Toyota's Real Moat Isn't a System. It's That Nobody Has the Patience to Copy It.

The Toyota Production System has been an open book - literally - since 1977, when four Toyota engineers laid out kanban and just-in-time in a peer-reviewed journal⁸. Eleven years later, Taiichi Ohno published the rest of it under his own name, in English, for anyone with the price of a hardcover¹. Every component is documented. Every principle is taught in business schools. Consultants will install it in your plant for a fee. And after nearly fifty years of this, no rival has matched it. That is the strange fact at the center of the most studied manufacturing system on earth: the recipe is public, and almost nobody can cook the dish.

The official story is that TPS is a set of brilliant tricks - the kanban card, the andon cord, the supermarket epiphany. Strip those away and you find the trick was never the point. The moat is not what Toyota knows. It is how long it took Toyota to know it in its bones, and the fact that knowing it in your bones cannot be bought by quarter.

The system isn't an idea. It's an accretion.

Pull a thread on any origin myth and it comes apart. The phrase 'just-in-time' was not Ohno's coinage - it was Kiichiro Toyoda's, the company's first president; Ohno operationalized the idea but credited the name to his founder in his own writing². The beloved tale that Ohno walked a Piggly Wiggly aisle and saw the pull system in a flash has no primary source behind it: the supermarket concept reached him through colleagues' photographs in 1951–52, kanban was already running by 1952–53, and Ohno did not set foot in a U.S. supermarket until 1956 - years after the system worked⁶. And TPS did not spring from retail at all. It was a magpie's nest: Ford's moving line, German aircraft makers' takt-time, a 1954 Lockheed logistics article, Japanese textile-mill jidoka⁷. The system was assembled, not invented - and that distinction is the whole strategic story.

Why does the myth-busting matter for strategy? Because a single epiphany is copyable and an accretion is not. If TPS were one insight, a competitor could license it and be done. Instead it is a synthesis that took decades to settle, and Toyota's own corporate history makes the timeline brutally clear: kanban was not adopted across every Toyota plant until 1963, and suppliers were not folded into the kanban loop until 1965³. That is roughly a decade between 'the system works in one shop' and 'the system runs the whole enterprise and its vendors.' Ten years of patient, unglamorous diffusion is the part no slide deck captures - and the part no rival on a quarterly cadence can stomach.

Why the open recipe stays uncopied

Here is the mechanism that makes a published system a durable moat. TPS rests on two pillars - just-in-time and jidoka, the practice of stopping the line the instant a defect appears². Both are cheap to describe and expensive to mean. Just-in-time means running with almost no inventory buffer, which only survives if every worker, every supplier, and every machine is reliable enough that a single hiccup doesn't halt the plant. Jidoka means giving a line worker the authority - and the cultural permission - to stop a multimillion-dollar line over one bad weld. A Western firm can install the andon cord in an afternoon. Getting a worker to actually pull it, and a manager to thank rather than punish them for it, takes years of demonstrated trust. The cord is hardware. The willingness to pull it is the moat, and the willingness is what compounds through kaizen - thousands of tiny improvements made by people who believe the system is theirs.

The numbers say the moat is real. In the fiscal year ending March 2024, Toyota turned ¥45.1 trillion in revenue into ¥5.35 trillion of operating income - an operating margin near 11.9%, with operating income nearly doubling from the prior year⁴⁵. For a mass-market automaker, where rivals routinely scrape by in low single digits, that is the financial signature of a production system competitors have been reading about - and failing to clone - for half a century. The advantage isn't secrecy. Sugimori and his colleagues gave it away in 1977⁸. The advantage is that everyone who read the paper still has to spend the decade Toyota already spent.

What if the moat is now the cage?

The honest objection is that this whole story is too flattering, and the flaw is hiding inside the strength. The very thing that makes TPS impossible to copy - that it is slow, embedded, and cultural - is exactly what makes it slow to change. A system optimized over seventy years to eliminate defects from physical assembly is a system tuned for a world where the car was the hard part. That world is ending. In the software-defined-vehicle era, the differentiator is no longer the fit of a panel but the speed of a code release, and speed-to-market punishes the patient. A culture that took a decade to roll out kanban does not, by temperament, ship a feature overnight. The same embeddedness that locks rivals out can lock Toyota in. The fair read is not that TPS has stopped being an advantage - the margin says otherwise - but that an advantage built on the slow accretion of physical discipline may be the wrong asset to be carrying into a race measured in software cycles.

Toyota's genius was never a card on a string or a flash of insight in a supermarket aisle that probably never happened. It was the unfashionable willingness to spend thirty years assembling other people's ideas into something only it could run - and to keep refining it long after the manual went public. That patience built a system rivals could read but not become. The open question is whether the same patience, pointed at a world that now moves at the speed of software, becomes the one thing Toyota studied too long to unlearn.