Why a technical wholesaler pays €98,000 every year for €396,000 of stock that has not moved in twelve months

The managing director of a technical wholesaler with 35 employees walks along with the annual stock count in January. The warehouse holds 12,000 stocked item numbers on 2,800 square metres, together €1.8 million in purchase value. By the racks at the back stands a pallet of fasteners in a coating the market has not asked for in two years, a row of boxes from a discontinued range of light fittings, two pallets in a packaging unit the manufacturer has replaced, and half a rack of size variants once bought in for a single customer. The warehouse manager says what he says every year: we have plenty of space, and it has already been paid for.

In the annual accounts that stock is carried at cost. The accountant has made no comment on it and the bank counts it towards its security. The company's gross margin is 32 percent, which counts as good for this wholesaler. The director looks for the reason profit is lagging in the sales department and in the higher interest rate, and in doing so looks past the answer. Stock that no longer sells sits on the balance sheet as an asset and behaves in the profit and loss account as a cost of around a quarter of its purchase value per year. Those costs are there, on the lines for interest, premises, insurance and warehouse staff, and that is why nobody sees them as the cost of that one pallet at the back.

Why stock that stands still costs money

Gurtu (2021) describes in the Journal of Risk and Financial Management (peer reviewed) that the literature usually assumes the cost of holding stock as a single percentage of item value, citing Waters (2003), who assumes about 20 percent per year, with a range of 16 to 20 percent among other authors. His finding is that this assumption does not hold for a broad range: a business with many different items has to calculate holding costs per item, because price, weight, volume and therefore storage costs per item differ too much for one average percentage. Berling (2008) had already reached a similar conclusion in the International Journal of Production Economics (peer reviewed), using an activity-based costing approach: the percentage of value rests on the idea that capital costs are the largest part, recent research suggests this is not necessarily the case, and there are situations in which the traditional method produces a cost increase of more than 15 percent.

For a wholesaler with a broad range, the difference lies in the space. Interest and insurance move with the value of an item. Warehouse rent moves with the number of locations, and an item that has not moved in twelve months occupies its location just as long as an item that is picked every week. If you allocate rent by value, you shift part of the rent of the dead stock onto the fast movers. At technical wholesalers we see that slow and dead items together occupy more than half of the locations, against a third of the stock value.

Obsolescence comes on top: a screw in a coating the market no longer asks for loses value every year, and so does a light fitting from a discontinued range. That loss appears nowhere until someone writes the stock down, and as long as that does not happen, the business pays tax on a stock value that no longer exists. How that mechanism works with rising purchase prices is explained in Why a hardware reseller pays tax on profit it does not have when memory prices rise; with falling market value it works the other way, and that is the subject of the section on the annual accounts below.

Why the owner leaves it where it is

Arkes and Blumer (1985) showed in Organizational Behavior and Human Decision Processes (peer reviewed) that people are more inclined to continue something once money has been put into it, and the same applies to effort or time, and that the justification for this is the wish not to appear wasteful. Those who had invested rated the chance of success of the same project higher than those who had not, and a previous economics course did not help. Ronayne, Sgroi and Tuckwell (2021) found in the Journal of Economic Behavior & Organization (peer reviewed), in an experiment, that 23 percent of participants held on to a lottery they had earned themselves while a demonstrably better lottery was there for the taking; the sense of ownership explained only about a third of this.

For the director, the pallet at the back is such a lottery. €60,000 was paid for it, and selling it to a stock buyer for €24,000 feels like throwing away €36,000. That the €60,000 is already gone, and that the only question is whether €24,000 now is worth more than what the pallet brings in over the coming years after deducting what it costs, is a calculation we rarely see made in a warehouse. At trading companies we see that the decision on dead stock is postponed every year at the stock count, with the same reason: there is plenty of space. At companies that have just extended their warehouse, that argument is strongest and the dead stock largest.

Worked example: €1.8 million of stock, of which €396,000 does not move

In this worked example, the wholesaler has revenue of €8.0 million and a gross margin of 32 percent, so a cost of sales of €5.44 million. The €1.8 million of stock therefore turns 3.0 times a year, and the company holds 121 days of stock on average. Of the 12,000 item numbers, 3,240 have had no sales line in the last twelve months. That is 27 percent of the item numbers and €396,000 in purchase value, 22 percent of the stock.

The overdraft facility is fully drawn. The bank charges 3-month Euribor plus a margin of 2.9 percentage points; 3-month Euribor stood at 2.62 percent on 18 September 2026 (market data), so the company pays 5.5 percent, rounded. With rent, energy, racking and maintenance, the warehouse costs €60 per square metre per year, €168,000. The company estimates insurance, counting, moving and the annual hauling around at stocktaking at 1.5 percent of value. For obsolescence, the example uses 6 percent per year on the dead stock: the part of the purchase value that disappears each year because an item leaves the manufacturer's range, a standard changes, the packaging unit is replaced or nobody is interested in the coating any more.

In this example the dead stock takes up 28 percent of the locations against 22 percent of the value, because each of the 3,240 item numbers occupies at least one location, whether it holds €2 or €200 per unit. That makes space the largest item, larger than interest, and at the wholesalers we see that is the rule. Of the €98,520, €74,760 is in the company's annual accounts, spread over interest, premises, insurance and wages, and reduces profit there by the same amount. The obsolescence of €23,760 appears nowhere, and it only shows up when the director decides to write down or sell.

In this example the wholesaler makes €380,000 profit before tax on €8.0 million of revenue, and a quarter of that goes on stock that has not generated a single euro of revenue in the past year.

What the average averages away

The company in the example uses 20 percent holding costs for its whole stock, the average Gurtu (2021) takes from the literature, and arrives at €360,000 a year. Split into four classes by turnover, the same stock looks different. Class A are the fast movers, B the regular stock, C the slow items with less than one turn per year, and D the items with no sales in twelve months.

Interest per class is 5.5 percent of value and insurance 1.5 percent; space is the share of locations times €168,000; obsolescence per class is 1, 3, 4 and 6 percent of value. At the bottom, the company arrives at 19.3 percent, close to the 20 percent from the literature. Per class it runs from 15.0 to 24.9 percent. Class A costs €108,000 a year and delivers €4.18 million of cost of sales; against the €98,520 of class D there is no revenue at all. Classes C and D together occupy 45 percent of the locations with 34 percent of the value, and carry €150,840 of the €347,640 in costs.

Teunter, Babai and Syntetos (2010) showed in Production and Operations Management (peer reviewed), on three large real-world datasets, that the usual ABC classification, by revenue value or volume with a fixed service level per class, is far from the lowest cost, and that a criterion that also weighs the criticality of an item performs better. A classification by last year's revenue automatically puts class D at the bottom of the list and leaves open how much money is in it. For that you need the columns stock in units, purchase value, last sale date and months of cover, and those four columns are available in the ERP systems we come across at trading companies.

Clear or keep: the calculation per batch

The pallet of fasteners at the back has a purchase value of €60,000 and fourteen months without a sales line. A stock buyer offers 40 percent, €24,000, now. The sales department thinks the batch will be gone within three years at a promotional price of 65 percent of purchase value, a third each year.

Keeping it then yields €13,000 a year for three years, together €39,000. Meanwhile the holding costs continue at the 24.9 percent from the first table, on stock that falls from €60,000 to zero: on average €50,000 in the first year, 30,000 in the second and 10,000 in the third, so 12,450 plus 7,470 plus 2,490 is €22,410. Net, the company keeps 39,000 minus 22,410 is €16,590 from keeping it, spread over three years. The stock buyer pays €24,000 within thirty days. The difference is €7,410 in favour of selling now, and that does not yet account for the chance that the batch is not gone in three years.

The break-even point: keeping only wins if the stock buyer offers less than 16,590 divided by 60,000 is 27.7 percent, and only if the assumption of a third per year holds. In the calculations we make with clients, that assumption is the weakest part. Van Jaarsveld and Dekker (2011) developed in the International Journal of Production Economics (peer reviewed) a method to estimate the risk of obsolescence per group of items from what happened to comparable groups in the past, based on demand data alone, without an expert's judgement. The do-it-yourself version of this is in step 4 below: look at what happened to the items that had also sat still for twelve months two years ago. At the wholesalers we see, less than half of those have been sold after two years, and most of that at a discount.

What the annual accounts and the tax authorities allow

The annual accounts set requirements for the value at which that pallet is carried on the balance sheet. Article 2:387 of the Dutch Civil Code (Burgerlijk Wetboek) (statutory text) requires in paragraph 2 that current assets are valued at current value if this is lower than the acquisition price at the balance sheet date, and in paragraph 1 that impairments are recognised regardless of the result for the financial year. A batch that cost €60,000 and for which a stock buyer offers €24,000 therefore does not belong on the balance sheet at €60,000, whether or not it was a good year.

The Dutch Tax Administration (government source) sets the bar differently for tax and writes that you normally value stock at cost price, that you may only take a lower value for obsolete goods, and that you may then report the difference between cost price and that lower value as a cost. The Tax Administration describes obsolete (incourant) as difficult to trade, and fixed percentages apply to textiles, clothing, furniture and upholstery. No fixed percentage applies to a technical wholesaler; the valuation falls under sound business practice (goed koopmansgebruik) with a consistent practice (Article 3.25 Income Tax Act 2001), so if you start writing down this year, you do it the same way next year.

In the worked example, a write-down from €60,000 to €24,000 reduces taxable profit by €36,000. In 2026, corporate income tax (vennootschapsbelasting, VPB) has two rates, 19 percent up to €200,000 profit and 25.8 percent above that; with a profit of €380,000 the write-down falls in the high rate, so corporate income tax for that year comes out €9,288 lower. That benefit is temporary: if the company later sells the batch for €39,000 after all, taxable profit in those years is higher. The write-down shifts tax from now to later, and for a company with a fully drawn overdraft that deferral saves 5.5 percent interest on €9,288 each year.

We do not give tax advice or legal advice; what is written above is what the law and the Dutch Tax Administration state. Ask your tax adviser and your accountant what evidence they need to classify a product group as obsolete, whether a systematic write-down per class (for example a percentage per month without sales) is accepted as a consistent practice, how the write-down in the annual accounts and in the tax return may differ from each other, and what happens to the write-down if part of the batch is later sold at cost price after all.

What is acceptable, and when

We know of no warehouse without dead stock. A wholesaler that keeps everything in stock and delivers the same day buys items for that purpose that sometimes wait a year, and a supplier with a minimum of 500 units ensures that an item that sells 80 times a year produces six years of stock.

The moment that yields the most is the purchase order, because that is where the €396,000 in the example came from. The 8 percent limit for class D is our own experience at technical wholesalers with a broad range; if you import own brands by the container, you are structurally above it, and the limit is then the trend and the amount per quarter.

Babai, Dallery, Boubaker and Kalai (2019) describe in the International Journal of Production Economics (peer reviewed) that the common forecasting methods for intermittent demand, such as Croston's, do not adjust the forecast downwards in periods without demand when an item becomes obsolete; their adapted method performed better in a simulation and on two datasets, from defence and the automotive industry, in many cases with obsolescence. The ordering system therefore works against you: the order proposal keeps proposing to reorder class C, even when the last sale was eight months ago. At companies where the buyer goes through the order proposal line by line, we see that class D consists mainly of items that were reordered two or three times after demand had already dropped away.

How to map the dead stock in your own warehouse

The rigorous method follows Gurtu (2021) and Berling (2008): a separate cost rate per item, built from interest on the purchase value, the actual location costs, handling per pick and an obsolescence percentage per product group that comes from your own history, as Van Jaarsveld and Dekker (2011) estimate per group. With that comes a monthly report that shows per class what went in and out, and a decision rule per batch. Setting that up in the ERP system and keeping it going is work we do for our clients, and so it is also how we earn our money.

The do-it-yourself version takes an afternoon with an export from the ERP system and gives you the first table of this article for your own warehouse.

Step 1: export per item number the stock in units, the purchase price, the sales in units over the last twelve months and the date of the last sales line. Multiply stock by purchase price; that is the purchase value per item. In the example: 12,000 lines, together €1.8 million. If you use the purchase price on the supplier's invoice, you calculate too low; how freight, import duty, currency and receiving belong in that price is explained in Why an importing wholesaler sees a 39 percent margin on an item that yields 25 percent after freight, import duty and currency.

Step 2: calculate the months of cover per item: stock divided by sales per month. Divide into four classes: cover under 3 months, 3 to 9 months, more than 9 months with some sales in the past year, and no sales in twelve months. Count the number of items and the purchase value per class. In the example, class D comes to 3,240 items and €396,000.

Step 3: fill in the four cost items. Interest: the rate on your overdraft facility times the purchase value per class. Space: count the locations per class and allocate the total warehouse costs over them. Insurance and handling: your insurance rate plus an estimate of the hours for counting and moving. Obsolescence: start with 1, 3, 4 and 6 percent for the four classes, as in the example, and replace them as soon as step 4 gives you your own figures.

Step 4: take the list of items that had no sales in twelve months two years ago, and look at what has happened to them since: what share was sold, at what price, what share was disposed of or returned, and what share is still there. That percentage is your own obsolescence rate for class D, and the proceeds as a percentage of purchase value is the figure you enter in the calculation per batch. If you do not have that list from two years ago, keep today's; in a year you will have your first data point.

Step 5: sort class D by purchase value and, for the top twenty batches, make the calculation from the section above: a stock buyer's offer or the return value at the supplier now, against the expected proceeds later minus the holding costs on the remaining stock. In the example, the twenty largest batches cover well over half of the €396,000.

Once you have done this, you know what the dead stock costs per year, and for the largest batches you have a figure that lets you take the decision at the next stock count in fifteen minutes. The monthly report per class, which shows which items are on their way to class D, comes after that.

When this worked example does not apply

The model assumes a wholesaler with a broad range, its own warehouse that is full, an overdraft facility that is in use and obsolescence of 6 percent per year on class D.

A company with an empty warehouse owned by the director-major shareholder (DGA) has no rent that changes when class D disappears. The space costs are still there, as depreciation, energy, maintenance and property tax, but they do not fall when part of the 3,240 locations becomes free. In the example, the largest item of €47,040 then largely disappears, leaving €51,480, 13 percent. That changes the calculation per batch: holding costs over three years drop from 22,410 to €11,700, and keeping then yields €27,300 against the offer of €24,000. As soon as the company grows and needs the freed-up space for items that do sell, the cost item returns.

For a company without bank debt, the interest rate is what the money would have earned elsewhere. For a DGA who puts it in a savings account, that is less than 5.5 percent, and it becomes more as soon as the company postpones an investment because the money is tied up in stock.

A service company that keeps parts for machines running at customers' sites has class D on purpose. A part that is needed once every five years and whose absence stops a customer for a day costs 24.9 percent a year and is worth it. There the question is which part of class D is deliberate and which is accidental, and Van Jaarsveld and Dekker (2011) wrote their method for that situation, with service parts as the case.

Obsolescence of 6 percent per year is on the high side for steel fasteners; electronics, lighting, battery tools and anything covered by a standard or a directive become obsolete faster. A packaging wholesaler that saw a directive make an entire product group unsellable in one go uses 100 percent for that group.

How a wholesaler brings down the cost of dead stock

Calculate the holding costs per class using the five steps above, and put the class D amount in the monthly report next to the gross margin. In the example that is €98,520 a year on €380,000 profit before tax.

Every purchase order should show the months of cover on screen, and an order proposal for an item whose last sale was more than six months ago does not go through without someone having looked at it. Babai and colleagues (2019) show why the system does not do this by itself.

Make the list of items without sales in twelve months every quarter, and decide per batch. In the example, the calculation for the €60,000 pallet is: €24,000 now against €16,590 over three years, a difference of €7,410.

Arkes and Blumer (1985) described why whoever made the investment rates the chance of a sale higher; so have the decision on the largest batches taken by someone other than the person who bought them. At wholesalers we see that a list assessed by the buyer stays the same length every quarter.

At the wholesalers we see, the supplier's returns arrangement is the cheapest way out and the least used: most suppliers have a take-back scheme for current items at a percentage of the purchase price, and that scheme often expires after twelve or eighteen months. If you make the list every quarter, you meet that deadline; at the annual stock count it has usually already passed.

Write down class D in the annual accounts under Article 2:387 of the Dutch Civil Code and discuss with your tax adviser what evidence the tax write-down needs. In the example that saves €9,288 corporate income tax in the year of the write-down, and the questions for this are in the section on the tax authorities.

Include the freed-up space and the freed-up money in the cash forecast. In the example, clearing the twenty largest batches yields well over €200,000 of purchase value in sales or returns; at 30 to 40 percent that is €60,000 to €80,000 in cash, and the locations of those batches become available for items that do sell. What interest on working capital costs a wholesaler at the current rate is explained in What the ECB rate rise costs a technical wholesaler that lets its customers pay in 52 days. Which part of your range carries the result and which part eats it is covered in The Pareto principle is also a risk gauge for your business, and why a growing trading company runs out of money while it makes a profit in Why a profitable SME still gets stuck on cash flow.

Frequently asked questions about dead stock and obsolete stock at a wholesaler

What does it cost to hold stock?

The literature cited by Gurtu (2021) assumes roughly 20 percent of the purchase value per year, with a range of 16 to 20 percent. In the worked example, a technical wholesaler arrives at 19.3 percent over its total stock, made up of 5.5 percent interest, the warehouse costs allocated by location, 1.5 percent insurance and handling, and obsolescence of 1 to 6 percent per class. Per class it ranges from 15.0 percent on the fast movers to 24.9 percent on the items that have not moved in twelve months.

How do I calculate the cost of dead stock?

Add up the purchase value of all items with no sales line in the last twelve months, and apply four cost items to it: the interest rate on your overdraft facility, the warehouse costs in proportion to the share of locations those items occupy, insurance and handling, and an obsolescence percentage from your own history. In the example: €396,000 times 5.5 percent interest is 21,780, plus 28 percent of €168,000 warehouse costs is 47,040, plus 1.5 percent is 5,940, plus 6 percent obsolescence is 23,760, together €98,520 a year.

When should I write off obsolete stock?

In the annual accounts, as soon as the current value at the balance sheet date is lower than the acquisition price; Article 2:387(2) of the Dutch Civil Code requires this, and paragraph 1 says it is done regardless of the result for the year. For tax purposes, according to the Dutch Tax Administration, you may only take a lower value than cost price for obsolete goods, with supporting evidence and a consistent practice. What evidence your situation requires is a question for your tax adviser; we do not give tax advice.

What percentage of my stock can be dead stock?

At technical wholesalers with a broad range, we see that less than 8 percent of stock value without sales in twelve months is workable, and that above 15 percent the problem lies in the purchasing process. In the worked example the company is at 22 percent. An importer of own brands that buys by the container is structurally higher, and then the quarterly trend says more than the percentage.

How do I calculate inventory turnover per product group?

Divide the purchase value of twelve months of sales by the average purchase value of that group's stock. In the example, class A turns 5.8 times a year and class C 0.65 times; across the whole warehouse it is 3.0. Months of cover is the reverse calculation: 12 divided by the turnover rate, so 2.1 months for class A and 18.5 months for class C. For deciding which item to clear, the last sale date per item is a better column than the turnover rate of the group.

Is it smarter to sell obsolete stock at a discount or to keep it?

Work it out per batch: today's offer against the expected proceeds later minus the holding costs on the remaining stock. In the example, a batch of €60,000 yields €24,000 now from a stock buyer, and if kept, €39,000 over three years minus €22,410 holding costs, €16,590. Keeping only wins with an offer below 27.7 percent, and only if the assumption that the batch is gone within three years holds. At the wholesalers we see, less than half of the items that sat still for twelve months have been sold after two years.

Can I write down obsolete stock for tax purposes?

According to the Dutch Tax Administration, you may only take a lower value than cost price for obsolete goods, and report the difference as a cost; fixed percentages apply to textiles, clothing, furniture and upholstery. In the example, a write-down of €36,000 reduces corporate income tax in that year by €9,288 at the rate of 25.8 percent on profit above €200,000, and that tax comes back as soon as the batch is later sold for more than the written-down value. Whether your evidence is sufficient and how you set up a consistent practice is something to discuss with your tax adviser.

Why do I keep putting off clearing old stock?

Arkes and Blumer (1985) showed that people are more likely to continue with something once money has gone into it, out of a wish not to appear wasteful, and that those who have invested rate the chance of success higher. Ronayne, Sgroi and Tuckwell (2021) found that 23 percent of participants in an experiment held on to a lottery they had earned while a better one was there for the taking. Selling a €60,000 pallet for €24,000 feels like throwing away €36,000; the calculation shows that in the example keeping it costs €7,410 more. The practical remedy: have the list assessed by someone who did not do the purchasing.

Further reading: Why an importing wholesaler sees a 39 percent margin on an item that yields 25 percent after freight, import duty and currency, Why a hardware reseller pays tax on profit it does not have when memory prices rise, What the ECB rate rise costs a technical wholesaler that lets its customers pay in 52 days and The Pareto principle is also a risk gauge for your business. New: Why a technical wholesaler takes a €600 discount on an MOQ of 500 units and loses €1,225 on it. New: What one day of working capital is worth to a metal supplier with €10 million revenue, and which lever to pull first.

← Back to knowledge base