The Neighbor Who Turned Out to Be a Power Plant

Maria Torres has lived in her Queens apartment for eleven years. The building is older, the wiring is older, and in recent summers the Con Edison bill has climbed in ways that do not track her own usage. She uses less air conditioning than her neighbors. She replaced her appliances with efficient models. Her bill went up anyway. What changed was not what she was consuming but what the grid was being asked to supply, and to whom.

Across New York, data centers have quietly become one of the largest and fastest-growing classes of electricity consumers. A single large hyperscale facility can draw as much power as a small city. They require cooling infrastructure that draws water. They require grid capacity that was built for a different consumption profile. They require, in short, public resources, a grid that ratepayers collectively built and maintain, in quantities that neither those ratepayers nor their elected representatives were asked to authorize.

New York’s reported move toward a statewide moratorium on new data center construction is, at its core, a decision about who controls the allocation of those public resources. It is the first statewide action of its kind in the United States, and it maps directly to a structural question that the founders argued about in 1787: what happens when private power grows large enough to absorb public goods without democratic accountability?

The Category 17 Mechanism

Centinel, the Anti-Federalist pen name of Samuel Bryan, writing in Philadelphia in 1787, argued that concentrated private wealth and power posed a structural threat to democratic governance distinct from the threat of government tyranny. The danger he identified was not merely that powerful private actors would influence government. It was that they would become infrastructurally essential in ways that made them difficult to constrain after the fact. Once a private actor controlled something that everyone needed, roads, mills, bridges, money, accountability mechanisms that worked in advance became nearly impossible to apply retroactively.

The modern analog runs through digital infrastructure. Data centers are not optional for the economy in which New York residents live. The computing that runs financial markets, medical records, e-commerce, streaming, and communication flows through them. But that infrastructural necessity does not automatically confer the right to absorb public grid capacity, water supply, and land in quantities determined by the private firms’ expansion plans rather than any democratic process.

The moratorium is a state government doing the thing that Category 17 of the Civic Engine describes: applying the accountability mechanism that exists, regulatory authority, when the market mechanism has not self-corrected and the federal framework has not acted. The question the moratorium poses is not whether data centers are good or bad. It is who decides how much public grid capacity private data infrastructure is entitled to consume.

The Grid Is Not Infinite

New York’s power grid was not designed for the current demand profile. The state has committed to aggressive renewable energy targets under the Climate Leadership and Community Protection Act, and it is simultaneously seeing demand growth driven primarily by AI and cloud computing infrastructure. These two trajectories are in tension. Renewable capacity is being added, but it requires years to permit, build, and connect. Data center demand is growing faster.

The practical consequence for residents is not theoretical. Grid operators in constrained markets manage demand through pricing signals and, in stressed conditions, through curtailment. When a grid is operating near capacity, the least reliable customers, residential consumers without backup generation, bear the most risk. The data centers, which operate mission-critical infrastructure and have contracts that prioritize their supply, are among the most protected. The apartment building without backup power is among the least.

The moratorium is, from one angle, a statement about whose reliability matters. From another angle, it is a statement about planning: that major new infrastructure consumption should require public authorization before it is locked in, not regulatory review after it has already been built.

Republican-governed states have taken the opposite position. Texas, Wyoming, and others have actively courted data center investment with subsidies and streamlined permitting. The structural argument is not that one approach is correct; it is that both approaches represent a governing choice about who controls the allocation of grid infrastructure, and that choice should be made explicitly rather than defaulted into. When a state government does not act, it is still making a choice, to allow private expansion to set the terms.

The Counter-Argument Deserves Honest Treatment

Moratoriums on economic development carry real costs. Data centers bring construction jobs, property tax revenue, and the economic activity that flows from digital infrastructure. New York is already a difficult environment for large capital investment; a statewide moratorium adds to that perception. Private firms facing regulatory uncertainty will site facilities elsewhere, and the economic activity follows them.

The market-corrects argument also has practical content: when grid power becomes scarce, electricity prices rise. Rising prices create an incentive for data centers to invest in efficiency, on-site generation, or locations with more grid slack. The market does generate pressure in the right direction, it simply does so slowly, unevenly, and without protecting the residents who bear the interim costs.

The structural response is not to dismiss these arguments. It is to ask what planning horizon and what distribution of costs they assume. A market that self-corrects over five to ten years while residential reliability degrades in the interim is a different thing from a regulated process that takes eighteen months to reach a decision. The moratorium is, at minimum, a statement that the existing planning process was not moving at the speed at which the problem was developing.

The Accountability Gap the Moratorium Exposes

What the New York moratorium reveals more than anything else is a gap in the existing federal framework. Data center siting is, in principle, regulated through a patchwork of local zoning, state permitting, and federal environmental review. In practice, the speed of the AI infrastructure buildout has outrun all three. Federal action has not materialized. Local governments lack the technical capacity and political leverage to negotiate meaningfully with major technology firms. The state is the accountability level that remains.

Whether the moratorium is the right instrument, versus mandatory grid impact assessments, capacity reservation requirements, or time-of-use demand restrictions, is a policy question. The structural point is different: a state government is doing what the Anti-Federalist framework predicted would sometimes be necessary. When private power accumulates faster than the federal regulatory framework can respond, the states are the remaining democratic check. Whether that check is strong enough, or arrives in time, is the open question.

Sources

CitizenFeedPress is an independent civic publication. Our editorial framework is drawn from the Anti-Federalist Papers, warnings written at the founding that map to structural failures visible in current events. We do not advocate for parties or candidates. We advocate for the citizen’s right to understand the system they live in.