Corva › Blog › Goodwill on the balance sheet
Balance sheet analysis
Every screener reports goodwill as a share of total assets. That is the wrong denominator. An impairment writes down against book equity, so equity is the number that tells you how much of the company survives one.
8 min read· Corva Research
The short version
Roper Technologies closed 2025 with $21.3B of goodwill and $19.9B of total stockholders' equity. The company puts those two figures side by side in its own risk factors, which is more than most analysis of goodwill ever does.
The standard screen would have reported Roper's goodwill as 62% of total assets and left it there. That number is true and it is close to useless, because total assets is not what a goodwill write-off consumes.
Equity is. So the ratio worth running divides goodwill by shareholders' equity, and it tells you something the assets version hides: how much of the company's book value is a price paid, rather than a thing owned.
Goodwill is the excess of an acquisition price over the fair value of the identifiable net assets bought. A company pays $500M for a business whose separately identifiable assets and liabilities are worth $300M net, and $200M of goodwill lands on the balance sheet as the plug.
It is not a valuation of a brand or a customer base. Those get identified and recorded separately, as intangible assets with their own useful lives. Goodwill is the residual: the part of the cheque that could not be attached to anything specific.
Two consequences follow, and both matter here. Goodwill is not amortised, so it sits at cost indefinitely and never declines with use. And it is only ever written down, never written up, because US GAAP prohibits reversing an impairment once taken.
So the balance sheet carries a number that can only fall. The question is what falls with it.
Search for how to assess goodwill and the answer is uniform. Corporate Finance Institute, MyAccountingCourse, Wall Street Oasis, InvestingAnswers and TradingView all publish the same formula: goodwill divided by total assets, with a rule of thumb that anything much above 5% to 10% deserves a look.
The formula is not wrong. It answers a different question from the one an equity holder is asking.
Total assets is the whole balance sheet, debt-funded and equity-funded together. When goodwill is impaired, the charge runs through the income statement, lands in retained earnings, and reduces shareholders' equity by roughly the full amount. Creditors are untouched. The equity absorbs all of it.
Worse, the tax code usually declines to help. Goodwill impairment on acquisitions structured as stock purchases is frequently non-deductible, so there is no offsetting tax benefit to soften the hit to book value. Scripps recognised $952M of goodwill impairment in 2023 and disclosed that $855M of it was a non-deductible expense.
Which means a company with goodwill at 30% of assets and 90% of equity is not a moderately intangible-heavy business. It is one full write-off away from having almost no book value at all, and the assets ratio never says so.
Both inputs sit on the same page of the balance sheet, so this takes under a minute.
Above 100% the arithmetic gets blunt: a full write-off leaves negative book equity. That is not a hypothetical for a small number of large, well-known companies, and it is the reason the ratio is worth having.
The other reason is that it is checkable in one company in about a minute. So check one.
The E.W. Scripps Company (SSP), a television broadcaster, filed its Form 10-K for 2025 on 27 February 2026. Three lines from the balance sheet and the goodwill note tell the story.
| Line | 2025 | 2024 |
|---|---|---|
| Goodwill | 1,918,334 | 1,968,574 |
| Total assets | 5,008,628 | 5,198,575 |
| Total equity | 1,246,092 | 1,318,014 |
| Goodwill / assets | 38.3% | 37.9% |
| Goodwill / equity | 153.9% | 149.4% |
Goodwill has been larger than the company's entire book equity in both years, while the assets ratio stayed under 40%. Source: E.W. Scripps FY2025 Form 10-K, Consolidated Balance Sheets and Note on Goodwill. Filed 27 February 2026. Ratios computed from the figures shown.
The assets reading places Scripps in ordinary territory for a media company. The equity reading says something else entirely: every dollar of book value, and then half a dollar more, is goodwill from past deals.
It gets more specific. Scripps splits its goodwill by reportable segment: $858.8M in Local Media, $1,052.4M in Scripps Networks, $7.2M in Other. The Scripps Networks balance alone is 84% of the company's total book equity.
And Scripps Networks has been impaired before. The same filing states that in 2023 the company "concluded that the fair value of our Scripps Networks reporting unit did not exceed its carrying value" and took $952M of non-cash goodwill impairment charges. The unit was written down once, and what remains still exceeds four fifths of the equity base.
None of which would appear on a screen sorted by goodwill as a share of assets. So what would tell you whether the next write-down is close?
Ordering by the conventional ratio makes the point faster than an argument does. These are five companies with December 2025 year ends and material acquisition histories, sorted by goodwill as a share of total assets, lowest first.
$ millions. All figures at 31 December 2025, taken from each company's Form 10-K for the year then ended.
| Company | Goodwill | Assets | Equity | GW/assets | GW/equity |
|---|---|---|---|---|---|
| CHTR | 29,710 | 154,213 | 16,054 | 19.3% | 185.1% |
| ELAN | 4,779 | 13,358 | 6,547 | 35.8% | 73.0% |
| SSP | 1,918 | 5,009 | 1,246 | 38.3% | 153.9% |
| TMO | 49,362 | 110,343 | 53,407 | 44.7% | 92.4% |
| ROP | 21,341 | 34,577 | 19,882 | 61.7% | 107.3% |
Sources, all Form 10-K for the year ended 31 December 2025: Charter, Elanco, Scripps, Thermo Fisher, Roper. Equity is the parent shareholders' line. Charter additionally reports $4,465M of noncontrolling interests; on total equity of $20,519M its ratio is 144.8%. Ratios computed from the figures shown.
Read the last two columns against each other. Charter has the lowest goodwill relative to assets in the group and by a wide margin the highest relative to equity, because its balance sheet carries a great deal of debt and the assets denominator is inflated by it. Elanco looks nearly twice as goodwill-heavy as Charter on the conventional measure and is less than half as exposed on the one that counts.
The ordering does not just weaken. It inverts.
Companies test goodwill for impairment at least annually, comparing each reporting unit's fair value against its carrying amount. The comparison sounds objective. It is not.
Fair value is usually estimated by management, from management's own forecast of the unit's cash flows, discounted at a rate management selects, sometimes blended with market multiples management chooses as comparable. A passed test means those inputs produced a number above carrying value. It does not mean an independent party would agree.
What is worth hunting for is the cushion, and some companies disclose it. Scripps does:
"Upon completing our annual test in the fourth quarter of 2025, we determined that the fair value of our Local Media reporting unit exceeded its carrying value by more than 20% and that the fair value of our Scripps Networks reporting unit exceeded its carrying value by approximately 5%."
Then, two sentences later, the sensitivity: "a 50 basis point increase in the discount rate used for the Scripps Networks reporting unit would reduce its fair value by approximately 6%."
Set those side by side. A 5% cushion, and a rate move of half a point that removes 6% of the fair value. The unit passed its 2025 test on assumptions that a modest change in one input would have reversed, and it holds goodwill equal to 84% of the company's book equity.
This disclosure is not mandatory in the way a balance sheet line is, which is why so few readers know to look for it. It exists because the SEC staff asks for it. In a 2013 comment letter exchange with Comtech Telecommunications, the staff instructed the company to "disclose, if true, in your critical accounting policies that none of your reporting units with significant goodwill is at risk of failing step one of the goodwill impairment test."
Where to look, in order:
A company that names no cushion has told you nothing either way. Silence is not a pass.
Here is the part that cuts against everything above. A large goodwill balance is not by itself a defect, and treating a high goodwill-to-equity ratio as a sell signal would be a straightforward error.
Goodwill measures what was paid. It says nothing about what was earned on the payment. A company that buys well and generates strong returns on the total capital deployed, purchase premium included, has created value, and the size of the goodwill line is simply the size of the programme.
Roper is the useful case. Goodwill at 107.3% of stockholders' equity means a complete write-off would take book equity from $19.9B to roughly negative $1.5B. It would also change nothing about the business, which generated $2,540.3M of cash from operating activities in 2025, up 6% on 2024. An impairment is a non-cash accounting event. It moves no cash, breaches no covenant on its own, and alters no customer relationship.
What an impairment does is confirm, late and in public, that a price paid in the past exceeded the value received. The information is real. The charge is not.
So the ratio is a sizing tool, not a verdict. It tells you how much is riding on the acquisition record. Whether that record is good is a separate question, answered by returns on invested capital, by organic growth in the acquired units, and by whether management has ever written anything down voluntarily.
Stated plainly, because a ratio sold without its limits is worse than no ratio.
Used properly it is a sizing question, asked before the analysis rather than instead of it: how much of this company's book value is a bet on deals that already happened?
There is no threshold that holds across industries, and anyone quoting one is guessing. What the ratio does give you is a clean reading of exposure: below about 30% a full write-off is survivable in book terms, and above 100% it produces negative book equity. Compare within an industry, because software, medical devices and media all run structurally higher than retail or utilities.
No. It is a non-cash charge that reduces reported net income and shareholders' equity. It can matter indirectly, through debt covenants written on net worth or reported earnings and through executive compensation targets, but no money leaves the business on the day the charge is booked.
Because the test depends on their own forecasts and discount rate, and because an impairment is a public admission that an acquisition disappointed. The incentive runs one way. This is exactly why the disclosed cushion, where it exists, is more informative than the pass or fail result.
Run it both ways. Goodwill alone is the cleaner measure, because acquired intangibles such as customer relationships and developed technology amortise on a schedule and decline without any impairment decision. But they are impaired under similar conditions, so the combined figure is the harsher reading and worth having next to the first.
Critical Accounting Policies and Estimates, inside Item 7, Management's Discussion and Analysis. Search the filing for "exceeded its carrying value" or "at risk". If nothing comes back, the company has chosen not to quantify it, which is common and is itself worth noting.
Closely enough for the ratio to work. Under IAS 36 goodwill is tested annually at the cash-generating unit level rather than the reporting unit level, and IFRS also prohibits reversing a goodwill impairment. The disclosure of key assumptions and sensitivity is generally more prescriptive under IFRS than under US GAAP.
Corva pulls goodwill, intangibles and equity straight from the filed statements, computes both ratios, and surfaces the impairment cushion language from the critical accounting estimates where a company discloses it. Figures are computed rather than guessed and cross-checked against SEC EDGAR. Where a number cannot be found, it says so instead of inventing one.
Research a company free →Live financials on any listed company are free. No card.
Corva is a research tool, not a broker or investment adviser, and nothing here is a recommendation to buy, sell or hold any security. Charter Communications, Elanco Animal Health, E.W. Scripps, Thermo Fisher Scientific, Roper Technologies and Comtech Telecommunications are named as documented examples of disclosed balance sheet figures and filing language, and for no other reason. All figures are taken from the companies' Forms 10-K for the years stated and from SEC correspondence on EDGAR; ratios are computed from those figures. Verify anything you intend to act on against the primary filing. See terms and disclaimer.