Opposition to data centers has become a threat to the computing capacity America’s AI buildout requires. In a Gallup poll conducted in March, 71% of adults opposed a data center near them — far more than the 53% who opposed a nearby nuclear plant. Data Center Watch counted at least 75 projects worth roughly $130 billion delayed or blocked in the first quarter of 2026, as many as in all of 2025.
Concerns about noise, water, power bills, and artificial intelligence itself matter. But the backlash also reflects a simpler failure: Local governments collect enormous data-center windfalls while residents struggle to identify what they received in return.
Recommended Stories
Ordinary development brings both revenue and service demands. New housing, for example, brings families who use schools and roads. Data centers employ relatively few people after construction, yet their buildings and equipment can generate immense tax revenue. That makes them unusually close to pure fiscal upside for local government.
Consider Northern Virginia. Data centers generated 38% of Loudoun County’s general-fund revenue in fiscal 2026. In neighboring Prince William County, data-center revenue was projected to reach $549.7 million in fiscal 2027 — about 28% of general-fund tax revenue.
Loudoun has cut its real-estate tax rate substantially over the past decade, a real benefit. But falling rates do not guarantee falling bills when assessments rise. For tax year 2026, the county kept the rate flat and said the average homeowner’s real-estate tax would increase by $141. Loudoun notes that a vehicle tax reduction should more than offset that increase for the average homeowner who owns vehicles. That is welcome — but it also shows why relief must be simple, durable, and visible.
Yet the proposed Prince William budget would have raised the average residential real-estate tax bill by about $181. The board’s final budget instead reduced it by just $56. That is less than five dollars a month in homeowner relief from a county expecting more than half a billion dollars from data centers.
The same dynamic helps explain why voters doubt the standard promise that new development will expand the tax base and ease their burden. If facilities that demand relatively little in public services and generate hundreds of millions of dollars cannot produce a conspicuous tax cut, residents may reasonably wonder what kind of development ever will.
This is the “flypaper effect”: Windfall revenue tends to stick where it lands. Rather than return much of the money, governments absorb it into new and recurring spending. Whatever the merits of each program, the political result is predictable. Organized interests see a new pot of money; individual taxpayers see no reason to welcome the facility producing it.
Officials can change that. Before approving a data center, they should commit a fixed share of the resulting revenue to automatic, visible tax relief. A dedicated homestead credit would appear directly on property-tax bills. A “data-center dividend” could put a check in residents’ hands. Where local law permits it, counties could reduce or eliminate grocery taxes, giving families savings at every checkout.
THEY’RE FIREBOMBING CEOS AND SHOOTING AT LOCAL OFFICIALS. WELCOME TO THE ERA OF TECHNO-TERRORISM
The key is to promise the relief before construction, place it beyond the annual budget scramble, and make it large enough to notice. A rate freeze or a token cut will not persuade skeptical voters.
The United States needs more computing capacity. The industry needs communities willing to host it. Hyperscalers should therefore favor jurisdictions that will share the fiscal gains visibly with residents — and local officials should recognize that tax relief is not a failure of imagination. It is how they turn a data center from somebody else’s windfall into a community asset.
Owen Voutsinas-Klose is vice president of policy and government affairs at Viohl & Associates and a student at Georgetown University Law Center. The views expressed are his own.
