Rural America is sitting on more leverage than it realizes. The right answer to a data center proposal isn't yes or no — it's a better bargain.

In 1992, I thought we were building warehouses.

Turns out, we had accidentally built on top of the internet.

I was a young commercial real estate guy working on shallow-bay warehouses next to airport cargo terminals. The thesis was simple: be close to FedEx and UPS. What we didn't know was that the big fiber-optic trunk lines being laid across the country followed railroad easements — and our Charlotte buildings sat right next to the Norfolk Southern line behind the airport.

The fiber came toward Charlotte. The airport was in the way. The fiber went behind the airport. Right next to us.

Soon the phone started ringing. Not from logistics tenants. From ISPs, Verizon, Sprint, and a string of companies with names nobody recognized yet. Within a couple of years we had leased hundreds of thousands of square feet to early data-center and telecom users.

We thought proximity to cargo was the whole play. The real play was proximity to invisible infrastructure.

That lesson is about to repeat itself across rural America — and most of the communities sitting on the new infrastructure don't know what they own.

---

Why this looks familiar

I have spent most of my career adjacent to this industry. I founded a company called Ariel focused on data centers and telco hotels. I helped lead East Coast telecommunications development and brokerage at Trammell Crow. I currently serve as an expert witness on a significant industry matter, and I'm doing active site selection work for a data center developer. For confidentiality reasons, I can't discuss either assignment. And I won't.

What I can tell you is what 30 years on the inside of this industry teaches you about what's happening right now.

These are not just technology questions. They are real estate questions. Land. Power. Fiber. Timing. Entitlements. Public trust.

And right now, America is badly underestimating that last one.

---

The Cloud is Moving to the Country

For years, most people thought of data centers as urban or suburban facilities. They sat near population centers, telecom hubs, corporate users, and existing power infrastructure.

That map is changing quickly.

According to Pew Research Center, the United States has more than 3,000 operating data centers and more than 1,500 additional facilities planned, under construction, or land-banked. Here is the important part:

67% of planned U.S. data centers are in rural areas.

By comparison, only 13% of existing data centers are rural. Pew also found that 39% of planned projects are in counties with no existing data centers. The South alone accounts for almost half of planned centers.

That is not a footnote. That is a geographic reset.

The cloud is leaving the office park and moving toward the soybean field.

Naturally, people are upset.

Gallup recently found that 71% of Americans oppose an AI data center being built in their local area, including 48% who strongly oppose it. That number is higher than Gallup's measured opposition to building a nuclear plant nearby.

And the opposition is not theoretical. Here in North Carolina, local governments are already pausing, studying, restricting, or fighting data-center projects. Apex. Wendell. Durham. Chatham County. Orange County. Cumberland County. Lee County. Edgecombe County.

WRAL has been tracking moratoriums and ordinance changes across central North Carolina. WUNC reported that data-center opposition has already become a political issue, including one Edgecombe County primary where a challenger ran directly against data centers and beat an incumbent.

So yes, the backlash is real. And frankly, some of it is justified.

---

NIMBYs are not crazy

It is easy for developers to dismiss opponents as NIMBYs. That is usually a mistake.

People are asking reasonable questions:

• Will this raise my power bill?
• Who pays for grid upgrades?
• How much water will it use?
• Will I hear cooling equipment all night?
• Are the permanent jobs real, or mostly press-release arithmetic?
• Why should a trillion-dollar industry receive local tax breaks?

Those are not dumb questions. Those are exactly the questions a community should be asking before any deal gets signed.

The problem is not that residents are asking them. The problem is that too many communities stop there. They see the costs. They do not always see their leverage. And they don't always see their potential to solve community problems close to home.

That is where this gets interesting.

“But isn't this just developer-speak for 'roll over and take the deal'?”

It is not.

Saying no is a perfectly good answer when the terms are wrong. The mistake is saying no before you have found out what the terms could be.

There is a difference between rejecting a bad deal and never sitting down at the table. The first is judgment. The second is forfeit.

---

Ugly buildings can have beautiful economics

Nobody drives past a data center and says, "Pull over, honey, I want a picture."

They are big. Plain. Secure. Usually windowless. Architecturally, they make a distribution warehouse look like the Biltmore.

But pretty is not the point.

A data center is a physical asset. It is expensive. It is taxable. And once built, it is very hard to move.

That matters.

Local governments spend enormous energy chasing employers, retailers, office users, and manufacturers that can expand, contract, relocate, automate, merge, or disappear. A data center is different. The servers may change. The equipment may turn over. But the land, power infrastructure, fiber connections, substations, cooling systems, and building improvements are not easily picked up and moved to the next county.

For a rural tax base, that can be meaningful.

This does not mean every data center is a good deal. It means the deal should be treated like a deal.

Too many communities respond in one of two bad ways. They either get starstruck and say yes too fast. Or they get scared and say no too fast. Both can destroy value.

The real data-center fight is not between Big Tech and small towns. It is between communities that know what they own and communities that give away leverage because they are either scared or starstruck.

---

Jobs are the wrong first argument

The standard economic-development pitch usually starts with jobs.

That is where data centers can get themselves into trouble.

Brookings recently reviewed data-center labor-market effects and found that they do create local employment gains, but fewer than industry boosters often imply. The strongest benefits depend heavily on the type of facility. Hyperscale campuses can produce broader local effects: fiber contractors, network operations, IT vendors, training pipelines, and supplier activity. Colocation facilities can be much thinner locally. One facility may not change much. A cluster can.

If a county commission hears "data center" and assumes all facilities are the same, it is already negotiating poorly.

A hyperscale AI campus, a cloud facility, a colocation building, a crypto-adjacent use, and an old-school telecom hotel are not the same animal.

So the better question is not, "Will this create thousands of permanent jobs?" Maybe it will not.

The better question is: What durable infrastructure, tax base, training, and local economic capacity can this project create if we negotiate correctly?

That is a much more adult conversation.

---

The power bill is the center of the fight

If I were advising a rural county, I would start with one question:

Show me how Grandma's power bill does not go up because your server farm moved in.

That may not be the technical wording. It is the political wording.

Electricity is the issue residents understand immediately. They may not know the difference between inference and training workloads. They may not care what chip is inside the rack. But they understand a utility bill. And as I've written before, power reliability and capacity have already become the binding constraint on this entire category of real estate — from Georgia to North Carolina to Virginia.

If a data-center developer cannot explain how it will power itself without punishing the neighbors, the deal is not ready to be approved.

That does not mean every project should be rejected. It means approval should come with a real mitigation plan: clear projected load, identified grid upgrade costs, ratepayer protection, dedicated or behind-the-meter power where appropriate, noise standards, water and cooling transparency, backup generation limits, tax transparency, local training commitments, and enforceable community benefits.

Not a glossy brochure. A bargain.

The industry should actually welcome that. Communities that understand the issues can approve faster and with less chaos. The worst environment for a developer is not a tough community. It is a confused one.

Confused communities panic. Sophisticated communities negotiate.

---

The window will not stay open forever

There is another uncomfortable truth.

The best sites will not sit around waiting for every county to finish studying them.

AI infrastructure spending is now being measured in hundreds of billions of dollars per year. Goldman Sachs recently framed the broader AI build-out as a multi-trillion-dollar capital cycle across chips, data centers, and power. Forecasts differ, but the direction is not subtle.

This is no longer a niche real estate category. It is becoming one of the physical engines of the economy.

And physical engines need a place to sit. They need land. Power. Fiber. Water strategy. Permits. Political permission.

That last item is becoming scarce.

If enough communities say no, the industry does not disappear. It moves to the communities that can say yes intelligently.

A community may think it is protecting itself by saying no to everything. Sometimes it is. Some sites are wrong. Some developers are too vague. Some projects should fail.

But sometimes "no" simply sends the tax base, infrastructure investment, and strategic relevance to the next county over.

---

Know what you are selling

Back in the early 1990s, we did not fully understand what we had at the Charlotte airport site until the phone started ringing.

We had land. We had buildings. We had airport proximity. But the real asset was the invisible line running near the property.

Today, rural communities may be in a similar position, only with higher stakes. They may not just own dirt. They may control the scarce intersection of land, power, fiber, water strategy, and public permission.

That is not something to give away casually. It is also not something to waste.

The right answer is not blind enthusiasm. No county should let a trillion-dollar industry show up with a rendering, a promise, and a tax-abatement request.

But the right answer is not reflexive opposition either.

The answer is a better bargain.

Data centers do not belong everywhere. They should not get a free pass. And if they cannot protect residents from real costs, they should not be approved.

But when the terms are right, these facilities can become part of the tax base, the power grid, the fiber map, and the next version of the American economy.

The towns that understand that will negotiate. The towns that do not will either get run over or left out.

I thought I was building warehouses in 1992.

I was standing next to the future and did not know it yet.

Rural America may be standing next to it now.

The question is whether it knows what to ask for before the future gets there.

---

Ready to think through a deal before you say yes or no?

At Cardinal, we use the Prepared to Win-Win™ Worksheet before every complex negotiation. It is the same structured tool we apply to corporate site selections, dispositions, and lease negotiations — and it works just as well for the kind of community-versus-industry conversations now unfolding across the South. The worksheet forces both sides to surface what they actually need, not just what they are asking for. That is where real bargains get made.

Cardinal Real Estate Partners | 704-900-0900 | www.Cardinal-Partners.com