Owner Earnings: Buffett's Favorite Cash Flow Metric Explained

In his 1986 letter to Berkshire Hathaway shareholders, Warren Buffett defined a number that he considered more honest than net income and more useful than reported cash flow. He called it owner earnings.

Forty years later, owner earnings remain one of the least-known and most useful metrics in fundamental analysis. Almost no broker shows it. Most financial data tools don’t compute it. And yet, if you only learn one new ratio this year, this is probably the one.

The Original Definition (Straight from Buffett)

Buffett’s 1986 definition is famous among value investors. In his own words, owner earnings equal:

Reported earnings, plus depreciation, depletion, amortization, and certain other non-cash charges, less 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.

In modern shorthand:

Owner Earnings = Net Income + D&A + Non-cash Charges − Maintenance Capex − Working Capital Investment

The key piece — and the reason most financial databases don’t compute it — is maintenance Capex. The number that distinguishes the Capex required to keep the business running from the Capex spent on growth.

Companies almost never report this split. Estimating it is part of the job.

Why Owner Earnings Beat Net Income

Net income is an accounting construct. It is what is left after a long series of choices about depreciation methods, revenue recognition, stock-based compensation, and tax timing. None of these choices are wrong. They just give management room to present a number that reflects an interpretation of reality, not the reality itself.

Owner earnings strip that interpretation away. The question they answer is the only one a long-term shareholder really cares about: how much cash could be taken out of this business each year, while leaving it as competitively strong as it was before?

That definition is precise. It does not include the cash a company spends to expand its store count, build new factories, or acquire competitors. It only includes the cash needed to keep what already exists running at full strength.

If a company can sustainably distribute that number to shareholders — through dividends, buybacks or debt reduction — the business is genuinely creating value. If it can’t, growth investment is masking deterioration.

Why Owner Earnings Beat Free Cash Flow Too

Most investors who go beyond net income stop at free cash flow (FCF):

Free Cash Flow = Operating Cash Flow − Total Capital Expenditure

FCF is great. It is the metric we recommend most retail investors learn first, and we have a whole article on FCF analysis. But it has one structural flaw: it treats all Capex the same.

Consider two companies, both reporting €100M of total Capex this year:

• Company A spent €40M maintaining its existing plants and €60M building a new one in Mexico that will start contributing revenue in 18 months.
• Company B spent the full €100M just keeping its existing factories from falling apart.

Both companies show identical FCF. But A is investing in growth on top of a productive base. B is sprinting just to stand still. The owner earnings of A are roughly €60M higher than B’s.

That distinction is the whole point of owner earnings.

How to Calculate Owner Earnings, Step by Step

Here is the practical version of the formula. We will use Apple as the working example because the data is public, the business is simple to model, and the result is informative.

Step 1 — Start with Net Income

This is the bottom of the income statement. For fiscal 2024, Apple reported approximately $93.7 billion in net income.

Step 2 — Add Back Non-Cash Charges

Pull from the cash flow statement:

• Depreciation and amortization
• Stock-based compensation (this is real — Apple actually issued the stock — but treated correctly only when reconciled against share count)
• Other non-cash items (impairments, deferred taxes, etc.)

For Apple in 2024, D&A is around $11.4 billion and SBC is around $12 billion. We add D&A back as a non-cash charge (the same charges EBITDA ignores entirely — with the risks that entails). SBC is debated — strict Buffett-style owner earnings adds it back, but you should mentally reserve it because dilution is real.

Step 3 — Subtract Maintenance Capex

This is the hardest step. Apple’s reported capital expenditure in 2024 was around $9.4 billion. But how much of that is just maintaining the current business?

Three common ways to estimate maintenance Capex:

1. Average it against depreciation. Over a long enough horizon, maintenance Capex should be roughly equal to depreciation expense. For mature, asset-light companies like Apple, this is usually close to true.
2. Use the “revenue stable” year heuristic. Look at a year where revenue was flat or declining. The Capex spent in that year is mostly maintenance.
3. Take management at their word — cautiously. Some companies (utilities, telecoms, REITs) actually publish a maintenance Capex figure. Most don’t.

For Apple, depreciation ($11.4B) is higher than total Capex ($9.4B), which tells us Apple is essentially in steady-state Capex mode. Maintenance Capex ≈ total Capex.

Step 4 — Adjust for Working Capital

Owner earnings need cash to fund growth in receivables, inventory, and other working capital. If the business is growing and tying up more cash in working capital, that cash is not really available to the owner.

For Apple in 2024, working capital movements were relatively modest. We will skip the line for simplicity, but in a growing business this can be material.

Step 5 — Add It All Up

Rough Apple 2024 owner earnings:

Net income:                 +$93.7B
+ D&A:                      +$11.4B
- Maintenance Capex:        -$9.4B
≈ Owner Earnings:           ~$95.7B

(Add or subtract SBC depending on philosophy. Apple’s true Buffett-style number is between roughly $84B and $96B depending on how you handle SBC.)

That number — call it $90B as a midpoint — is what Apple could in principle return to shareholders each year while leaving the business unchanged. Apple’s actual capital return program (dividends + buybacks) was on a similar order of magnitude. The company is, in Buffett’s framework, distributing approximately its owner earnings.

Owner Earnings vs Reported Numbers: A Comparison Table

The pattern is clear: as you move down the table, the metric gets harder to compute but closer to the economic reality.

When Owner Earnings Are Most Useful

Owner earnings shine in three situations:

1. Capital-Heavy Businesses

Railroads, telecom infrastructure, utilities, refiners. These businesses report large Capex, but a significant portion is genuinely maintenance. Looking at FCF makes them look mediocre. Looking at owner earnings shows their real cash-generative power.

2. Serial Acquirers

Companies that grow by acquisition show high “investing” outflows that are growth, not maintenance. Owner earnings clear that fog. They tell you what the underlying organic business generates, ignoring the empire-building.

3. Companies with Lumpy Capex

Some businesses build a major asset every 5 or 10 years. In the build year, FCF collapses. In the years between builds, FCF looks artificially excellent. Smoothing maintenance Capex over the full cycle gives you a much truer owner earnings figure.

Want fundamentals close to the companies you already follow? Join STOK Terminal free and let STOK Terminal connect company research with your watchlists and portfolio.

When Owner Earnings Are Less Useful

Owner earnings have limits. Use them carefully when:

• The company is early stage. A high-growth software business genuinely should be spending nearly all its cash on growth. Owner earnings will look terrible — and that may be correct.
• Maintenance Capex is genuinely unknown. Some industries change so quickly that yesterday’s maintenance Capex doesn’t keep you competitive tomorrow. Tech infrastructure in the AI era is a current example.
• You need a fast comparison across many companies. Owner earnings require company-by-company judgment. For initial screening, FCF Yield or ROIC is faster.

How to Use Owner Earnings in Practice

A simple workflow:

1. For the 5–10 companies in your portfolio (or the ones you are seriously considering), compute owner earnings yearly for the last 5–10 years.
2. Compare owner earnings to reported net income. Persistent gaps are signal.
3. Divide owner earnings by enterprise value. This is the “owner earnings yield” — the most honest version of an earnings yield Buffett would recognize.
4. Compare that yield to the 10-year Treasury. If a great business is yielding 6% in owner earnings against a 4% risk-free rate, you have a margin of safety. If a mediocre business is yielding 3% against the same 4%, you don’t.

A Worked Comparison: Coca-Cola vs an Average Beverage Company

To make the metric concrete, imagine two beverage companies of similar size:

• Coca-Cola is essentially a brand and concentrate business. Maintenance Capex is small relative to depreciation. Reported FCF and owner earnings track closely.
• An average bottler owns trucks, plants, and warehouses. Capex is large. A significant chunk is just keeping the fleet alive. FCF looks healthy until you separate maintenance and growth — and owner earnings drop noticeably below FCF.

Same beverage industry. Same revenue. Very different owner earnings profile. This is exactly the difference Buffett was trying to capture when he chose to invest in the concentrate business.

Conclusion

Owner earnings are not a number you will find pre-computed on most websites. They take 15 minutes of work per company. They require you to make judgment calls about maintenance Capex.

That is precisely why they are useful. A metric that anyone can compute in two clicks is, almost by definition, priced in. A metric that requires real work is where the edge is.

If you are serious about fundamental investing, learn this one — ideally as part of a broader repeatable process. It will change how you read a cash flow statement forever.

Frequently Asked Questions

Owner earnings become more useful when read beside the full financial history. Explore the value investing research workflow in STOK Terminal.

What are owner earnings? Buffett’s 1986 definition of a company’s true earning power: net income, plus depreciation and other non-cash charges, minus the maintenance Capex needed to keep the business competitively strong, adjusted for working capital. It is the cash an owner could take out each year without weakening the company.

How do owner earnings differ from free cash flow? FCF subtracts all Capex; owner earnings subtract only maintenance Capex — the part required to sustain the current business. A company investing heavily in growth can show weak FCF while its owner earnings remain strong.

How do you estimate maintenance Capex? Three common approaches: use depreciation as a long-run proxy, look at Capex in a year when revenue was flat, or take a management-published figure where available (utilities, telecoms, REITs). It is an estimate — a judgment call, not a line item.

Are owner earnings pre-calculated anywhere? Rarely. You compute them yourself from the income statement and the cash flow statement — about 15 minutes per company. That work is exactly why the metric still offers an edge.

See Cash Flow Trends Without Spreadsheet Gymnastics

STOK Terminal is being built to make multi-year fundamentals — cash flow, Capex, margins, ratios — easy to inspect alongside the companies you actually follow and own.

👉 Join STOK Terminal free — because owner earnings, like all serious analysis, only works when the data is in front of you.