Owner Earnings: Buffett's Formula for What a Business Really Makes

Reported earnings are an accountant's opinion. Cash is a fact. Somewhere between the two sits the number Warren Buffett considers the true measure of what a business produces for its owners: owner earnings. He introduced the concept in an appendix to his 1986 letter to Berkshire Hathaway shareholders, and no better definition of business value creation has replaced it since. Owner earnings answer a deceptively simple question: after paying every bill and spending whatever is required to keep the business competitive, how much cash could the owners take out this year without harming the company?

That question sounds like it should be answered by net income or free cash flow, but neither quite does the job. Net income is distorted by accrual accounting; free cash flow punishes companies for investing in growth. Owner earnings sit between them, and understanding all three makes you sharply better at reading a cash flow statement and estimating intrinsic value. This guide works through the full calculation, the hard problem of maintenance capex, and the earnings-quality red flags the framework exposes.

Buffett's 1986 Definition

In the appendix to his 1986 shareholder letter, Buffett defined owner earnings as: "(a) reported earnings plus (b) depreciation, depletion, amortization, and certain other non-cash charges... less (c) the average annual amount of capitalized expenditures for plant and equipment, etc. that the business requires to fully maintain its long-term competitive position and its unit volume." He added that if the business needs additional working capital to maintain its position and volume, that increment belongs in (c) as well. And he was unambiguous about why it matters: owner earnings, not GAAP figures, are "the relevant item for valuation purposes — both for investors in buying stocks and for managers in buying entire businesses."

The definition's most important word is maintain. Buffett does not subtract all capital expenditures — only the average annual amount required to preserve the company's competitive position and unit volume. Money spent to expand — new factories, new stores, capacity for future customers — is an investment decision, not a cost of this year's earnings. That single distinction separates owner earnings from every standard cash flow metric, and it is where all the analytical work lives.

Owner Earnings vs. Net Income vs. Free Cash Flow

Net income is the bottom line of the income statement, built on accrual accounting. Revenue is booked when earned, not when cash arrives; depreciation spreads past capital spending over future years on a schedule that may bear no relationship to the true cost of staying competitive. A company depreciating 30-year-old equipment at historical cost can report handsome earnings while facing replacement bills far above its depreciation charge. Accruals are not fraud — they are the point of accounting — but they open a gap between reported profit and economic reality that owner earnings are designed to close.

Free cash flow closes the accrual gap from the other direction: it starts with operating cash flow and subtracts all capital expenditures — maintenance and growth alike. That makes FCF wonderfully objective (every input is on the cash flow statement) but economically harsh on companies investing heavily to expand. A retailer opening 50 new profitable stores this year reports depressed FCF, yet its owners are getting richer, not poorer. Owner earnings subtract only the maintenance portion of capex, so for a growing company, owner earnings ≥ free cash flow, with the difference equal to growth capex. For a static, no-growth business the two converge.

So the three metrics form a spectrum. Net income is what accountants say the business earned. Free cash flow is what cash was left after every investment. Owner earnings are what the business generated after mandatory spending only — the number an owner who chose to stop growing could actually pocket. Buffett's argument is that for valuation, the third number is the one that counts.

The Hard Part: Estimating Maintenance Capex

Companies almost never split their capex line into maintenance and growth, so you have to estimate. Three practical methods exist, and careful analysts use at least two as a cross-check. First, the D&A proxy: assume maintenance capex roughly equals depreciation and amortization. It is fast and often reasonable for mature businesses, but it fails when inflation makes replacement cost exceed historical cost, or when acquired intangibles inflate amortization with charges that require no cash reinvestment at all. Second, the historical capex-to-sales method: compute the ratio of capex to revenue over a full cycle (five years or more), take the average, and multiply by current sales — the logic being that supporting a given level of revenue requires a fairly stable level of capital spending. Third, management disclosure: some companies state maintenance capex directly in annual reports or on earnings calls, and when they do, that figure — sanity-checked against the other two methods — is the best starting point.

Whatever method you use, remember Buffett said average annual maintenance capex. Capital spending is lumpy — a plant refit lands in one year, then nothing for four. Using a single year's figure will whipsaw your owner earnings; smooth it over a cycle.

A Full Worked Example

Now assemble the whole calculation for one company and compare all three metrics side by side. Every input below comes straight from the financial statements except maintenance capex, which we estimate as shown above.

The ordering tells a story. Owner earnings come in $20M below net income because true maintenance spending ($70M) plus the working capital build ($15M) exceed the $65M of non-cash charges added back — accrual accounting was flattering Harborview slightly. But owner earnings sit $40M above free cash flow because Harborview chose to spend $40M expanding capacity. An investor valuing the company on FCF alone would implicitly punish it for growing; one valuing it on net income would overstate its distributable cash. Owner earnings split the difference on principled grounds.

Red Flags: Testing Earnings Quality

The owner-earnings framework doubles as an earnings-quality detector, because computing it forces you to reconcile reported profit with actual cash. The classic warning sign is the accruals gap: net income persistently running above operating cash flow. In any single year the gap can be innocent — a big receivable, an inventory build ahead of a launch. Sustained over three or more years, it means reported profits are made of promises rather than cash, and companies later revealed as manipulators almost always showed this pattern first.

Beyond the accruals gap, four checks belong on every quality checklist. One-time items that recur: restructuring or impairment charges labeled "non-recurring" but appearing every single year are ordinary costs wearing a costume — add them back to nothing. Capitalizing what peers expense: a company that capitalizes software development or customer-acquisition costs while competitors expense them shifts cost off the income statement and into capex, inflating earnings per share today at the price of amortization tomorrow. Receivables and inventory outrunning revenue: the Brightvale pattern above. Pension and stock-based compensation adjustments: aggressive pension return assumptions manufacture accounting income, and SBC is a real cost to owners even though it is non-cash — when computing owner earnings, do not add SBC back without subtracting the dilution it causes. A systematic screen like the Piotroski F-Score automates several of these checks.

Owner Earnings Yield: A Quick Valuation Shortcut

Once you have an owner earnings figure, the fastest way to use it is a yield: divide by market capitalization. This treats the stock like a bond whose coupon is the cash an owner could extract each year, and lets you compare it directly against fixed-income alternatives — with the crucial difference that a good business's coupon grows.

The comparison with FCF yield in that example is the whole argument in miniature: for companies reinvesting heavily, FCF-based yields understate the underlying earning power, and owner earnings yield corrects the distortion. Buffett has applied exactly this bond-versus-business framing throughout Berkshire Hathaway's history — asking what "equity coupon" a business offers relative to riskless alternatives, and paying up only when the spread compensates for the uncertainty.

Using Owner Earnings in a DCF

Owner earnings are not just a screening ratio — they are a legitimate cash flow definition to discount. In a standard discounted cash flow model you project free cash flow; substituting owner earnings changes the meaning of the exercise. An FCF-based DCF values the company as it currently behaves, growth spending included. An owner-earnings DCF values the company's earning power — what it generates for owners assuming growth capex is a choice that earns its keep separately. For a company whose growth investments earn roughly their cost of capital, the two approaches converge in theory; in practice the input you choose can move the answer dramatically.

That sensitivity is a feature, not a bug: it forces you to form a view on reinvestment quality before trusting the output. If you want to see the mechanics, our guide to how DCF models work walks through every stage — and you can run both versions side by side by plugging in owner earnings as the starting cash flow instead of FCF.

Limitations: Where Owner Earnings Break Down

Buffett himself flagged the central weakness in the same 1986 letter: because maintenance capex must be estimated, owner earnings will never be a precise figure. Two honest analysts can look at the same company and land 20% apart, and a motivated one can land wherever they like — classify enough capex as "growth" and any capital-hungry business looks like a cash machine. The estimate is most fragile exactly where it matters most: capital-intensive industries where maintenance spending is huge, lumpy, and entangled with upgrades that are part replacement, part expansion. Second, owner earnings are not comparable across capital intensity without care — an asset-light software firm and a railroad can show similar owner earnings yields while carrying utterly different reinvestment risk, because the same percentage error in the maintenance estimate swings the railroad's answer far more. Third, the metric says nothing about balance sheet risk: a leveraged company's owner earnings belong partly to its creditors' peace of mind. Use owner earnings as a lens alongside FCF and net income, never as a solitary oracle.

How to Put Owner Earnings to Work

A practical routine: start with five years of financial statements. Check the accruals gap first — if net income persistently exceeds operating cash flow, resolve why before doing anything else. Estimate maintenance capex two ways (D&A proxy and the capex-to-sales average) and take a conservative figure. Compute owner earnings, compare the yield against bonds and against the company's own history, and only then run a DCF — once with FCF, once with owner earnings — treating the spread between the two answers as a measure of how much your thesis depends on the quality of the company's growth spending.

FairPriceIndex's valuation model uses free cash flow rather than owner earnings in its DCF component — deliberately, because FCF is objective and consistently computable across 37,000+ stocks, while maintenance capex estimates require company-by-company judgment no model can automate honestly. That makes our fair values a conservative baseline for growing companies: if a stock already looks undervalued on an FCF basis, an owner-earnings view will usually make it look cheaper still. Screen for candidates across 37,000+ stocks, then stress-test your own owner-earnings assumptions in the DCF calculator — Buffett's framework works best when you do the last mile of thinking yourself.