Data Centers in New York City Come To A Halt. What Will It Change?

Share This Article
A landmark bill was passed halting new data centers in New York City and the state. Dr. McNees shares his insight on what it will actually change.
From the viral town hall meetings on social media to the NAACP filing a federal lawsuit against xAI’s Colossus, there is a lot of backlash against data centers these days. And, for good reason too. There are an 1215 operational data centers in the U.S with 1732 more planned projects. In tandem with this growth is data coming to light about their colossal energy use, greenhouse gas emissions, water consumption, air pollution, and more negative costs to local communities. Governments are vying to keep up with both data center growth and communities’ reservations. On July 14th, New York Governor Kathy Hochul launched the nation’s first statewide moratorium to pause growth on hyperscale data centers in New York City (and the state) and allow time to address concerns about the fast growth of these large facilities and their impact on the environment and local communities.
“New York has always been at the forefront of innovation and change, but we’ve also always guaranteed that New Yorkers benefit. As data center development threatens to hike up utility bills, deplete our natural resources, and create uncertainty for New Yorkers, it’s my responsibility to take action and lead. New York will lead the way in creating the strongest standards in the nation for data center development.”
Governor Kathy Hochul
“New York’s executive order pauses state environmental permits for new hyperscale data centers for up to one year,” Dr. Mark McNees, Director – Social and Sustainable Enterprises – Jim Moran College of Entrepreneurship at Florida State University, explains. But, he cautions, “It does not answer the question of who pays.” That remains to be seen over the course of the next year. The pause halts the approval of new large-scale data centers while the state creates a better framework to balance tech growth with protecting electricity ratepayers, the environment, and the power grid. The goal being that once these new standards are in place, the state will lift the moratorium and allow projects to move forward if they meet the new, stricter requirements.
With an economic-first background dealing with data centers, Dr. McNees spoke to The Wellness Feed about proactive planning, how key states like New York are closing the gap between who pays and resource burden, and the best public resources for us to be involved in these decisions.
How will New York’s one-year moratorium impact the growth of data centers?
A permitting pause is not a cost rule. It stops the clock. It does not answer the question of who pays. Those are two different instruments, and readers who want to advocate effectively need to know which one they are asking for.
The instruments that decide who pays look different. New Jersey signed the Data Center Fair Share Act on July 7. It puts large data centers into their own rate class and requires them to guarantee they will take at least 85 percent of the power they request for ten years, with deposits up front covering new transmission. That is a cost rule. It has teeth on the question of who carries the bill.
Florida is on a similar clock with SB 484. Utilities file large-load pricing rules with the Public Service Commission by October 1.
So on impact: a one-year pause changes the timeline and buys New York room to build a framework. Whether it changes who pays depends entirely on what the state builds during that year. The pause is the beginning of the question, not the answer to it.
What’s the difference and impact between a hyperscale and a regular data center?
The honest answer is that the line is drawn differently depending on who is drawing it, and that ambiguity is itself part of the transparency problem.
The most useful way to think about it is not square footage or company size. It is electrical load. A facility that draws a few megawatts sits inside the capacity a utility already has. A facility that draws hundreds of megawatts does not. It requires new generation, new transmission, or both, and something has to be built to serve it.
That threshold is where the economics change completely, and it is why the laws now being written define data centers by megawatts rather than by type. New York’s legislation uses a one megawatt definition for a data center and a twenty megawatt threshold for the pause it contemplates. The executive order that actually took effect is a separate instrument and sets its own threshold. New Jersey’s law targets large load customers. Florida’s SB 484 works the same way.
Once infrastructure gets built, someone pays for it over decades, and that is the decision that lands on a community, whether or not the community was ever asked.
What are the solutions to ensure transparency and hold companies accountable?
Three things, in order of how much they actually change:
First, disclosure has to be a condition of the deal, not a request after it. Once a facility is approved and built, a community has almost no leverage to obtain information it did not require up front. Non-disclosure agreements are common in the negotiating phase, and local officials frequently sign them before they know what they are agreeing to. The moment to require reporting on water and energy use is the moment before approval, and that moment does not come back.
Second, the numbers have to be tied to money. A disclosure requirement with no consequence produces a filing nobody reads. A cost allocation rule produces a number somebody has to defend, because a dollar figure is attached to it. This is why the state-level cost rules matter more than they sound like they should. When New Jersey requires a data center to guarantee it will take 85 percent of the power it requested for ten years, that creates a documented, enforceable number about actual consumption. Transparency arrives as a byproduct of the accounting.
Third, the venue matters and almost nobody knows where it is. Utility rate decisions happen at state public service commissions, in dockets, on the public record, with comment periods open to anyone. Those proceedings are where cost allocation is actually decided, and they are typically attended by utility lawyers and almost no one else. A public comment filed in a docket carries more weight per hour of effort than almost any other form of participation, precisely because so few people file one.
The underlying pattern is the same one I see across sectors: a company internalizes the profit and passes the cost to whoever is not in the room. The remedy is being in the room early, while the terms are still being written.
There is a Florida example worth knowing. The original version of SB 484 prohibited state and local agencies from signing non-disclosure agreements that hide data center plans from the public. That prohibition was stripped by a House amendment before final passage. The law as enacted permits NDAs that keep siting decisions confidential for up to a year. The bill’s own sponsor, Sen. Bryan Avila, publicly expressed reservations about the final text. A state cannot regulate what it cannot see.
What is a good resource to learn about data centers in your area?
The honest starting point is that most people do not have a good local resource, and that gap is part of the problem. There are national efforts worth knowing:
FracTracker maintains a US data center map focused on proposals and community response. DataCenterTracker tracks community action and legislation. Both are useful for seeing whether something is proposed near you.
For the money question specifically, I maintain a resource at MarkMcNees. It includes a twelve-state tracker of what states are doing on large-load cost allocation, with a primary source link for every claim, and a reporter FAQ. There is also a page written for county commissioners and residents facing a proposal in their own community, with the questions worth asking before a vote.
Beyond that, two local sources- your state public service commission or public utility commission website publish dockets and large-load tariff filings before rates change. Your county commission agenda is where siting and incentive decisions appear, often with very little notice. Both are public. Both are where the decisions actually happen.
XAI was sued this year for by the NAACP over air pollution. What proactive measures can we take to prevent similar situations?
In May, I published a piece in The Invading Sea called “Industrial redlining: Data centers repeating historic pattern of harm.” The short version. Boxtown, on the southern edge of Memphis, was founded by freedmen in 1863. In 1960, the city proposed annexing it, and the Memphis Housing Authority’s stated logic was direct: bring a power plant and nearby refineries into the tax base, and use the community as an industrial labor force.
Sixty-five years later, Boxtown sits inside one of the most polluted ZIP codes in the country. Seventeen facilities tracked through the Toxics Release Inventory surround it. Cancer risk from industrial sources runs 4.1 times the EPA’s acceptable threshold. Ninety-five percent of residents are Black. In 2024, the Colossus data center was built there. The community learned about the project through media reports.
This follows the path of least resistance. Developers look for cheap land, available power, permissive zoning, and a local government unlikely or unable to impose conditions. Communities with less political capital and fewer resources to hire their own counsel and engineers score well on that list every time. The pattern is a product of the search criteria.
On what prevents it:
• Conditions attached before approval. Almost all of a community’s leverage exists before a vote, and almost none exists after. Emissions limits, water use reporting, cost guarantees, and decommissioning terms are negotiable in advance and nearly impossible to add later.
• Cumulative impact analysis. Before any large industrial facility receives a permit, the state should require a documented analysis of what the new facility adds to the burden that is already there.
• Technical capacity on the community’s side of the table. State-funded technical assistance for local governments evaluating large projects would change more outcomes than most of what gets proposed.
• Statewide rules rather than local ones. As long as terms are set county by county, developers can shop for the weakest terms available. That competition is a race to the bottom, and the communities with the least capacity lose it. A statewide rule removes the incentive to go looking for the community least equipped to say no.
How can we balance tech expansion while preserving our natural resources and communities?
The current arrangement is not a balance. It is a subsidy. When a data center triggers grid capacity, water demand, and local infrastructure costs, and does not pay the full cost of what it triggered, the difference does not disappear. The buildout should carry its own costs.
Nobody has to agree about climate change to agree that a profitable company should pay its own electricity bill. Communities are preserved by writing the terms before the shovels arrive. Not by stopping growth, and not by trusting that it will work out.