A tractor does not know what it is planting. It runs the same number of hours whether it is pulling a planter through corn ground or hauling a manure spreader between the barn and the vegetable beds, and it sends no invoice that says which enterprise owes for which hour. Every farm that runs more than one enterprise off the same iron runs into this problem, and every farm that has tried to solve it has discovered the same thing: there is no fact of the matter waiting to be found. There is only a basis you choose, and the discipline of writing down which one.
Call it the tractor problem, because a tractor is the cleanest version of it, but it is really the same question asked of every piece of equipment a farm owns once: the sprayer, the combine, the loader, the irrigation pump. This page is about how to pick a defensible basis for splitting a shared machine’s cost across the enterprises that use it, and why the basis matters more than the arithmetic that follows it.
This is not tax or accounting advice. How you depreciate equipment, and what allocation methods your tax return will accept, is a question for your accountant, and the answer can legitimately differ from the management allocation described here. This page is about building a number you can use to decide which enterprise is carrying its own machinery cost — not about how that same equipment is treated on your books or your return.
The tractor does not know what it is planting
Every direct cost on a farm resolves itself the moment it is incurred. Seed bought for the north field is a cost of the north field; there is no argument to have. Shared equipment breaks that pattern completely. The same tractor plants the corn, sprays the beans, and hauls feed to the cow herd, and none of those jobs generates a separate invoice for its share of the machine. The cost exists at the level of the whole tractor, for the whole year, and has to be pushed back down to the enterprises that actually used it.
This is not a bookkeeping inconvenience. It is the same structural problem that shows up wherever one resource serves several enterprises — the same one covered in enterprise budgeting more broadly, and it is worth naming plainly rather than working around, because the temptation is always to skip it by treating the tractor as an overhead line that touches the whole farm evenly. Evenly is the easiest answer and usually the wrong one.
It is also worth being clear about what counts as a shared piece of equipment in the first place, because the answer is broader than the obvious candidates. A tractor and a combine are the easy cases. A truck that hauls feed one week and produce to a farmers’ market the next is a shared asset in exactly the same sense, and the same basis-and-write-it-down discipline applies to it, whether the farm runs a cow-calf herd or a mixed vegetable operation off the same yard.
Pick a basis, not an average
Splitting the tractor’s annual cost in half because there happen to be two enterprises produces a number that looks precise and carries no information. The enterprises did not use the tractor equally just because there are two of them. A defensible split starts from a measure of actual use — hours run on each enterprise, tracked from an hour meter, or acres worked where the work each acre demands is comparable across enterprises. Either can be right. What makes a basis defensible is not which one you pick, it is that you can explain it to a stranger and that the explanation survives the question “why this and not something else?”
Say the tractor runs four hundred hours in a season. If two hundred and fifty of them were on the row crop and a hundred and fifty were on hay ground and hauling for the cow herd, the tractor’s annual cost — fuel, repairs, depreciation, interest, insurance, the shed — splits roughly five-eighths to the row crop and three-eighths to the rest. That is an example built to be hypothetical, not a number to expect on your own machine, but the method is the point: the split follows measured use, not the number of enterprises on the farm.
Ownership cost is usually bigger than the fuel bill, and easier to forget
It is natural to think of a tractor’s cost as what it burns and what it breaks — fuel and repairs, both of which show up as cheques written during the season. Depreciation, interest on any loan against the machine, insurance, and the cost of the shed it sits in idle six months a year are just as real, and often larger, but they do not arrive as a bill during planting or harvest, so they are the first thing left out of an enterprise cost built in a hurry.
Say the tractor costs some amount a year to own, once depreciation, interest, and insurance are added up — a figure that is entirely hypothetical here and specific to your own machine, financing, and age of equipment in practice. That ownership cost gets split by the same basis as the operating cost, hours or acres, and added to it before the total is divided among enterprises. Skip the ownership half and every enterprise on the farm will look cheaper to run than it actually is, in exactly the same amount, which at least does not distort the comparison between enterprises — but it will badly distort the comparison against buying, renting, or hiring custom work instead of owning the machine at all.
This is also where the split needs to hold still long enough to mean something. A tractor traded for a newer model partway through the season, or a loan refinanced mid-year, changes the ownership cost without changing anything about how the hours were actually used. Keep the two apart: update the ownership figure when the facts underneath it change, but do not let a financing decision quietly rewrite the usage basis you have been tracking all along, or next year’s comparison to this year stops being a comparison of farming and starts being a comparison of financing.
Write the basis down where the number lives
The number that survives a year, a lender’s question, or your own memory next winter is not the allocated dollar figure — it is the rule that produced it. “Tractor cost split by hours run, logged from the hour meter at each job” is a sentence a lender can accept or push back on. “It seemed about right” is not a sentence at all by the time next winter arrives, because by then it will not even be a memory. Write the basis next to the number, every time, in the same record where the cost lands.
This is also what makes the basis changeable on purpose rather than by accident. If the mix of work shifts — more acres in the row crop, less hauling for the herd — hours run will shift with it and the split will update honestly next year. A basis you cannot recall cannot be revisited; it can only be repeated by habit, whether or not it still fits how the tractor is actually used.
Hours logged at the job beat hours reconstructed in December
An hour-meter reading taken when a job starts and again when it ends, noted against the field or group the tractor worked, is a fact. The same figure rebuilt in December from memory of a season is a guess, and guesses reconstructed under time pressure tend to guess in whichever direction confirms whatever the farm already believed about which enterprise is doing well. The gap between those two versions of the same number is the gap between a cost of production and a plausible story, which is the same latency problem covered in recording work at the point of work. Log the hours when the tractor moves from one job to the next, not when the spreadsheet gets built.
Farm40 records operations and the hours or transactions tied to them against a specific enterprise, and its cost-of-production export rolls a shared piece of equipment’s tagged costs up by whichever enterprises you assigned them to. Its limit sits exactly where this page has been pointing: it does not choose your basis, and it cannot tell the difference between an hour logged at the job and a number typed in from memory months later. The export totals what was tagged. Whether what was tagged reflects what actually happened at the tractor is a discipline no export can supply.
A farm that logs hours at the job and writes down its basis will get a real number out of a notebook. A farm that skips both steps will get a confident number out of the best software available, and the confidence will not survive the first hard question about how the split was made.
The tractor problem is really just one instance of the general discipline laid out in cost of production: shared costs do not resolve themselves, and the honesty of the whole exercise lives in the basis you choose and the willingness to write it down.