Executive Overview
The explosive growth of artificial intelligence and cloud computing has triggered a seismic shift in global energy markets. As hyperscale operators and enterprise AI facilities erect sprawling data centers capable of devouring hundreds of megawatts—quantities once reserved for heavy manufacturing or entire mid-sized municipalities—policymakers are scrambling to manage the fallout. The tension between soaring digital infrastructure demands and constrained electrical grids has officially moved from regional utility boardrooms to the halls of Capitol Hill.
In August 2026, U.S. Representative Andrea Salinas (D-Ore.) introduced the Data Center Community Reinvestment Act of 2026 (H.R. 10102), a landmark piece of legislation that proposes a federal excise tax on electricity consumed by qualifying data centers. Under the bill, facilities with a maximum rated power capacity or total peak load exceeding one megawatt (MW) would face a 1-cent-per-kilowatt-hour (kWh) federal excise tax on all electricity consumed. The legislation directs the resulting revenue—projected to be roughly $1.76 billion annually—toward five distinct federal funds.
While proponents argue that the measure ensures local communities and the broader public share in the windfall and bear fewer economic burdens from escalating grid demands, industry experts warn of deep disconnects. Critics point out that a blunt, nationwide flat consumption tax fails to solve the core economic problem of grid cost causation. By charging data centers based purely on kilowatt-hours without accounting for localized infrastructure strain, transmission bottlenecks, or operational flexibility, the bill risks imposing heavy financial penalties on operators while failing to protect everyday utility ratepayers from localized capacity price spikes.
This in-depth investigative report examines the mechanics of H.R. 10102, explores the staggering financial implications for data center operators at scale, evaluates the ongoing debate over grid cost causation, and weighs the broader legislative outlook for the digital infrastructure sector.
Detailed Chronology of the Legislative Push
The Escalating Infrastructure Strain
The introduction of H.R. 10102 does not occur in a vacuum. Over the past three years, the rapid deployment of large language models (LLMs), deep learning clusters, and massive cloud storage networks has driven an unprecedented surge in electricity demand. Regional transmission organizations (RTOs)—most notably PJM Interconnection, the largest grid operator in the United States—have issued repeated warnings that soaring data center growth threatens grid reliability and is driving up capacity market clearing prices.
As communities across Virginia, Ohio, Texas, and Oregon grapple with the prospect of new fossil-fueled generation plants or accelerated nuclear plant retirements being delayed to feed AI loads, public backlash has intensified. Citizens and consumer advocates have raised urgent questions regarding who should ultimately pay for the multi-billion-dollar transmission upgrades, substation expansions, and generation buildouts required to support the AI boom.
Moving from State Houses to Capitol Hill
State-level regulators have attempted to grapple with these realities first. Earlier in 2026, Virginia became the pioneer jurisdiction when it approved a state-level electricity consumption tax explicitly targeting data centers. Set at 1.1 cents per kilowatt-hour, Virginia’s tax applies to utility-supplied power, retail providers, and behind-the-meter self-generation, with state budget documents projecting roughly $600 million in annual revenue for the state’s general fund.
Building upon this regulatory precedent, Rep. Salinas took the concept national. On August 13, 2026, she formally introduced the Data Center Community Reinvestment Act in the House of Representatives. The bill was immediately referred to three primary committees:
- The House Ways and Means Committee (due to its tax-writing jurisdiction),
- The Energy and Commerce Committee (oversight of energy markets and utility regulation), and
- The Science, Space, and Technology Committee (oversight of technological infrastructure and federal energy technology funds).
Supporting Context & Metrics: What Operators Would Pay at Scale
To understand the weight of H.R. 10102, one must look closely at the math. The bill targets any facility that primarily contains electronic equipment used to process, store, or transmit digital information and possesses a maximum rated power capacity or total peak power load exceeding 1 MW.
This threshold ensures that the legislation captures far more than massive hyperscale AI campuses built by tech giants; it reaches deep into the colocation, enterprise, and institutional data center markets. Furthermore, the tax applies broadly, encompassing both grid-supplied electricity and behind-the-meter generation procured onsite.
Financial Models at Scale
Neil Osnato, founder of Persistence Analytics Group, has mapped out the financial impact of the 1-cent-per-kWh excise tax across various operational scales, assuming continuous full load (24/7/365 operation at maximum capacity):
- 100 MW Facility (Continuous Full Load): Incurs approximately $8.76 million annually under the tax.
- 500 MW Facility (Continuous Full Load): Incurs approximately $43.8 million annually.
- 1 GW Facility (Continuous Full Load): Incurs approximately $87.6 million annually.
When evaluated at a more realistic 90% annual load factor (accounting for maintenance downtime, variable computing loads, and efficiency curves):
- 100 MW Facility: Approximately $7.9 million per year.
- 500 MW Facility: Approximately $39.4 million per year.
- 1 GW Facility: Approximately $78.8 million per year.
Impact on Energy Procurement Economics
How significant is a 1-cent levy relative to underlying power prices? Osnato notes that the impact depends heavily on a facility’s baseline energy costs:
- If a data center’s underlying wholesale or retail electricity price is 5 cents/kWh, the proposed federal tax adds a staggering 20% surcharge to the energy component.
- If the underlying price is 10 cents/kWh, the tax adds a 10% surcharge.
While these figures represent tens of millions of dollars in operational overhead, Osnato cautions that for the largest AI and hyperscale builds, other variables—such as access to executable megawatts, transmission availability, interconnection timing, and long-term development certainty—frequently outweigh the marginal cost of a federal tax when deciding where to site a campus.
Official Statements and Industry Perspectives
Proponents: Ensuring Community Shared Benefit
Rep. Salinas defended the legislation as an essential safeguard to protect local taxpayers from absorbing the hidden infrastructure costs of rapid digital expansion, while ensuring that the immense economic value generated by AI is shared.
"Every community deserves to drive a hard bargain with data center developers, and at the end of the day, every community deserves to share in the benefits from data center growth," Salinas said in an official press statement.
By capturing an estimated $1.76 billion nationwide based on current data center electricity consumption patterns, the bill seeks to channel corporate digital revenues back into public programs.
The Revenue Breakdown: Five Federal Funds
Rather than funneling the proceeds into a generic state treasury like Virginia’s model, H.R. 10102 divides the anticipated $1.76 billion annual revenue evenly into five separate federal destinations (roughly $352 million per year for each fund):
- The Energy Technology Trust Fund: Designed to support federal loan guarantees for clean energy, advanced nuclear power, and next-generation grid infrastructure projects.
- Four Additional Federal Funds: Dedicated to broader public investments and community-oriented infrastructure initiatives (as outlined in the broader legislative text).
Critics and Consumer Advocates: The Missing Link of Grid Cost Causation
Despite the noble intentions behind funding clean energy and community projects, industry analysts and consumer advocates argue that the bill misses the forest for the trees.
David S. Lapp, a prominent People’s Counsel, told Data Center Knowledge that while the tax might increase federal revenues for public programs, it fundamentally fails to protect existing utility customers from surging electricity bills.
"Those risks arise largely from existing PJM rules for the administration of its capacity market and transmission planning that, without necessary reforms, will force residential customers to subsidize massive data center growth," Lapp explained.
Lapp emphasized that true ratepayer protection requires structural interventions rather than a blunt federal tax. Effective measures, he argues, must include:
- Mandating that large data center operators bring their own dedicated generation online or accept curtailment agreements during grid emergencies.
- Reforming regional transmission organization (RTO) rules for assigning transmission costs directly to the loads that cause them.
- Implementing rigorous large-load tariffs with robust safety nets for everyday ratepayers.
The Flaw of "Megawatts Are Not Interchangeable"
Echoing these sentiments, Neil Osnato highlights the fundamental mismatch between a flat consumption tax and the engineering realities of electrical grids.
Two distinct 500 MW data centers can consume identical amounts of energy while imposing radically different economic and physical costs on the electrical grid. Factors such as geographical location, the precise timing of peak demand coincidence, required local substation upgrades, local transmission constraints, and operational load flexibility mean that no two grid footprints are alike.
"A flat federal consumption tax does not distinguish among any of those conditions," Osnato remarked. "Policymakers could collect money from the right industry while still allocating the underlying grid costs incorrectly. Megawatts are not interchangeable."
Osnato argues that robust public policy must distinguish between a facility’s energy consumption and its true grid cost causation. Customer-specific transmission and interconnection costs should be borne directly by the causer, while genuinely shared system costs should be distributed proportionally based on systemwide benefits. Because the federal excise tax directs its proceeds into general federal funds (with only one of the five funds having a direct energy connection via loan guarantees), the tax acts more as a corporate revenue generator than a precise tool for electrical engineering equity.
Future Outlook
As the Data Center Community Reinvestment Act of 2026 sits before the House Ways and Means, Energy and Commerce, and Science, Space, and Technology committees, its path forward remains uncertain.
The political appetite for reigning in Big Tech and managing the environmental and economic externalities of the artificial intelligence boom is undeniably bipartisan. However, the mechanics of H.R. 10102—particularly its imposition of a flat nationwide tax regardless of local grid impact—will face intense scrutiny from industry lobbyists, tech sector representatives, and lawmakers representing states eager to attract high-tech economic development.
Whether Congress ultimately chooses to advance a national electricity excise tax or pivots toward RTO-level market reforms and localized tariff structures, one reality is clear: the era of unchecked, frictionless expansion for digital infrastructure is over. As AI computational clusters continue to push electrical grids to their absolute limits, the question of who pays for the wires, the power plants, and the community impacts has become the defining energy policy debate of the decade.
