Cost of Goods Sold: What Goes In and What Stays Out

Cost of goods sold is what the goods that were actually sold cost to get ready for sale. The figure is built from opening inventory plus purchases less closing inventory, adjusted for anything that was incurred to bring the goods to their present location and condition. Every rupee moved across that line changes gross margin without changing profit by a single rupee, so what the figure excludes matters as much as what it includes.

Here is what sits underneath that. A business that sells a physical thing spends money in two quite different ways, and everything about the subject turns on telling them apart. Some of the money is spent getting the thing into a state where somebody will pay for it. The rest is spent running the place that finds the buyer, raises the invoice, chases the payment and keeps the lights on. Only the first kind attaches to the goods. The second kind belongs to the year, not to the notebook.

The distinction is easiest to see on a scale small enough to hold in the head, well before any statement of profit and loss. A woman sells vada pav from a handcart outside a station. The pav, the potatoes, the besan, the oil and the gas in the cylinder are what one vada pav cost to exist. The municipal licence she pays for once a year, the phone she uses to take bulk orders, and the auto that carries the empty cart home at nine at night are not. Every rupee left the same purse on the same day. Only one set of them can be traced to a vada pav somebody bit into. Cost of goods sold is the first set, counted only for the items that actually left the cart, and the entire difficulty of the subject is that reasonable people draw the boundary between the two sets in different places.

The build comes first, term by term, then the test that sorts a cost never seen before, then the reason an Indian statement of profit and loss carries no line called cost of goods sold and how one is rebuilt from the lines it does carry, and finally what moving a single Rs 5,00,000 cost across the line does to gross, operating and net margin.

What is cost of goods sold, and what is it actually measuring?

Everything in the subject is built on a three-term identity. Stock held at the start of the year, plus stock bought during the year, less stock still held at the end: what is left is what went out of the door. Opening inventory plus purchases less closing inventory. The word doing all the work in the phrase is sold. A business can spend an enormous amount of money in a year on goods that are still sitting in its own godown on the last day of March, and not one rupee of that spending is a cost of anything yet.

The word sold is where most first attempts go wrong, so sit with it for a second. If Anjani Stationers, an invented maker of school notebooks, buys Rs 1,57,50,000 of paper in a year and only cuts and binds part of it, the paper still stacked in the godown has not become a cost. The paper has become an asset. It sits on the balance sheet at what it cost, waiting, and it will become a cost in whichever year it finally leaves as a notebook. Subtracting closing inventory is not a rounding adjustment or a prudence device. The subtraction is the mechanism that stops the profit statement from charging this year for goods this year did not sell.

The three-term identity is the skeleton, and the real build has more bones than that. Goods rarely arrive in a saleable state. Carrying paper from the mill to the godown costs money. Cutting, stitching and binding it costs wages. The line it is bound on needs power, floor space and somebody supervising it. All of those were incurred turning a ream into a notebook, so all of them attach. The complete build therefore runs: opening inventory, plus purchases of material, plus freight inwardThe transport and handling cost of bringing bought goods in to the purchaser's own premises. The buyer pays it as part of what the goods cost to acquire., plus direct labourWages of the people whose hands are on the product. On a binding line, the cutting, stitching and folding staff, not the office., plus production overheadCosts of running the place where the goods are made, such as factory power, factory rent and line supervision, which cannot be traced to one unit but are still part of making it., less closing inventory.

Which costs get in, and what is the test?

Once it is settled that some costs attach to the goods and some do not, a test is needed that works on a cost never seen before. No list will ever be long enough. The test used across accounting practice, and the one written into the inventories standard, is a question about the goods rather than about the money: was this cost incurred to bring the goods to their present location and condition? If the answer is yes, it attaches and it sits above the gross line. If the answer is no, it belongs to the period and it sits below.

The test is always about location and condition, never about who spent the money, how large the amount was, or whether it felt necessary. That is worth saying twice, because three wrong tests are extremely tempting and all three are commonly used by accident. The first wrong test is size: a Rs 5,00,000 cost feels important enough to be a cost of sales and a Rs 8,000 cost feels like an overhead, and the amount has nothing to do with it. The second wrong test is necessity: the accounts clerk is absolutely necessary to the business, and necessity is not the question being asked. The third wrong test is who authorised it. Authorisation says something about the organisation chart and nothing about the notebook.

Work six costs that people genuinely argue about. Freight inwardTransport paid to bring purchased goods in to the buyer's own premises, such as the lorry that carries reams from a paper mill to a godown. of Rs 3,00,000 attaches. Nobody could cut a ream until it had reached the godown. Freight outwardTransport paid to deliver finished goods to a customer. The goods are already saleable before freight outward is paid, making it a cost of selling rather than a cost of making., the Rs 4,00,000 of delivery van running costs at Anjani Stationers, does not. The notebook was already finished and saleable before it was loaded. Storage before conversion is still part of getting the goods into a condition somebody will pay for, so the godown that stores raw paper attaches. Nothing further happens to a notebook while it sits in the Rs 5,00,000 warehouse awaiting despatch, so that warehouse generally does not. The binding line supervisor's hours are spent turning paper into books, so her Rs 2,50,000 attaches. Not one minute of the accounts clerk's time reaches a ream, so his Rs 1,80,000 does not.

The two warehouses are worth pausing on. Same kind of building, same kind of rent, same landlord for all anyone knows, and they fall on opposite sides of the line purely because of what is inside them and what stage the goods have reached. Opposite answers for two near-identical buildings is the test doing its job rather than a rule being applied mechanically, and it is why the test can be taught while a list can never stand in for it.

Why does an Indian statement of profit and loss have no cost of goods sold line?

Here is the discovery that catches almost every reader who learned the concept from an international textbook and then opened an Indian set of accounts looking for it. There is no line called cost of goods sold, and there is not going to be one. The prescribed format for the statement of profit and loss under Schedule III to the Companies Act does not carry that caption at all. The format carries instead, as separate mandatory lines, the cost of materials consumedThe value of raw materials actually used up in production during the year, being opening material stock plus material purchases less closing material stock, a materials figure only., purchases of stock in trade, and changes in inventoriesA single line capturing the movement in finished goods, work in progress and stock in trade across the year. A rise in stock reduces the charge, a fall increases it. of finished goods, work in progressGoods that have been started but not finished on the reporting date, such as sheets already cut but not yet stitched into a notebook. and stock in trade. Employee benefits, depreciation and other expenses are their own separate lines below.

An Indian reader who wants a cost of goods sold figure has to build it, and the building involves at least one judgement. Two competent analysts can therefore produce two different figures from the same published accounts and both be right. Understand why the format is like this rather than treating it as an inconvenience. The prescribed format is built around the nature of a cost, so every rupee of salary appears in employee benefits whether the person was standing on a binding line or sitting in an accounts office. A cost of goods sold caption is built around the function of a cost, and function cuts the same rupees a completely different way. The format chose nature, and once it did, function-based subtotals stopped being available on the face of the statement.

The reconstruction is mechanical up to a point and then it is not. The mechanical part is adding the three inventory-related lines together to get what materials and traded goods cost. The statement does not split employee benefits or other expenses between production and selling and administration, so the judgement part is deciding how much of each was spent on production. Where a business describes its own policy in the notes, that description governs. Where it does not, the split has to be estimated, and the honest thing is to state the estimate as an estimate.

How is Anjani Stationers' own figure built from the published lines?

Now put real quantities on it. Anjani Stationers opened year two holding 10,000 reams of paper carried at Rs 190 a ream, the Rs 19,00,000 that appeared as closing inventory a year earlier. During the year it bought three lots, 30,000 reams at Rs 205, 30,000 reams at Rs 220 and 15,000 reams at Rs 200, together 75,000 reams and Rs 1,57,50,000 of purchases. Opening stock and purchases together make 85,000 reams available at a total cost of Rs 1,76,50,000. At the year end 14,000 reams were still in the godown, valued at Rs 28,00,000, so 71,000 reams were consumed and the cost of that consumption was Rs 1,48,50,000.

Anjani Stationers' published cost of materials consumed of Rs 1,48,50,000 is the three-term identity and nothing more, so the published gross profit of Rs 1,21,50,000 and gross margin of 45.0 per cent are a materials margin rather than a full cost of goods sold margin. A materials margin is not a criticism of the presentation. The prescribed format produces exactly that figure, and every earlier figure in this business's accounts has been drawn on it. But applying the location and condition test properly leaves one item sitting on the wrong side for this purpose: the binding line supervisor's Rs 2,50,000. The format puts that salary inside employee benefits along with every other salary. Added back, the location and condition build gives Rs 1,51,00,000, gross profit of Rs 1,19,00,000, and a gross margin of 44.1 per cent.

Two things in that table deserve a second look. The first is that the closing 14,000 reams are valued at Rs 200 rather than at any of the other three prices. The reams still in the godown are taken to be the newest ones bought. Which reams are treated as sold and which as remaining is the cost formula, a separate subject with its own arithmetic, set out under FIFO and weighted average cost. The second is that the two totals differ by only Rs 2,50,000, 0.9 per cent of revenue, and the gap looks trivially small. Hold on to that feeling. A number that small moves a margin that people compare businesses on.

What happens to every margin when one cost crosses the line?

The classification question earns its trouble the moment one cost crosses the line. Suppose an analyst looking at Anjani Stationers decides that the Rs 5,00,000 spent on the warehouse holding finished notebooks is a cost of getting the goods ready and belongs above the gross line. Moving the warehouse above the line is a defensible view, held by real practitioners, particularly where goods need conditioning or repacking in store. Nothing dishonest is happening. What happens to the numbers?

Cost rises from Rs 1,48,50,000 to Rs 1,53,50,000. Gross profit falls from Rs 1,21,50,000 to Rs 1,16,50,000. Gross margin falls from 45.0 per cent to 43.1 per cent, a drop of 1.9 percentage points. On a business whose gross margin has been the one stable thing in the accounts, a drop of 1.9 points is a very loud movement. And other operating costs fall by exactly Rs 5,00,000, from Rs 26,00,000 to Rs 21,00,000. The rupee did not disappear. It moved. A reclassification moves a cost from one line to another inside the same total and changes nothing whatever about how much money the business made, so earnings before interest and tax stay at Rs 41,50,000, profit before tax stays at Rs 38,00,000 and profit after tax stays at Rs 30,00,000.

The consequence, stated plainly, is the single most useful fact about a gross margin: it can move for reasons that have nothing to do with how the business traded. Gross margin fell 1.9 points here without one extra rupee being spent, without one notebook being sold differently, and without one paise of profit being lost. A gross margin read without knowing what the business put above the line is a number whose movement cannot be attributed. The margins that are immune are worth naming. Every subtotal from operating profit downwards sits below all the lines that were shuffled, and every one of them is untouched.

The readings worth writing down run as follows. The published position is the narrowest defensible one plus freight inward: Rs 1,48,50,000 and 45.0 per cent. Adding the binding supervisor, as the location and condition test requires, gives Rs 1,51,00,000 and 44.1 per cent. Adding the finished goods warehouse on top gives Rs 1,53,50,000 and 43.1 per cent. Pushing every one of the five above the line, the outer edge that nobody sensible would choose, reaches Rs 1,61,80,000 and 40.1 per cent. Pulling all five below, including the freight, reaches Rs 1,45,50,000 and 46.1 per cent. The full spread of defensible and semi-defensible gross margins for one business, one year and one unchanged set of transactions is six full percentage points, from 40.1 to 46.1, and profit after tax is Rs 30,00,000 at every point on that range.

What is cost of goods sold not?

Three things get mistaken for cost of goods sold often enough to be worth naming individually, and Anjani Stationers happens to produce four different figures in year two that show all three confusions at once.

Cost of goods sold is not cash paid to suppliers. Anjani Stationers bought Rs 1,57,50,000 of paper and paid the mill Rs 1,50,50,000. It started the year owing Rs 15,00,000 and ended it owing Rs 22,00,000, so Rs 7,00,000 of the year's purchases had not been settled by the last day of March. Cash paid answers a question about timing and a bank statement. Cost of goods sold answers a question about consumption and has nothing to do with when anybody was paid.

Cost of goods sold is not purchases. Purchases of Rs 1,57,50,000 include the 14,000 reams that were still in the godown at the year end and had never been near a school. Charging purchases to the year would charge this year with next year's notebooks, the exact error the closing inventory subtraction exists to prevent.

And it is not automatically the same as the cost of materials consumed, even though for Anjani Stationers the two happen to coincide at Rs 1,48,50,000. The two figures agree here only because this business holds no finished goods and no work in progress at either year end and because the published line carries materials alone, and in a business with a large finished goods stock or significant production labour they would not agree at all. Treating them as synonyms is a habit that works until the first time it does not, and it fails silently rather than loudly.

Who builds this figure in practice, and what do they do with it?

Step out of the classroom. Three people open the same statement in the same month and none of them is admiring the arithmetic.

A lender uses the cost figure to test whether a stock valuation is credible, an equity analyst rebuilds it so that two businesses can be set side by side, and Vaidehi Rao, Anjani Stationers' finance controller, uses it to explain to her board why gross margin held at 45.0 per cent while every margin below it fell. Watch each of them work. The same number is doing three different jobs. The lender is deciding how much to advance against Rs 28,00,000 of paper in a godown. The cost figure tells him what a ream is carried at, and dividing the closing inventory by the year's consumption tells him how many months of production is sitting there. The months of cover, and not the balance itself, set his advance rate. The equity analyst has two businesses and one comparison to make, and the first thing she does is stop trusting either reported gross margin until she has read what each one puts above the line. If she cannot find out, she moves the comparison down the ladder to operating profit, where the difference has washed out.

Vaidehi Rao's use is the most practical of the three, and it is the one a student can most easily picture. She is not comparing anything. She is being asked why profit fell, and the cost figure is how she proves that the answer is not in the cost of paper. Gross profit rose from Rs 1,08,00,000 to Rs 1,21,50,000 and gross margin held at exactly 45.0 per cent, so the paper, the freight and the binding are all doing what they did last year. Everything that went wrong went wrong below the gross line, and she can say so with a straight face only because she knows precisely which costs are above it and that the same ones were above it last year. Consistency of the build across years is the only thing that makes this year's margin comparable with last year's, and that consistency turns a classification choice from a problem into a tool.

Anjani Stationers Private Limited, Chitra Binding Works Private Limited, Vaidehi Rao, the Sunrise Public School group, Kalyani Paper Products, Girija Stationery Mills, Bhavani Exercise Books, Deodhar Notebooks and Sarita Print Works are invented.
Educational material. Not advice on any investment, tax, budget or market position.