Who Will Own the New Data Center?

When a developer submits a conditional use permit application for a new data center, city officials typically spend weeks debating setbacks, traffic studies, and noise ordinances. What often receives far less scrutiny is the question that matters most to the surrounding community: who actually ends up owning and operating the facility — and for how long?

That question isn't as simple as it sounds. The entity that breaks ground isn't always the entity that flips the switch on opening day, and the operator running the facility five years later may have no relationship to either one.

Data Center Ownership Is Rarely a Straight Line

The public tends to imagine a data center as something a company builds and owns, much like a manufacturer builds a factory. The reality is considerably more layered.

Most large-scale data center projects involve at least three distinct parties: a developer who secures land and entitlements, an operator who manages day-to-day facility functions, and a tenant — often a hyperscaler like Microsoft, Google, or Amazon — who actually fills the building with servers. These roles can be held by the same company or by three entirely different organizations. Ownership can shift through sale-leaseback arrangements, REIT acquisitions, or private equity recapitalization — sometimes before a single rack gets installed.

The developer who shows up at city hall for a conditional use permit hearing may have every intention of selling the asset the moment it reaches stabilization. That's not a flaw in the system; it's often the explicit business model.

Real estate investment trusts like Equinix, Digital Realty, and Iron Mountain have built multi-billion-dollar portfolios by acquiring stabilized data center assets from developers who specialize in ground-up construction. Meanwhile, hyperscalers increasingly build their own campuses, cutting out the intermediaries entirely. Both models exist simultaneously, which means the ownership question has no universal answer — only project-specific ones.

What Drives Ownership Decisions

Regulatory Considerations

Zoning and permitting requirements shape ownership structures in ways most people don't anticipate. A conditional use permit, by definition, attaches conditions to a specific use of the land. Those conditions — covering everything from power draw limits to operational noise thresholds — transfer with the property. A future buyer inherits them.

This matters because institutional buyers doing due diligence will scrutinize permit conditions as carefully as the financial model. Overly restrictive conditions can reduce asset marketability, which influences how developers negotiate with municipalities at the permitting stage. It also means communities have more leverage than they typically exercise: the permit hearing is one of the few moments when residents and officials can shape long-term operational parameters, regardless of who eventually owns the facility.

Most municipalities treat conditional use permits as a development checkpoint. Experienced communities treat them as a long-term operating agreement.

Financial Structure and Capital Stack

Data centers are capital-intensive in ways that dwarf most commercial real estate. A hyperscale campus can run $1 billion or more per building, and even a modest colocation facility might require $50–150 million in upfront investment. That scale demands a specific capital stack — typically a combination of construction debt, equity from institutional investors, and pre-leasing commitments that de-risk the project for lenders.

Who provides that capital shapes who controls the asset. A project funded primarily by a private equity firm operates under a different ownership logic than one developed on a corporate balance sheet. PE-backed developers face fund lifecycles that create built-in sell timelines, usually five to seven years. Corporate developers — think the hyperscalers building their own campuses — have indefinite holding periods. Community stakeholders asking "who will own this?" should also be asking "how is it financed?" because the two questions have the same answer.

The Conditional Use Permit Process: More Powerful Than It Looks

The conditional use permit process is where ownership questions intersect with community interests in the most direct way. Cities grant CUPs to allow land uses that require case-by-case evaluation — data centers fall into this category almost universally because their infrastructure demands (power, water for cooling, fiber connectivity) are abnormal relative to standard industrial or commercial uses.

The process typically runs 60–180 days from application submission to approval, though contested projects in environmentally sensitive areas can drag well beyond that. During that window, planning commissions and city councils can attach conditions that follow the land in perpetuity.

Conditions commonly tied to data center CUPs include caps on power consumption (measured in megawatts), restrictions on cooling water usage — an increasingly contentious issue as data centers can consume millions of gallons annually — noise limits for cooling equipment, and requirements for local hiring or community benefit agreements.

What many communities don't realize is that a well-structured CUP can function as insurance against the risks of ownership transfer. If the original developer sells to a buyer with different operational priorities, the permit conditions remain binding. The mechanism only works, though, if those conditions are specific, measurable, and enforceable — vague commitments to "be a good neighbor" don't survive ownership changes.

Community Impact: The Economics Don't Tell the Whole Story

Data centers generate real economic activity. A facility employing even 50–100 full-time workers pays competitive salaries, often in the $70,000–$120,000 range for skilled technical roles. Property tax revenues can be substantial — a single hyperscale campus can generate tens of millions of dollars annually for local governments. Construction phases create temporary employment that benefits regional contractors and trades.

But the economic case deserves scrutiny alongside the enthusiasm. Data centers are not labor-intensive relative to their physical footprint. A million-square-foot campus might employ fewer people than a mid-sized office park. The tax revenue argument is real, but some states offer aggressive data center tax exemptions on equipment purchases — sometimes eliminating the primary economic benefit communities assume they're receiving.

The infrastructure demands are the other side of the ledger. A large data center can consume 50–100 megawatts of power continuously, placing demands on local utility infrastructure that other ratepayers may ultimately subsidize through rate increases. Water consumption for cooling is a growing concern in drought-prone regions. And unlike a manufacturing plant, a data center generates minimal retail foot traffic, which limits spillover economic benefits for surrounding businesses.

None of this means data centers are bad community partners. Many are not. But communities negotiating from a position of knowledge — understanding what they're giving and what they're getting — consistently secure better outcomes than those dazzled by headline investment numbers.

Where Data Center Ownership Is Heading

The ownership model for data centers is under genuine pressure from several directions simultaneously.

Hyperscalers are internalizing more of the development function, reducing dependence on third-party developers and operators. At the same time, the AI infrastructure buildout has created such intense demand for power-dense facilities that a new category of specialized developer has emerged — companies focused specifically on building GPU-optimized data centers for AI workloads, often with different ownership and financing structures than traditional colocation developers.

Sovereign wealth funds and pension funds have become significant data center investors, attracted by the asset class's long-term contracted cash flows. This institutional capital tends toward longer holding periods than private equity, which may actually stabilize ownership patterns over time.

Perhaps most significantly, power availability has become the binding constraint on data center development in many markets. Utilities in Northern Virginia — home to the largest data center concentration on the planet — are projecting demand growth that strains existing generation capacity. This dynamic is pushing developers toward markets with abundant renewable energy, which in turn changes which communities are fielding these permit applications.

The communities that will navigate this era best are the ones that ask the hard questions early: Who owns this facility at year one, year five, and year ten? What happens to permit conditions if ownership transfers? How does this facility's power and water consumption interact with existing infrastructure capacity? And what community benefit commitments are legally enforceable versus aspirational?

A conditional use permit hearing is a brief window. The facility that gets approved during it could be operating under different ownership — answering to investors in a different city or country — for the next 30 years. Communities that treat the permit process as a formality are leaving leverage on the table they can't get back.

[INTERNAL LINK: data center ownership]

[INTERNAL LINK: conditional use permits]

[INTERNAL LINK: community impact of data centers]

Ready to learn more about data center ownership and its implications for your community? Explore our resources at [InfraSale Marketplace](https://infrasale.com/marketplace).