Goodwill and Impairments

Canonical data. Wexford figures tie to the canonical page (USD millions). The impairment scenario is an explicit hypothetical layered onto Wexford — it did not occur and does not alter the canonical statements.

It is not a machine, a patent, a building, or a contract. It arises for one reason and one reason only: a company bought another company and paid more than the identifiable pieces were worth. The accounting question is what to do with that difference — and the investor's question, which the accounting cannot answer, is whether the difference bought anything.

Where goodwill comes from

When one company acquires another, the purchase price is allocated across everything acquired at fair value. Machines, inventory, receivables, and assumed debts are measured. Then identifiable intangibles that were invisible on the target's own balance sheet — customer relationships, brands, developed technology, licences — are recognised and valued, precisely because they can now be identified through a transaction. Whatever price remains unallocated becomes goodwill.

Goodwill is therefore a residual, not a measurement, and that residual is a mixture of several quite different things: genuinely valuable items that cannot be separately identified, such as an assembled workforce and reputation; expected synergies from combining the businesses; the control premium paid to persuade owners to sell; and — if the acquirer paid too much — the overpayment. All four land in the same line, and the accounting makes no attempt to distinguish them. This is the single most important thing to understand about the number.

The allocation itself carries judgement worth noticing. On the illustration below, a 300 acquisition of a business with 180 of identifiable net assets produces a premium of 120. Allocate 50 of that to identifiable intangibles with a ten-year life and 70 to goodwill, and the acquirer charges 5.0 a year of amortisation for a decade. Allocate the entire 120 to goodwill instead and the annual charge is nothing — because goodwill is not amortised. Reported profit is 5.0 a year higher for ten years, from a decision about how to label the same premium. Both allocations must be supportable and are reviewed by auditors; the point is that the line between them is an estimate, not an observation.

The impairment test, and what it inherits

Because goodwill has no determinable life, it is not amortised. It is tested — at least annually, and whenever events suggest a problem. The test compares the carrying value of the business unit the goodwill attaches to against its recoverable amount; if carrying value is higher, the difference is written off through the income statement.

The recoverable amount is an estimate of what the unit is worth, which means the impairment test is a valuation exercise — and it inherits every uncertainty that valuation carries. Projected cash flows, growth assumptions, and a discount rate all feed in. Pillar 22 established the sensitivity precisely: a cash-flow stream of 100 growing at 2% is worth 1,428.57 discounted at 9% and 1,250.00 at 10% — a 12.5% swing in value from a one-percentage-point change in an assumption. An impairment test is that arithmetic applied to a business unit, which is why whether goodwill is impaired in a given year is genuinely a matter of judgement, and why rising interest rates mechanically increase impairment pressure across whole markets without anything changing inside the businesses.

Two rules complete the picture. An impairment cannot be reversed — goodwill written down stays down, even if the acquired business recovers, under both major frameworks. And impairment is entirely non-cash: no money moves.

What a write-down actually tells a reader

The most useful reframing in this article: an impairment is not the moment value was destroyed. It is the moment the accounting caught up. If a company overpaid for an acquisition, the loss occurred when the cash left — possibly years earlier. The write-down is a backward-looking acknowledgement, which has three consequences.

It is information about the past, not a forecast. A large impairment says a previous acquisition did not work out; it says little about current operations, and nothing about future cash generation. Operating cash flow is unaffected, which is why a company can report a catastrophic loss and a perfectly healthy cash statement in the same year.

But it is real information, and dismissing it as "non-cash" goes too far. It is evidence about the acquirer's capital allocation — management paid a price the assets did not justify — and that is a track record worth knowing, because the same team is likely to make the next acquisition decision. Non-cash does not mean unimportant; it means the cash already went.

And the timing of write-downs follows documented patterns. Impairments cluster in years that are already bad, and around changes of senior management — a new team has little incentive to defend a predecessor's acquisition prices. This is widely observed and is not in itself improper: an impairment requires a triggering reassessment, and a new management team reassessing is exactly such an event. It does mean a reader should treat the announcement date as weak evidence about when the value was lost.