Inventory in Due Diligence

A large share of smaller businesses expense inventory straight through cost of goods sold as they buy it, rather than capitalising it on the balance sheet and releasing it as it sells. For tax purposes that is attractive and entirely understandable. It also means reported earnings are understated for as long as the business is growing its stock.

Then the business goes to market, and the question arrives: can we add all of that back?

Usually something can be, but rarely the amount being asked for. Getting this right is the single most valuable piece of inventory work in a transaction, and it is where we see sell-side adjustments fail most often.

Here is the error, and it is common enough to be worth stating plainly.

Where a business has been expensing purchases through cost of goods sold, the EBITDA adjustment is the change in inventory across the period, not the closing balance. A company that began the year holding a certain amount of stock and ended it holding substantially more has understated its earnings by the increase. It has not understated them by everything sitting on the shelf.

We have reviewed sell-side analyses that attempted this adjustment using the full amount rather than the movement between opening and closing. The intent was legitimate. The number was not, and it does not survive contact with a buyer’s advisor.

There is also a threshold below which none of this matters much. Where stock turns over in about ten days, the timing difference is small enough to be immaterial. Once a business is holding inventory beyond roughly thirty days and expensing it as purchased, the distortion compounds year over year and the magnitude becomes significant.

The assumption that this adjustment always runs in the seller’s favour is wrong, and testing the direction is part of the work.

If a business deliberately buys less in the run-up to a sale and works through older stock, cost of goods sold falls, margins look better than the underlying trade supports, and EBITDA is overstated rather than understated. That is the mirror image of the first problem and it is the one a buyer should worry about.

So the question is never simply whether to add something back. It is which direction the movement runs, whether it is consistent with the ordering pattern and margins we can see in the general ledger, and whether the explanation the company gives matches what the records show. On many files the honest answer turns out to be close to neutral.

This one is subtle and it catches careful people.

Not every inventory movement passes through the profit and loss. Where a purchase has been posted to accounts payable and then straight to the inventory account on the balance sheet, it never touched earnings at all, so it cannot be added back. It has already been captured.

On one engagement a portion of the inventory build had been recorded that way while the rest ran through cost of goods sold. Taking the full balance sheet movement as an EBITDA adjustment would have double counted the part that never hit the profit and loss, and we had to back it out to get to the real impact, which was materially smaller than the headline figure. We also confirmed directly with the company’s own accountant which items she had already included in her calculation, because two people adjusting the same inventory independently is exactly how a number gets counted twice.

Very few smaller businesses apply one method consistently. The realistic picture is a company that does not really use its inventory account, except when it does.

We have worked with ledgers where most purchases were expensed but occasional transactions were posted into inventory instead of cost, so the account was neither a true stock record nor unused. That is not an error anyone made deliberately. It does mean the treatment has to be established transaction by transaction rather than assumed from the account name, and it is a large part of why inventory-heavy files take longer.

Where more than one adviser is involved, inventory is where the numbers diverge, and understanding why matters more than deciding who is right.

On one transaction our figures and another firm’s began from an identical starting count, then separated over the following months. The cause was method rather than error. They had derived later balances using a purchasing and cost of goods sold calculation; we reconciled to the company’s own physical counts and then rebuilt back to those counts using the same purchasing method, so the two approaches could be compared directly. The residual differences traced to a stock write-down and an invoice exchange in a single month. Timing contributed as well, since purchase orders raised in one month sometimes appeared in the accounting system in another.

The post-close warning is the part worth carrying away. If working capital is pegged using one party’s inventory figure, and the company then runs its own count after closing using its usual method, the buyer sees a large apparent difference and concludes stock was run down before the sale. Often nothing of the kind happened. The two numbers were produced by different methods. Whatever method sets the working capital peg must be the method applied at the true-up, and that needs to be written down while everyone still agrees.

Those first two are independent, and conflating them is a recurring source of argument. An adjustment that increases EBITDA does not automatically change the peg, and a change to the peg says nothing about earnings. Add-backs that do not survive diligence covers the wider pattern.

One genuinely favourable case is worth naming: inventory held on consignment, where the business can draw on stock it has not paid for. That is a real cash flow advantage and it should be recognised as one rather than treated as a complication.