John Deere Sold You the Tractor. It Kept the Key to the Engine.

A farmer can buy a combine the size of a small house, sign the title, drive it off the lot, and own it outright - and still be unable, legally, to make certain repairs to it in his own shed. Not because he lacks the wrench. Because he lacks the software. The full-function diagnostic tool, Service ADVISOR, stays with the dealer. When a sensor trips and the machine throttles itself down in the middle of harvest, the wrench is useless without the key, and the key is on a different farm, behind a different door, billing by the hour.¹

The official story is that John Deere sells tractors. It does. But the more durable business is the one that begins after the sale - the repair that can only happen through the dealer, on Deere's terms, because Deere kept the one tool that finishes the job. You bought the machine. They kept the engine's key.

The sale ends. The dependency doesn't.

Here is the mechanism, worked down. A modern Deere machine is a computer that happens to harvest corn. Many repairs aren't mechanical anymore - they're diagnostic: a fault code has to be read, a part has to be re-paired to the machine's electronics, the software has to bless the fix before the machine will run at full power again. That final, blessing step requires the fully functional Service ADVISOR tool. The FTC's complaint alleges Deere withheld that tool from farmers and independent shops, restricting it to authorized dealers - and that, for the repairs requiring it, this produced a complete grip on the work.¹⁴ The lock-in isn't a contract clause anyone reads at purchase. It's a missing tool you discover only when the machine stops, at the worst possible moment of the year, with the weather closing in. Proximity to that moment is the whole asset. The dealer doesn't have to be cheaper or better - it has to be the only door that opens.

That is the thesis, stated plainly: Deere isn't only an equipment maker that happens to run dealerships. It built an ecosystem in which the sale is the easy part and the repair is the recurring one - and the software is the gate that keeps the recurring part flowing back through its own network. The genius, and the legal vulnerability, are the same thing.

Why a handshake didn't close the toolbox

The pressure didn't start in court. In 2023 Deere signed a Memorandum of Understanding with the American Farm Bureau Federation, pledging to give farmers and independent mechanics the same tools and software as dealers. It read like a concession. It wasn't much of one. The MOU had no enforcement mechanism, and Deere could walk away from it with thirty days' notice; in exchange, the Farm Bureau agreed not to support Right to Repair legislation.³ Read it closely and the structure is revealing: a voluntary promise that costs nothing to break, traded for a real political favor - the suppression of laws that would have made the promise mandatory. A gate you can re-lock in a month is not an open gate. It's a gate with better public relations.

Then the binding thing arrived

On January 15, 2025, the FTC and the attorneys general of Illinois and Minnesota sued Deere under the Sherman Act and the FTC Act, alleging the company maintained a monopoly in the repair aftermarket for large tractors and combines by withholding Service ADVISOR.¹ Then, in April 2026, the separate class action that had been grinding since 2022 settled: Deere agreed to pay $99 million into a fund and - this is the part that matters more than the dollars - to make digital repair tools available for ten years.² A handshake the company could exit in thirty days had become a court-supervised commitment measured in a decade. That is the difference between a press release and a contract.

Notice what the $99 million is and isn't. It is not a penalty - the settlement contains no finding of wrongdoing, and Deere continues to deny liability.² The fund also covers attorneys' fees and litigation costs before a dime reaches farmers, and PIRG worked out the residue at roughly seventy-nine cents per affected acre across the period covered.⁸ Set that against $7.1 billion of net income in fiscal 2024, on net revenues of $51.7 billion, and the number stops looking like punishment and starts looking like the price of a strategy that worked for years.⁷ The settlement isn't the cost of the lock-in. It's the cost of finally writing down the terms.

The data moat everyone assumes - and nobody can prove

There's a tempting second chapter to this story: that the real prize was never repair revenue but data. Every connected machine feeds the Operations Center, Deere's opt-in cloud platform that centralizes machine performance and agronomic data, with JDLink telematics uploading automatically over cellular.⁶ The narrative practically writes itself - lock the farmer into the ecosystem, harvest the field and the data, sell software back as a high-margin recurring service. It's a clean story. It's also, on the public record, unproven. Deere does not break out Operations Center or precision-ag software revenue as a discrete line in its filings; it folds the activity into 'Production & Precision Agriculture' broadly.⁶⁷ Any specific 'data revenue' figure you see attributed to Deere is an outside estimate, not a number you can trace to a filing. The repair lock-in is documented and litigated. The data moat is, for now, a thesis in search of a disclosure.

The honest case for the defense

The fair objection is that 'monopoly' is the plaintiff's word, not a settled fact. The FTC voted only 3-2 to sue, and both Republican-appointed commissioners dissented - one calling the complaint a product of 'brazen partisanship' filed without sufficient evidence.⁵ Deere calls the suit meritless and has admitted no wrongdoing anywhere in the record.⁵² And the FTC's framing of total dominance applies specifically to the narrow band of repairs that require the full Service ADVISOR tool - not to tractors at large, where reasonable estimates of Deere's share vary by source and methodology. The honest read: this is a contested case, not a proven crime, and a company is entitled to argue that gating complex software is about safety and quality, not extraction. But the steelman has a ceiling. Deere fought the disclosure for years, signed a voluntary promise it could revoke in a month, and only committed to ten years of tool access inside a binding settlement it still won't call an admission. A position defended that hard, conceded only that grudgingly, tells you what the position was worth.

Deere built one of the most durable businesses in heavy industry by understanding something subtle: the moment of maximum leverage isn't when you sell the machine - it's when the machine breaks. It stood in that doorway with the only key, and for years it collected the toll of being the only door that opened. The lawsuits didn't prove the strategy was illegal. They proved it was valuable - valuable enough to litigate for years and settle for a sum that barely registers against a year's profit. The tool is opening now, ten years at a time, by court order rather than goodwill. The lock-in was never in the hardware. It was in the one piece of software they declined to hand over - and they spent a very long time discovering exactly what that refusal was worth.