Waiting to sell pays only if the price you get on the sale date clears the price to beat. That price is today’s price plus storage and interest for the months held, divided by the share of the lot left after weight loss (0.985 when 1.5% is lost). You can work it out before deciding, from five figures the farm fills in itself: today’s cash price, the months you plan to wait, the storage charge per month, the interest rate per month and the percentage of weight lost. The sixth figure, the price you expect, is the only one that depends on where the market goes.
Written on a one-page decision sheet for the lot, with a date to reopen it, that number ends the weekly argument about whether to sell. Without it, the lot sits because the last offer seemed low, the question comes back every week, and the cost of waiting keeps running the whole time.
Take a lot worth 1,200 per tonne today, in the farm’s own currency, held five months at a storage charge of 6 per tonne a month, with interest at 1.2% a month and 1.5% of the weight lost. Holding it costs 121.20 per tonne: 30 of storage, 72 of interest and 19.20 of lost weight, valued at the sale-date price of 1,280 rather than today’s 1,200. A sale-date price of 1,280 looks like a gain of 80, a rise of 6.67%, yet once the 121.20 comes off, it ends 41.20 per tonne worse than selling today. The price to beat is 1,321.83, a rise of 10.15%, and working it out takes no forecast at all.
Rurivia calculation from a hypothetical lot of 1,200 per tonne, held five months at 6 per tonne a month in storage, 1.2% a month in interest and 1.5% of the weight lost.
Why does interest belong in the account even when you owe the bank nothing?
Money you did not collect at harvest is either paying down a loan or sitting where it could earn something, and either way it has a rate. Janzen, writing for the Department of Agricultural and Consumer Economics at the University of Illinois in September 2024, splits the cost of holding grain into “physical costs related to the handling, maintenance, and insurance of physical crop inventories, and the opportunity cost of giving up revenue from grain sales at harvest in favor of a higher price later”. The second half of that sentence is the line many farms never subtract, and it is usually the larger one.
Which rate to use depends on what the money would otherwise be doing. Janzen prices the same decision at two rates: 4.5%, “roughly equal to the current return on high-yielding savings accounts”, and 8.5%, “or roughly the current interest rate on new farm operating loans”. A farm carrying an operating loan is paying the loan rate to hold its own grain, so that is the rate to use.
Run Janzen’s two rates on the example lot, in place of its 1.2% a month, and the answer changes sides. At 4.5% a year, which is 0.375% a month, the cost of holding falls to 71.70 per tonne and waiting ends 8.30 better than selling today. At 8.5% a year, which is 0.7083% a month, the cost is 91.70 and waiting ends 11.70 worse. Same grain, same storage, same five months, same expected price of 1,280: only the interest rate moved, and many storage calculations leave it out.
Hurt, in the Purdue Agricultural Economics Report of June 2019, also charges interest when the grain sits in the farm’s own bins: “for on-farm storage, only weekly interest costs are subtracted as a cost of storage”. He counts it by the week, this article by the month. A farm with its own bins gets no storage bill and still pays that interest. For a farm that does get a bill, the bill covers under a third of what one more month costs: of the 20.40 each extra month adds on the example lot, 14.40 is interest and 6.00 is storage. Lost weight is not in that 20.40, because you enter it once, for the whole wait.
Which number on the decision sheet needs no forecast?
The price to beat comes from the five figures you already know, and it goes at the top of the decision sheet. It is not the break-even price, which another article in this library works out against what the crop cost to grow; the price to beat is measured against selling today.
Writing it down turns an argument into a comparison. The question stops being whether the market feels firm and becomes whether the bid on the table is above or below one written number, which a phone call answers in ten seconds. The sheet also carries the date it will be reopened and the name of the person who may reopen it, so the lot is not re-decided every Tuesday by whoever looked at a screen.
The price to beat also moves with the length of the wait. Two farms can hold the same lot at the same expected price and land on opposite sides of the answer because one planned three months and the other nine.
There is also a month past which waiting stops paying, and it is the second number to write on the sheet. The gross gain is 80. The loss takes 19.20 of it, since the calculation enters the loss once, the same for any length of wait. That leaves 60.80 to cover the 20.40 each month costs, and 60.80 divided by 20.40 is 2.98 months. At an expected price of 1,280, a three-month wait already ends 0.40 per tonne worse than selling today, and each month beyond adds 20.40. One number says how far the price has to rise; the other says how long you can wait for it.
Where does the loss go, and at which price is it valued?
The weight that goes missing would have been sold on the sale date, so this article values it at the sale price, not at today’s. The choice changes the numbers. At today’s price, the example lot costs 120.00 to hold and the price to beat is 1,320.00. At the sale price, it costs 121.20 if sold at 1,280, and the price to beat is 1,321.83, because there the loss is valued at that price itself. At a 5% loss, today’s price gives 162 in cost and 1,362.00 to beat; the sale price, 166 and 1,370.53. Today’s lower price would make waiting look cheaper, and waiting is what the sheet is testing, so the article uses the stricter figure.
Loss here means weight that disappears, not grain that drops a grade. Kumar and Kalita, reviewing grain storage in developing countries for the journal Foods in 2017, report insect weight loss in maize after six months in traditional granaries in Togo estimated at 0.2% to 11.8%: not your bins, but a sign that a borrowed percentage can be far off.
So the figure to enter is the weight that went in minus the weight that came out, from the farm’s own scale tickets for that lot. A grade discount at the elevator on the way out also costs money, and the calculation above leaves it out: write it on the sheet as a line of its own, under the price to beat. Janzen’s costing has no loss line at all. This article adds one because grain does lose weight in storage, and says at which price it values that loss so you can make a different choice.
Why does the market pay you to store, and how does the payment show up in the basis?
Buyers who need grain later in the year, rather than at harvest, pay for someone to hold it, and you can see that payment before deciding. Basis is the local cash price minus the futures price. Dhuyvetter, writing for the Kansas State University extension service in 1992, points at where the payment shows: “The improvement in basis from harvest until the contract expires in July represents the market’s payment for storing soybeans”. Measuring basis against a later futures contract, not the nearest one, is how you see, in his words, “if the market is offering returns to storage”.
The signal works both ways. When basis is stronger or narrower than usual, “the market is providing a financial incentive to make cash sales”; when it is wide or weak, the same mechanism favors storage. A wide basis at harvest is an offer to pay you for the months, and you can set that offer against what a month of holding costs your lot, 20.40 in the example.
Basis is also easier to predict than price. Dhuyvetter reports two Kansas wheat years that were $1.42 a bushel apart in price and only $0.17 apart in basis, and concludes that “basis levels tend to be more predictable than cash or futures prices”. A farm that keeps a price record with a source, a dated list of the bids it was offered and where each came from, can build a basis column worth using within a year or two.
That column gives the most solid way to set the expected price: the futures price for the month you plan to sell, plus the basis you expect for that month, taken from your own record. It can still miss, but it misses on top of a number the farm measured, not on top of a feeling about the market.
Why are holding a priced lot and holding an unpriced lot different decisions?
Holding a lot whose later sale is already priced and holding a lot with no price look the same from the road, but they are different decisions. The difference in price between two futures delivery months is called the spread, and Janzen writes that futures spreads “provide a guaranteed return for those willing to hold the commodity to that later date”. The market calls that return the carry, “since it represents the return to the act of ‘carrying’ or storing the grain over time”.
The return is guaranteed only for whoever locks it in, which the trade sums up as “you can’t capture the carry unless you sell the carry”: you collect it only by selling for the later date now. Holding grain with no price attached “is simply speculating on the prospect that the commodity will be worth more in the future than the current price”. The market’s payment for storing can show up in two places, the carry between futures months and the local basis, and selling a later futures month locks the carry while the basis stays open until delivery.
The 121.20 is the same either way. Janzen is explicit that “these costs of storage are incurred whether the commodity holder captures the carry or is speculating on unpriced inventory”. A farm paying the full cost of holding while leaving the lot unpriced pays for the position and leaves its return uncollected.
So write on the sheet which of the three cases the lot is in: priced forward, hedged with futures, or unpriced. Each leads to a different next question. For a lot sold forward, the question is whether the contract price beats a market you cannot see yet; the article is locking the price now worth it makes that comparison. For a futures position, the article what price you keep after hedging works out the price you keep after the basis settles at delivery. If nothing is priced, holding is a deliberate bet on the price: write that down, with how much of the crop is already priced, a count the article how much is already priced shows how to make.
Why is the expected price the weakest figure?
The expected price usually comes from how prices moved in past years, and that pattern can shift. In Brazil, Souza, Silveira and Ballini found in 2023 that, as second-crop maize planted after soybeans grew, the months of lowest maize prices moved from January through April in 1996-2001 to July through September in 2011-2019; your market differs, but a remembered pattern can go stale the same way.
None of this makes the expected price useless. It makes it the figure to hold most loosely: the price to beat goes on the sheet in ink, the expected price in pencil.
What do the other five figures hide?
The other five figures carry choices too, and three of them are worth writing down beside the price to beat.
Storage goes in as a monthly charge, and many tariffs also have a one-off intake fee. How to work out the monthly figure, for a commercial warehouse or for your own bins, is covered in what a month of storage costs, the library’s article on the cost of storage line by line. One point belongs here: an intake fee already paid is gone whatever you decide, so it stays out for a lot already in store. For a lot still on the truck it goes in once, outside the monthly charge, and leaving it out makes waiting look cheaper than it is.
In the calculator the loss does not change when you change the months; in the bin, it does. The loss you enter in the calculator covers the whole period. If you compare five months with nine, storage and interest grow and the loss stays put, so estimate the loss again for the longer wait instead of reusing one percentage.
Interest is simple and runs on today’s price. It does not compound. At 1.2% a month over nine months, compounding would add about half a percentage point of the lot’s price: small over five months, not negligible over a whole marketing year. It is stated here so you can correct for it yourself.
What changes on each side of the result?
If waiting comes out worse than selling today, the lot calls for either a sale or a written reason other than price. There are sound reasons to hold anyway: a harvest truck line at the elevator too long to join now, a delivery window already committed, cash the farm does not need yet. Janzen also notes that “profitable storage generally happens when the firm has available capacity both in terms of its physical space and financial ability to defer revenue”. Holding because nobody wrote anything down is not one of those reasons. A named reason can be checked in three months; an unnamed one leaves the lot in the bin next year with the same excuse.
A negative result can also be changed without the market, and interest, the largest line, is where to look. While an operating loan is still open, money from selling this lot would go to pay it down, so holding the lot costs the loan rate. Paying off only part of the loan does not change that, because the rest of the balance still charges the same rate.
Once the loan is fully paid with other money, such as another lot or other income, cash from this lot would sit in a deposit instead, and the rate drops from the loan rate to the deposit rate. With Janzen’s two rates, 8.5% and 4.5% a year, in place of the example’s 1.2% a month, that cuts 20 per tonne from the cost of holding the example lot and turns 11.70 worse than selling today into 8.30 better. If it takes this lot’s own sale to clear the loan, there is no lot left to hold at the lower rate.
The storage charge is a price someone quoted, and a quoted price can be negotiated when the contract comes up for renewal, so note that date. Paying off the loan and renegotiating the storage charge both change the number directly, which is worth more than another month of hoping the market will.
A positive result raises one question: is the return being collected or only hoped for? If waiting pays on paper, the way to collect it is to price the later sale now, since the carry is paid for committing and is not a forecast that prices will rise. If the farm has a written selling plan, the document where the trigger and the buyer for each part of the crop are set before the first sale, holding a lot past the date in that plan is a new decision, to be made on purpose.
Hurt’s figures on past corn storage returns add that the time you price matters too, and that it differs by who owns the storage. He found “the best time to price out of commercial storage was in late-February and early-March”, and for on-farm storage “the highest returns came from pricing in May and early-June”. In his conclusions he qualifies the first finding, setting late February against May for commercial storage: “these two windows were roughly equivalent for commercial storage while May and early-June was superior for on-farm storage”.
Where to start
Allow an hour for the first lot, with the operating loan statement, the warehouse tariff and the scale tickets at hand. The sell now or store calculator does the arithmetic and prints the decision sheet, and the steps below follow it.
File the sheet with the farm’s other selling records; more articles on selling grain are in the marketing section of the library. With the sheet in hand, the question is no longer whether the price is good but whether the bid clears the price to beat (1,321.83 in the example), and it gets asked on the date the sheet names, not every morning. Deciding once, on paper, and checking the decision on a set date is one concrete piece of running a farm as a business.