The $26 Million No: What Data Centers
Keep Getting Wrong About Land

In 2007 I stood in a gravel driveway in York County, South Carolina, trying to buy a strip of right-of-way from a man who had lived in the same trailer his whole life. I had a good buyer, real money, and a more than fair price. He listened politely, thanked me for driving out, and said no. I would hear that word a lot that year. It took me almost two decades to understand what he was actually telling me.

A number alone, even a really good one, is not a complete offer.

---

Twenty-six million reasons

On Sunday the Wall Street Journal ran a story about a Kentucky mother and daughter who turned down $26 million for their farmland. The buyer, later reported to be Meta, wanted 534 acres for a data center the size of two Central Parks. The women signed. Then they learned what the land was for, revoked the agreements, and told the buyer's representative to "kick rocks".

Most of the commentary treated them as stubborn "country hicks" (their words, not mine). I read it differently. They made two decisions, both rational. They agreed to sell under the belief the land would be taken anyway. When the facts changed, they changed their minds. What they were handed was a purchase price, an unnamed buyer, an unstated use, and a clause on page seven telling an 83 year old widow not to discuss the deal with her neighbors. Under that much uncertainty, no is the only answer you can take back. So no is what a sensible person picks.

I've been Meta. Same playbook, same result, smaller zeros.

---

The Seed Farm

Bowater, the paper and timber giant, owned roughly 1,034 acres on the Catawba River near Rock Hill that everyone called the Seed Farm: an experimental plot for growing the perfect newsprint tree, and a duck hunting retreat for executives. Miles of river, rolling topo, minutes from Rock Hill's mall. One of the finest tracts I have ever walked. I represented Bowater when it sold to Newland Communities in 2007, and that part went the way a well run disposition should.

Then Newland hired me for the harder job. They needed roughly another thousand acres around the tract, and they needed a corridor to extend Dave Lyle Boulevard from Rock Hill out to the property. That meant assembling land from the people who already lived there. Many of them lived in trailers. Almost all of them had been on that ground longer than I'd been in the business.

So I knocked on doors. I sat at kitchen tables. I explained the buyer was reputable and the price was above market and the check would clear. And I got no. Then no again. Then a version of no that came with a glass of tea and an invitation to come back anytime, which is the nice Southern way of saying "go kick rocks" or "pound salt".

I could not finish the assemblage. Newland never got Dave Lyle Boulevard extended. Without the road, the master plan couldn't pencil, and the company eventually sold the Seed Farm to York County for a park. One of the finest development tracts on the Catawba became public land, not because the vision was wrong or the money was short, but because a few families on the route to it were never given a reason to say yes.

For a long time I told myself the usual things: they were unrealistic, they were emotional, they didn't understand what the land was worth.

Here is what I understand now. The number was never the constraint. The number was all I had. I couldn't tell those families what would sit next to them, what the road would sound like, whether their well would still run, or what would be different for them the morning after closing except that they'd be gone. I was asking them to trade certainty for cash, and cash was the one thing most of them didn't need badly enough to take the risk.

---

The Wave

The North Carolina figures that follow come from Hamlet ("Hamlet, Sightline record, as of Aug 13, 2026"), whose Sightline team gave me access to their record of what North Carolina actually said about data centers: every mention in the public meeting transcripts they hold, read and verified, each quote linked to the video of the moment it was said. Not sentiment scores, not survey data; the record itself.

Charlotte City Council voted 11 to 0 in June for a 150 day moratorium on new data centers. Not a fringe. Not a close vote. Every ideology on Council landing in the same place.

That vote is one data point in a pattern. In North Carolina alone, 28 jurisdictions now have a data center restriction adopted and in force, with five more taking formal steps. Nationally it's 87 places across 29 states. And here's the detail that matters: most of these aren't bans. They're pauses with end dates, 150 days in Charlotte, a year in Asheville and Boone. Towns aren't saying never. They're saying not on these terms.

The pace of this change is the story here, not the moratoriums themselves. Twenty-eight North Carolina jurisdictions in roughly nine months. Durham County wrote its pause to line up with the city's. This is a cascade, and cascades don't happen because one town got angry. They happen because a template exists and the cost of adopting it has dropped to near zero.

The kicker is Statesville. In the entire North Carolina record, Statesville produced more statements in favor of data centers than any other town. On August 10, Statesville adopted its own 150 day moratorium. Staff framed it as proactive good governance: pause, write the rules, then reopen the door. Even the friendliest town in the state wanted its standards on paper first.

---

What they're actually objecting to is not what you think

Of 161 recorded objections in the state's meeting record, nearly half are about land use and where these buildings go. Water is one in eight. Noise is one in fifty. People aren't objecting to what data centers consume; they're objecting to what shows up next door without asking.

That number reframes the whole debate. The industry has spent two years arguing about consumption of water and power. Water usage per megawatt, cooling loop efficiency, comparisons to golf courses and almond farms.

Peter Diamandis has made the point on his Moonshots podcast that a golf course can draw more water than a modern data center, and that agriculture dwarfs both. True enough. And beside the point. You can win the water argument in every hearing room in the state and still lose, because water was never what filled the room.

---

Who fills the room

According to Hamlet's research, the meeting record runs 161 statements against to 19 for. That's not a poll; it's turnout. The people who bear the disruption come to the meeting; the people who'd quietly benefit stay home. Which is exactly why a number aimed at the seller can't fix this: the seller was never the one filling the council chamber.

Economists have a name for this. Concentrated costs and diffuse benefits. The family whose back fence will face a 200 acre building has every reason to drive to city hall on a Tuesday night. The county taxpayer who will pay slightly less because the assessed value doubled doesn't know it yet, and wouldn't drive across town for it if he did.

Sunday's WSJ Kentucky story shows the same split in miniature. The landowners with 200 acres and dividend portfolios said no. The bar owner across the road with twelve kids and regulars stretching a drink said hell yes. The people who hold land as identity resist. The people who hold a job as survival welcome it. That is not a fight about artificial intelligence. It is a fight about who bears the disruption and who captures the windfall.

---

"Fine, but these towns are leaving money on the table."

I hear you, and I'm not going to pretend otherwise. I've worked on data centers in Charlotte since the early 1990s, back when they were called switch sites and nobody outside a utility cared where they went. I serve as an expert witness in data center litigation today. I have done business with data center operators. I am, by temperament and by trade, pro-development, and I still think most of these pauses are the right call.

Here is the honest version of the trade. Data centers are the best fiscal deal in local government: enormous assessed value, almost no demand for schools, police, or social services. They are also among the weakest job creators per acre of any major land use. Both of those things are true at the same time. A serious pro-growth case for data centers has to say that out loud rather than paper over it.

And a pause to write siting standards for a genuinely new land use, arriving at unprecedented scale, is not obstruction. It's the thing I've argued for in every other context: rules set in advance, applied evenly, so deals get decided by standards instead of by whoever shouts loudest at the hearing. Ambiguity plus outrage is what produces one year bans. Clear standards produce pipelines.

---

What consent costs to earn

The good news is that the industry's problems are unusually self-inflicted, which means they are unusually fixable. Data center developers can be their own worst enemy. The buildings are ugly. They arrive under NDA. They tell a town what they need before they ask what the town needs.

If I were advising a developer today, I would stop bidding and start designing. A building that doesn't announce itself, sited and buffered so the neighbors don't see it or hear it. Power the project generates on site, so ratepayers are never the backstop for someone else's load. Water people can watch recirculate, not a compliance memo about it. And benefits that land on the households living beside the project, not just the landowners cashing out. In Kentucky, the money went to the sellers and the town got promises. Turn that around and you change who shows up on Tuesday night.

---

Back to the driveway

If I could go back to that gravel driveway in York County, I wouldn't bring a bigger check. I'd bring answers. Who is coming, what it will look like, what you will hear from your porch, and what will be different for you the day after we close that you would actually want. That is an offer. What I brought in 2007 was a number.

The developers who figure this out first will find towns on the other side of the moratoriums waiting to recruit them. Move the benefit next door, in a form the neighbors can see from the porch, and you change who drives to city hall on Tuesday night. When turnout flips, the political math flips with it. I've watched every kind of infrastructure make that trip, from railroads to interstates to the switch sites nobody wanted in 1993. Data centers will make it too, but only for the developers who understand that consent is an input, not an obstacle.

---

The tool

This is the question at the center of The Negotiation Worksheet™, the same tool we use on every Cardinal engagement. It forces you to write down not what you want from the other side, but what the other side is trying to protect. The families on Dave Lyle Boulevard were protecting the only certainty they had. Meta's neighbors in Kentucky were protecting the right to know who was moving in. In every case the number was the least interesting thing on the table.

If you're heading into a negotiation where the other side keeps saying no and you can't figure out why, give us a call or send an email. We'll walk you through the Worksheet and help you find the question you haven't asked yet.

John Culbertson | 704-900-0900 | jculbertson@cardinal-partners.com | www.Cardinal-Partners.com

---