There is no single break-even price for what your farm sells. There are at least two, both of them honest, and they answer different questions: one covers what leaves your bank account this cycle, the other covers everything, land included. The distance between them is wide enough to reverse a decision, which is why the number has to be written with the list of what is inside it.
A buyer calls with a price and the answer goes back down the phone, off a sense of cost that nobody ever wrote down. Ask two people on the same farm on the same day what the lowest price is, and two figures come out. One of them counted the rent and the other did not. One counted the family’s draw and the other did not.
One counted the combine harvester’s depreciation and the other treated the machine as paid for. Neither is wrong. They are answers to different questions, and nobody said out loud which question was being asked. Months later, when the argument is whether the price was any good, there is no referee, because no figure with a date on it existed at the moment the phone was in your hand.
Two floors, and the third one behind them
The formulas are one line apart in the same paragraph. The Handbook on Agricultural Cost of Production Statistics, published by FAO in February 2016 to tell national statistics offices how to compile this data, writes them out as a pair: “Break-even price to cover variable costs (or gross margin): Total variable costs ÷ expected yield = USD/ unit produced”, and “Break-even price to cover total costs (or net margin): Total costs ÷ expected yield = USD/unit produced”. Same denominator, two numerators, two prices. The handbook then recommends that the indicators compiled be plural on purpose: returns over variable costs, returns over total costs, and returns over total costs excluding imputed costs for owned inputs.
Three is what the handbook shows in practice, in a country far from any of the budgets quoted below. Its Zambian example of maize costs is compiled in local units, a 50 kg bag, and the sentence introducing the table names what it reports: “total cash expenditures; total cash expenditure plus household labour and owned assets (excluding land); and total cost including land cost”.
| Cost aggregate, as the handbook builds it | What is inside it | The question it settles |
|---|---|---|
| Total cash expenditure | What left the bank this cycle | Does the sale cover what was actually paid out |
| Cash plus household labour and owned assets, land excluded | Adds the work of the family and the machinery already owned | Is the work and the machine being paid for |
| Total cost including land | Adds a charge for the ground, rented or owned | Is the activity paying for the land it stands on |
How far apart the two floors sit
Far enough to change the answer. The 2025 Illinois Crop Budgets put both figures side by side for the same grain in the same year: “Break-even corn prices to cover non-land costs range from $3.17 to $3.66 per bushel”, and “break-even corn prices increase to $4.60 to $4.66 per bushel for all costs”. For soybeans the pair runs “from $6.53 to $8.38 per bushel” against “$11.01 to $11.56 per bushel for all costs”.
The gap is what a land charge costs, and the budgets are explicit about how they measure it, using an average cash rent value that varies by region. Illinois draws its lower line in a different place than the sheet above does: its first floor covers every cost except the land, machinery and labour included, not only what left the bank this cycle. That is the argument rather than an objection to it. The line has to be drawn somewhere, and what a sheet owes its reader is the sentence saying where.
Those figures describe grain farms in Illinois, in US dollars per bushel, which is the unit those budgets report in, and they are not your floor. What travels is the shape: a price that clears the first floor and misses the second is a price at which the crop pays for the diesel and does not pay for the ground. The publication names each side of that line, since “Operator and land return equals gross revenue minus non-land costs” and “Subtracting out a land cost results in net farmer return”.
Your own pair comes from the same place theirs does, a public research or extension budget for your market, read with one question: which lines went in. A floor only settles an offer if the offer is written down beside it, which is what a price record with a source keeps. The figure is built out of costs and spent in a sale, which is why it sits with the rest of the marketing axis and not with the bookkeeping.
What the second column has to cover
Four lines, named one by one. The 2025 Purdue Crop Cost and Return Guide names what has to fill the gap between the two floors. It works with a figure it calls the contribution margin, “obtained by subtracting total variable cost from market revenue”, and states in the next sentence what that figure is for: it “is used to cover overhead costs such as machinery ownership costs, family and hired labor, and cash rent”.
Machinery ownership, family labour, hired labour, land. Those four are the contents of the outer column, and this sheet does not work any of them out. It receives them already worked out, from the invoices, the payroll and the lease the farm already keeps, and puts each one in a column. Working out what a machine costs to own is a different piece of work in the finance axis, and it starts from the useful life you declared. Doing it here would turn a two-hour job into a month. What sits above the outer column is a further declaration, since a declared margin target is written against this floor and never in place of it.
Why your neighbor’s floor is not yours
Because the same line is a cost on one farm and invisible on the next. A study of Czech farm accounts published in 2022 makes the mechanism plain: the standard income indicator counts wages paid, rent paid and interest paid, and “The final indicator does not take into account the costs of unpaid labor, the owned land, and the equity capital, which are available to agricultural holdings”.
The neighbor who rents his ground has a rent line. You, who own yours, have none, and your floor comes out lower without anybody deciding it should. As the authors put it, the costs of a farm’s own factors “are equivalent to external factors which, unlike own factors, are already considered in the final management indicator”.
Nobody has settled how to price those own factors, and the study says so twice: “The different authors lean towards different approaches for determining wage costs that are used to value unpaid labor”, and approaches to the land charge “also differ”. That is a five-year run of European farm accounting data from one country, and what it is doing here is the accounting, not the figures.
So this piece will not tell you which lines belong in which column. It tells you the split has to be written at the top of the column and has to stay the same across any two cycles you compare, which is exactly what the FAO handbook demands of a statistician deciding how to split a shared cost between two products: “the first requirement is to (a) establish a uniform algorithm and (b) make transparent the one that has been used”. Whether the family’s draw is a cost, and whether owned land carries a charge, is a call for you and the accountant who files where you file.
The line that says what did not go in
Written under the two figures, in words, with a reason each. The 2026 Illinois Crop Budgets carry a sentence that a farm sheet almost never carries: “These break-even price calculations do not include other revenues such as support payments.” A publication covering an entire state declares what its own number leaves out. A single farm has less reason to hide that, not more.
One line is left out so often that it deserves naming here rather than on the list: what it costs to hold the product after harvest. A floor built at harvest stops being the floor the day the lot goes into the bin, because every month held adds interest, shrinkage and handling to the same denominator, and what a month of storage costs is worked out per unit and per month for exactly this reason. The floor six months into storage is not the floor at harvest.
The exclusions run on the production side too. The FAO handbook is specific that the denominator is not everything the field grew, since “the production unit should reflect only the marketable output by excluding waste”. Grain that went to the bin and never left it, calves that did not make weight, fruit that graded out: each is a decision, and each one moves the price per unit without ever showing up as a cost line.
The number you divide by
Expected production, and it is a declaration like any other. The FAO handbook solves the same three quantities for the other unknown and calls it break-even yield, “Total costs/expected price = unit produced (minimum yield required to cover all costs)”. Cost per unit falls when the denominator rises, so an optimistic production figure lowers your floor on paper and lowers nothing in the bin. Write the figure, and write where it came from.
The floors also move between one cycle and the next, which is the argument against writing the sheet once. The same authors published a new set of floors for 2026, for the same state and the same two grains, built the same way, and they are not the floors they had published a year earlier. A floor is a statement about one cycle’s costs and one cycle’s expected production, and both of those change.
The same sheet where there is no grain
Break-even is not a crop idea. A scoping review published in the Journal of Applied Farm Economics in 2025 screened 372 articles on farm-level financial methods in beef cow-calf operations and kept 61, identifying eight methodologies in use, among them break-even analysis, gross margin, enterprise analysis and partial budgeting. The review argues for models that integrate “key decision points, such as break-even prices”, which is the same sheet with a weaned calf in the denominator instead of a bushel.
What that review counted was published studies, not farms, so it tells you the method exists and is named in the literature of cattle. It does not tell you how many operations run it, and no source in this piece does.
Who signs it, and the date it comes back
A name and a date, at the bottom, in handwriting if that is what is available. The Purdue guide ends on the instruction rather than the figure: “Producers are encouraged to create crop budgets and, in general, improve their record-keeping.” A sheet with no name on it cannot be questioned, because there is nobody to ask which way a line went, and the person who made it will not remember in eleven months.
Consistency is the part that only shows up later. The FAO handbook, facing the same problem when one field carries two crops at once, requires that once a method is chosen “it is important to apply it consistently and ensure transparency for the data user”, and warns that with several ways available to value unpaid family work, “consistency, transparency and regularity in the application of the chosen method” is what makes the estimates comparable across time. On a farm that means the correction is written beside the line that changed, the old sheet is filed rather than replaced, and the next cycle starts from a criterion somebody can read instead of one somebody has to reconstruct.
Declaring a figure, holding a decision against it and rebuilding it with a name attached is what the four functions of management look like when they are applied to one number. Of the two figures, the one that travels is the second, the floor that covers everything, because that is the one a declared price assumption in a three-year budget is built from, projected total costs over expected production, and the one every trigger sits above in a written selling plan. Whichever of the two a farm sends across, it has to be the same one in all three documents, with the same criterion written beside it. If the three carry three numbers, two of them are wrong.
Where to start
Two to three hours, with last cycle’s invoices and the lease on the table.
A floor that later proved too low, written down with its two lists and its date, is worth more than the figure in your head that happened to be right, because only the first one can be argued with by someone who was not in the room when it was set. A public university prints the sentence saying what its own number leaves out, and prints a new number every year. The farm that keeps its floor in memory gives a slightly different answer to every caller, and finds out which answer it gave only once the load has already gone.