A margin target counts as a standard only when it was written before the cycle started, with the formula beside it, the date it was written, and its own limits on the same sheet. Written afterwards, it is not a target. It is a description of what happened, and a description cannot be missed.
The margin usually turns up at the end, when the accountant closes the year or when somebody adds up the invoices. No figure was written before it, so the figure that turns up gets accepted as the figure that was there to be had. A good year becomes merit and a bad year becomes weather, and both explanations are assembled after the result is known, which is exactly what makes them impossible to get wrong.
The conversation about dropping an activity, renting more ground or leaving a field out next cycle then runs with no declared floor under it, and each person at the table defends the option with whichever figure suits them. Next cycle, the same thing, because nothing was left on paper saying what was intended.
The instrument for this is not new, and it is not two instruments. The Manual de Organización y Gestión de la Empresa Agropecuaria, published for the agricultural schools of Buenos Aires province by its agriculture ministry with content coordinated by INTA, the Argentine national agricultural research institute, puts the point in a footnote: the gross margin can be used to plan, before the fact, and to evaluate results already in, after the fact, and in the first case it serves to budget while in the second it serves to control. The same subtraction, run twice, a cycle apart. That is only true if the subtraction was written down the first time.
Why the figure has to exist before the result does
Because your memory of your own forecast moves toward what happened. Cassar and Craig, in the Journal of Business Venturing in 2009, found it in people who had left a record to be checked against. Working from the Panel Study of Entrepreneurial Dynamics, they took 198 people who had been starting a business, who had stated during the process how likely they thought it was to become an operating business, and who later quit. Asked afterwards to recall the figure they had given, they came in well below what they had actually stated at the time. The recall was scattered, and under the scatter it leaned one way: downward, toward the outcome they now knew.
Those were people starting businesses, and what they were recalling was a likelihood, not a margin. What carries across is narrower than the study and enough for a farm: the memory of a figure you gave yourself drifts toward the result. The authors put the recall question only to the ones who quit, so the work says nothing about how people who succeeded remember their own forecasts. Time did not turn out to be the culprit either, since at least eleven months had passed for everyone in the sample and more months did not make the drift worse.
What happens to the explanation once the result is known
It narrows to one. Roese and Vohs, reviewing the research on how people judge events whose outcome they already know, in Perspectives on Psychological Science in 2012, describe what people do with an outcome they already have: they recall selectively the information that fits it, and they build an account that makes the sequence look like it had to go that way. Two consequences they name are a myopic attention to a single causal understanding of the past, at the cost of other reasonable explanations, and a general overconfidence in the certainty of one’s own judgment.
A farm with no figure written before the cycle has nothing that resists that. Comparing a declared standard against what actually happened is the control function, the fourth of the four functions of management, and it needs two numbers. One of them has to have been written while the outcome was still open.
What the formula on the sheet actually says
Income from the activity, minus the costs charged directly to that activity. The Argentine manual states it as the global gross margin, calculated by subtracting the direct costs of the activities from the total net income of the activities. FAO’s farm management extension guide, written by David Kahan and published in Rome, arrives at the same subtraction in different words: the gross margin of a crop or livestock product is obtained by subtracting the variable costs from its value of production, and profit is the total gross margin of all enterprises minus the total fixed costs.
Two documents, two continents, one subtraction, and no agreement on what to call the lines being subtracted. One says direct costs, the other says variable costs, and the Irish tables split variable from fixed and price them apart. Nothing rides on which word a farm picks. Everything rides on which lines it puts underneath, because two people working from the same books and two different lists produce two different margins, and neither can show the other is wrong.
| Source | What it describes | What it subtracts from income | What it calls those lines |
|---|---|---|---|
| Manual de Organización y Gestión de la Empresa Agropecuaria, Buenos Aires province | Teaching manual for agricultural schools, content coordinated by INTA | The costs charged directly to each activity, taken from the net income of the activities | Direct costs |
| Economics for Farm Management Extension, FAO | Guide written for extension work, not tied to one country | The variable costs, taken from the value of production | Variable costs |
| Crops Costs and Returns 2024, Teagasc | Irish tillage crops, in euro per hectare | Variable costs, with fixed or overhead costs listed apart | Variable costs, and fixed or overhead costs |
Which lines are yours to subtract, and why no table hands them to you
Because the costs sitting outside the direct list belong to the individual farm. Crops Costs and Returns 2024, compiled by Ciaran Collins and Shay Phelan for the Crops, Environment and Land-Use Programme at Teagasc in Oak Park, says so about its own tables: fixed costs are largely unique to each individual farm, and all farmers should calculate their own costs rather than using standard industry figures.
It prints the figures anyway, in euro per hectare: an average for specialized tillage farms out of the Teagasc National Farm Survey, and beside it a much wider spread out of eProfit Monitor results that it puts down to each individual situation. So the document that has the number is the one telling the reader not to lift it, and the heading it sits under leaves out interest, machinery and land rental besides. Anyone who wants to see the Irish figures can open the table. What they are not is the number that belongs on your sheet.
The publication also declares what it is, in its own opening line: an indicative guide to crop margins, with land suitability, rotation, risk avoidance and husbandry skills still to be weighed by whoever reads it. A document that prices margins for a whole country refuses to hand any single farm its figure. A sheet written on one farm, for one activity, has less reason to borrow one. One of those lines has no invoice anywhere to point at, which is exactly why it is the one that gets left out: what the office costs the farm builds that total out of twelve months of statements, before it can be subtracted from anything.
What the number does not show, written on the same sheet
Almost everything the farm owns. The Argentine manual sets out the limits in the paragraph right after the formula, and they are not small: the gross margin is a tool of exclusively economic analysis, so it does not take financial matters into account, and it does not examine how the business works as a production system, because it evaluates activities in isolation. Then comes the line that earns its place on a farm sheet: the gross margin is not showing you all the goods, resources and people that went into that production.
FAO’s guide reaches the same gap from the labor side. Family labor is an important input for most farmers, and most of all where the system is only partly mechanized or not mechanized at all; the gross margin calculation already carries hired labor inside the variable costs and nothing for the rest; and comparing two activities without putting a cost on family labor is not a comparison. One line on the sheet, in the farm’s own words, saying what the figure leaves out. A number with its limits written next to it can be argued with. A number without them gets read as the result of the year, which it never was.
Why the year is not the explanation you are looking for
Because farms in the same year land in different places. Wilson’s decomposition of dairy profitability, in The Journal of Agricultural Science in 2011, works from 228 dairy enterprises in England in one accounting year, 2007/08, drawn from the Farm Business Survey. The average economic return over the whole sample was 52 British pounds per cow. The quarter with the best net margin per cow returned 335 pounds per cow, and the quarter with the worst lost 361 pounds per cow, a spread near 700 pounds per cow inside a single accounting year, one country, one sector. Wilson reports total fixed costs per cow as almost identical between the top and the bottom quarter.
These are English dairy figures from one survey year, and they are not a target for anybody. The target on your sheet is the one you write, in your unit, for your activity. What the spread does rule out is the explanation that the year decided the result, because the accounting year was the same for all 228 of them and the results were not. Comparing your margin against other farms is a separate exercise with separate rules, and this sheet does not do it: the only thing this margin gets compared against is what this farm said it intended.
The number you write, and the floor it stands on
A target margin needs a floor underneath it, or it is a wish with a figure attached. That floor is the price below which the activity does not cover what it costs to run, and it is worked out in the break-even price, which is the piece that belongs beside this one in the marketing axis. The margin target sits above that floor by an amount the farm decides and nobody else can.
Write it in the shape the rest of the farm already uses for objectives, which is a goal with a number and a deadline: a quantity, a unit, a date. The unit matters more than it looks, because it has to be the unit you will actually be able to compute at the end, per hectare, per head or per tonne.
The income half of the formula carries a decision that the unit alone does not settle: FAO’s guide counts the value of production wider than the sale, because the total value of production includes produce sold, produce consumed by the farmer’s family and produce stored, so a farm holding product in the shed has to say on the sheet whether that product is inside the figure or outside it, and say it the same way at both ends of the cycle.
And the figure has to survive contact with the rest of the paperwork: if the farm already carries a declared price assumption in a medium-term budget, and the margin target implies a different price, one of the two documents is wrong and this is the cheap moment to find out which.
The back column, and the date somebody puts on the calendar
The sheet is half a document until the realized margin sits beside the target. The income half of that calculation depends on the average price the farm actually got, which is what the sales against the plan works out, and the cost half depends on the same lines you named at the start and no others. Substituting a different list at the end is how a farm ends up comparing two things that were never comparable.
Set the date now, with a name against it. Without a scheduled review date and somebody who convenes it, the back column stays empty and the whole cycle is a page in a drawer. On that date, the realized figure goes in, and under it one short line saying what explains the difference.
Roese and Vohs review one way of pushing back that has held up in testing, which is to consider and explain how outcomes that did not occur could well have occurred, and they put a cap on it: it is most effective if restricted to consideration of no more than two or three alternative explanations, because once generating more of them starts to feel difficult, the difficulty can get taken as a sign that they were implausible, and that hardens the first account instead of loosening it. Two or three lines, and then the correction, or the documented decision not to correct, and the sheet goes into the folder where next cycle’s sheet gets written.
Where to start
A declared margin target is the only figure on the farm that can be shown to have been wrong. What comes out at the end is never wrong; it is whatever happened, and a number that cannot be wrong teaches nobody anything. That is the whole trade. An afternoon of writing buys the farm the right to be contradicted by its own books, and being contradicted by your own books is the only thing that makes a second cycle different from a repeat of the first.