Capex and Free Cash Flow: The Number Owners Actually Eat
4 steps · one page
In short
Free cash flow asks the question every other measure dodges: after the business has paid its costs, its interest, its tax, and the capital spending it needs to keep operating — what is actually left?
Canonical data. Figures tie to Wexford Instruments (USD millions).
That residual is what can fund dividends, buybacks, debt repayment, or acquisitions without raising new money. It is the most useful single figure in the statements, and it has no standardised definition — which is the first thing to know about it.
The calculation, and why definitions differ
The common form is operating cash flow less capital expenditure. Both inputs come straight from the cash-flow statement, which makes it reproducible.
But variants abound and they are not equivalent. Some deduct only maintenance capex, on the reasoning that growth spending is discretionary — which sounds right and runs into the problem the capex article identified: the split is not disclosed, so any such figure is an estimate. Some deduct dividends, treating them as committed. Some add back acquisitions or lease payments. When a company reports its own free cash flow figure, the definition is the company's choice — so the first question about any such number is what was subtracted to get it.
The plainest version — operating cash flow less all capital expenditure — is the most conservative and the most comparable, which is why it is the version used throughout this portal.
What it reveals that profit does not
Profit can be large while free cash flow is small, and the gap is the point. A profitable company that must reinvest heavily to stay competitive generates less spendable cash than its earnings suggest — and that is not an accounting quirk, it is the economics of a capital-hungry business.
Two comparisons make it useful. Free cash flow against net income shows how much of reported profit becomes money the owners could actually take. Free cash flow against what the company distributed shows whether returns to shareholders were funded from the year's cash or from somewhere else — and when distributions exceed free cash flow, the difference came from the balance sheet: cash reserves, borrowing, or asset sales. None of those is improper, and all of them are finite.
One caution. Free cash flow is lumpy. A year of heavy investment can produce a low or negative figure at a perfectly sound company that is building something, and a year of deferred spending can flatter it. A single year says very little; the multi-year pattern says a great deal.
Worked example
Worked example: Wexford's free cash flow, and what it had to cover (canonical figures, USD millions). Operating cash flow of 98.3 less capital expenditure of 78.0 gives free cash flow of 20.3 — a free cash flow margin of just 2.0% of revenue, and 32.6% of net income. Roughly two-thirds of Wexford's reported profit did not survive as spendable cash. Now the test that matters. The company paid 18.0 in dividends and repurchased 12.0 of stock — 30.0 returned to shareholders against 20.3 of free cash flow, which is 147.8% of it. Wexford distributed roughly half as much again as it generated. Where the difference came from. The shortfall is exactly 9.7 — and that figure appears elsewhere: it is precisely the amount by which net debt rose during the year, from 264.0 to 273.7. The gap between what the company generated and what it handed to shareholders was funded from the balance sheet, and the balance sheet records it to the decimal. And the counterfactual. Had capital expenditure run at the depreciation rate of 42.0 instead of 78.0, free cash flow would have been 56.3, and the same 30.0 of returns would have been comfortably covered at 53% of it. Nothing here is improper — but the arithmetic shows that this year's distributions and this year's investment could not both be funded from this year's cash. (Canonical figures; independently verified. The 9.7 tie holds because free cash flow less distributions less net debt repayment equals the change in cash: 20.3 − 30.0 − 20.0 = −29.7, and net debt is debt less cash.)
Frequently asked
6 questions
How is free cash flow calculated?
Most commonly operating cash flow less capital expenditure, both taken straight from the cash-flow statement. Variants deduct only maintenance capex, or dividends, or add back acquisitions — and they aren't equivalent, so the first question about any figure is what was subtracted.
Why doesn't it have a standard definition?
Because it isn't a required statement line — it's a derived measure. When a company reports its own figure, the definition is the company's choice. The plainest version is the most conservative and the most comparable.
Why can't maintenance capex just be deducted instead?
Because the split between maintenance and growth isn't disclosed and is genuinely hard to determine even internally. Any figure claiming to deduct only maintenance capex contains an estimate, and it should say so.
What does it mean when distributions exceed free cash flow?
The difference came from the balance sheet — cash reserves, borrowing, or asset sales. On the illustration, 30.0 was returned against 20.3 generated, and the 9.7 shortfall matches the year's increase in net debt exactly. None of that is improper, and all of those sources are finite.
Is low free cash flow a bad sign?
Not by itself, and it's lumpy. A year of heavy investment can produce a low or negative figure at a perfectly sound company building something; a year of deferred spending can flatter it. A single year says very little.
How much of profit typically becomes free cash flow?
There's no typical figure — it depends entirely on capital intensity. On the illustration it's 32.6%, so roughly two-thirds of reported profit didn't survive as spendable cash. For an asset-light business the proportion would be far higher.
References
Educational and informational only — not investment advice, a recommendation, or an offer to buy or sell any security. Investing involves risk, including the possible loss of principal. Worked examples use fictional companies and figures.