Asset Impairment: When Carrying Value Stops Being Supportable

Impairment is the admission that an asset will not deliver what its carrying amount claims. Depreciation is planned and spread; impairment is unplanned and lands at once. A rational holder takes the better of two courses, so recoverable amount is the higher of what the asset would fetch if sold and what it would generate if kept. The test compares that figure against carrying amount.

Here is what sits underneath that. A depreciation schedule is a forecast made on the day an asset arrives: eight years, nil residual value, straight line, and the charge runs on unchanged whatever happens afterwards. The schedule cannot notice that a machine stopped earning, that a customer left, or that a newer machine made this one unwanted. Something has to stand outside the schedule and ask a different question: not how much of the cost has been used up, but how much the asset is still good for. That second question is the impairment test, and it can be asked of any long-lived asset the business is carrying.

Most of the parts are already in place. Carrying amount and the gross block are covered under depreciation. The instinct came from inventory, where stock is carried at the lower of what it cost and what it will realise, so a ceiling on an asset's carrying value is not a new idea at all; it is the same instinct pointed at a machine instead of a stack of paper. A charge that sits above operating profit moves every line beneath it, so the margin ladder shows where an impairment charge will surface. The test itself has four parts: a recoverable amount computed without the error that catches most people, a charge that lands in one place on the statements and moves next year's return ratios, a rule for when the charge can be put back, and a reading of what a large write-down actually shows.

What is impairment, and how is it different from depreciation?

ImpairmentWriting an asset down because the amount at which it is carried can no longer be recovered, either by selling it or by using it. The write-down is recognised when the shortfall is identified, not spread over time. and depreciation both reduce the value of an asset on the balance sheet, and that shared outcome is where the confusion starts. Everything else about them is different. Depreciation is a plan: it is decided when the asset is bought, it runs at a rate nobody revisits most years, and it says nothing about whether the asset is doing well. Impairment is a correction: it is triggered by something happening, it is measured on the day it is identified, and it says a great deal about whether the asset is doing well.

The shape is easier to feel at household size, so take a household example first. A woman buys a sewing machine for Rs 40,000/- and tells herself it will last ten years, so she counts Rs 4,000/- of it as used up each year. The Rs 4,000/- a year is depreciation, and it is a plan she made on day one. Four years in, the machine is on her books at Rs 24,000/-. Then the tailoring shop she supplied closes and no other shop nearby takes that stitch. She could sell the machine second hand for perhaps Rs 9,000/-, or keep it and earn perhaps Rs 11,000/- from occasional work over the years it has left. Neither route gets anywhere near Rs 24,000/-. The Rs 4,000/- a year plan was never wrong as a plan. It simply had no way of knowing that the shop would close, and that gap is exactly the one the impairment test exists to close.

Now say it in the balance sheet's own words. Carrying amountWhat an asset stands at in the accounts today: its original cost less all the depreciation or amortisation charged against it so far, and less any write-down already taken. is cost less everything charged against it so far. The impairment test asks whether that carrying amount can still be recovered. If it can, nothing happens and the depreciation schedule simply carries on. If it cannot, the asset is brought down to what can be recovered, and the difference is charged against profit in one go. Notice the asymmetry built into that: the test can only ever push the carrying amount down. There is no matching rule that writes an asset up when it turns out to be worth more, so what appears on the balance sheet is a ceiling rather than a valuation.

Why is recoverable amount the higher of two figures and not the lower?

The higher of two figures is the whole mechanism in one sentence, so it is worth slowing down. Recoverable amountThe most a business can still get out of an asset, measured as the better of two courses: selling it, or keeping it and using it. is the higher of two separately measured figures. The first is fair value less costs of disposalWhat the asset would sell for between willing parties, after deducting what it would cost to get the sale done, such as removal, transport, commission and legal fees., the amount selling the asset would actually net after the cost of getting the sale done. The second is value in useThe present value of the cash the asset is expected to generate if the business keeps it and goes on using it, discounted to today., the present value of the cash the asset is expected to generate if it is kept and used.

Ask why the higher rather than the lower. The answer is not an accounting convention at all. The higher of two is a statement about what a business would do. A holder facing two available courses takes the better one, so an asset is only impaired when both routes fall below the carrying amount, and taking the lower of the two is the single most common error on this subject. Picture a woman with a wedding sari she paid Rs 30,000/- for. A resale dealer offers Rs 8,000/-. She reckons wearing it to weddings over the coming years is worth more to her than Rs 8,000/-, so she keeps it. She was never going to take the Rs 8,000/- offer, so the offer never measures the sari's value to her. Recoverable amount works exactly like that: the low route is available, it is simply not the one a sensible holder would choose.

Two practical consequences follow from the word higher, and both are easy to miss. The first is that only one of the two figures usually has to be computed. If fair value less costs of disposal already exceeds the carrying amount, the asset cannot be impaired whatever value in use turns out to be, and the same holds the other way round, so the work stops as soon as either route clears the bar. The second is that the test is not symmetric with the balance sheet: an asset whose recoverable amount is far above its carrying amount is not written up by so much as a rupee. The higher of two is a rule for finding the ceiling, and the ceiling only ever comes down.

What makes a business test an asset in the first place?

Running the test on every machine every year would be an enormous amount of work for an answer that is almost always no, so nobody does. Instead the business looks for indications that an asset might be impaired, and runs the full test only where it finds one. Six indications come up repeatedly, and each has an ordinary shape in a business that makes school notebooks.

An external fall in the value of the asset itself is the first: second-hand binding machines of that generation start changing hands for much less because a newer cutting head became the standard. Physical damage is the second, and it needs no explanation beyond a monsoon that reaches the shed floor. Obsolescence or a change in the way the asset is used is the third: schools move to a paper size the old machine cannot cut without a second pass. A decision to dispose of the asset before the end of its useful life, or to stop using it, is the fourth. Internal evidence that the asset is performing worse than the business expected is the fifth: output per shift has fallen and rework has climbed for four quarters running. A rise in market interest rates or required returns is the sixth. The same expected future cash is worth less today when discounted at a higher rate.

Five of those six indications are about the asset and one is about the outside world moving under it. So an impairment can appear in a year in which nothing at all went wrong inside the business. That possibility is worth holding on to before any write-down is read as a verdict on management. There is also one exception to the indication-first approach. Neither goodwillThe amount paid for a business above the value of the identifiable assets and liabilities acquired with it. Goodwill sits on the buyer's balance sheet, and no schedule ever amortises it. nor an intangible asset with an indefinite useful life carries a depreciation schedule that would bring it down on its own, so both are tested every year regardless of whether any indication exists.

Where does the charge land, and what does it leave untouched?

The arithmetic is short. Carrying amount less recoverable amount is the charge, it goes to profit for the year, and the asset comes down by the same amount. Where on the statements the charge shows up, and where it conspicuously does not, matters far more than the arithmetic.

The charge sits above operating profit. So it reduces earnings before interest and tax (EBIT), it reduces profit before tax, it reduces profit after tax, and every margin computed from any of those falls with it. The charge is not a cost of making anything, so it touches neither revenue nor gross profit. And here is the part readers get wrong most often: an impairment moves no cash whatsoever, so a business can report a collapse in profit and a completely unchanged operating cash flow in the same year, and an impairment charge is the usual reason. The cash left years earlier, on the day the asset was bought. All that has happened now is that the accounts have stopped pretending the money will come back.

No cash moved, so an impairment is added straight back in the operating section of the cash flow statement, in exactly the same place and for exactly the same reason as depreciation. Both are charges against profit that never involved a payment. Anjani Stationers Private Limited reported operating cash flow of Rs 36,30,000/- in year two. A write-down of the old binding machinery would have pulled earnings before interest and tax down by Rs 3,00,000/- and left that figure standing at Rs 36,30,000/-, to the rupee. Two numbers on the same set of accounts, moving in completely different ways, for one reason.

Why do the return ratios improve the year after a write-down?

Careful readers get caught the year after a write-down. They are watching the ratios rather than the raw numbers, and that is precisely what catches them. A write-down cuts the asset. The asset sits in the denominator of every return and turnover measure there is. So the year after an impairment, with trading completely unchanged, return on capital employed rises, asset turnover rises, and fixed asset turnover rises. Nothing improved. The bottom of the fraction got smaller.

Work it on Anjani Stationers Private Limited, holding everything else rigid. Published capital employed for year two is Rs 1,52,00,000/- against earnings before interest and tax of Rs 41,50,000/-, a return of 27.3 per cent. In the teaching case, the Rs 3,00,000/- charge takes the numerator to Rs 38,50,000/- and the denominator to Rs 1,49,00,000/-, so the return for that year falls to 25.8 per cent. Then the following year arrives. Suppose trading is identical, so earnings before interest and tax would have been Rs 41,50,000/- again. The denominator is still Rs 3,00,000/- smaller, and the return reads 27.9 per cent against the 27.3 per cent it would have shown without any write-down.

There is a second push in the same direction, and it is the one that surprises people. A written-down asset carries less depreciation for the rest of its life. The machinery would have been carrying Rs 9,00,000/- over three remaining years, Rs 3,00,000/- a year. Written down to Rs 6,00,000/- over the same three years, it is Rs 2,00,000/- a year, so next year's depreciation charge is Rs 1,00,000/- lower and next year's earnings before interest and tax are Rs 1,00,000/- higher at Rs 42,50,000/-. Divide that by the smaller Rs 1,49,00,000/- and the return reads 28.5 per cent. Return on capital employed goes from 27.3 per cent to 28.5 per cent, a gain of 1.2 points, with not one extra notebook sold and not one rupee of cost avoided, purely because a write-down shrank the denominator and lightened the depreciation charge above the numerator.

What would a write-down of the old binding machinery actually do?

Set out the case in full. The original binding machinery cost Rs 24,00,000/- and runs on an eight year straight line schedule at Rs 3,00,000/- a year with nil residual value assumed, an assumption worth naming because it is an assumption rather than a measurement. Five annual charges have been taken, so accumulated depreciation on that machine stands at Rs 15,00,000/- and its carrying amount at the year two balance sheet date is Rs 9,00,000/-, with three years of useful life remaining. Every figure so far is the actual asset ladder. The indication and the two routes that follow are hypothetical.

Suppose an indication appeared: newer cutting heads became standard and second-hand machines of that generation started changing hands for much less. Vaidehi Rao, the finance controller, runs the test. Selling the machine would net Rs 5,40,000/- after costs of disposal. Keeping it and using it for the three years left would generate Rs 6,00,000/- of discounted cash. Recoverable amount is the higher, so Rs 6,00,000/-, and the write-down is Rs 3,00,000/-.

Look hard at the last row. Taking the lower route writes off Rs 3,60,000/- rather than Rs 3,00,000/-, Rs 60,000/- more than the shortfall actually is. The business can simply go on using the machine for the three years it has left rather than selling it, so the extra Rs 60,000/- is value it can still recover. The error is not conservative, it is wrong, and it produces a machine carried at Rs 5,40,000/- that the business itself expects to yield Rs 6,00,000/-. It also plants a Rs 60,000/- profit in a later year, when the machine outperforms the amount it was written down to.

Now the effects, every one of them belonging to the hypothetical and none of them to the published accounts.

Read the last two rows as a pair. Separately they mislead; together they teach. The return drops by 1.5 points in the year of the charge and then rises by 1.2 points above where it would otherwise have been in the very next year, on identical trading throughout. Nothing about the business changed in either direction. One write-down produced both movements.

The panel gives four readings worth writing down, all of them hypothetical. Pushing value in use to Rs 9,00,000/- or beyond makes the write-down disappear entirely, and it stays at nil however much further the figure goes. An asset that can recover its carrying amount is not impaired, and there is no rule that writes it up. One clearing route is enough on its own, so setting the sale route to Rs 9,00,000/- and dragging value in use all the way to the floor still leaves the write-down at nil. The flat stretch where nothing is written off is the largest region of this panel, and it is the reason most tests end with no charge at all. Now switch the rule to the lower. With a sale route of Rs 9,00,000/- and a use route of Rs 6,00,000/- the correct answer is nil and the wrong rule produces a charge of Rs 3,00,000/-, a write-down invented out of nothing more than a misread definition. Push both routes down to the floor and the panel reaches its worst state: a write-down of Rs 9,00,000/-, earnings before interest and tax of Rs 32,50,000/- and the machine carried at nil.

Can an impairment ever be put back?

For most assets, yes. If the circumstances that caused the write-down have gone, or the estimates used to measure recoverable amount have genuinely changed, the earlier charge is reversed and the asset goes back up. The reversalPutting back an earlier write-down when the reason for it has gone. The reversal is capped: the asset can never end up carried above where the original schedule would have left it. is credited to profit in the year the recovery is recognised, which is the mirror image of the original charge.

But there is a cap, and the cap is the whole of the rule. A reversal can never carry an asset above the amount it would have stood at had the impairment never happened, so the write-down can be undone but the depreciation cannot be un-run. Take the machinery in the case hypothetical one year further on. Written down to Rs 6,00,000/- with three years left, it depreciates at Rs 2,00,000/- a year and stands at Rs 4,00,000/- a year later. Had nothing ever been written down, it would have stood at Rs 9,00,000/- less Rs 3,00,000/-, so Rs 6,00,000/-. The Rs 6,00,000/- the schedule alone would have left is the ceiling. If recoverable amount has recovered to Rs 7,00,000/-, the reversal is Rs 2,00,000/- and not Rs 3,00,000/-, and the last Rs 1,00,000/- of the recovery stays unrecognised until the asset is sold.

Goodwill works differently, and the difference is deliberate rather than accidental. An impairment of goodwill can never be reversed, in any circumstances, however clearly the value has come back. The reasoning is that once goodwill has been written down, any recovery in the value of the business is indistinguishable from goodwill the business generated itself through its own trading, and internally generated goodwill may never be recognised at all. Allowing the reversal would let a company put internally generated goodwill on the balance sheet through the back door, using the earlier write-down as the doorway. So the door is nailed shut. The asymmetry is worth carrying as a fact about the accounts being read: a machine can bounce back on the balance sheet, and goodwill never can.

How to Interpret Impairment Charges and Asset Write-Downs

Meeting an impairment charge in a set of accounts prepared by somebody else calls for five steps, run in order. Each step changes what the next one means.

Step one, establish what was impaired and where it came from. A write-down of something the business bought and a write-down of something the business built are two entirely different kinds of news. Goodwill impairment says a price paid for another business turned out to be too high. A write-down of machinery says a spend on capacity did not produce what was expected. An intangible write-down often says a project was abandoned. The three findings are not interchangeable, and the notes to the accounts will usually say which asset was involved.

Step two, check whether the charge is the first or one of a series. A single large charge in an otherwise clean run is one kind of event. A charge in each of four consecutive years is a different kind entirely. It suggests either that the estimates keep needing revision or that the underlying deterioration has not stopped. The years are counted before the size is interpreted.

Step three, check what the ratios do the year after, and recompute them on the pre-impairment base before reading them. Most misreadings happen at this step. Step four, look at what the notes disclose about the assumptions behind value in use and at how those assumptions moved. A recoverable amount rests on estimates of future cash and a discount rate, and where those are disclosed, a change in them between years tells more than the charge itself does. Where they are not disclosed, the honest statement is that they are not disclosed, rather than an assumption that they were reasonable.

Step five, and this is the one people skip, ask what the impairment says about the decision that created the asset. Somebody approved that purchase or that plan. The write-down is the accounts catching up with how it turned out. An impairment is information about the past and only weakly about the future, and a reader who treats it as a forecast has read a history note as a projection. It shows what a past decision was worth. The charge does not show what this year's trading was worth, and it does not show what next year holds.

Who reads an impairment charge, and what do they do with it?

Step away from the mechanism. Four different people open the same set of accounts in the same week, and none of them is admiring the arithmetic.

A lender reads a write-down to find out how much less security it now has, an analyst reads it to strip a one-off out of the trend, an investor reads it as a verdict on a spending decision, and Vaidehi Rao reads it as a signal to look again at how the asset is being used. Watch each in turn. The lender's covenant tests often run on tangible net worth, and Anjani Stationers Private Limited's stands at Rs 1,38,00,000/- being equity of Rs 1,42,00,000/- less the Rs 4,00,000/- of software. A write-down of Rs 3,00,000/- would carry that to Rs 1,35,00,000/-, and if a covenant sat anywhere near that level the charge would matter enormously to a lender even though not a rupee of cash moved. Breaching a covenant can make a loan repayable, so a covenant is the one place where a non-cash charge can have a very cash consequence.

The analyst's use is arithmetic and separation. A charge is pulled out of the year it landed in so the trend can be read without it, and then it is put back somewhere else. The money really was spent, and pretending otherwise flatters every year in the series. The honest treatment is to note the charge as an outcome of an earlier decision rather than as a cost of this year's trading, and then to check whether it recurs. An investor is doing something closer to bookkeeping on management: a purchase was made, a price was paid, and the charge is the accounts reporting how it went. Vaidehi Rao, inside the business, can see the machine, so she has the most direct use of all. She knows whether output per shift fell, whether a format changed, whether the second machine took over work the first one used to do. The published charge is the last step of a conversation she started months earlier.

One boundary belongs here rather than in a footnote. An impairment establishes that a carrying amount could not be supported. It never establishes whether the original decision was reasonable when it was made. Plenty of sound purchases are followed by write-downs when the world moves. Plenty of poor purchases never produce a charge. The estimates behind value in use were generous enough to keep the carrying amount supportable. Anyone converting an impairment charge into a judgement about the quality of management has taken a measurement rule somewhere it cannot go.

Anjani Stationers Private Limited, Chitra Binding Works Private Limited and Vaidehi Rao are invented.
Educational material. Not advice on any investment, tax, budget or market position.