What the Footnotes Told Me: A Case-by-Case Breakdown of Disclosures That Changed My Valuation
The income statement is what management wants you to see. The footnotes are what the accounting standards require them to show you.
There is a moment in reading a 10-K that I have come to look for deliberately.
It happens in the footnotes, usually somewhere between page 60 and page 120, in a section written in dense technical language that most investors never reach. The sentence structure is passive. The font is the same size as everything around it. There is nothing to signal that what you are reading is significantly more important than the paragraph before it or the paragraph after.
But something has changed. A revenue recognition policy that was described one way last year is described differently this year. A pension discount rate assumption has been revised in a direction that happens to improve reported earnings. A contingent liability that did not appear in last year’s filing has been quietly added, characterized as remote but present.
These are the moments where reading a 10-K stops being a research exercise and starts being an actual edge. Not because the information is hidden, it is all disclosed, technically, but because the vast majority of investors are not in the room when it happens. They are reading the earnings release and the slide deck and the transcript of the call where management speaks in round numbers and forward guidance.
The footnotes do not lie. They also do not make it easy.
This piece is a breakdown of eight specific footnote patterns, drawn from real situations across the history of publicly traded companies, and from the kind of filings I have encountered working through my own analyses, that materially changed how I valued a business. In each case, I will describe what the disclosure said, what it meant in practice, and what it implied for the valuation I was building at the time.
The 10-K reading framework has been covered in a prior piece in this archive. This one is not about the methodology. It is about what the methodology finds when it actually works.
A Note on How These Examples Are Structured
Each case below follows the same pattern: the disclosure itself (what it said and where it appeared), the analytical implication (what it meant for the actual economics of the business), and the valuation impact (how it changed the numbers and the conclusion).
Some cases come from well-documented historical situations where the footnote disclosures are part of the public record, situations where the gap between what the footnotes showed and what management was saying became clear only in retrospect. Others come from patterns I have encountered repeatedly across different analyses. In both types, the lesson generalizes: the disclosure pattern matters more than the specific company, because these same structures appear in different businesses in different industries in different decades.
The goal is pattern recognition, not a list of historical curiosities.
Case 1: The Revenue Recognition Change That Inflated a Year of Growth
The disclosure: In the revenue recognition footnote, the company announced it was adopting a new method of recognizing revenue on multi-year contracts, shifting from recognizing revenue at delivery milestones to recognizing it ratably over the contract period. The change was described as a voluntary adoption of updated accounting guidance. No prior-year restatement was required. Management mentioned it in a single paragraph of the accounting policies note.
What it actually meant: The timing of revenue recognition is one of the most powerful levers available to management within the rules of GAAP. Shifting from milestone-based to ratable recognition changes when revenue appears in the income statement without changing the underlying economics of a single contract. For this company, the effect in the transition year was to pull forward revenue from future periods, inflating reported growth in year one of the new policy at the expense of years two and three.
The earnings release that quarter showed 14% revenue growth. Adjusting for the accounting change, underlying growth was closer to 8%. Both numbers are technically accurate. Only one reflects what the business actually did with its customers during the year.
The valuation impact: I was building a DCF with a base case of 12% revenue growth, which felt conservative relative to the reported 14%. When I stripped out the accounting change, the base case dropped to 7%, below the rate needed to justify the multiple. The investment case required a different conclusion at that number. The thesis was not dead, but the entry price that made sense was meaningfully lower than what I had been working with.
The pattern: any year in which a company changes its revenue recognition policy requires you to reconstruct the prior-period comparables on the new basis before drawing any conclusions about growth trends. Management is not required to do this restatement for you. You have to do it yourself.
Case 2: The Pension Assumption That Was Doing Great
The disclosure: The pension and post-retirement benefits footnote disclosed that the company had revised its assumed long-term rate of return on plan assets upward, from 6.5% to 7.5%, and had also adjusted its discount rate assumption in a direction that reduced the present value of future obligations. Both changes were described as reflecting updated actuarial assumptions.
What it actually meant: Pension accounting is one of the most fertile areas for earnings management within GAAP because the assumptions are genuinely uncertain and the income statement impact of changing them is immediate. A higher assumed return on plan assets reduces pension expense in the current year. A higher discount rate reduces the recognized obligation. Both changes improve reported earnings without any change in the underlying economic reality of what the company owes its retirees.
In this case, the combined effect of the two assumption changes added approximately 8% to reported operating income for the year. The company’s headline earnings growth was 11%. Adjusting for the pension assumption changes, earnings growth was closer to 3%.
More importantly, the assumed return of 7.5% was materially above the forward-looking return expectations for the asset mix the pension fund actually held. The company was booking income today based on returns it would not earn tomorrow. This does not create a permanent problem, eventually, actual returns diverge from assumptions and the accounting catches up. But in the interim, reported earnings overstate economic reality.
The valuation impact: Pension assumption engineering is a classic late-cycle signal. Companies whose core businesses are under pressure often reach for accounting levers to smooth the reported trajectory. When I encounter aggressive pension assumptions, return assumptions above 7%, discount rate assumptions well above the current yield on high-quality corporate bonds, I adjust the reported earnings down to a normalized figure before applying any multiple. In this case, the adjustment reduced the apparent earnings power by roughly $0.40 per share on a $4.50 earnings figure. Not catastrophic, but enough to shift the fair value estimate by 8–10% and change a borderline buy into a clear pass at the current price.
Cases 1 and 2 cover the two most common earnings-inflating footnote patterns: revenue recognition timing and pension assumption management. The six cases that follow go further, into the disclosures that change not just the earnings multiple but the entire investment thesis. Case 3 covers the goodwill impairment footnote, where the assumptions used for testing can overstate net asset value by hundreds of millions of dollars in plain sight. Case 4 covers the contingent liability that becomes the whole story, and the specific language signals that distinguish genuinely remote from strategically characterized as remote. Cases 5 through 8 cover off-balance-sheet lease obligations, related-party transactions, SBC gaps between adjusted and GAAP margins, and the reclassification that only appears when you read two filings side by side. The synthesis section at the end draws the common thread across all eight and explains what it means for how conservatively to underwrite any business where the footnotes tell a different story than the income statement.
