Pallet size lock-in
A Socratic walk-through of pallet size lock-in — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why does the size of a wooden pallet still decide the shape of trucks and warehouses?
A pallet is a slab of softwood with gaps in it. In North America the common one measures 48 by 40 inches; in much of Europe it is 1200 by 800 millimetres. Neither number was derived from anything fundamental. They are conventions, settled decades ago, and they are not even the same convention — the international standard for pallet footprints, ISO 6780, recognises six sizes rather than one.
Yet those numbers keep turning up in far more expensive objects. Trailer interiors are about the width of two pallets side by side. Racking bays are sized in pallet positions. Forklift forks, carton dimensions, shelf depths, the pitch of a warehouse's columns: all of them refer, eventually, back to the slab of softwood. Why should the cheapest object in the chain be the one nothing else is allowed to contradict?
Reasoning it through
REASONING #Turn the question around and ask what the first mover actually chose. In the 1930s and 1940s, when forklifts and pallets spread together, somebody had to pick a footprint. Any of several would have worked. The choice was close to arbitrary — constrained loosely by what a person could handle and what early trucks could take, but not determined.
Now ask what happens to the second decision. Once a firm owns pallets of a given size, the useful trailer is the one that fits them cleanly. Two 48-inch pallets set side by side come to 96 inches, which slots into a standard trailer's interior of roughly 98 inches with a little to spare. That is not a coincidence and it is not physics; it is the trailer being built around the pallet that already existed.
The third decision inherits both. Racking is bought to hold that pallet in that trailer's flow. Cartons are dimensioned to tile the pallet's face without overhang, because overhang crushes and voids cost money. Each new investment is cheapest when it conforms, and each one that conforms makes the next deviation dearer.
This is the engine worth naming. The advantage of the standard is not that it is good; it is that it is shared. Every additional adopter raises the payoff to adopting and raises the penalty for not. Economists call this increasing returns to adoption, and where it operates, an early lead compounds into something no later competitor can dislodge on merit.
So consider what switching now would cost. Not the pallets — they are the cheap part, and a fleet could be replaced in a season. The cost is everything sized around them: the racking, the trailers, the automated storage cranes, the carton tooling, the load-planning software, and above all the coordination, since a shipper who changes alone simply becomes incompatible with every partner who has not. The stock of complementary assets is what holds the standard in place, and it is worth orders of magnitude more than the standard itself.
Here is the observation that makes the diagnosis firm rather than a story. If the surviving sizes were the ones best fitted to the task, the world should have converged on one. It has not. Different regions locked in different footprints, for local reasons — the European pallet's dimensions trace back to a railway-wagon convention adopted through the international rail union in the early 1960s, and the North American grocery trade standardised separately around its own. Both persist. Multiple stable outcomes, selected by history rather than by merit, is precisely the fingerprint of path dependence, and it is not what you would see if the pressure of efficiency were doing the choosing.
One honest complication: the sea container arrived from a different lineage, sized for ships and cranes, and its interior does not tile neatly with either main pallet footprint. So the world carries permanent dead space in its containers — a cost nobody chose and nobody can unilaterally remove.
The analogy
THE ANALOGY #Think of the rails under a tram network in an old city.
The gauge was fixed once, early, by whoever laid the first track. Nothing about that width was optimal. But every tram bought since has axles cut to it, every depot has pits spaced for it, every junction and points motor assumes it. The rails themselves could be relaid over a summer. The system on top of them could not — and while any single line is relaid, its trams can run nowhere else.
A tram gauge is a physical interlock that simply will not accept a wrong-sized vehicle, whereas an odd-sized pallet does fit on a standard trailer — badly, wastefully, with lost cube and crushed corners — so the pallet standard is enforced by accumulated cost rather than by outright impossibility, which is why exceptions exist at the edges and gauge exceptions do not.
Clarifying the model
THE MODEL #Lock-in is not proof that the standard is bad. Path dependence says the outcome was selected by history rather than by optimisation; it does not say the result is inferior. The 48 by 40 footprint is perfectly serviceable. The claim is narrower and more interesting: we cannot infer from its survival that it is the best available, because survival here is explained by adoption, not by fit.
The pallet is not really the standard — the interface is. What is locked in is a shared expectation about a footprint, and the wood is only its cheapest physical expression. This is why plastic and composite pallets replaced wooden ones in many chains without disturbing anything: they changed the material and kept the interface. Changing the interface is the hard thing; changing the object is not.
Coordinated switches do happen, which shows the lock is economic rather than absolute. Pallet pooling schemes converted large parts of some national markets to a common footprint by moving all the participants together. That is the general shape of an escape: someone must make the simultaneous move worth everyone's while, because no individual move is.
The falsification test. If the footprint persists through increasing returns rather than through fitness, then a market that industrialised later, without an installed base to inherit, should be free to have chosen differently — and should show a different size. Regional footprints do differ along roughly those lines, which is evidence for the lock-in account; a world converged on one size would have been evidence against it.
A picture of it
THE PICTURE #How to readRead every relationship as "was dimensioned around". The pallet sits on the one side of nearly every line: the crow's-foot entities each adopted its number, and each is far costlier than the pallet. The container is the exception — a different lineage that never agreed, so the mismatch is permanent. The pool is the only entity acting on all pallets at once, which is why a coordinated switch must come from something shaped like it.
What became clearer
WHAT CLEARED #The pallet governs the warehouse not because it is important but because it was early. Once a shared dimension exists, every later and costlier object is cheapest when it conforms, and the accumulated conformity — not the wood — is what cannot be moved. That tells you where change is possible: never at the standard itself, always at the level of everyone depending on it, moving together.
Where to go next
ONWARD #- Whether automated warehouses, buildable to any pitch, loosen the lock or re-lock it elsewhere.
- The same pattern in rail gauge and keyboard layout.
- How pallet pools finance a coordinated switch, and why most such attempts fail.
Key terms
TERMS #| Term | What it means |
|---|---|
| Path dependence | when the outcome of a process is determined by its history, so that early and possibly arbitrary events select among several stable end states. |
| Increasing returns to adoption | when the value of using a standard rises with the number of others using it. |
| ISO 6780 | the international standard that recognises six pallet footprints rather than mandating one. |
| Complementary assets | the surrounding equipment and processes built to fit a standard, whose replacement cost is what makes switching expensive. |
| Pallet pool | a shared rental fleet circulating one common pallet type among many participants. |
Every term the collection defines is gathered in the glossary.